Research in Landscape Architecture: Methods and Methodology 9781315396897, 1315396890

Citation preview

www.allitebooks.com

Research in Landscape Architecture Defining a research question, describing why it needs to be answered and explaining how methods are selected and applied are challenging tasks for anyone embarking on academic research within the field of landscape architecture. Whether you are an early career researcher or a senior academic, it is essential to draw meaningful conclusions and robust answers to research questions. Research in Landscape Architecture provides guidance on the rationales needed for selecting methods and offers direction to help to frame and design academic research within the discipline. Over the last couple of decades the traditional orientation in landscape architecture as a field of professional practice has gradually been complemented by a growing focus on research. This book will help you to develop the connections between research, teaching and practice, to help you to build a common framework of theory and research methods. Bringing together contributions from landscape architects across the world, this book covers a broad range of research methodologies and examples to help you conduct research successfully. Also included is a study in which the editors discuss the most important priorities for the research within the discipline over the coming years. This book will provide a definitive path to developing research within landscape architecture. Adri van den Brink is Professor and Chair of Landscape Architecture at Wageningen University, the Netherlands. Diedrich Bruns is Professor and Chair of Landscape Planning and Land Use at Kassel University, Germany. Hilde Tobi is Associate Professor and Coordinator of the Research Methodology Group at Wageningen University, the Netherlands. Simon Bell is Professor and Head of the Department of Landscape Architecture at the Estonian University of Life Sciences and Associate Director of the OPENspace Research Centre at the University of Edinburgh, United Kingdom.

www.allitebooks.com

Landscape architecture is a potentially powerful profession and discipline: a field poised to transform the planet for the better. This possibility will only be realized through a more robust research agenda. The authors of Research in Landscape Architecture have produced just such a framework. They present a helpful, thoughtful roadmap for landscape architecture scholars. — Frederick Steiner, Dean and Paley Professor, School of Design, University of Pennsylvania, USA

As a practice-led discipline, landscape architecture faces a challenge when trying to impose methodology on a somewhat theory-resistant subject. This new book presents cases of landscape architecture research in their methodical context. We learn how landscape architecture research questions are formulated and how evidence for answering them can be found. We live in an era of ever increasing complexity on the one hand and strong specialisation on the other. Where to position the holistic perspective of this domain? This book will give valuable orientation for anybody looking for systematic knowledge production in landscape architecture. It will inspire especially early-career researchers. — Ellen Fetzer, Nürtingen-Geislingen University, Germany, International Master of Landscape Architecture (IMLA)

www.allitebooks.com

Edited by Adri van den Brink, Diedrich Bruns, Hilde Tobi and Simon Bell

Research in Landscape Architecture Methods and methodology

www.allitebooks.com

First published 2017 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 711 Third Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2017 selection and editorial matter, Adri van den Brink, Diedrich Bruns, Hilde Tobi and Simon Bell; individual chapters, the contributors The right of Adri van den Brink, Diedrich Bruns, Hilde Tobi and Simon Bell to be identified as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data Names: Brink, Adri van den, editor. Title: Research in landscape architecture : methods and methodology / edited by Adri van den Brink, Diedrich Bruns, Hilde Tobi, and Simon Bell. Description: New York, NY : Routledge, 2017. | Includes bibliographical references and index. Identifiers: LCCN 2016026543| ISBN 9781138020924 (hardback : alk. paper) | ISBN 9781138020931 (pbk. : alk. paper) | ISBN 9781315396903 (ebook) Subjects: LCSH: Landscape architecture--Research--Methodology. Classification: LCC SB469.4 .R47 2017 | DDC 712.072--dc23 LC record available at https://lccn.loc.gov/2016026543 ISBN: 978-1-138-02092-4 (hbk) ISBN: 978-1-138-02093-1 (pbk) ISBN: 978-1-315-39690-3 (ebk) Typeset in Frutiger by HWA Text and Data Management, London

www.allitebooks.com

Contents

List of figures List of tables List of contributors

vii ix x

Introduction

1

ADRI VAN DEN BRINK, DIEDRICH BRUNS, SIMON BELL AND HILDE TOBI

PART I: RAISING AWARENESS 1 Advancing landscape architecture research

9 11

DIEDRICH BRUNS, ADRI VAN DEN BRINK, HILDE TOBI AND SIMON BELL

2

A process approach to research in landscape architecture

24

HILDE TOBI AND ADRI VAN DEN BRINK

PART II: SETTING THE STAGE 3 The role of theory

35 37

IAN H. THOMPSON

4

The relationship between research and design

54

SANDA LENZHOLZER, INGRID DUCHHART AND ADRI VAN DEN BRINK

5

The challenge of publication

65

MAGGIE ROE

6

Assessing research priorities and quality

85

JURIAN MEIJERING, HILDE TOBI, ADRI VAN DEN BRINK AND DIEDRICH BRUNS

PART III: SELECTED APPROACHES AND METHODS 7 Case studies

103 105

SIMON SWAFFIELD

v …

www.allitebooks.com

Contents

8

Landscape biography

„

120

JAN KOLEN, HANS RENES AND KOOS BOSMA

9

Social media

136

RON VAN LAMMEREN, SIMONE THEILE, BORIS STEMMER AND DIEDRICH BRUNS

10

Virtual environments

161

SIGRID HEHL-LANGE AND ECKART LANGE

11

Walking

179

HENRIK SCHULTZ AND RUDI VAN ETTEGER

12

Design guidelines

194

MARTIN PROMINSKI

PART IV: ADDRESSING SOME OF THE GRAND CHALLENGES 13 Cultural landscape meanings and values

209 211

KEN TAYLOR

14

Landscape and health

235

CATHARINE WARD THOMPSON

15

Thermally comfortable urban environments

263

ROBERT D. BROWN AND TERRY J. GILLESPIE

16

The urban water challenge

285

ANTJE BACKHAUS, OLE FRYD AND TORBEN DAM

Index

307

vi …

www.allitebooks.com

Figures

1.1 1.2 1.3 1.4 5.1 6.1 6.2 7.1 7.2 7.3 7.4 7.5 9.1 9.2 9.3 9.4 9.5 10.1 10.2 10.3 10.4 10.5 10.6 10.7 10.8 11.1 11.2 12.1

Three aspects of doing research Research process overview Conceptual research design Technical research design Overlapping fields of value and sources of theory in landscape architecture The general procedure of a Delphi study Overview of the systematic review sampling process Theoretical cases – different types of small town on a tourism circuit Within-case and cross-case comparisons Multiple perspective analyses of a single complex case – two examples Embedded geographical cases Cases based on types Transdisciplinary framework with social media interface Extended transdisciplinary framework with interaction typology Detail of a map of public input to landscape assessment with descriptive texts Identified hotspots (above) and tourist route densities (below) Effect of spatial granularity on level of detail The path in the virtual model Still images of the status quo and the three scenarios and five viewpoints Filmstrip representing the animated walkthrough Participant’s ratings of four different landscape scenarios, dynamic walkthroughs versus static images Long-term forest management plan indicating the year when felling takes place Setup of the participatory workshop using 3D visualisation Static images of the four landscape models represented through five images along the path Stakeholders’ ratings of different representation media Interplay of three walking modes The schematic model of the CSM method ‘River. Space. Design.’ Design guidelines are organised in five groups of design strategies

15 17 19 21 69 87 92 110 111 112 113 114 139 140 143 150 152 165 165 166 167 170 171 172 174 184 188 198

vii …

www.allitebooks.com

List of figures

12.2 ‘River. Space. Design.’ Three design tools from the design strategy ‘A1-Linear spatial expansion’ 12.3 The design guideline system of high-bay racking in the project on shrinking cities 12.4 Diagram of the research method or process to develop design guidelines 13.1 Planning model for heritage conservation management policy 13.2 Location of Wingecarribee Shire 13.3 Wingecarribee Study Model 13.4 Throsby Park, a painting by Conrad Martens 13.5 Present day view of Throsby Park 13.6 Paddock showing nineteenth century ridge and furrow plough marks 14.1 Quality of Life (QOL) predicted by various factors 14.2 Number of participants in each part of the study 14.3 Logic model of a longitudinal approach to study design 14.4 An example of two of the option tasks in the questionnaire 15.1 A small LI-COR pyranometer can be used to measure incoming solar radiation in open areas 15.2 A Kipp & Zonen CNR1 15.3 A radiation shield should always accompany temperature and humidity sensors to eliminate radiation error 15.4 An anemometer with three cups 15.5 A typical set-up of microclimate instruments 16.1 Sketch by Sven Ingvar Anderson (1994) showing the spiral of the design process 16.2 Conceptual framework, outlining variations in research approaches relevant to research on the urban water challenge 16.3 Diagram of the design process 16.4 Set-up of the design experiment 16.5 How the three studies presented in Chapter 16 differ in focus and research approach 16.6 The different roles of the researcher in the three studies presented in Chapter 16

„

199 201 203 220 225 227 229 229 230 239 246 249 251 268 270 272 276 278 287 289 295 297 300 301

viii …

www.allitebooks.com

Tables

1.1 4.1 5.1 5.2 5.3 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9 9.1 9.2 9.3 13.1 13.2 13.3 15.1 16.1 16.2 16.3

Research proposal checklist Overview of types of RTD Standards for assessing the quality of research Methods reporting: survey of papers in 2009 published in Landscape Research Emerging themes from the Landscape Forward Symposium, 18 March 2015 Percentage of experts that selected each research domain as ‘most important’ or ‘most useful’ in round 3 of the Delphi study Search query applied in SCOPUS Percentage of empirical papers reporting a research objective, question or hypothesis Percentage of empirical papers reporting a research objective or question focused on a retrospective, prospective or present timeframe Percentage of empirical papers explicitly reporting the use of a case study, experiment or other study design Percentage of empirical papers reporting the use of a random or non-random sampling method Percentage of empirical papers reporting the use of various data collection methods Percentage of empirical papers reporting the quantitative and qualitative analysis of data Percentage of empirical papers related to a particular research domain Semantic analysis of the messages Distribution of the harvested photos Overview of invalid Flickr photos Examples of heritage value typologies Example landscape characteristics for rural landscapes Example landscape characteristics for urban landscapes Information for estimating temperature increase due to UHI Main characteristics of the three studies presented in Chapter 16 Five key parameters selected for project analysis and comparison Key themes and recommendations identified from the authors’ general knowledge of LSM design aspects

14 60 71 74 79 89 91 94 94 94 95 96 97 98 146 149 151 216 223 223 274 291 292 298

ix …

www.allitebooks.com

Contributors

Antje Backhaus is assistant professor in landscape architecture at the University of Copenhagen, Denmark. Her research and teaching interests include green urban infrastructure, water sensitive cities and specifically the use of participatory- and design-based research methods on the interface between research and practice. Besides her work at the University, Antje is a partner in the landscape architecture office gruppe F Landschaftsarchitekten in Berlin. Simon Bell is professor and head of the Department of Landscape Architecture at the Estonian University of Life Sciences and associate director of the OPENspace Research Centre in the Edinburgh School of Architecture and Landscape Architecture (ESALA), University of Edinburgh, UK. He is also president of the European Council of Landscape Architecture Schools. His research interests currently focus on the relationship between landscape and health, especially forests and blue space. Koos Bosma (1952–2015) was professor of architectural history and heritage studies at VU University Amsterdam, the Netherlands. He was an expert on the history of urban space and European planning history. His other interests included the role of heritage in spatial design. He sadly passed away on 10 September 2015, but his personality, attitude and work will inspire his colleagues in the Netherlands (and abroad) for many years to come. Robert D. Brown is professor of landscape architecture at Texas A&M University. His research focuses on microclimatic design. He models and measures urban microclimates and applies this work to understanding how design of urban environments modifies microclimates, and how microclimates affect the thermal comfort of urban residents. Diedrich Bruns is professor and chair of landscape planning and land use at Kassel University, Germany. He has many years of experience in planning practice and research. Current teaching and research interests are in participatory planning methods, particularly with respect to inclusiveness and early public involvement. Torben Dam is an associate professor in landscape design at the University of Copenhagen, Denmark. His teaching and research interests are landscape architecture, design and detailing with a focus on the design process and the exchange between design and knowledge.

x …

List of contributors „

Ingrid Duchhart is a senior researcher at Wageningen University, the Netherlands, with longstanding teaching experience in (strategic) landscape planning and design and with practical experience in transformative co-design and network governance, landscape architecture research methods, emerging economies and disaster management, adaptivity and nature-based solutions and climate change. Ole Fryd is a lecturer in urban and environmental planning at the University of Melbourne, Australia. His teaching and research interests include water sensitive cities, collaborative knowledge creation and urbanisation in developing countries. Terry J. Gillespie is professor emeritus in the School of Environmental Sciences at the University of Guelph in Ontario, Canada. His research and teaching interests centre on the measurement and modelling of microclimates. He applies his interests in microclimatology to agriculture and to methods of evaluating human comfort in outdoor environments. Sigrid Hehl-Lange is a research fellow in the Department of Landscape at the University of Sheffield, UK. Through her research projects, including a senior Marie-Curie fellowship, she has gained considerable experience in using virtual landscapes in a decision-making context. Jan Kolen is professor of landscape archaeology and cultural heritage at Leiden University, the Netherlands, and the director of the Centre for Global Heritage and Development of Leiden University, Delft University of Technology and Erasmus University Rotterdam. His research interests include the long-term history of landscapes (‘landscape biography’), the evolution of human niche construction, and the value and role of environmental heritage in landscape design and urban planning. Eckart Lange is a professor at the Department of Landscape at the University of Sheffield, UK. His research focuses on how digital methods in landscape and environmental planning can influence anthropogenic landscape change. Sanda Lenzholzer is an associate professor in the landscape architecture group of Wageningen University. She was trained as a landscape architect and urban designer in international practice and specialised in urban climate matters during her academic career. In her teaching and research she focuses on climate responsive design and the relation between research and design, specifically research through designing (RTD). Jurian Meijering is a PhD candidate and lecturer within the research methodology group at Wageningen University, the Netherlands. As a lecturer he has taught various courses in research methodology. His PhD project focuses on the use of the Delphi method as well as the application of the method to the development of urban sustainability rankings. Martin Prominski is professor and chair of Designing Urban Landscapes at Leibniz University Hannover, Germany, and a registered landscape architect. For many years he has been working on the interfaces between research, theory and practice. His current interests are in design

xi …

List of contributors

„

research strategies, qualification of urban landscapes, and concepts of nature and culture in the Anthropocene. Hans Renes is historical geographer at Utrecht University and professor of heritage and planning at VU University Amsterdam, the Netherlands. He has a background in human geography (Utrecht University) and rural planning (Wageningen University). His main interests are landscape history and the relationships between heritage and planning. He has published extensively on the history of the cultural landscapes of Europe. Maggie Roe is a senior lecturer at the School of Architecture, Planning and Landscape at Newcastle University, UK. Her research focus is generally on landscape planning and sustainability, particularly related to participatory landscape planning, cultural landscapes and landscape change. She is an editor of the international peer-review journal Landscape Research and has considerable experience in reviewing research proposals, grants, papers and books. Henrik Schultz is co-founder of Stein+Schultz, a lecturer at Leibniz University Hannover and a researcher at Studio Urbane Landschaften. He has many years of experience of research and design projects for sustainable landscapes and of methods of exploring and understanding landscapes. His current main interest is the role of bodily engagement in landscape transformation. Boris Stemmer is professor and chair of landscape and recreation planning at the University of Applied Science Ostwestfalen-Lippe, Germany. He earned his PhD with a thesis on landscape assessment methods in Germany. His current research interests are participatory planning methods especially for renewable energies as well as landscape as a resource of happiness, well-being and health. Simon Swaffield is professor of landscape architecture at Lincoln University, New Zealand. His research and scholarship is focused upon governance and planning of rural and peri-urban landscapes undergoing rapid change and upon the development of theory and research methodology in landscape architecture. Ken Taylor is an adjunct professor in the Centre for Heritage and Museum Studies, at the Australian National University, Canberra; emeritus professor of landscape architecture, the University of Canberra; and visiting professor at Silpakorn University, Bangkok. He has been teaching and researching in the field of cultural landscapes since the mid-1980s and more recently with a particular focus on urban cultural landscapes, particularly in Asia. Simone Theile is a lecturer and researcher in the Department of Landscape Planning and Land Use at Kassel University, Germany. Her research focus is on planning theory, participatory planning methods and on communication and protest organisation in social media in the context of planning projects. Ian H. Thompson is reader in landscape architecture in the School of Architecture, Planning and Landscape at Newcastle University, UK, where his research has mostly been within the fields of landscape design history and theory.

xii …

List of contributors „

Hilde Tobi is associate professor and coordinator of the research methodology group at Wageningen University. She has many years of research experience in medical, social and environmental sciences. Her current main research and teaching interest is methodology for interdisciplinary research. Adri van den Brink is professor and chair of landscape architecture at Wageningen University, the Netherlands. Before that he worked as a landscape planner, research manager and policy advisor in public service, and as a professor of spatial planning, also at Wageningen University. In his research and teaching he explores the relationship between design and research. Rudi van Etteger is assistant professor of landscape architecture at Wageningen University, the Netherlands. Trained both as a landscape architect and a philosopher, his research is aimed at exploring the theoretical foundations of landscape design. His main focus is on the aesthetics of designed landscapes. Ron van Lammeren is associate professor of geo-information science at the Laboratory of Geoinformation Science and Remote Sensing at Wageningen University, the Netherlands. In his research and teaching he pays attention to the role of geo-information in planning and design of landscape and environment, especially the nature and impact of visual representations of locational data ensembles in participatory planning and design processes. Catharine Ward Thompson is professor of landscape architecture and director of OPENspace Research centre at the Edinburgh School of Architecture and Landscape Architecture (ESALA), University of Edinburgh, UK. Her research on inclusive access to outdoor environments and environment–behaviour interactions has informed government, public agencies and the World Health Organization (WHO). She teaches a master’s programme on landscape and well-being.

xiii …

This page intentionally left blank

Introduction Adri van den Brink, Diedrich Bruns, Simon Bell and Hilde Tobi

THE GENESIS OF THE BOOK Landscape architecture is a discipline – or a disciplinary field – which developed from a very practical basis in park and garden design. The name was coined in the 19th century when it became a profession along with that of architecture. Landscape architects were originally trained as pupils or apprentices to established practitioners, as was common for many professions. The ideas and theories which the original practitioners applied derived from a range of sources such as architecture, gardening, engineering, agronomy and dendrology. Formal education in universities or colleges of higher education only started in the early 20th century. The first landscape architecture programme in the USA was founded in 1900 at Harvard University, followed in Europe at the Norwegian Agricultural University at Aas near Oslo in 1919, the Technical University of Berlin Germany in 1929, the University of Reading in the UK in 1932, the University of Lisbon in Portugal in 1942 and Wageningen University in the Netherlands in 1948, and other institutions followed. The teachers of the early programmes were a mix of horticulturalists, dendrologists, architects, agronomists and engineers, perhaps with some other disciplines, depending on the character of the institution. Studio courses were interspersed with lectures on scientific subjects along with practical work. Studio tutors often came from offices and many professors had offices themselves so there was a strong connection between theory and practice. In the second half of the 20th century there was a growth in the number of schools, in the range of work undertaken by the profession and in the professional status of landscape architects. The original multi-disciplinary character of the field continued to expand and specialisations gradually developed. However, until recently there was little in the way of academic research taking place which could be considered as strictly belonging to landscape architecture. Professors with doctorates tended to obtain these in associated or neighbouring subject fields and it is still the case that in a number of places it is difficult to study for a doctoral degree in landscape architecture. Of course, all landscape architects are familiar with the idea of carrying out research and many, whether in offices or universities, undertake projects commissioned by various clients which have a research component. Equally, all planning or design projects need some research, for example into the history of a site, using methods such as archival study or the dendrological analysis of ancient trees. This kind of research is important and it familiarises practitioners with a range of basic research skills but it is different from the kind of academic research undertaken in many other disciplines such as natural or social sciences, humanities, art history and the like. Perhaps it takes a special kind of

1 …

Adri van den Brink, Diedrich Bruns, Simon Bell and Hilde Tobi

„

person to move away from the world of site design and the satisfaction of seeing a project built and maturing over time, serving people and conserving the environment towards the laboratory or library in order to carry out research which will not directly result in concrete plans or designs. We do not suggest that a landscape architect should be either a planner/designer or a researcher – far from it, and many combine both aspects. Be that as it may (and we will explore the nature of academic research in comparison with project-based research in Chapter 1) there is a desperate need for much more research in landscape architecture to equip the profession with a solid evidential base for its practice, to help it to deal with many of the challenges facing society and the environment, and in order to hold its own in the competitive academic world, to name but three compelling reasons. The multi-disciplinary origins of landscape architecture and the diverse range of institutions where it is taught have persisted and this is what characterises landscape architecture: concurrent diversity. It tends not to be a field where research can excavate deeply into a narrow seam in the way that much scientific, art and humanities research can – focusing on a single gene, the work of a single author or a narrow historic period. Instead, landscape architecture research must keep its feet firmly on the ground and its head out of the clouds for the very simple reason that landscape architects want new knowledge in order to solve complex problems. Indeed, owing to the interaction between the physical environment and people, which is essentially what landscape architecture concerns, problems frequently, if not always, involve research both on people’s surroundings and on the people together (and possibly some further dimensions). As research within landscape architecture has developed it has tended to borrow shamelessly from methods developed and tested in many other fields. In 2011 the European Council of Landscape Architecture Schools (ECLAS) published an edited volume entitled ’Exploring the Boundaries of Landscape Architecture’ (Bell et al. 2011) in which a range of experts in some of the ’neighbouring disciplines’ summarised a number of key theoretical foundations and research methods used in each of these subject fields. ECLAS has also carried out some surveys to see how much academic research is undertaken across Europe within university departments and using which methods (Clewing and Jørgensen 2006; Bell et al. 2010; Fetzer 2011; see also van den Brink and Bruns 2014). The results tend to show that a diversity of subjects are studied and many methods are used but that it is very difficult to identify any methods which can strictly be said to belong solely to landscape architecture. Frequently, mixed methods are used, selected from a kind of á la carte menu drawn from the wide range of neighbouring disciplines. So, to contrast the deep enquiry in a narrow seam, much landscape architecture academic, doctoral or commissioned research tends to look broadly at a problem from many angles but cannot (and often does not need to) dig so deeply into them. In its efforts to develop and strengthen landscape architecture as a discipline (or disciplinary field), and as a result of surveys such as those noted above, ECLAS aims to improve the understanding of the character and nature of landscape architecture research, to raise awareness and to increase the understanding of the specific nature of landscape architecture research and how to approach research design and methodological development among academics and professionals, teachers and students. This book is an attempt to capture and present some key aspects which ECLAS believes are fundamentally important for anyone coming into academic research, whether as a doctoral student or an early career researcher post-doc joining a research team for a project or, for that matter, more senior academics coming from a practice background who wish to pursue their own research. We have tried to present a possible scenario faced by a new researcher in developing their research in order to demonstrate how the reader might approach the book.

2 …

Introduction „

SO YOU WANT TO BE A LANDSCAPE ARCHITECTURE RESEARCHER? Consider this scenario: you have just been accepted as a researcher to join a team in a well-known landscape architecture department. Your first task is to develop one or more research proposals and to apply for research grants in order to build up the academic reputation of the department. You report to a senior professor who happens to be a renowned landscape architect with an impressive project portfolio and who has already attracted several young researchers who, like yourself, are eager to help to build up an innovative research group. You soon discover, since your professor’s experience is mainly in project design and little in academic research, that your need for advice about what research approach and which methods to include in a project application is not met. You look for help and turn to the other researchers in your group. These include, for example, doctoral candidates, each of whom is pursuing a research topic that is different from yours and is applying methods which seem inappropriate for what you want. None of your immediate peers is able to provide useful assistance. You are getting worried. What should you do? Where and to whom can you turn for help and guidance? You should bear in mind that it has mainly been through design solutions that landscape architects have succeeded in the past, and continue to succeed, in advancing their field. The proactive planning and design way of thinking has contributed to major innovations but in almost all cases they tend to be area- or site-specific and, although their use as precedents is common, there is also a desire among creative people to show originality and uniqueness, as opposed to generating solutions which may be transferred more widely. Do not underestimate the extent to which creativity has been a feature of some of the major scientific breakthroughs, so that we should not feel that scientific research approaches are necessarily achieved through mechanical data analysis and logical reasoning alone! All the while you continue having difficulties in defining your research question, in designing your study and in selecting your study methods. You probably have a pretty good idea of your topic, whether it be addressing a particular climate change issue, contributing to urban sustainability, dealing with landscape impacts of renewable energy transition, understanding the relationship between health and green infrastructures, or ‘something’ on water and biodiversity. However, even formulating a good title and set of research questions, a task that initially appeared to be a straightforward thing to do, may be soon turn out to be rather complicated. You are finding yourself diving deeply into the available scholarly literature on the topic of your choice. You are finding how, like many other design fields, landscape architecture presents itself as a discipline integrating and synthesising knowledge from several other fields, some of which you now need to start learning more about. You are also finding that landscape architects not only take broad views on the world as it is; they also have views on how the world should be. As practitioners your senior staff and thesis supervisors may not easily free themselves from normative thinking. They may, in fact, be confronted with the same difficulties as you. Your research project, however, should be approached differently from solving a design project commissioned by a client needing spatial or procedural solutions. Your study addresses, first and foremost, the requirements of academic research. You and your supervisor need to respond to standards and expectations set by the academic community at large. Academic quality standards are common to all disciplines – often ones with whom you will be collaborating – as most of the grand landscape challenges of our time require multi-, inter- or trans-disciplinary approaches.

3 …

Adri van den Brink, Diedrich Bruns, Simon Bell and Hilde Tobi

„

Both you and your supervisor may feel overwhelmed by the need to meet such scholarly quality standards. What are the do’s and the don’ts, what steps do you need to take before you can actually start your research, and to find methods that are suited best to meet your research project’s objectives and answer your research question(s)? While attending many conferences and doctoral seminars, during the last 20 years or so, we the editors of this book, feel the scenario described above might be true for many academic landscape architecture research groups. This is in part because of the relatively low numbers of experienced landscape architecture researchers and doctoral students across large numbers of departments, so that there is no long tradition and culture of doing academic research. There are few experienced publishers of top-quality research papers or even of doctoral supervision or of work in major European Union (EU) or nationally funded research projects. It has also been our experience in various conferences, including the annual ECLAS conferences and those of other organisations where landscape architecture research may be presented, that when listening to oral papers, a surprisingly large number of supposedly research presentations are found wanting a proper research question. Also surprisingly large is the number of researchers having difficulty in describing why they chose a specific theme, what is the problem they are addressing, what is their objective and, significantly, how they selected particular methods and how they applied them. However, the number of researchers, senior and junior, who are clear in reporting how they went about collecting and analysing data and in extracting from them meaningful conclusions and finding robust answers to the questions is small (no surprise if the questions were ill-posed, of course!). It is the latter, who, in our view, should – and in several cases already do – take the lead in advancing the landscape architecture research community. Of course, all of the academic community must take up the call. We all have obligations to both the current and the next generation of researchers. The job for senior researchers is to develop strategies, to give direction and, ultimately, to identify the route towards research excellence. This book on research methods is but one contribution to the many aspects that are needed.

THE PURPOSES OF THE BOOK This book is sponsored and published under the aegis of ECLAS, which considers landscape architecture to be the discipline concerned with the conscious shaping of people’s surroundings. ECLAS defines the tasks of landscape architecture as the planning, design and management of landscapes to create, maintain, protect and enhance places so as to be functional, beautiful and sustainable (in every sense of the word), and appropriate to diverse human and ecological needs (see http://www.eclas.org). Academics and professionals increasingly recognise how good practice and training both depend on excellent education and that all benefit from top quality research. Practitioners in a range of fields with which we interact look for evidence upon which to base design (such as which landscape design aspects contribute to human health and wellbeing). To develop the connections between research on the one hand and teaching and practice on the other, and to be able to define landscape architecture as a discipline that relies on its own body of knowledge, it is important to build a common framework of theory and of research practice. Any research, after formulating objectives and research questions, needs an appropriate overall approach and a methodology. A methodology is about selecting the most appropriate suite of methods for answering the research question(s). This book is firstly about the routes towards selecting the

4 …

Introduction „

methods that landscape architects (and others) may use in and for research and secondly about demonstrating some of the methods most appropriate for certain kinds of research. It is, thirdly, also about the choices researchers make: at the beginning and during designing and conducting their research, which methods to select and how to apply them. These choices – at both collective and personal levels – are profound. The first purpose of this book is therefore to provide some guidance concerning the rationales needed for selecting methods to meet specific research needs. The second purpose of the book is to provide the discipline with a resource on how to frame research and how to develop an appropriate research approach and methodology. It is written to be used in courses at master’s and especially at doctoral level where landscape architecturespecific approaches and methods are likely to be applied and critically reflected upon. In addition to educating the future generation of landscape architecture researchers, the book also aims to support teachers and supervisors. Where this book is likely to find a readership outside academia it will mainly be with research-orientated landscape consultants and staff members of research institutes. Many landscape architecture design and planning firms carry out research, such as that commissioned by government agencies and international institutions. Professionals thus engaged in research may have an interest in keeping up to date about landscape architecture research approaches and methods. Landscape architecture researchers will, particularly after successfully completing a study, feel the urge to contribute to building bodies of knowledge. Results from studies need to be published in international peer-review journals and high quality books; these journals and books also provide a platform for landscape architecture researchers to take part in the scholarly debate on design research. In this respect it should be noted that landscape architecture across the globe still needs to catch up in advancing the scholarly dimension of the field when compared with other design disciplines such as architecture and industrial design. A number of peer-review journals explicitly devoted to research exist within the larger field of design and advance the scholarly dimension of the field. Examples include Frontiers of Architectural Research, Journal of Design Research, Design Studies and Journal of Urban Design. While these generally focus on a broad and interdisciplinary understanding of design and design processes, including those of landscape architecture (Journal of Urban Design) and ‘landscaping architecture’ (Frontiers of Architectural Research), landscape architecture researchers rarely publish in any of these journals. We hope that, in the future, landscape architecture researchers will be publishing in these journals much more. Similarly, landscape architects also need to consider contributing to book projects. For example, although many chapters in The Routledge Companion to Design Research (Rodgers and Yee 2015) are relevant for landscape architects and landscape architecture researchers, none includes contributions specifically from the field of landscape architecture. In the future, books with a focus on landscape may be authored and co-authored by landscape architects much more. Researchers need not restrict their focus on books and journals with the term ‘landscape’ in their title, such as Journal of Landscape Architecture (JoLA), Landscape Research, Landscape Journal and Landscape and Urban Planning. Indeed, these journals regularly publish articles written by landscape architects, but the number of articles is relatively small in total, although they cover a broad variety of subjects. Moreover, apart from JoLA these journals do not qualify as typical landscape design journals. Nonetheless, they provide a welcome forum for publishing the results of landscape architecture research. It should be noted, however, that landscape architecture partly overlaps with the broader area of landscape studies, which draws from fields such as hydrology, soil sciences, landscape ecology,

5 …

www.allitebooks.com

Adri van den Brink, Diedrich Bruns, Simon Bell and Hilde Tobi

„

geography and environmental psychology, to mention a few. Through planning and designing, landscape architecture also reaches beyond landscape studies and represents an academic field in its own right. Hence, the third purpose of this book is to support the further development of this academic field by, for example, addressing a number of topics that we believe to be highly relevant for scholarly landscape research. The fourth and final purpose of the book is to contribute to help raising scholarly standards. Researchers may find, for example when submitting a manuscript to a peer-review journal, that manuscripts are not readily accepted by editors of those journals. Rejection and revisions are to be expected and they are not the end of the world, as it were. As explained by Maggie Roe in Chapter 5 of this book, both must be seen as part of the educational process that all, junior researchers and experienced professors alike, profit from in the long run. If rejected, in particular, one may initially feel misunderstood. After all the hard work that went into reporting what we are convinced is a great study, a rejection might turn into disappointment. Little by little, though, after taking the trouble of finding out what may have gone wrong, one might learn how to improve, first on the submitted manuscript, and later on future research and their publication. This book, therefore, also aims to help people understand what good research is and how results should best be reported in order for researchers to successfully take part in scholarly debate, and to publish successfully in the best journals that exist in the market. Viewed from both inside as well as outside, landscape architecture may not always have been recognised as a research discipline (e.g. Gobster et al. 2010; Deming and Swaffield 2011; Brown and Corry 2011). However, over the last 15–20 years, landscape architecture research has steadily been advancing. We are not yet out of the woods, so to speak, and much still needs to be done.

THE STRUCTURE OF THE BOOK The book has four parts. The following is a brief overview of the chapters in each of these parts. Part I is about raising awareness. It presents our views on academic research in landscape architecture and provides the foundation of this book. By relating the worlds of research and design, Chapter 1 discusses different types of landscape architecture studies, and, using examples from this book, explains how a landscape architecture research study might be designed. It also outlines the research process, the conceptual research design and the technical research design as well as addressing some practical aspects of actually doing research. In Chapter 2 the process approach to research in landscape architecture is discussed in more detail. The views expressed in these chapters are those of the editors and we do not expect everyone to agree with these – in fact we hope they will stimulate a debate about the nature of landscape architecture research and lead people to question their current views. Part II is setting the stage for discussing research methods. Chapter 3 provides thoughts on the role of theory in landscape architecture and it addresses whether theory is an input to or output from research. Chapter 4 includes reflections on the special role that designing has in research and in landscape architecture in particular. The act of designing is conceptualised here as an intrinsic part of doing research. Chapter 5 offers a journal editor’s view on how to improve the quality of research papers in landscape architecture and on key areas of future research. Based on two studies Chapter 6 reports on the state of research in landscape architecture, and it offers thoughts on enhancing research transparency and developing the research agenda. Part III discusses a number of selected approaches and methods. When designing research projects strategic decisions have to be made. Decisions include deliberations on research questions

6 …

Introduction „

and methodological strategies and also on choosing specific methods. Chapter 7 argues for making well-informed decisions for using a case study and selecting cases to meet criteria such as being a case of something specific or representative. This is important for landscape architecture researchers because the option of ’doing a case study’ often presents itself and forms a basis for many projects. The following chapters of Part III offer a selection of the broad palette of methods from which researchers may choose when planning their research. For example, for the purpose of informing landscape-changing designs, it is important to study former and current changes and also to find out what people perceive and think of landscape transformations. Chapter 8 presents the landscape biography approach and points to different methods for understanding the history of landscape. Chapter 9 discusses social media as vehicles for knowledge acquisition through collaborative processes, such as social discourse and public participation, and it also presents methods for analysing data shared on social media sites. With a focus on virtual environments, Chapter 10 discusses methods of landscape visualisation appropriate for use in research, including ’real-time virtual landscape models’ and ’virtual walks’ through landscapes. Chapter 11 also puts the focus on landscape perception and discusses research methods related to walking the ’real landscape’. Chapter 12 returns to the question, hinted at above, of how to develop transferable study results when design is itself dealing with singular and specific situations. This important question is addressed by presenting examples of design research that lead to guidelines aimed at informing landscape design practice. In Part IV we address some of the grand challenges that we consider important in landscape architecture research. Researchers are encouraged to add their own examples to the list. Starting with heritage management, a growing challenge in a rapidly changing world, Chapter 13 speaks about research that puts the focus on intangible landscape meanings and values and on their role in landscape design. Challenged by current health crises, particularly in the western world, Chapter 14 reports on research and methods that help to generate the knowledge needed for designing ’healthy landscapes’. Aimed at informing designs to create microclimates that ameliorate the effects of urban heat island intensification Chapter 15 is about researching thermally comfortable urban environments. Finally, Chapter 16 explores how landscape architecture research contributes to generating knowledge that helps to solve urban water challenges. With the aim to help readers navigate through the book and easily find where different parts connect, we have inserted a number of cross-references at several places in the book; they generally read as (author(s) – Chapter x). Making cross-references are not the responsibility of chapter authors but of the editors alone. We believe we have covered a wide range of relevant aspects related to research in landscape architecture in this book. Nevertheless, we needed to make choices, such as, for example, restricting ourselves to a selection of grand challenges in Part IV. We have also made choices about the research methods presented in Part III. Needless to say, there are many more research methods that are regularly being applied in landscape architecture research studies. As a number of good books exist about such methods we kindly refer readers to these. More generally we refer to text books on research methods that are highly relevant for landscape architecture research, such as Cresswell (2014) and Kumar (2014). We would encourage readers to consult such sources in combination with this book when designing a research project. We also refer readers to the ‘Suggested further reading’ sections that are included in most (not all) of the chapters. These sections include sources that help researchers to dive deeper into the subject matters discussed by chapter authors.

7 …

Adri van den Brink, Diedrich Bruns, Simon Bell and Hilde Tobi

„

ACKNOWLEDGEMENTS We would like to thank all the authors who contributed to this book. We are tremendously grateful for their support, which helped make this book possible. Developing the book was a fascinating exploratory journey during which we felt encouraged by the expertise chapter authors were generously willing to share. We would also like to thank our colleagues at Routledge, Sade Lee and Louise Fox, who never lingered to respond to our many questions and who greatly helped in the book production process. At Wageningen University we would like to thank the following colleagues for their important contributions to the book. Adrie van ‘t Veer did a fantastic job in (re-) drawing all figures and tables, for which he cannot be thanked enough. We also thank Monique Jansen for preparing all photos for publication and Marit Noest for designing the research process figures in Chapter 1, and for her assistance in preparing the manuscript.

REFERENCES Bell, S., Stiles, R. and Jørgensen, K. (2010) LE:NOTRE Two Output report: Research and teaching (‘Teaching for Research – Research into Teaching’). European Council of Landscape Architecture Schools, Internal report [online], available: http://www.le-notre.org. Bell, S., Sarlöv Herlin, I. and Stiles, R. (eds) (2011) Exploring the Boundaries of Landscape Architecture, Abingdon: Routledge. Brown, R.D. and Corry, R.C. (2011) ‘Evidence-based landscape architecture: The maturing of a profession’, Landscape and Urban Planning, 100, 327–329. Clewing, C. and Jørgensen, K. (2006) LE:NOTRE Output year 3 report: European PhD in landscape architecture, LE:NOTRE Outputs, Year 3. Oslo, Norwegian University of Life Sciences and European Council of Landscape Architecture Schools, internal report [online], available: http://www.le-notre.org. Cresswell, J.W. (2014) Research Design: Qualitative, Quantitative, and Mixed Methods Approaches, Los Angeles, CA: Sage. Deming, E. and Swaffield, S. (2011) Landscape Architectural Research: Inquiry, Strategy, Design, New York: John Wiley. Fetzer, E. (2011) Development of a European doctoral programme in landscape architecture. Selected results from LE:NOTRE survey on doctoral education. European Council of Landscape Architecture Schools, internal report [online], available: http://www.le-notre.org. Gobster, P.H., Nassauer, J.I. and Nadenicek, D.J. (2010) ‘Landscape Journal and scholarship in landscape architecture: The next 25 years’, Landscape Journal, 29(1), 52–70. Kumar, R. (2014) Research Methodology: A Step by Step Guide for Beginners, Los Angeles, CA: Sage. Rodgers, P.A. and Yee, J. (eds) (2015) The Routledge Compendium to Design Research, Abingdon and New York: Routledge. van den Brink, A. and Bruns, D. (2014) ‘Strategies for enhancing landscape architecture research’, Landscape Research, 39(1), 7–20.

8 …

PART I RAISING AWARENESS

This page intentionally left blank

Chapter 1: Advancing landscape architecture research Diedrich Bruns, Adri van den Brink, Hilde Tobi and Simon Bell

THE RELATED WORLDS OF RESEARCH AND DESIGN In all design disciplines, research and design go hand in hand. They are two sides of the same coin. The founders of landscape architecture, icons such as André le Nôtre, Lancelot ‘Capability’ Brown and Frederick Law Olmsted, as well as their successors to the present day, took thorough note of the needs of the people who were going to use and enjoy the landscape to be created. In addition, they studied, at every design location, land use, elevation and drainage, slope and light, soil and vegetation, views, possible walking routes and so on. They systematically acquired sitespecific knowledge on which to base firm proposals for landscape interventions. Through their work influential practitioners have inspired, and continue to inspire, generations of landscape architects. Two things are specific to a designer’s knowledge about a particular location and already emerged as those pioneers developed their practice. First, practitioner knowledge is defined as ’embodied’, that is it is the knowing of what to do under particular circumstances. Second, the knowledge is situational, that is it is difficult to transfer to other places and situations, and also time. It is not for nothing that ‘Capability’ Brown received his nickname: for seeing the capabilities in a specific piece of land which made a unique design possible while using a relatively restricted palette of materials. However, it is also true that the early practitioners did not carry out any fundamental research nor make especially systematic observations about different phenomena as a result of their work which could have informed the development of their ideas and designs. They came from a different tradition in an era when science was in its infancy and scientists were amateur gentlemen who could afford to indulge their private interests. Early landscape architects were quick to spot the possibilities offered by the new technologies which emerged during the industrial revolution, such as steam engines, cast iron and glass as well as profiting from the plant collecting expeditions and botanical research which enriched the parks and gardens. Scientific breakthroughs in pest control, fertilisers and plant breeding impacted practice. Later, during the 20th century, social science theories and empirical research also came to be applied. Landscape architects have long since been the beneficiaries and recipients of the results of research carried out by other related disciplines. Today, the relationship between research and design is becoming increasingly intense. As the remit of landscape architecture has expanded from the design of parks and gardens to much larger scales and with links to many more disciplines than architecture, horticulture, engineering and planning, the nature of design problems has become multi-faceted and more complex. The rather straightforward commissioner–designer or client–designer–contractor relationship common in the

11 …

Diedrich Bruns, Adri van den Brink, Hilde Tobi and Simon Bell

„

past has grown into complex societal tasks, combined with processes in which designers can play different roles at the same time. Ongoing technological development has opened a vast range of possibilities for design, representation and construction but has also meant that the range of knowledge needed to solve problems has become too great for any one person to be an expert in. These are some of the reasons why landscape architects are increasingly engaging in academic research. Landscape architects need much more than embodied and situational knowledge. The need for a wide variety of background research related to specific projects, sites or situations remains and is a necessary part of what a practicing landscape architect does in their day to day life. This, however, is not the same as academic research and should not be confused with it, even if the landscape concerned is broad in extent and rich in aspects which need to be explored before planning, design or management can begin. There is an ongoing open debate in the design disciplines in general, and not only in landscape architecture, about what might be meant by ‘research’ (e.g. Davis 2008). In this chapter we aim to clearly define how we as editors of this book perceive research in landscape architecture from a mainly academic perspective (though set in the context of the wider activities often described as ‘research’, including those aspects alluded to above). In doing so we provide some classifications and definition of terms (which some readers may find inclusive or exclusive depending on their point of view). To start with, the relationship between research and design can be grouped according to three categories (Lenzholzer et al. – Chapter 4). The first category is research on design that includes studies about the products of design, in our case, for example, historic gardens, modernist new towns or baroque parks. The second category is research for design that covers all types of research supporting the design process and the coming into being of the design product. Examples here might include research from the natural or social sciences which informs and provides evidence supporting design decisions. The third category is research through designing that includes all research and studies that actively employ designing as a research method. This may be comparable to developing prototype solutions to particular problems which are tested against certain criteria and in which the active, reflective process of design plays a fundamental role. A distinction is often made in science between fundamental or basic research and applied research. Fundamental research includes work which tries to understand the basic structure of the universe, for example, but for which the results have no specific and direct application to the solution of a problem, although the knowledge gained might lead to key technological breakthroughs later. In the context of landscape architecture the term ‘applied research’ may need clarification. Landscape architects strive to make an impact on society (including to make places more habitable, healthier or diverse), regardless of their role as researcher, teacher, professional designer or a combination of all three (which is widely accepted as a profound way of linking academia with innovative practice). Some may speak about ‘project-based research’ rather than ‘applied research’ and the term ‘project’ includes all real-world landscape design projects. However, this may mean that the research done for some projects only yields situational or embodied research results. If we are trying (as we editors believe we are) to raise the game and to move forward in depth and quality of research we must try to make a clear distinction. Retaining the term ‘applied research’ is more valid as it reflects the fact that this is what we do. The difference is that, in aiming to carry out much more evidence-based design, the quality of the evidence and its applicability must transcend the specific site. While it may not be possible to generate generalisable results which can be applied absolutely universally (as do the laws of physics – except in the vicinity of black holes)

12 …

Advancing landscape architecture research „

the results should be generalisable within a broad but limited set of conditions and much more so than being applicable to a single situation. Thus, many of the results of research for landscape architecture are very widely applicable within, for example, similar cultural settings, similar climates or similar ecosystems. Many findings about, for example, how access to nature enables us to recover from stress, have been found to extend very broadly and with few limitations. It is important not to confuse ‘project research’ with a ‘research project’, a term used to describe any sort of study including, for example, a discrete doctoral research project. Project research is research for a specific planning or design project and embodied or situational research. A research project starts with a problem which may or may not be associated with the physical landscape and if it is, may not be concerned with planning or designing that landscape but with understanding phenomena by using it (or more than one) as an exemplar or case study. This is not to say that the results may be less interesting or even valuable, it is just that they are different in kind and in purpose and this difference is crucial for anyone moving from practice into academia or for anyone commissioning research. Nor does it matter whether the research project is commissioned or noncommissioned. Non-commissioned research may be what an individual academic does as part of their daily academic life – pursuing an interesting and rewarding topic out of personal interest, or it may be what many PhD students do (as long as they can get a grant for it!). Much applied research is commissioned by national or international funding agencies where the problem is presented to the academic world who have to compete to win the funding. Doctoral and post-doctoral research can – and in many cases will – be conducted as part of or in the context of commissioned research. We also think it might be helpful to distinguish between academic research on the one hand and non-academic research on the other in landscape architecture. This distinction refers to principles that academics must observe individually, amongst each other, and in their responsibility towards society, when engaging in research. Examples of such principles are scrupulousness, reliability, validity and impartiality. These principles can be read as general notions or norms of good academic practice, in other words: quality standards about what research should be and how it should be conducted in order to qualify as research that contributes to enhancing any disciplinary knowledge base, in this case that of landscape architecture (many countries and disciplines have a good research practice declaration with these norms). This is the kind of research this book is about. Such research is primarily conducted by academic, university-based researchers and sometimes by academic researchers employed by independent institutes. Examples are doctoral and post-doctoral studies or as senior researchers working on medium to large-scale commissioned projects (e.g. funded by the European Union (EU)), or else academics and professional landscape architects doing research as part of their job and publishing their work in peer-review journals or as high quality, refereed reports. Such research also complies with scholarly standards in terms of its relationship to the existing body of research in the specific field, its theoretical grounding and its methodological soundness. Writing for popular and professional design journals and inquiries at the project level that inform designing can include some kind of research. Searching databases or literature for information and writing up some conclusions to inform policy is frequently referred to as research and companies may employ people to do this as ‘researchers’ but they are not the same as our use of the term in academia. Although such work can be very important and useful, it is not the kind of academic research this book addresses. The scholarly principles mentioned above are strongly related to the way research is conducted, including the methodology and methods applied. Valid and reliable knowledge, for example, is

13 …

Diedrich Bruns, Adri van den Brink, Hilde Tobi and Simon Bell

„

Table 1.1 Research proposal checklist

knowledge that is the outcome of methodologically sound research. The choice of research question, the research set-up, the choice of method and the references to sources used all need to be accurately documented in a form that allows for verification of each step taken in the research process. These are necessary conditions for research to consistently and reliably contribute to the body of knowledge in landscape architecture. However, research may also be about ‘seeing things’ that others don’t see, such as formulating challenging research questions that lead to new knowledge (which is, by the way, one of the hardest aspects of doing research), and about creatively exploring new paths to study these questions. This is where research methodology and research methods come in to play their vital role in enhancing the landscape architecture knowledge base. In order to qualify as research in our terms, researchers might find it useful to test a proposal against a set of simple questions which are in effect the logic of academic research and which can be mirrored in the development of a research project from a practical standpoint (see Table 1.1).

TYPES OF LANDSCAPE ARCHITECTURE STUDIES While making indispensable design contributions landscape architects must also continue broadening the role they are playing in society. They must contribute by generating new knowledge which can be applied more generally, although usually within some limits – it is difficult to expect that we can come up with universal laws! This may be, for example, knowledge about the materiality and concepts of landscapes, and about the processes that shape them. This may also be about the perception and experience of landscapes, about design methods and about the way in which knowledge from other disciplines may be combined with design knowledge. Confronting the grand challenges society is facing from a landscape architecture perspective will contribute our special approaches to tackling problems. We believe we have a lot to offer, by producing research and also by converting and melding scientific results into practical solutions. The latter can be a viewed as a kind of post-research activity (producing planning or design guidelines can be one such role). The challenges of climate change, energy transition, urbanisation, health, food security and others can successfully be addressed only if the driving processes and the spatial and human dimensions are considered together. Landscape architects provide and will continue to make important contributions while society is undergoing the necessary social and physical transformations of land, space and environment (ESF/COST 2010; van den Brink and Bruns 2014).

14 …

Advancing landscape architecture research „

Landscape architecture research is designed to expand our knowledge related to the shaping of landscapes and to processes of landscape intervention at various scales. For the purpose of this book we define landscape by (a) the interaction of the human and non-human, and (b) the human perception of the resulting material phenomena, that is, features and processes (Roe 2013, p.401). A landscape is both about the ‘phenomenon itself and our perception of it’ (Wylie 2007, p.7). A particular area may be studied in purely physical terms, but, since each area also carries multiple meanings ‘that emanate from the values by which people define themselves’ (Greider and Garkovich 1994, p.1) landscapes are also studied with respect to all kinds of cultural and social practice, including symbolic representation, memory and so on. An integrated approach aimed at yielding knowledge through landscape architecture research will, especially when pursued in the depth appropriate for academic research, often deal with three aspects in a triangle combination: image, structure and action (Figure 1.1). Image is all about perception, symbolism and the communication content of landscapes. Structure deals with the fabric of the landscape spatially, materially and in depth, for example the layers which constitute it. Action can include processes occurring in the landscape, taking place in it (e.g. human activities) or external forces affecting it. The integrated nature comes about when there are mutual dependencies which need to be understood through research: for example, how people in a city make use of a park for various activities (action) depends both on the design and layout of the park (structure) and what they think of it – its safety or attractiveness (image). Likewise, the research can be organised to include methods which capture some aspects of each dimension and one of the dimensions can act as the entry point to addressing the problem. Missing one of the dimensions may lead to inadequate or incomplete results which cannot be applied successfully. One of the hallmarks of landscape architecture research therefore is its multi-faceted and multi-disciplinary nature. Landscape architecture research involves studies that belong to a wide range of different types. Case studies have been identified as very popular modes of research; cases feature in the majority of published peer reviewed articles and are often very prominent in conference presentations (Swaffield – Chapter 7). Some researchers have attempted to study multiple cases within a single research project and, through systematic comparison, succeeded in accumulating generalisable knowledge (Prominski – Chapter 12). In addition to case studies, there is wide range of study designs one

Figure 1.1 Three aspects of doing research

15 …

Diedrich Bruns, Adri van den Brink, Hilde Tobi and Simon Bell

„

may use for a specific type of study and to plan and undertake a research project (Deming and Swaffield 2011). For example, if research aims to inform practice about the design solutions which are most effective for flood risk management, it might be prudent for researchers to know how much confidence they can place on the study findings; a priority would thus be to consider the ‘level of evidence’ one might be able to obtain when deciding between particular types of study. A completely different strategic challenge is determining the long-term balance of resources and benefits needed for the research to achieve its objectives. What size should a landscape architecture study be, and for how long should it be conducted? For example, a researcher may need to choose between cross-sectional surveys or longitudinal studies (Ward Thompson – Chapter 14). In a short term cross-sectional study expenses could be high because the number of survey subjects might need to be large and several research assistants may be needed all at once. Conversely, in a long-term (observational) study investigators might strive to assess design outcomes in groups of participants. For example, investigators may observe a group of adults to learn more about the effects that specific designs or environments have on people’s wellbeing after one, three and five (or even more) years. One research assistant might be needed who pays annual visits to conduct interviews with participants, and the same questions would need to be asked of the same study participants for at least five years. If no resources for research assistants can be made available after three years, and if no changes are observed between one and two years, the decision might be made to save money and stop the study prematurely. What happens though, if the really interesting effects present themselves after five years, or even later? Funding periods do not usually even span periods of five years. In a perfect world, landscape architecture research would strive to identify all sorts of planning, design and management strategies to improve landscapes for the benefit of people and their environment. It would study all possible effects, beneficial or otherwise, that designs may produce over long periods of time. Conducting such large and long-term studies requires planning for several years of investigation and including large numbers of participants, possibly in representative populations. Alas, in landscape architecture this is neither practical nor affordable. The reality of landscape architecture research is that most studies that have been carried out to date are small in size and short in duration. In particular, studies carried out in the context of a PhD are usually done by one person and not by a large team of investigators, their timelines are set by university regulations and their financial and other resources are limited to their stipend or grant plus some small costs (if they are lucky).

DESIGNING A RESEARCH PROJECT In this section we explain our understanding of (a) the research process, (b) the conceptual research design and (c) the technical research design. Referring to Figures 1.2, 1.3 and 1.4 we present and discuss at a generic level the different components of a research project. These are nested figures like matryoshka dolls. Examples are given to make explicit what we are talking about. The choice of examples is based on those described in the book, so that readers may look at the relevant chapters to get a clearer view of how, for instance, a ‘conceptual framework’ and a ‘study design’ might actually look. In Chapter 2 Tobi and van den Brink discuss our process approach to research in landscape architecture in more detail.

16 …

Advancing landscape architecture research „

Research process Figure 1.2 illustrates our understanding of the research process as a whole. One or more ‘research questions’ are identified. A ‘research project’ is developed, consisting of three stages: designing the research project, doing the research and reporting on findings/results (e.g. publishing a paper or a report). Research findings represent a knowledge gain which fills a knowledge gap. A knowledge gap is usually the starting point for developing a research project. This structure clearly maps onto the logic of research presented earlier.

Figure 1.2 Research process overview

17 …

Diedrich Bruns, Adri van den Brink, Hilde Tobi and Simon Bell

„

Identifying gaps in existing bodies of knowledge and generating relevant research questions are not easy tasks. To be original and innovative, a research question (or problem) must be identified that has been answered insufficiently or not at all in landscape architecture or another field of study. Of course, one could simply read a number of published articles and, by examining the parts which include ‘suggestions for future research’, draw up a list of interesting looking topics and choose one of them. One could also try and ask experts to offer advice and opinion. One could also use research websites and tools such as ‘Google Trends’ to find out about the general interest in certain topics that way. However, creative landscape architecture researchers might rather choose a more ambitious approach. If so, we suggest that researchers identify and try addressing a ‘challenge’. A challenge may be different things. It may be calling existing assumptions into question; in the field of landscape architecture one could include questioning and thus disrupting established ways of designing (for example by using social media, as discussed by van Lammeren et al. – Chapter 9). A research challenge could also be based on or answering to a real-world design challenge by simulating them as in virtual reality (as discussed by Hehl-Lange and Lange – Chapter 10). Then there are societal challenges that need addressing. The aim of a research project that responds to such a challenge would be to generate or (dis)confirm knowledge in order to help inform design solutions that would be addressing this challenge. Included in this book are examples where researchers responded to the need for publicly demonstrating intangible values pertaining to cultural heritage and their landscape-based meaning (Taylor – Chapter 13), to health related challenges (Ward Thompson – Chapter 14), also to the effects of urban heat island intensification (Brown and Gillespie – Chapter 15) and to urban water management challenges (Backhaus et al. in Chapter 16). The examples collected in this book merely represent the tip of the iceberg, so to speak, and many more challenges present themselves to be taken up by the landscape architecture researcher. Sustainable mobility (‘green transport’), demographic change, biological and cultural diversity and inclusive governance are just a few of the many globally important challenges. In addition there are the challenges and priorities set by national governments and international organisations such as the EU’s Horizon 2020 research framework which sets out what are considered to be the big challenges facing society and the planet – not dissimilar to the challenges discussed in this book – but with the added incentive of large amounts of money and the disincentive of extreme competition for this largesse. Nevertheless, the calls for research projects set out a series of problems and look for innovative ways of addressing them through research. The real-world problems that trigger the demand for new knowledge provide both background and purpose (i.e. solving this problem) for the research, and they are formulated in terms of one or more research objectives. The extent to which these objectives are reached after the research is done and the findings are reported is a measure for the contribution of the research to filling the identified knowledge gap or to informing landscape design practice.

Conceptual research design Figure 1.3 puts the focus on designing the research project. This stage of the process consists of (a) the ‘conceptual research design’, and (b) the ‘technical research design’. The conceptual research design addresses the why and the what of the research. The why focuses on the identification of the ‘need to know’, that is the knowledge gap regarding a specific challenge (as discussed above). The what is about the research questions (RQs) that relate to the identified knowledge gap

18 …

Advancing landscape architecture research „

Figure 1.3 Conceptual research design

and challenge. A general research question (GRQ) might be formulated first and then divided into a limited (usually not more than three to five) number of specific research questions (SRQs). For example, when researchers are generally asking how urban areas must be designed so that ‘they will provide thermally comfortable environments and minimise the impact of urban heat islands’, Brown and Gillespie (Chapter 15) suggest defining two specific (interrelated) questions. The first question is ‘How do built environments affect the macroclimate to create microclimates?’ The second question is ‘How do humans perceive their thermal comfort under different microclimatic conditions?’ (Also note that these relate to the structure and image aspects of the triangle mentioned earlier – Figure 1.1 – and may lead to consequences for the action part in terms of the interaction of physical condition with perceived comfort.) In her chapter on researching links between landscape and health, Ward Thompson (Chapter 14) first asks how landscapes must be designed that ‘support health in mind and body’ and, by putting the focus on the local community scale, then poses several GRQs that each again leads to two or three SRQs. For example, staying with the neighbourhood scale and targeting a specific age group, the GRQ ‘How should environments be designed that support outdoor access in an ageing society?’ leads researchers to specifically ask ‘Do older adults’ perceptions of the qualities of their physical neighbourhood environment predict their quality of life?’ and ‘What qualities of a local park or open space are associated with older people’s varying levels of outdoor activity and quality of life?’ Both the why and the what of the research project are informed by a conceptual framework that describes how all or parts of the research questions are conceptualised. The conceptual framework consists of relevant definitions and theories (for the role of theory see Thompson – Chapter 3). Basic to landscape architecture are definitions of ‘landscape’ (see above in this chapter), ‘cultural

19 …

Diedrich Bruns, Adri van den Brink, Hilde Tobi and Simon Bell

„

landscape’ (Taylor – Chapter 13), and of ‘designing’ (Lenzholzer et al. – Chapter 4). How people perceive and understand landscape may be conceptualised through phenomenological theory (Schulz and van Etteger – Chapter 11), and also by constructivist theory, while discourse theory helps to explain how ideas about landscape are constructed through communication among different actors (van Lammeren et al. – Chapter 9). Research specifically responding to a grand challenge includes, in the case of urban climate studies, the ‘concept of energy budgets’ used for modelling microclimates as well as human thermal comfort in outdoor environments (Brown and Gillespie in Chapter 15). In addition to referring to established theory researchers may also decide to develop a conceptual framework that specifically meets the needs of their study. For example, in her research pertaining to landscape and health in an ageing society Ward Thompson (Chapter 14) builds on existing theory and develops a concept that specifically links environmental attributes of people’s perception of them in relation to activities of older people. In the case of analysing and exploring the history of landscapes Kolen et al. (Chapter 8) are developing concepts that are integral parts of the landscape biography approach.

Technical research design Figure 1.4 puts the focus on the technical research design and on finding answers to the how-theresearch-will-be-done question. In Figure 1.4 each SRQ corresponds to one ‘research module’, but it may also be possible to respond to more SRQs in one research module. For example, Backhaus et al. (Chapter 16) combined methods of site observation with photographic documentation, semi-structured interviews and document analysis to find answers to the one specific question of how stormwater management projects are ‘performing’ several years after being constructed. Similarly, Taylor (Chapter 13) suggests combining a number of different methods to find answers to the one specific challenge of heritage landscape assessment. Ward Thompson (Chapter 14), by measuring ‘aspects of activity, wellbeing and quality of life’ and relating these to ‘outdoor environmental attributes’ is responding to two SRQs by a ‘mixed methods approach’ (see Chapter 2). This involved several phases, ‘starting with focus groups and then a wider survey based on a questionnaire’. In some instances researchers might have to choose between several different approaches that all may appear useful to answer one SRQ. For example, aiming to identify the thermal comfort of a person in an outdoor environment, a common approach is to ask people. This approach has, as Brown and Gillespie (Chapter 15) pointed out, the disadvantage, at least regarding perception of thermal optima, of being subject to cultural effects that might lead to considerable variance between responses. One approach without this disadvantage would be ‘to observe where people choose to spend time in a landscape’, assuming that humans will seek out thermally comfortable locations. A third approach would be to take measurements of people’s core and/or skin temperature. Again, both alternative approaches come with advantages and disadvantages, and well-informed decision making is required. The same applies to operationalising concepts used in the research question. For example, as a concept thermal comfort is probably understood differently by different people (volleyball players, people sitting on a bench). Researchers must define all concepts as input or outcome factors. One needs to know what is meant by a statement of ‘better or worse thermal comfort’ (e.g. too hot for ball players, too cold for bench sitters), particularly where value judgement is included such as in heritage values, and in aesthetic judgements.

20 …

Advancing landscape architecture research „

Figure 1.4 Technical research design

For each research module the researcher must find a way to describe (a) the tools, or ‘instruments’, by which one collects data, and (b) the sampling strategy, the strategy by which one selects the units from which the data will be obtained. Both go together and are part of the research design. For the purpose of measuring solar radiation, for instance, Brown and Gillespie (Chapter 15) explain how data on solar radiation are collected by means of a pyranometer, a measuring instrument that senses the strength of solar radiation. The output (signals) from the pyranometer sensors is read and recorded using a data logger. When the study design is experimental, protocols must be clear and

21 …

Diedrich Bruns, Adri van den Brink, Hilde Tobi and Simon Bell

„

rigorous application of rules carefully observed, for responses to be reliable (Hehl-Lange and Lange – Chapter 10). Measuring, reading and recording of data are not always straightforward. Schulz and van Etteger (Chapter 11) use walking to study experiences made from the first-person point of view. Sensory impressions are collected and detailed landscape information gathered. They obtained data at points that are fixed along a route at equal distances from one another. Recorded data were translated into graphs and maps. In his chapter on heritage and landscape, Taylor (Chapter 13) explains how data are collected for the purpose of conducting a historic landscape assessment study. When unable to obtain and record empirical evidence through measuring things themselves, researchers will systematically search for information on landscape history in existing sources, and also by on-site landscape examination. While measuring radiation is done by following operating instructions, few generally accepted standards exist on how to extract data from a walk, or from private and public records, family archives, newspapers, historic plans and maps. (There are guidelines on standardised record extraction and especially on how to document and report on it.) Researchers will need to read up on exemplary cases and also take guidance from experienced colleagues and experts from other fields. This is particularly true in cases where, in search of evidence about people’s ‘history of life and work’, researchers are developing new methods for gathering local landscape related knowledge, including memory about ‘old practice’, anecdotes and so on (Kolen et al. – Chapter 8). Standards do exist for preparing and using several of the methods landscape architecture researchers regularly borrow from the social sciences. For example, a questionnaire is often listed as a data collection ‘instrument’ by which data are obtained and interviews are processed by transcribing. Sources from empirical social sciences are used to prepare researchers to employ these methods that are not traditionally, but are increasingly becoming part of, the landscape architecture research repertoire, as can be seen in several chapters of this book. As pointed out by Ward Thompson (Chapter 14), it is ‘always advisable to use existing, well-tried tools wherever possible’ such as, for example, existing questionnaire modules that have successfully been used and tested.

CONCLUSIONS In this chapter we have tried to clarify what we see as the precise nature of research within landscape architecture. At the starting point research meant collecting background information about a specific landscape for purposes of planning or design – embodied and situational – tasks which are of course still vital in landscape architecture practice. Then, an evolution took place where the scope and scale of research developed away from the site specific and practical to the more academic and generalisable. However, we still find confusion about where the boundary lies and people still describe research for a project as a research project in an academic sense. Academic research in all aspects of landscape architecture as a disciplinary field has been growing. Indeed, the rate of publication in academic journals as well as the synthesis of research into high quality books has substantially increased over the last 10–20 years. Our endeavour in this book is to help to stabilise the position of what academic research is, to strengthen the understanding of what research practice involves and to demonstrate (some of) the range of research methods which are suitable for addressing the kinds of challenges facing us.

22 …

Advancing landscape architecture research „

REFERENCES Davis, M. (2008) ‘Why do we need doctoral study in design?’, International Journal of Design, 2(3), 71–79. Deming, E. and Swaffield, S. (2011) Landscape Architectural Research: Inquiry, Strategy, Design, New York: John Wiley. ESF/COST (2010) Landscape in a changing world. Bridging divides, integrating disciplines, serving society, Science Policy Briefing, October, No. 41. Greider, T. and Garkovich, L. (1994) ‘Landscapes: The social construction of nature and the environment’, Rural Sociology, 59(1), 1–24. Roe, M. (2013) ‘Animals and landscape’, Landscape Research, 38(4), 401–403. van den Brink, A. and Bruns, D. (2014) ‘Strategies for enhancing landscape architecture research’, Landscape Research, 39(1), 7–20. Wylie, J. (2007) Landscape (Key Ideas in Geography), London: Routledge.

23 …

Chapter 2: A process approach to research in landscape architecture Hilde Tobi and Adri van den Brink

INTRODUCTION The research-as-a-process approach is at the foundation of this book. In Chapter 1 the editors of this book presented their view on academic research in landscape architecture by means of three nested figures outlining the research process (Figure 1.2), the conceptual research design (Figure1.3) and the technical research design (Figure 1.4) respectively. We advocate a process approach to doing research because the actual process involves developing research questions, developing the methodology, collecting data, analysing data and reporting results. The process approach is especially well-suited for conducting empirical research – research that generates evidence and derives knowledge from such evidence rather than from theory or belief – although of course there ought to be some theoretical foundations too. Our purpose here is to set out the process approach and some fundamental aspects of research methodology in the social and natural sciences as well as in the arts and the humanities. Research methodology is the discipline that systematically studies the full research process from designing a research project to reporting of results, including research methods or techniques and research instruments (Kampen and Tobi 2011). Studying research methodology lies at the heart of every academic discipline. This chapter should help readers of this book to develop a common understanding of what research and research methodology are and how methods can be applied, developed and critically assessed in landscape architecture research. We have chosen, in this chapter, to introduce the research methodology perspective for two reasons. Firstly, when talking about research methods the diversity of terminology seems to make the topic rather complex. This is less the case if the discipline of research methodology with its own common language is used. The second reason is that enhancing landscape architecture research may call for the critical assessment of well-known research methods and instruments as well as the development and the assessment of new ones. This may happen, for example, when new domains of study are being explored. These new methods need to be embedded in a clear understanding of research methodology. Of course, the reverse may also be true, as we think it likely that the study of research methods in the context of landscape architecture can also contribute to research methodology and the use of research methods in other domains. As landscape architecture research is generally multi- or interdisciplinary and perhaps even transdisciplinary in nature (e.g. Deming and Swaffield 2011), building a common research language and methods set is a challenge. After all, the natural sciences, the social sciences, the arts and

24 …

Research in landscape architecture „

humanities have very different cultures (Kagan 2009) and different notions of ‘scientific research’ and how it should be done. A literature review of the collaboration between natural and social scientists (Fischer et al. 2011) identified two barriers relevant for interdisciplinary domains, such as landscape architecture, that we think are best countered by means of a process approach. First, the differences in paradigms or epistemologies (Kuhn 2012 [1962]) used by current mono-disciplinary sciences hinder the interpretation of the results and approaches from, for example, the natural sciences to the social sciences and vice versa. This may be overcome by collaboration. A process approach to interdisciplinary research focuses on what actions need to be planned and how they can best be undertaken to answer the research question(s) concerned. To take full advantage of this the researcher needs to think outside the box of a single dominating disciplinary epistemology to embrace others. The second barrier is the skills and competences of the researchers involved who, for example, fail to understand each other’s academic vocabulary or jargon. The process approach facilitates mutual learning through meaningful communication because the emphasis is placed on research design, research methods and research outcomes and less on (differences in) epistemologies. We believe that acquiring a common methodological language is vital for landscape architecture if it is to advance as an academic discipline that relies on its own body of knowledge rather than theories or beliefs. However, generating such a common language may be both challenged and enriched by the fact that landscape architecture research is interdisciplinary in nature. As noted in the Introduction and Chapter 1, it draws on a variety of different knowledge areas and their related methods of inquiry as well as its own: such an eclectic collection of methods may prohibit a common understanding of methodology. Acquiring a common language may be enriching if it facilitates making connections within and between different research cultures and opens aspiring researchers to approaches from which inspiration and experience can be derived. This chapter also aims to identify many of the implicit and explicit decisions that researchers make in the course of their research to increase transparency of the research process. We also provide some hands-on advice for making decisions on, for example, the study design and the data collection methods to be employed. With every one of these decisions ethical considerations must be taken into account, for example about which data to collect and which not to collect. It is important that these decisions and considerations are thought over and reported in a way that is as explicit and transparent as possible. This chapter is structured as follows. After describing the process approach to research methodology in more detail we discuss five major steps in research design. These steps are: formulation of the research question, study design, data collection and sampling, data analysis and reporting. Each of these will be briefly described, terminology will be (re)introduced and their relevance to the research process as a whole will be explained. Readers might like to refer to Chapter 1 for examples from this book for each of the steps explained below. Here we are addressing them from a methodological perspective.

RESEARCH METHODOLOGY AND THE PROCESS APPROACH Research methodology takes a pragmatic stance. It is the research question(s) that determines the choice of study design, methods of data collection, and instruments used or developed for that data collection and so on, and not the researcher’s preference or previous experience. Consequently, the distinction between quantitative research (roughly characterised as research in which numbers are

25 …

Hilde Tobi and Adri van den Brink

„

collected and analysed, e.g. statistically), and qualitative research (in which words, images, sounds, etc. are collected and not statistically analysed) becomes blurred. In interdisciplinary research practice the classic divide between ‘traditional’ (i.e. quantitative) research methods and ‘holistic’ (i.e. qualitative) research methods hardly exists anymore. Many studies in natural science, traditionally focused on quantitative methods, may nowadays also incorporate qualitative methods. For example, a study into the outcomes of a newly introduced technology may start off with interviews in order to decide what indicators to use for testing its success or, at the end, might include interviews on why the new technology did (or did not) meet expectations. At the other end of the spectrum, humanities such as linguistics and law (traditionally known for relying on qualitative text analysis) also use quantitative approaches, for example a quantitative analysis of text using computer simulations in linguistics or the weighting of certain characteristics in the assessment of legal plans. Interviews and texts but also talks and visual materials may be analysed through both quantitative and qualitative methods. Because of the range of options it is therefore essential to distinguish between the data that will be collected and the method with which this data will be analysed. Landscape analyses that aim to understand the relevant social– ecological interactions in the landscape under study are typically based upon such a mix of research methods from each of the ‘three cultures’ or ‘empires’ as Thompson calls them (see Chapter 3) and also to fulfil the needs of combining image, action and structure as noted in Chapter 1. Research methodology, particularly in the social sciences, evolved from the study of quantitative approaches, via the inclusion of the study of qualitative approaches, to the quantitative and qualitative combination referred to nowadays as ‘mixed methods’ (Creswell and Plano Clark 2011; see also Adamson 2005). The introduction and popularity of mixed methods is based on the complementarity of both quantitative and qualitative methods. Obviously, this complementarity requires readers of, for example, Cresswell and Plano Clark’s (2011) textbook Designing and Conducting Mixed Methods Research, to be familiar with both quantitative and qualitative research concepts. This has probably helped to reduce the qualitative–quantitative divide within the domain of research methodology. This may be illustrated by Creswell’s (2014) bestseller, Research Design, in which the three approaches (i.e. qualitative, quantitative and mixed methods approaches) are taken together in the subtitle. The main title emphasises the importance of research design. Kumar (1999) was one of the first authors to organise a methodology textbook around research design as a process. He distinguished three phases: deciding on what to research, planning a research study and conducting a research study. In the planning phase, or as we call it the technical research design phase, decisions on study design, method of data collection, instruments used for data collection, sampling or selection strategies, and data analysis are distinguished and made. When conducting the study, the field work may demand a change of plan, but the fact remains that a departure from the plan is better than no plan at all. Then the process approach requires an explicit and transparent description of the changes made to allow a critical assessment of the research and methods used, regardless of the outcomes. The advantage of applying a process approach may be illustrated by the following example. Traditionally, a choice of either a qualitative or a quantitative approach seemed to lead automatically to the choice for a case study yielding qualitative data or an experiment or survey yielding quantitative data. The process approach forces the researcher to put the research question at the very centre of the research design. Consequently, the process approach may lead to the choice for interviews as the data collection method within an experiment, or quantitative questionnaires within a case study.

26 …

Research in landscape architecture „

As shown elsewhere in this book (see Meijering et al. – Chapter 6), many topics in landscape architecture research call for collaborative research efforts. An open, though critical, mind for methods is required to meet the academic challenges of the discipline. When the research question is leading, instead of personal preference or experience, choices on methods need to be made explicitly and shared in scientific reports. This openness on the choice and use of methods enables systematic reviews to identify and enhance the landscape architecture knowledge base. This would obviously benefit both researchers and practitioners within landscape architecture as well as those from neighbouring disciplines and helps the development of evidence-based landscape architecture (Brown and Corry 2011).

FORMULATING A RESEARCH QUESTION Usually, a research project starts with a topic: a very rough idea that originates from a real-life problem, a gap in knowledge brought to the attention of the researcher, a debate or professional curiosity. This topic is then screened to get a feel of context and constituting parts, and to enable identification of relevant scientific and professional literature. Traditionally, a literature review was prone to bias due to the researchers’ network and lack of awareness of, or access to, relevant publications and the human tendency to cherry-pick. The introduction of computerised bibliographic databases such as Web of Science and Scopus have enabled systematic approaches to finding relevant articles, reports and books, regardless of the aim of the literature review. Systematic literature review methodology was developed a few decades ago for the medical sciences to provide academics, professionals and policy makers with information on treatments, taking as much scientific evidence as possible into account. More recently, systematic review methodology has also found its way in the social sciences. Systematic review methodology provides a range of tools to identify, select and synthesise research articles and reports (e.g. Gough et al. 2012; Petticrew and Roberts 2006). In this book an example of such a systematic literature review is presented by Meijering et al. in Chapter 6, another example in landscape architecture is the systematic review of benefits of urban parks for the IFPRA (Konijnendijk et al. 2013). Once a knowledge gap has been identified, either a set of research questions or a set of testable hypotheses is constructed. A testable hypothesis is characterised by its potential to be falsified. For example, a theory may contend that the frequency with which people visit a neighbourhood park depends on the distance between their home and that park. This hypothesis would have to be rejected if the researchers find no association between distance to the park and frequency of visiting and they would have to use the data and its analysis to construct an alternative explanation (theory) to replace the failed one. Whether it is possible and meaningful to formulate and test hypotheses depends on what is already considered known as well as the purpose of the research. In the past, landscape architecture researchers appeared to be mostly interested in exploring and describing phenomena, less in testing and predicting them. This is confirmed by Meijering et al. (see Chapter 6). A next step also helped by the literature review is the development of a conceptual (theoretical) framework to study the research question. Such a framework may be built upon theories and concepts from different disciplines. Examples include substantive theories, such as the Theory of Island Biogeography (MacArthur and Wilson 1967) that formed the basis for studying and implementing ecological networks (e.g. Bakker et al. 2015), and procedural theories, such as the theory of communicative action (Habermas 1981) that inspired Healy (2006 [1997]) and many others to develop

27 …

Hilde Tobi and Adri van den Brink

„

their thoughts on collaborative planning. An example from the humanities is the Aesthetic Creation Theory that is championed by Zangwill (2007). This theory holds that works of art have aesthetic functions that are essential to them, but also allows that they have other, non-aesthetic functions, for example practical or ecological ones. It thus removes the strict distinction between fine art and the useful arts to which works of landscape architecture can be counted (van Etteger et al. 2016). The conceptual framework should be coherent and appropriate for framing and answering the research question. It can be seen as the lens through which the researcher views the particular part of the real world demarcated by the research question. It also explains how these lenses are related to theories and evidence provided in the literature, what is known about the key concepts used in the research question and how these key concepts are related to each other. It should be noted, however, that the lenses are a tool to study another object and not an object of study themselves.

RESEARCH DESIGN AND STUDY DESIGN Research design is the design of the research project as a whole. Study design is the design of the set-up of data collection (e.g. cross-sectional study, panel study, pre-test–post-test design) within the research project. Three characteristics of the research question lead to the choice of study design: the level of control required, the number of data collection waves, and the reference period for data collection. The fit between study design and research question shapes the internal validity of the research. When the research looks for the description of a phenomenon (e.g. tourism in nature reservations), co-occurrences (e.g. tourist and wildlife numbers), or the identification of a set of possible post hoc explanations (decreasing wildlife numbers results from increasing tourist numbers), an observational study may be appropriate. Generally, an observational study is not suitable for establishing causation, although, in an observational study design, the use of a particular theory can make a causal claim plausible (Yin 2014; Swanborn 2010) but this claim cannot at the same time be used as proof of the general theory, as this would be circular reasoning (Swanborn 2010). When knowledge claims on causation are aimed at, one of the many experimental or quasiexperimental study designs is required (Campbell and Stanley 1963). These designs help to ensure that the cause precedes the effects and that the study of such a possible causal relation is not disturbed by a third factor (e.g. a reduction in the number and size of floods in the nature reserve that may explain the decrease in wildlife and the increase in tourist numbers). In an experimental design the researcher has full control over the when, what and to whom of the exposures under study and randomly assigns study participants to the experimental conditions. Exposures can be related to behaviour (e.g. sunlight to the growth speed of plant species) or an individual attribute (e.g. particular tree formations to the aesthetic value according to participants). The only differences between a quasi-experimental design and an experiment is that the researcher cannot randomise participants to exposures in a quasi-experimental design or have complete control over many of the variables such as weather. Because experimental designs often require a laboratory setting, quasiexperimental designs may, in many instances, be more useful in landscape architecture research than experimental study designs. Overall, threats to the internal validity of causality statements (i.e. the correct claim of a causal relationship) are better controlled in experimental than in quasi-experimental designs, whereas threats to external and ecological validity (i.e. the possibility to generalise the results of the study) may be better controlled in quasi-experimental designs. The required number of data collection

28 …

Research in landscape architecture „

waves (i.e. the number of times data are collected) is also determined by the research question. This also applies to the timeframe of the study. For example, a study on how a polder was reclaimed in the 1960s requires a retrospective study design, whereas the evolution of the same landscape since then under a range of changing rural development and conservation policy schemes would call for a retrospective–prospective study design. In general, research questions with a longitudinal component always demand multiple data collection waves. In conclusion we would like to emphasise that, in presenting their study design, researchers will need to be more precise than simply referring to container concepts such as ‘experiment’ or ‘case study design’. Container concepts are not sufficiently informative of the structure of data collection. Some case studies have one data collection wave whilst others have multiple data collection waves. Some include an experiment or quasi-experiment whilst others are observational. The only characteristic that all case studies seem to have in common is the aim to study a phenomenon in a particular context, with attention to multiple perspectives and data sources. Examples include multiple public and private stakeholders in an energy landscape development project and multiple data sources (archives, historical maps, etc.) to turn to when composing the biography of a specific landscape (Kolen et al. – Chapter 8).

DATA COLLECTION AND SAMPLING The key concepts in the research question should also direct decisions on the type of data collection method to be adopted, such as observation, interviews and questionnaires. For example, in research that aims to describe the actual behaviour of people in an open and frequently hot city square, data are best collected by means of observation, whereas data on the preferences of these people is best collected by means of questionnaires or interviews. The population under study, for example human beings or plant species phenomena, plots of land or water bodies, further reduces the number of possible data collection methods as only humans can fill in questionnaires and participate in interviews. While in chemistry, physics and biology detailed measurement procedures and measurement instruments are generally accepted (BIPM 2012), in social sciences and neighbouring disciplines this is not the case (Kampen and Tobi 2011). This means that operationalisation is required (Tobi 2014). Operationalisation is the iterative decomposition of the key concept until identifiable characteristics or measurable variables are reached in an explicit and auditable way that is usually structured by theory. For example, if one wants to study the effects of air quality on urban liveability, one needs to specify which chemical elements (e.g. sulphur oxides, nitrogen oxides or particulates) will be measured and how. One would also need to specify how the concept of urban liveability would be made operationally meaningful to investigate the relationship with air quality. Here the conceptual framework is directive: it would determine whether the definition of urban liveability includes only life expectancy and mortality rates or also includes other aspects such as social well-being. Only after operationalisation can appropriate scales and questionnaires be recognised as such or constructed. The construction and assessment of data collection instruments is a full-grown branch of research within the academic discipline of research methodology. Some guidance on questionnaire construction and survey design can be found in Dillman (2007) and Czaja and Blair (2005), on scale and test construction in Shultz et al. (2014) and DeVellis (2012) and on interviews in Rubin and Rubin (2012). Operationalisation and data collection methods together lead to the decision who (or what) are eligible as study objects or study participants. For example, a researcher could observe characteristics of children’s play in a natural playground (e.g. operationalised as the number and type of children

29 …

Hilde Tobi and Adri van den Brink

„

interactions with different tools and with other children). However, to obtain information on how well the children slept afterwards the researcher would need to ask either the children or the parents, dependent on the children’s age. After deciding who (or what) are the study participants (or objects), the researcher can look for a sampling frame (e.g. list of children, consumers, garden plots, historical sites) to draw a sample in a way that best reflects the theoretical population (i.e. the population to which one aims to generalise) and the requirements of the research question. The advantage of a random sample is that it allows for statistical inference (e.g. Kumar 1999). The alternative would be to choose a non-random sampling method. Non-random sampling draws on the theory used and under investigation, to select a particular site or particular informants (e.g. Carter and Henderson 2005). The selection of cases for a case study is usually purposive, non-random (see also Swaffield – Chapter 7). It should be noted that, depending on the research question, it may make perfect sense to draw a random sample and refrain from statistical analysis because the data collected is of a qualitative nature. Finally, in landscape planning and design studies the claim of ‘representative sampling’ seems to lead a life of its own. Simply calling something ‘representative’ is meaningless unless information is given on what it is representative of and how this representativeness resulted from the sampling frame (see also Swaffield – Chapter 7). In conclusion, based on the key concepts in the research question and the population concerned, data collection methods and data collection instruments are chosen or designed. Study objects (such as sites) or study participants (such as people) are either sampled or selected. To allow the reader to assess and learn from the strategy chosen, the sampling frame or the selection scheme must be presented in scientific reports.

DATA ANALYSIS A distinction is usually made between data processing and data analysis. Data processing precedes data analysis. Data processing includes the transfer of data from one medium to another (say paper to computer), the identification of erroneous data due to typing errors or technical problems, the handling of missing data (e.g. the development of imputed data sets), the identification and handling of very extreme data, the so-called outliers, and the linking of different data sets (e.g. linking questionnaire data of individuals to geo-social data on the neighbourhood they live in). Also simple standardised calculations (e.g. in health sciences, calculating body mass index based on length and weight as provided by the study participants) can be seen as data processing as can be statistical checks on variance, representativeness and so on. The outcomes of data processing are the data sets used for analysis. Decisions on the handling of missing or incomplete data and the handling of outliers will also affect results. To protect the analysis from researcher bias and incorrect inference these decisions should be taken prior to analysis and shared in the research reports. When proposing and especially when reporting research there must be a section that describes the data analysis, planned or performed, to answer the research question. The data analysis section in the research proposal will help to reduce the tendency to cherry-pick, or to fish or mine the data for a statistical significant result or a particular ‘juicy’ quote (although what is actually done may differ considerably from what is proposed owing to a range of practical and other factors). Primary and secondary analyses are often distinguished. Primary analyses provide descriptive information and a general answer to the research question. Secondary analyses, undertaken conditional on the outcome of the primary analyses, are more in depth or provide greater detail (e.g. a subgroup analysis).

30 …

Research in landscape architecture „

Another common distinction made in data analysis uses the type of data being analysed: qualitative data analysis and quantitative data analysis. It should be remembered that mixed methods research also requires both qualitative and quantitative data to be analysed separately, with the order and timing of analyses and synthesis set by the mixed method design (Creswell and Plano Clark 2011). A distinction in quantitative research is between descriptive and inferential statistical analyses. Much landscape architecture research has been restricted to descriptive statistics, for example frequencies of park use, as opposed to regression analysis to explain who is likely to use parks the most, given certain conditions, for example accessibility. In quantitative data analysis, the statistical analyses ought to be described at a level of detail that is sufficient to allow other researchers to repeat them. Often the software package used is also stated in the report. The combination of the statistical techniques used and the estimated statistical proficiency of the readership of a journal will make it appropriate either to refer to statistical textbooks and other sources, or to describe the method used in some detail. In qualitative data analysis the method most informative to the research question should also be chosen and described. Different strategies and procedures can be distinguished in qualitative data analysis methods, for example metaphor analysis and domain analysis (Coffey and Atkinson 1996), discourse analysis and membership categorisation analysis (Silverman 2006). Each of these methods makes assumptions about the data (e.g. spontaneous speak or interview exchange) and has its own strengths, weaknesses and challenges. All qualitative data analyses involve some sort of identification and labelling of relevant parts, the so-called coding, and can be supported by software (e.g. Atlas.ti or QSR nVivo). A top-down coding approach refers to the attribution of pre-defined codes to the data. Usually these codes are based on the theories used. In bottom-up coding the codes are not pre-defined but based on the words used in the data (e.g. in the conversation transcript or the legal document). Here too, the researcher must make pertinent choices explicit (Silverman 2006; Flick 2006), first of all about the coding (top down or bottom up), and describe the analytical strategy chosen. Coding enables the reader to evaluate the suitability of the analytical strategy chosen for answering the research question with the particular kinds of collected data. It also allows other researchers to do similar analyses. In conclusion, data processing needs to be distinguished from data analysis. In data processing many pertinent decisions are usually taken on which data will be included in the analysis. These decisions need to be made explicit. Numerous data analysis methods are available. In the case of qualitative data the analysis may be supported by software.

REPORTING The importance of good quality reporting cannot be overstated. Reporting aims not only at sharing research with the field at large in such a way that readers can assess the trajectory that lead to the results reported and learn from the research choices made. In addition, the report is a constituent part of the research process – results are no use, however good, sitting on a shelf or in a computer – they must be reported and set in context. An academic article needs more than a convincing story to be scientifically sound: it needs to inform the readers about how the study was done and should provide sufficient detail to allow an assessment of the methodological rigour within the theoretical framework provided. It cannot be emphasised enough that the writing of the report or article should start as early as possible. Writing is just another way of thinking; writing helps to reflect continuously on the research process.

31 …

Hilde Tobi and Adri van den Brink

„

A research report or article typically starts with an introduction in which the research objective and question is introduced and put into the context of, for example, one of the grand landscape challenges or a design assignment. Generally, a second section provides further details on the conceptual framework, it includes the theories used for the operationalisation of the relevant concepts. The third, methods section, obviously plays a pivotal role in light of the theme of this book. This explains the design of the research and gives details of different aspects of the study design. It also provides details of the data collection methods and the measurement instruments used, the sampling or selection of study subjects and informants, how the data were processed, and what methods for data analysis were used and how (e.g. with the help of what software). When a new data collection instrument is being constructed, details of the construction and test phase of the instrument must also be included. The next section presents the results of the study. It gives information on basic characteristics of the study participants (e.g. species, demographics) and outcomes of the analyses. The report or article is concluded by a discussion and conclusion section (ideally divided into two sections). No new information from the study is given here but the main results are discussed in the light of previous and future research, referencing the literature review and considering the implications of the results. The discussion section also needs to include a reflection on the methodological strengths and weaknesses of the study, how these may have affected the study results and which lessons can be learnt from it. The conclusion should, ideally, formally answer the research question(s) (with the limitations or uncertainties to these answers made explicit). Too many reports somehow forget what the research questions(s) was and fizzle out with limited or rather banal conclusions.

RESEARCH INTEGRITY In Chapter 1 several principles of good academic practice were mentioned. These principles belong to the field of research integrity, a wide and expanding field that, in this context, we can only scratch the surface of. In general, codes of conduct for scientists should be followed in order to protect both the study object/subject and the researcher. In most countries the national scientific community (e.g. universities and/or research institutes) has endorsed such codes of conduct. These codes set out the core principles of research integrity and ethics to which researchers should adhere. All refer to academic standards that are internationally shared and that, in some form, are also practised by peer review journals. We recommend readers to inform themselves about the content of their national code and also about how research integrity is warranted at their own institution and where to seek further advice on specific research integrity issues. When planning any type of data collection several ethical issues should be considered. Data collection always costs someone something, not only to the researcher (time, effort, money), but also to other people involved in the study, for example informants or study participants. Therefore, as a first responsibility, the researcher must be sure that the data collection is truly necessary. To help warrant the well-being and safety of humans involved in research, basic ethical principles, such as respect for persons, beneficence, justice and informed consent (i.e. people under study need to be fully informed about and voluntarily agree with the purpose of the study and to take part), need to be taken into account. One of the consequences of these principles is that neither the data collection method nor how the results are published must violate the privacy of those who cooperated in collecting the data.

32 …

Research in landscape architecture „

For audit purposes, data collection procedures should be recorded and both the raw data and the data sets used for analyses should be securely stored and be accessible. These data include those of, for example, (quasi-)experiments, questionnaires and interviews, modelling results and modelling software. Data storage is the individual responsibility of the researcher. Universities and research institutes across the world have data management policies that should help researchers to develop their own data management plan. Editorial boards of peer review journals increasingly demand that authors of articles submit statements about data management and research integrity.

REFERENCES Adamson, J. (2005) ‘Combined qualitative and quantitative designs’, in Bowling, A. and Ebrahim, S., eds. Handbook of Health Research Methods: Investigation, Measurement and Analysis, Oxford: Open University Press, 230–245. Bakker, M.M., Opdam, P.F., Jongman, R.H.G. and van den Brink, A. (2015) ‘Model explorations of ecological network performance under conditions of global change’, Landscape Ecology, 30: 763–770. BIPM (2012) International Vocabulary of Metrology: Basic and General Concepts and Associated Terms (VIM). Available at: http://www.bipm.org/utils/common/documents/jcgm/JCGM_200_2012.pdf. Brown, R.D. and Corry, R.C. (2011) ‘Evidence-based landscape architecture: The maturing of a profession’, Landscape and Urban Planning, 100, 327–329. Campbell, D.T. and Stanley, J.C. (1963) Experimental and Quasi-Experimental Designs for Research, Chicago, IL: Rand McNally College Publishing Company. Carter, S. and Henderson, L. (2005) ‘Approaches to qualitative data collection in social science’, in Bowling, A. and Ebrahim, S., eds. Handbook of Health Research Methods: Investigation, Measurement and Analysis, Oxford: Open University Press, 215–229. Coffey, A. and Atkinson, P. (1996) Making Sense of Qualitative Data: Complementary Research Strategies, Thousand Oaks, CA: Sage. Creswell, J.W. (2014) Research Design: Qualitative, Quantitative & Mixed Method Approaches, 4th edition, Thousand Oaks, CA: Sage. Creswell, J.W. and Plano Clark, V.L. (2011) Designing and Conducting Mixed Methods Research, Thousand Oaks, CA: Sage. Czaja, R. and Blair, J. (2005) Designing Surveys: A Guide to Decisions and Procedures, Thousand Oaks, CA: Sage. DeVellis, R.F. (2012) Scale Development: Theory and Applications, Thousand Oaks, CA: Sage. Deming, E.M. and Swaffield, S. (2011) Landscape Architecture Research: Inquiry, Strategy, Design, Hoboken, NJ: John Wiley. Dillman, D.A. (2007) Mail and Internet Surveys: The Tailored Design Method: Update with New Internet, Visual, and Mixed-Mode Guide, Hoboken, NJ: John Wiley. Fischer, A.R.H., Tobi, H. and Ronteltap, A. (2011) ‘When natural met social: A review of collaboration between the natural and social sciences’, Interdisciplinary Science Reviews, 36(4), 341–358. Flick, U. (2006) An Introduction to Qualitative Research, Thousand Oaks, CA: Sage. Gough, D., Oliver, S. and Thomas, J. (2012) An introduction to Systematic Reviews, Thousand Oaks, CA: Sage. Habermas, J. (1981) The Theory of Communicative Action (two volumes), Boston, MA: Beacon Press. Healy, P. (2006 [1997]) Collaborative Planning, Shaping Places in Fragmented Societies, 2nd edition, Basingstoke: Macmillan. Kagan, J. (2009) The Three Cultures: Natural Sciences, Social Sciences, and the Humanities in the 21st Century, Cambridge: Cambridge University Press. Kampen, J.K. and Tobi, H. (2011) ‘Social scientific metrology as the mediator between sociology and socionomy: A cri de coeur for the systemizing of social indicators’, in Burt, C.M., ed. Social Indicators: Statistics, Trends and Policy Development, Hauppauge, NY: Nova Science, 1–26.

33 …

Hilde Tobi and Adri van den Brink

„

Konijnendijk, C.C., Annerstedt, M., Nielsen, A.B. and Maruthaveeran, S. (2013) Benefits of Urban Parks: A Systematic Review, a report for IFPRA, Copenhagen and Alnarp. Available at: http://curis.ku.dk/ws/ files/44944034/Ifpra_park_benefits_review_final_version.pdf. Kuhn, T.S. (2012 [1962]) The Structure of Scientific Revolutions, Chicago, IL: University of Chicago Press. Kumar, R. (1999) Research Methodology: A Step-by-Step Guide for Beginners, 4th edition, Thousand Oaks, CA: Sage. MacArthur, R.H. and Wilson, E.O. (1967) The Theory of Island Biogeography, Princeton, NJ: Princeton University Press. Petticrew, M. and Roberts, H. (2006) Systematic Reviews in the Social Sciences: A Practical Guide, Oxford: Blackwell. Rubin, H.J. and Rubin, I.S. (2012) Qualitative Interviewing: The Art of Hearing Data, Los Angeles, CA; London; New Delhi; Singapore; Washington, DC: Sage Publications. Shultz, K.S., Whitney, D.J. and Zickar, M.J. (2014) Measurement Theory in Action, New York; London: Routledge. Silverman, D. (2006) Interpreting Qualitative Data: Methods for Analyzing Talk, Text and Interaction, London; Thousand Oaks, CA; New Delhi: Sage Publications. Swanborn, P.G. (2010) Case Study Research: What, Why and How?, Thousand Oaks, CA: Sage. Tobi, H. (2014) ‘Measurement in interdisciplinary research: The contributions of widely-defined measurement and portfolio representations’, Measurement, 48, 228–231. van Etteger, R., Thompson, I. and Vicenzotti, V. (2016) ‘Aesthetic creation theory and landscape architecture’, Journal of Landscape Architecture,11(1), 80–91. Yin, R.K. (2014) Case Study Research: Design and Methods, Thousand Oaks, CA: Sage. Zangwill, N. (2007) Aesthetic Creation, Oxford: Oxford University Press.

34 …

PART II SETTING THE STAGE

www.allitebooks.com

This page intentionally left blank

Chapter 3: The role of theory Ian H. Thompson

Academics often exhort their students to declare their theoretical perspectives, so it is right that I should declare mine before plunging into the difficult topic of ‘theory’. Prior to becoming a landscape architect, I studied philosophy as an undergraduate. I was taught by Wittgensteinians and took an optional course in Wittgenstein’s philosophy of language. As a result I carry around with me certain ways of thinking, particularly about words and meanings, which are, so to speak, tools in my theoretical toolbox. The most useful of these is the idea that the meaning of a word is to be found by considering its use. If we encounter a word which puzzles us, the way to understand it is to look at the many and various ways that the term is used.

A TROUBLESOME WORD ‘Theory’ can be just such a troublesome word. When we think we have got a handle on it, we discover that someone else is using it in a completely different way. If someone asks us ‘do you have a theory about this?’ they are probably asking for an explanation. On another occasion, someone might remark ‘oh, but that’s only a theory’ suggesting that something proposed is merely a supposition, and doesn’t have the sort of authority that might come, for example, from rigorous observation or scientific testing. People also talk about ‘making contributions to theory’, ‘adding to theory’, ‘developing theory’ and so on, all of which suggest that theory can be accumulated. Similar phrases are used about ‘knowledge’ and indeed people do sometimes talk about ‘theory’ as if it is a synonym for ‘knowledge’. On the other hand, people say things like ‘what theoretical perspective are you taking?’ (see my first sentence!) or ‘try looking at this through the lens of (this or that) theory’. In landscape architecture (and I’m sure in many other fields) one often hears the complaint that we don’t have enough theory (as if it is a kind of stuff that we can pile up like gold) or that we don’t have sufficient theory that is truly our own (we have only borrowed it from other disciplines and this is somehow shameful). There is a commercial aspect to this latter grumble. If a profession can claim to have a body of knowledge to which it has an exclusive entitlement, and if this is the basis for its expertise, then its practitioners have a form of monopoly in the marketplace, but this is a matter of business or sociology rather than epistemology. Attempting to corral all the theory that pertains to landscape architecture is probably impossible, although a decade ago Michael D. Murphy had a good stab at it with his Landscape Architecture Theory: An Evolving Body of Thought (2005). He had the good sense to set some limits, focussing specifically upon ‘the body of knowledge required to inform design thinking and on ways to apply

37 …

Ian H. Thompson

„

that knowledge to improve the human/landscape condition and enhance quality of life through design performance’ (Murphy 2005, p.vi). Notice, incidentally, that ‘theory’, ‘knowledge’ and ‘thought’ are used virtually as synonyms here, and this is not unusual. However, when considering the place of theory in research design, rather more precision is required. Another troublesome word: ‘landscape’. To make matters even more difficult, the word ‘landscape’, which we might expect to denote the object of our enquiries, is pretty slippery too. Murphy deals with it rather too briskly by opting for what he calls the ‘traditional definition’, that is ‘an area of the earth’s surface that has been modified by human activity’, which he takes from J.B. Jackson’s Discovering the Vernacular Landscape (1984), but in the light of more recent scholarship, we cannot make do with this anymore. We cannot even say that a landscape is a complicated object because, as many theorists have pointed out, landscape is something mental as well as something physical. Cultural geographers, such as Denis Cosgrove and Stephen Daniels (1988; see also Cosgrove 1984), pointed out that ‘landscape’ is not a neutral term, but an ideologically charged ‘way of seeing’. Kenneth Olwig, on the other hand, responded to the ‘scenic’ understanding of landscape by focussing upon landscape as a set of customary practices bound by precedent and law (Olwig 2002). Don Mitchell has delivered a bracing materialist critique of the notion of landscape by revealing the unjust social and economic relations involved in the production of agricultural landscapes (such as the strawberry fields of California) (Mitchell 1996, 2003, 2007). To shake things up further, a raft of influential work has come out of phenomenology, non-representational theory and performance studies, which John Wylie has admirably summarised in his short book Landscape (2007). Perhaps the most challenging idea to emerge from this recent work, for landscape architects at least, is the notion that the landscape, far from being designed, is performed, that is made and remade by a succession of repetitive actions, often guided by custom and precedence. How these customary actions relate to designed interventions by landscape architects, planners and managers has yet to be fully explored. Although I am reluctant to draw a distinction between researchers in landscape architecture and researchers who approach landscape from some other perspective, that of an archaeologist, say, or a geologist, ecologist, environmental psychologist or cultural geographer, the fact is that landscape is a transdisciplinary concept. This book may be written primarily for landscape architects who want to do research, but such readers need to be aware of where their own research might fit into the much broader field of landscape studies. It is worth asking ‘What particular topics might a researcher who is a landscape architect be better placed to tackle than someone from a different discipline?’ Perhaps these topics include such things as: the landscape design process in general, the design practices of particular landscape designers or offices, the history of designed landscapes, the education of landscape architects, the aesthetic, social and ecological values underlying landscape interventions, evaluations of the effectiveness of design interventions and so on. Murphy (following Ndubisi 1997) draws a distinction between ‘substantive theories’, which ‘promote a better understanding of the landscape as the interface between human and natural process and are descriptive and predictive’, and ‘procedural theories’ which ‘originate from design practice and the academic development and technical application of knowledge in a social setting’ (Murphy 2005, p.27). However, many aspects of natural science, which one might have expected to be included under substantive theory, including geology, ecology, climatology, soil science, hydrology and so on, are left to a later chapter which deals with ‘the biophysical environment’, while his chapter on substantive theory bundles together sustainable development, environmental psychology and systems theory. I confess that

38 …

The role of theory „

I don’t know what was in his mind here. Knowledge in all of these fields is certainly relevant to landscape architecture, but his criteria for identifying theory, let alone landscape architecture theory per se, are not very clear. Murphy’s chapter on procedural theory is much more straightforward. Under this rubric come such matters as the design process, design programming, data gathering and analysis, landscape planning and landscape suitability analysis. He might also have included the material concerning design practice and design collaboration which appears towards the end of the book. All of these topics are closely related to the profession of landscape architecture and if one were looking solely for theory that has developed out of the practice of landscape architects themselves, these would probably be the best places to look.

DISCIPLINE AND PROFESSION There isn’t space to fully examine the complex relationship between the academic discipline1 of landscape architecture and the profession that goes by the same name. Landscape professionals have to compete for business with other professionals, such as engineers and architects, and the institutional apparatus of accredited training, examinations, continuing professional development and so on was created as much to secure territory as to guarantee expertise. It isn’t quite the same for academics, who may be more open to overlaps and transdisciplinary ways of thinking, but there would be no discipline (no schools, no teachers and no researchers) without the profession. Theory is often invoked in relation to practice, where it means something like ‘a statement of method’ or ‘knowledge of general principles’. A couple of examples spring to mind. One is the Survey-AnalysisDesign (S-A-D) methodology which was a staple of landscape architecture, in Britain anyway, in the 1970s and ’80s and was based upon the idea of Survey-Analysis-Plan model originally developed by the Scottish town planner Patrick Geddes (1854–1932). S-A-D was later criticised for being too deterministic and inimical to imaginative design, but it held sway for a long time and, to the extent that tutors still ask students to undertake surveys and analyses, it is still with us. Interestingly this is not mentioned in Murphy’s book – perhaps it was not taught in the US – although similar models, such as the ‘six-step design process’, do appear (Murphy 2005, pp.63–64). The second example, also from landscape planning, is Carl Steiniz’s Landscape Change Model (1994), which he developed further under the label Geodesign (2013). Steinitz, who is now emeritus professor of landscape architecture at the Graduate School of Design, Harvard University, was a pioneer in the use of computer technology in landscape planning. He was particularly concerned with the methods landscape professionals use to analyse large areas of land and make important design decisions (see also Lenzholzer et al. in this book – Chapter 4). He produced a flowchart which organised the landscape planning process around a series of models, each associated with a question. For example, ‘representational models’ of the landscape were associated with the question ‘How should the landscape be described?’, ‘process models’ with the question ‘How does the landscape operate?’ and so on, through six stages. In the full version of the process, each of these steps must be gone through three times, first to establish the context and scope of the enquiry, second (in reverse order) to specify the project methodology and a third time to actually perform the study. Steinitz’s research amounts to a formulation or codification of data gathering, analysis, testing and decision procedures which might otherwise be undertaken in a haphazard or unsystematic manner. As such, it appeals to those who wish to bring rigour to their practice, though it probably

39 …

Ian H. Thompson

„

has little appeal to intuitive designers and those who place themselves at the more artistic pole of the landscape profession. Another important figure is Ian McHarg, author of the seminal book Design with Nature (1969), who developed the technique of suitability analysis while Professor of Landscape Architecture and Regional Planning at the University of Pennsylvania. In terms of Murphy’s distinction between substantive and procedural theory, both Steinitz and McHarg can be seen to have made major contributions to procedural theory. It is worth noting that both of these thinkers have been borrowed or even claimed by other disciplines, largely on the basis of their contributions to the development and use of geographical information systems. Here, at least, are instances of landscape architecture exporting theory, rather than importing it.

BETWEEN THREE EMPIRES Some of landscape architecture’s difficulties with theory arise as a consequence of its relationship to the three great empires of academia: the natural sciences, the social sciences, and the arts and humanities. My use of the word ‘science’, it is worth mentioning before going any further, follows the anglophone conventions in which its extension does not include the arts and humanities. Anglophone readers are sometimes startled to find that their conference papers on garden history or modernist design principles are being considered by a ‘scientific committee’ which conjures up a vision of investigators in white lab coats. In much of Europe, ‘science’ can mean any sort of scholarly activity. I use the metaphor of ‘empire’ to hint at the rivalry and territorial friction between these blocs, but Jerome Kagan, emeritus professor of psychology at Harvard University, chose the title The Three Cultures for his revealing discussion of the differences between three ways of investigating and understanding the world (Kagan 2009). In this he was following the usage coined by C.P. Snow (2012) whose provocative book The Two Cultures and the Scientific Revolution, was first published in 1959. Snow, both a physical chemist and a novelist, had been concerned to reveal the extent to which scientists were ignorant of the arts, but – more seriously for Snow – the extent to which highly educated British elites, educated in the humanities, were illiterate in science. The social sciences, however, he entirely ignored, despite their growing influence after the Second World War, when, for a few decades at least, it seemed that they might hold the answers to problems as wideranging as alcoholism, school failure and mental illness. Kagan’s book gives them their place, but as a psychologist he is aware of the borderline position of his own discipline. Some psychologists, he notes, gravitated towards the study of brain activity and were embraced by biology. Others, more humanistic in their approach, stayed with social science. It is also worth noting that ideas born of humanistic psychology, such as the theories of Freud, Jung, Klein and later Irigaray, Kristeva and Lacan, have been hugely influential in the humanities. Landscape architecture occupies a similar position in the borderlands and researchers have a similar choice of paradigms. Murphy contrasts two positions within landscape architecture, one exemplified by the American landscape architect Garrett Eckbo (1910–2000), who took a creative stance and argued that it was the profession’s role to create fresh and innovative ways for people to relate to their physical environment, the other epitomised by McHarg’s approach, which emphasised the scientific approach to ecologically sound landscape planning. Murphy, optimistically perhaps, thinks that the discipline has transcended this apparent battle between the empires of art and science (he does not suggest a champion for social science) and says that all landscape architects are holists

40 …

The role of theory „

now (Murphy 2005, pp.26–27). Nevertheless, it is clearly important, when setting out on a piece of research, to know which of the three cultures it most closely relates to, because, as I will show, there are different norms, values and even language in each. At present, within the academic world as a whole, there is much emphasis on collaborative and cross-disciplinary working, which suggests that methods from the three cultures might be creatively brought together. Landscape architects ought to be well placed to facilitate or participate in such collaborations. Landscape architects are sometimes inter-disciplinarians par excellence, able to bring together and synthesise very different perspectives and forms of knowledge, and equally capable of talking the language of aesthetes, agronomists or archaeologists. At other times this in-between position can be a headache. Certainly when considering theory it can be a problem, since ‘theory’ tends to mean different things within these three domains.

THEORY IN NATURAL SCIENCE In the natural sciences (physics, chemistry, biology etc.) a theory is an explanation which is generally accepted to be true. A hypothesis, on the other hand, is an informed guess, based upon observation. The philosopher of science Karl Popper suggested that hypotheses can never be proved, only disproved, arguing that if a hypothesis, or a group of linked hypotheses, survives repeated experimental testing, it may come to be accepted as a theory (Popper 2002 [1934]). A theory offers an explanation of how nature works. Important theories are named and often come to be well-known outside the confines of their disciplines. Some have great predictive power and offer abundant benefits to humanity. A good example is the Germ Theory of Disease which replaced the earlier idea that diseases were spontaneously generated, thus opening up the way to lifesaving medical procedures such as sterilisation. In some scientific disciplines, particularly physics and chemistry, one also finds ‘laws’ and it is reasonable to ask what is the difference between a theory and a law. The term ‘law’ tends to be used to predict what nature will do in certain conditions and this is often expressed mathematically. An example from chemistry would be Boyle’s Law which states that the pressure exerted by a gas held at a constant temperature varies inversely with the volume of the gas. Biology, perhaps the scientific field with the most bearing upon landscape architecture, has produced many theories but few laws and this is probably an indication of the complexity of the living world, which is not easily reduced to simple mathematical formulae. Natural science, which for our purposes can also be taken to include applied science, medical research and all varieties of engineering, is so useful and has such commercial potential that it is generally very well-funded and institutionally secure. Part of the appeal of McHargian landscape theory was that it sought to place landscape architecture on a scientific basis, which, it was felt, would improve the discipline’s standing in the academy and the profession’s offer to potential clients. Landscape architecture has, therefore, often aligned with natural science. This accounts for the interest in environmental psychology which came to the fore in the 1980s through the work of researchers like Rachel and Stephen Kaplan at the University of Michigan (Kaplan and Kaplan 1989; Kaplan 1995b). As Kagan notes, psychology, because of its central interest in human behaviour, is often classified among the social sciences, but Kagan considers ‘investigators who study the biological bases for, or evolutionary contributions to, animal or human behaviour as natural scientists’ (Kagan 2009, p.4). The Kaplans’ Information Processing Theory offered an experimentally based explanation of landscape preference, which builds upon evolutionary theory. The researchers

41 …

Ian H. Thompson

„

were psychologists working within the School of Natural Resources and Environment, which is also the home of a long-standing landscape architecture programme. Later work looked at the restorative potential of natural environments (Kaplan 1995a), a topic also explored by Terry Hartig of the Department of Psychology at Uppsala University and his collaborators (Hartig et al. 1991, 2003). Catharine Ward Thompson and her multidisciplinary research team at Edinburgh University called OPENSpace, which included the psychologist Peter Aspinall, looked at the influence of outdoor environments on people’s physical activity, particularly walking (Ward Thompson and Aspinall, 2011; see also Ward Thompson in this book – Chapter 14). One might say that all of these examples rightly belong to the discipline of psychology, but landscape architects have been integral to many of these research projects and the research findings are directly relevant to the activities of landscape design and planning. This work can therefore also be considered as part of the body of theory which pertains to the discipline. Another example is landscape ecology, which can be considered a sub-discipline of ecology, although, once again, the fuzziness or porosity of boundaries allows us to say that it also belongs within the disciplinary field of landscape architecture. It took off with the publication of landmark books by Naveh and Lieberman (1984), and Forman and Godron (1986). Arthur Lieberman is an emeritus professor of landscape architecture who collaborated with the agronomist Nev Naveh to write Landscape Ecology: Theory and Application which was the first English language monograph on the transdisciplinary science of landscape ecology. Richard T.T. Forman is an ecologist working within the Graduate School of Design at Harvard University, which also hosts one of the world’s preeminent master’s programmes in landscape architecture. Landscape ecology held the promise that landscapes could be designed and planned to achieve ‘an optimal spatial arrangement of ecosystems and land uses to maximize ecological integrity’ (Forman 1995, p.522). Robert Riley, who edited Landscape Journal from 1987 to 1995, argued for a narrow, science-based definition of ‘theory’ that would limit its use to ‘knowledge that explains some real-world phenomenon’ (Swaffield 2002, p.2). ‘Anything concerned with what to do or why to do it, instead of how to do it, is proudly proclaimed as theory. This is not theory; this is pseudotheory’ (emphasis in the original) Riley proclaimed, in an argument which is reminiscent of the logical positivists’ strictures against discussions of aesthetics and ethics (Riley 1990, p.48). Logical positivism was a mid-twentieth-century school of philosophy which argued that the only propositions which had meaning were either analytic (i.e. statements in logic and mathematics that were tautologically true by the meaning of their terms) or those that could be empirically verified, such as those of natural science. Everything else was, quite literally, meaningless, including for instance statements about landscapes being beautiful or democratic decision-making being a good thing. Riley did not go quite so far, but he thought that discussions about the place of landscape architecture in society, for example, should be labelled ‘frameworks’, not dignified by the term ‘theory’. But logical positivism’s influence has waned and attempts to regulate language in the way Riley proposed are generally doomed to failure. If landscape architecture were simply a natural science, Riley’s directive might have had some chance of sticking, but as the discipline overlaps with the arts and social sciences it simply could not hold. In large swathes of the academy, positivism is now out of favour, indeed we are in the midst of a kind of intellectual war between two camps, characterised by the philosopher Simon Blackburn as ‘Objectivists’ versus ‘Subjectivists’ (Blackburn 2005). Among the objectivists are traditionalists, modernists, rationalists, universalists and natural scientists. In the ranks of the subjectivists are relativists, postmodernists, social constructionists and contextualists, including many in the social

42 …

The role of theory „

sciences and humanities. Riley and those who think along similar lines clearly belong with the objectivists, but it is not clear that they remain in the ascendant. Deming and Swaffield (2011, p.3) suggest that there are three epistemological positions: reality may be independent of its relationship to the investigator, it may be dependent upon that relationship or it might be interdependent with it. Only the first of these follows the positivist model of (most) natural science (when we get to relativity and quantum physics, things get more complicated). The second concerns the sort of insight that might be delivered by an artist or by a social scientist using subjectivist methods (see the section below ‘Theory in the arts and humanities’ on phenomenology). The third involves what Deming and Swaffield, following Crotty (1998), call ‘constructionist’ methods. These presume ‘that knowledge is generated through the interaction between the investigators (and their society) and a reality (or realities) that exists but can never be known independently of the presumptions of the investigators’ (2011, pp.8–9). Students and early career researchers need to be clear about which of these models they are employing when designing their research.

SCIENTISM In the Tractatus Logico-Philosophicus (1922) Wittgenstein, wrote ‘even if all possible scientific questions be answered, the problems of life have still not been touched at all’ (6.52). A healthy respect for natural science and the benefits it can bring should be distinguished from scientism, which is the belief that empirical science alone can answer every question that has ever troubled humanity, including matters of morality, aesthetics, consciousness and religion, which have historically been seen to lie beyond its reach. Critics of scientism often accuse its proponents of reductionism, which can involve the translation of complex phenomena, such as the appreciation of art, the appeal of music or ideas of justice, into simple physical processes like the firing of neurons, or may offer sociobiological explanations which interpret the development of these human capacities in terms of evolutionary biology. Some biological explanations of landscape preference, such as the Savannah Theory advanced by Gordon Orions, do indeed belong in sociobiology. The theory is mentioned in E.O. Wilson’s The Diversity of Life (2001 [1992]). Wilson, now an emeritus professor of biology at Harvard, contributed to the development of the theory of island biogeography which was one of the launch points for landscape ecology. He has also been a prominent advocate for biodiversity and conservation. However, he is also regarded as the father of sociobiology on the strength of his controversial book Sociobiology. The New Synthesis (1975). Critics have accused sociobiology of being biologically deterministic, and an instance of the naturalistic fallacy whereby something is considered good or right because it is natural. Theories of this stamp continue to appear in the landscape literature. For example, Barrett et al. (2009) argue for the conceptualisation of landscape aesthetics as an ‘economy essential to survival’. Even today it remains controversial whether humans have any behavioural traits which are near universal biological adaptations. Opponents of sociobiology would argue for the significance of learning and culture.

THEORY IN SOCIAL SCIENCE Across the border in the social sciences, the word ‘theory’ has different connotations. Landscape architects probably do not regard themselves as social scientists, but town planners (or ‘urban planners’ or ‘spatial planners’ – or whatever they currently prefer to be called) customarily do. Gert

43 …

Ian H. Thompson

„

de Roo, an eminent Dutch planner, recently addressed a workshop of PhD students at Newcastle University and remarked ‘of course, you are social scientists, so you don’t prove anything’. His point was that social science offered frameworks of understanding and ways of interpreting evidence, but it cannot provide certainty. Indeed much of De Roo’s own work has been concerned with the ways planners can operate in conditions of complexity, fuzziness and unpredictability (e.g. de Roo and Silva 2010; de Roo 2011); these are just the same conditions faced by landscape architects. ‘Certainty’ and ‘truth’ have almost become dirty words in some parts of the academy, but nevertheless some social scientists still seek to emulate natural scientists, developing quantitative methods for use in disciplines such as sociology, political science, economics and psychology to produce knowledge that could be considered on a par with the findings of physics or chemistry. Even in postmodern times, some social scientists retain a positivist stance, investigating observable social phenomena and employing measurement, statistical techniques and mathematical modelling to test hypotheses. Social psychologists, for instance, run experiments and produce theories. Among the research that bears upon landscape architecture there are, for example, theories of personal space (or proxemics) and of territoriality (for example: Hall 1966; Sommer 1969; Becker and Mayo 1971). Other researchers work with large data sets, such as the information gathered for national census. It is fair to say that landscape is seldom foregrounded in such work, although the economic geographer Danny Dorling’s work on the UK housing market has profound implications for the future of the countryside (he dissents from the popular view that there is a shortage of houses in Britain, arguing that there is no need to rip up the green belt) (Dorling 2014). ‘Big data’, as it is often called, may turn out to be very important for landscape architecture, not just census data, but also the data produced by social media. This ties in with initiatives under the rubric of ‘smart cities’ to provide urban planners and managers real time information on a whole range of indicators which could include traffic and pollution levels through to data on park usage or refuse collection. Such work is still relatively new and its implications for landscape research have yet to be fully explored. An example is a Wageningen University MSc thesis on running routes in Amsterdam, for which the students used the data from two apps, representing in total some 11,000 runs. They could analyse this data to show how the city was used from a runner’s perspective, and then propose design interventions to improve running possibilities (Reiling and Dolders 2015; for other examples, see van Lammeren et al. – Chapter 9). Their analysis relied on numbers and, hence, is considered quantitative research. Alternatives to quantitative research are qualitative research and mixed methods research (see also Tobi and van den Brink – Chapter 2). In qualitative analysis the data are analysed by means of words without using numbers or statistics. Analogously to statistical analyses there is a range of qualitative data analyses, and their suitability depends on the research question and the paradigm of the researcher. Data gathered using such methods as unstructured or semi-structured interviews, focus groups and participant observation are often suited for qualitative analysis. The sample size is typically small, but researchers are often able to access deeper attitudes, opinions and motivations than would be possible using a purely quantitative approach. Qualitative and quantitative methods may be combined in mixed methods research, and it is not uncommon for exploratory qualitative work to prepare the way for quantitative work. Qualitative research generally involves interpretation and intuition and this places it at some distance from the methods of natural science and much closer to the sorts of activities which take place in the arts and humanities. Closely reading the transcript of a focus group is not so different from trying to elucidate the text of a novel or the script of a play. Indeed, we might be tempted to redraw our territorial map of the academic

44 …

The role of theory „

empires, following David Macey, the author of The Penguin Dictionary of Critical Theory, and talk instead of ‘the modern human sciences’, an area which covers ‘the domains of literature, philosophy, psychoanalysis, film and the visual arts, historiography and sexual politics’ (Macey 2000). It may be the case that there is more flexibility than there used to be in the choice of methods available, but researchers still need to be clear about the paradigm within which they are working. The term ‘critical theory’ means something very different from the sort of theory produced by natural science. It is a confusing term because it has two distinct origins, one in social science and one in the arts and humanities. Max Horkheimer, a founding member of the neo-Marxist Frankfurt School of social scientists in the 1930s, saw critical theory as the kind of social theory that was aimed at critiquing and changing society, rather than just understanding it. In the arts and humanities, as we will see later, the same term means something closer to ‘theory of criticism’ and encompasses a wide range of tools and perspectives for examining cultural products, such as films and plays (and indeed landscapes). Marxism provided the basis for the Frankfurt School and can thus be considered the original ‘critical theory’ and it illustrates the way in which theory can be used as the basis for social critique (Macey 2000, p.139). In its original form it was both an explanatory theory which offered an account of why certain conditions and injustices existed and an ethical theory which suggested what should be done about them. The theory predicted that certain contradictions within capitalism would bring the system down; the ethical injunction was to bring this about more swiftly through revolution. Many cultural geographers, including Cosgrove, Daniels and Mitchell, all mentioned earlier, have employed Marxian understandings of social and economic relations. In his seminal book The Postmodern Condition: A Report on Knowledge (1984) the French philosopher Jean-François Lyotard labelled Marxism a ‘grand narrative’, a universal and totalising explanation of all aspects of society, indeed all societies, throughout the span of history. Religious worldviews also purport to offer complete accounts, as do the narratives of Enlightenment reason, scientific progress and the spread of liberal democracy. In many instances, grand narratives serve to legitimate and reinforce existing social norms and power relations, but the Marxist story seemed to be different, in that it offered emancipation and issued a call to arms. Taking a Marxist position allows one to consider any activity or aspect of life and evaluate it in terms of the extent to which it helps or hinders the project of remaking society along more just lines. A complication is that most versions of the Marxist mission involve revolutionary upheaval, and anything which delays the revolution or makes it less likely to happen must be condemned. The nineteenth-century urban parks movement, which was in many ways the foundation upon which landscape architecture developed, could be a target for this sort of criticism, since parks were provided paternalistically to improve conditions for working people and to lessen tensions between different social classes. Intuitively these seem like good things to do, unless one takes the ideological position that they delayed the revolution and thus locked in systemic injustice. In its baldest form, the Marxist narrative can seem a crude one about the clashing interests of two blocks in society: workers and bourgeoisie. At the heart of the theory was the difference in the relationship of each group to the means of production, but it was an oversimplification or reduction, since it overlooked the myriad differences that characterise pluralistic society and the many and varied ways in which particular groups – women, ethnic groups, homosexuals, the disabled, the elderly and so on – come to be excluded, marginalised or disadvantaged. Lyotard suggested that the postmodern world had become incredulous about grand narratives, so he did not propose

45 …

Ian H. Thompson

„

one of his own, arguing instead for ‘little narratives’ (or ‘language games’, a term he borrowed from Wittgenstein). Typically postmodern theorists see ‘truth’ as something limited, situated and contingent. When words like ‘truth’ and ‘knowledge’ are used in postmodern discourse, they are often given quotation marks (‘scare quotes’) to indicate that no claim is being made to universality, or they are discussed in the plural – truths, knowledges – to show that the author knows that no group or individual has a monopoly on truth. Elizabeth Meyer, who describes herself as a feminist landscape architect, has stated that ‘theoretical work should be contingent, particular, and situated. Grounding in the immediate, the particular, and the circumstantial—the attributes of situational criticism—is an essential characteristic of landscape architectural design and theory. Landscape theory must rely on the specific, not the general’ (2002 [1997]). Meyer then says that design and theory must be based on observation and experience, the immediate and sensory (all of the senses, not just vision) and that landscape architectural theory is situational, ‘it is explicitly historical, contingent, pragmatic, and ad hoc’. Meyer clearly belongs to the camp which Blackburn calls the subjectivists. Balance is needed here. Postmodernism has been a fecund source of new perspectives and critical tools and it has provided a counter-weight to the cool, distanced, objectifying gaze of science. Recognising the myriad differences between human beings has been a much-needed corrective to the homogenising tendencies of modernism. However, when Meyer says that grounding in the immediate is an essential characteristic of landscape architectural theory, she stumbles upon a difficulty. How can anyone so committed to the contingent and situational say anything about an ‘essence’ (i.e. something immutable)? She wants to say that all knowledge is situated and relativistic, but wishes to exempt her own statements about the character of landscape architecture theory which she thinks are essentially (i.e. universally and fundamentally) true. The postmodern (or post-structural) turn has sensitised researchers and landscape practitioners to the differences between people and this has undoubtedly had important benefits such as the development of new techniques for community consultation and participatory design. If modernism led to the cult of the ‘expert’, who was generally a ‘landscape outsider’ in the sense suggested by the geographer Edward Relph (1976), postmodernism has favoured the local knowledge which can be provided by ‘landscape insiders’, that is people who live and work in a particular place. However, if all knowledge is local knowledge, as some would assert, how are we to identify mumbo-jumbo and irrational ideas? How are we to apply knowledge gained in one geographical or historical context in any other? Moreover, there will be instances when an outside expert, an ecologist, say, or a hydrologist, actually does know more about a particular topic than anyone who lives in the area. The imperative is to reconcile these viewpoints, but it can be a difficult work of mediation. The integration of local and scientific knowledge has become a research topic in its own right: see for example Failing et al. (2007); Raymond et al. (2010).

THEORY IN THE ARTS AND HUMANITIES The easiest way to distinguish between theory in the social sciences and theory in the arts and humanities is to consider their objects. If a theory offers an explanation or critique of the way society functions, then it belongs in social theory. If it offers an explanation or critique of cultural production or cultural products, then it belongs to cultural theory. That said, it is by no means a clear or simple distinction. How could it be? Consideration of cultural products, whether they are symphonies, detective novels, plays, video installations or gardens and designed landscapes, involves

46 …

The role of theory „

consideration of the societies in which they were produced. Indeed, criticism of such artefacts or cultural practices can provide a way into critique of the societies which produced them. The Marxist critic, Raymond Williams (1921–1988), understood the relationship between the economic organisation of a society and its cultural production as the relationship between a determining substructure and a determined superstructure. Arts of production – and we can include the production of designed landscapes – are generally an articulation of the dominant culture, and embody meanings and values (Williams 1959). But, as the Italian Marxist Antonio Gramsci asserted, this ‘determination’ cannot operate in a straightforward way. Gramsci disputed Marxism’s claim to be ‘scientific’ and objectively, ahistorically true, arguing instead that the dominant class in any society rules not just through physical coercion, but by persuading everyone in that society (or at least the vast majority) that the prevailing arrangements are natural, normal and valid. The word Gramsci used to refer to this form of control was ‘hegemony’ (Gramsci 2011). For the Marxist cultural geographers, Cosgrove and Daniels, ‘a landscape park is more palpable but no more real, no less imaginary, than a landscape painting or a poem…’ (1988, p.1). In other words, a landscape, whatever else it may be, is always a symbolic representation, a cultural product. Their argument depends upon a conception of landscape as visual, a view over land, rather than as a tract of land. Raymond Williams, they observe, suggested that ‘a working landscape is hardly ever a landscape’, because landscape is a way of seeing (ibid.). The ploughmen and harvesters are too busy with their labours to look at the land in an objective and aestheticising way. Picturesque aesthetics have had a particularly hard time at the hands of cultural critics. They have been accused of aestheticising and therefore legitimising poverty (and it is true that picturesque paintings sometimes feature hovels, beggars and poor barefoot children). Picturesque aesthetics are said to be distancing: they set up a picture plane between the landscape and the observer: they are thus the antithesis of an immersed or engaged aesthetic. They represent the commodification of land and the unjust distribution of wealth; Mr and Mrs Andrews, in Gainsborough’s portrait of 1750, look out smugly over their well-tended fields (though there is not a labourer in sight). The critic John Berger thought that ‘their proprietary attitude towards what surrounds them is visible in their stance and expressions’ (Berger 1972, pp.106–107). I have dwelt upon theory derived from Marx at length because of its prominence in critiques of landscape produced by cultural geographers, but the realm of critical theory has expanded vastly beyond its origins to include a plethora of -isms, many of which also have a bearing upon landscape architecture. Freudian psychoanalytic theory and its derivatives, such as Kleinian and Lacanian psychoanalysis, are often said to be the other major source of critical theory, but they have not featured prominently in landscape theory, though the British landscape architect, Sir Geoffrey Jellicoe (1900–1996) elaborated a theory which attempted to explain the design process and the apparent power of some designed landscapes by an appeal to the Jungian notion of the collective unconscious. Feminism, however, has employed psychoanalytic concepts to critique the masculine gaze upon landscape. The geographer Gillian Rose, for example, has considered the way that particular sets of power relations, in particular those between men and women, have structured the meanings of images of landscape (Rose 1993, 1996). Rose identifies a dualism between masculine forms of geographical knowledge, which she identifies with rational approaches in the social sciences, and the aesthetic approach to landscape associated with the arts, which feminises places. Within this dualism it has traditionally been the masculine form of knowledge which has occupied the dominant position. In a similar vein the philosopher Caroline Merchant has combined

47 …

Ian H. Thompson

„

feminist and ecological critique to examine the culture–nature binary. The earth, according to Merchant, was seen as a wild but beneficent mother until the emergence of modern science. Merchant expressly links the exploitation of nature to the oppression of women (Merchant 1980, 2005). A feminist ethics of care has emerged from environmental ethics which has a direct bearing upon notions of stewardship in landscape architecture (see also Plumwood 1993). Literary studies was the seedbed for postcolonial theory, another productive strand within critical theory; it seeks to understand the impacts of European colonialism upon the rest of the world (NB it is a slightly misleading title, because it studies the colonial period, not just what happened after countries acquired their independence). In terms of landscape theory, research has often been focussed upon the imposition of European ways of seeing and controlling landscape at the expense of indigenous understandings. For example, Paul Carter in The Road to Botany Bay (1987), examines the way in which picturesque tropes were employed to entice settlers into making the land their own, while Allaine Cerwonka has studied the impacts of importing European garden plants into the Australian landscape (Cerwonka 2004). Jacky Bowring (1997) has shown how the language of the picturesque permeated the development of the landscape architecture profession in New Zealand, through a close reading of the published discourse of the New Zealand Institute of Landscape Architects. The smorgasbord of cultural theory is laden with -ologies and -isms, so many in fact that a complete review is not possible within the confines of this chapter; I will have to leave it to readers to decide whether queer theory, semiology, biopolitics, communicative action, deconstruction, hermeneutics, existentialism, structuralism, post-structuralism, speciesism, surrealism and the rest have a bearing upon landscape architecture. However, there is one -ism which stands out as being particularly pertinent to research in landscape research. It approaches investigation from an angle which is so radically different from positivist science that it can seem antithetical to it. This, of course, is phenomenology and the emphasis it places upon subjective knowledge and the knowing subject. When writing conventional scientific papers it is usual for authors to efface themselves in order to achieve a properly objective tone. If they have to mention that the research was carried out by living human beings, they adopt some distancing epithet like ‘the researchers’ or ‘the author’. In phenomenological writing, whether in the humanities or the social sciences, there is no such imperative and first-person accounts are quite normal. On the battlefield of Blackburn’s truth wars, phenomenologists would be found in the subjectivist camp. Indeed, the adjective phenomenological is sometimes used rather lazily as a synonym for subjective. However, if we consider the origins of phenomenology in the writings of the philosopher Edmund Husserl (1859–1938), we find that his approach was rooted in the rational scepticism of Rene Descartes. Like Descartes, Husserl wanted to build our knowledge on rock-solid foundations. Sceptics, such as Descartes, have wondered whether there is an external world and whether it matches up with our experiences of it. Husserl deliberately side-stepped this problem. The things we can be absolutely certain about, he argued, are the contents of our conscious awareness, which include perceptions, thoughts, bodily awareness, memories, emotions and volitions. We have direct access to these and can study them. In this respect the phenomenological enterprise is empirical, something it shares with science. However, Husserl’s version of phenomenology features less in landscape theory than that of his successor, Maurice Merleau-Ponty (1908–1961). Whereas Husserl, true to his Cartesian inclinations, regarded the conscious subject as transcendental and dis-embodied, Merleau-Ponty took an altogether different line, regarding the body as the seat of perception. For Merleau-Ponty our knowledge comes through our bodies (Merleau-Ponty 2012 [1945]). We could

48 …

The role of theory „

not have a world, if we did not have a body. This is much closer to common-sense ideas of our relationship to our environment, and for those, like landscape architects, who are interested in human–environment interactions, this notion of embodied consciousness has much to recommend it. The contrast between positivist and phenomenological methods of investigation can be illustrated by the work of one of my PhD students who has been investigating the microclimate of public parks in Cairo, Egypt, her home city. Her initial inclination was to use the measuring instruments of the climate scientist, but she ultimately rejected this in favour of a phenomenological approach which focussed upon subjective accounts of climate as experienced by a number of subjects in the park (including herself). Her respondents provided rich data about how they experienced sun and shade within the park and how the microclimate influenced their behaviour there. Arguably this sort of data is more useful to would-be park designers than the objective charts and tables of scientific investigation, or if not more useful then at least as useful, but in a different way (see also Schultz and van Etteger – Chapter 11). The shift from Husserl to Merleau-Ponty has focussed attention upon the materiality of landscape. The anthropologist Tim Ingold published an essay entitled ‘The temporality of landscape’ (Ingold 1993) which stressed that a landscape was a lived-in world, rather than just a way of viewing land. There are strong resonances here with the work of another philosopher, Martin Heidegger (1889–1976), who asked what it meant to ‘dwell’ upon the earth (as opposed to merely existing or surviving) (Heidegger 1971). Many writers, including the American geographer J.B. Jackson, have laid emphasis upon the often repetitive landscape practices (ploughing a field, trimming a hedgerow, tending an allotment, walking a footpath etc.) which create everyday landscapes, and interest in these practices now extends to artists and specialists in performance studies.

THE RIGHT SORT OF THEORY FOR THE JOB One of the most interesting things about Wittgenstein’s later philosophy was that it was supposed to be therapeutic. By showing philosophers how their misuse of everyday language led them into perplexity, Wittgenstein hoped that he could dissolve, rather than solve, the problems that had perplexed them for centuries. This was hugely radical, in its way, but at the same time it was conservative. It left everything alone and didn’t call for any drastic revision of everyday language. If we didn’t push words into uses for which they were not suited, everything was fine – we all knew our way about. I feel the need to say something similar here. The word ‘theory’, as we have seen, is used in a great variety of contexts. We sometimes muddle ourselves when we think we are dealing with one sense of ‘theory’ and it turns out that our interlocutors are using another. We have seen that attempts to police the use of ‘theory’ within the discipline of landscape architecture (by Riley and others) have failed. I suggest that we do no tinkering at all and that we leave things pretty much as they are. Landscape architecture is never going to move wholesale into the natural sciences, nor will it be absorbed by the social sciences or swallowed by the arts. Its position, uncomfortably perhaps, will always be at the borders, and as long as that remains so, competing notions of theory are likely to persist. We might as well get used to this and cultivate our awareness of the ways words are used within our neighbouring empires. However, knowing our way about becomes even more important. One thing I have not tried to do in this chapter is to tabulate the various types of theory encountered along the way. In part this is because it would replicate efforts already undertaken by Elen Deming and Simon Swaffield in their

49 …

Ian H. Thompson

„

book Landscape Architecture Research: Inquiry, Strategy and Design (2011), where they categorise research strategies against two axes, one labelled Inductive-Deductive (with ‘Abduction’ forming a third possibility between) and the other labelled ‘Objectivist-Subjectivist’ and then come up with nine ‘strategies of inquiry’. For the most part, their categorisation is useful, though phenomenology is oddly absent. Rather than spending this entire chapter critiquing or elaborating their framework, I wanted to show, in a more discursive way, how the puzzlement about the role of theory in landscape architecture might have come about. James Corner has suggested that a theory can be both a ‘stabilizer’ and a ‘disruptive mechanism’ (1990). In the sciences, a good theory can provide an explanation for a diverse range of data or phenomena, gathering them together, so to speak, and making them intelligible. However, even a well-established theory can be overturned, if it is found wanting in explanatory power. In what Thomas Kuhn has called a ‘paradigm shift’, the old explanatory model can be rapidly superseded by a new one. The example always mentioned here is the way that the Theory of Relativity displaced Newtonian mechanics. So, even within objective science, a new theory can be radically disruptive. However, there are species of critical theory, both social and cultural, which find their raison d’ être in disrupting the status quo. A fashionable though ugly word for this is ‘problematising’, which we seem to owe to Michel Foucault. Problematising occurs when a critical thinker calls into question a piece of common knowledge or a commonplace practice. This problematisation generally involves the revelation of some hidden operation of power in the everyday routines. Thinkers who would regard themselves as progressive are often enthusiastic about the disruption of settled forms of knowledge because it seems to offer opportunities for new consciousness, hope or action to emerge. More pragmatically the budding researcher needs to understand how theory (and what sort of theory) is relevant to her research design. It has a direct and immediate bearing upon the framing of research questions and the selection of methods. To return to the example of my Egyptian research student, her reading of phenomenological theory, particularly Gernot Böhme’s work on atmospheres and aesthetics (Böhme 1993, 2005), led her to reject her original ideas about gathering climatic data with scientific instruments, in favour of ‘walk-along’ interviews with subjects in a Cairo park, recording and later analysing their perceptions of microclimate and place. Böhme is a philosopher with interests in anthropology, another borderland discipline sometimes classified among the social sciences, but sometimes considered a humanity. The methods this student selected came from social science, but they were qualitative rather than quantitative. Another student (from Thailand) wished to undertake largely historical research into open space in Bangkok. From the outset, this suggested archival research, but the investigation was shaped by an understanding of both Henri Lefebvre’s theories concerning the production of social space and also postcolonial theory. Thailand was never colonised, so the interesting question was ‘why not?’; the answer, as this researcher was able to demonstrate, was that the Thai monarchy was able to present the country as a modern European-style state. The design and use of public spaces contributed to the projection of this image. These two, very different, examples show that the researcher needs to engage with existing theory from the outset and to be prepared to find it anywhere. As it happens, neither of these theses drew upon theory produced from within the discipline of landscape architecture, though both students were landscape architecture graduates. If theory is important from the outset, as a kind of framing input into the research process, does research also produce theory? The answer must

50 …

The role of theory „

be sometimes, but not necessarily. All PhD level work ought to make a contribution to knowledge, although it does not need to be an earth-shattering one to earn the degree. The research might test a theory in new circumstances. For example, another of my students took well-tried European methods of collaborative design and sought to discover whether they could be applied in Malaysia. His thesis added to the evidence which showed that these methods worked, while identifying a number of particular difficulties which needed to be addressed when they were used in Malaysian society. Sometimes PhDs do add to theory (as opposed to knowledge) because they offer a new theoretical framework for looking at a particular topic or issue. My own thesis was of this kind. I was interested in the values and motivations of landscape architects, so I went and interviewed a number of them in-depth. On the basis of these interviews and a wide-ranging literature review, I suggested that their values fell into three areas: ecology and environment, social and political, and creative and aesthetic. I also suggested that there were inevitable clashes between these values, which each practitioner had to resolve in their own way. I was able to develop this framework further in the book Ecology, Community and Delight (Thompson 2000). One might say, I suppose, that the empirical research (i.e. the interviews) provided a snapshot of the way British landscape architects considered their vocation at the end of the twentieth century, but I think that it was the framework, the theoretical innovation, that readers found more interesting. Theory does, of course, move on, and this is true whichever of the three cultures it belongs to. We have noted how paradigms in natural science can be overthrown by new evidence. The social sciences and the humanities are more prone to enthusiasms, as particular critical thinkers become fashionable, then fall from favour. We saw this with Derrida and Deconstruction, for example, with the Parc de la Villette as an unusually concrete monument to an academic craze. Then the mantle passed to Deleuze and Guittari, who provided the intellectual underpinning (or maybe just the gloss) for Landscape Urbanism. This is just the academic weather, but scholars and researchers to be able to read it. The newest idea is not necessarily the best and seemingly dead theory can sometimes be reanimated, albeit in a slightly new guise. I was originally approached to write a chapter with the title ‘Theory: Chicken or Egg?’ I think the thought was: ‘what comes first, the theory or the method’? For the most part, I believe it is the theory, though theory may be stretched or otherwise modified by the outcome of the research. In the humanities and the social sciences, theory provides frameworks or lenses through which to consider phenomena, and sometimes to define and evaluate them. It is different when we are thinking about procedural research. As we saw, when thinking about Steinitz and McHarg, in these instances the research starts from consideration of existing practices and results in some recommended codification of these procedures. Perhaps the question was ‘Is theory an input or an output?’ I hope that I have shown that it can be either (or both).

NOTE 1 I note that elsewhere in this volume my colleague Maggie Roe argues that landscape architecture is not a ‘discipline’ but a ‘disciplinary field’. I think she does this in order to acknowledge the breadth of research which informs landscape architectural practice and the degree to which that knowledge is shared with others. The Oxford English Dictionary gives as one of its definitions ‘discipline: a branch of learning or scholarly instruction’. I don’t have much difficulty with this. There are things which need to be passed on to future generations of landscape architects, but I agree with Roe that there need not be anything exclusive about this.

51 …

Ian H. Thompson

„

REFERENCES Barrett, T.L., Farina, A. and Barrett, G.W. (2009) ‘Aesthetic landscapes: An emergent component in sustaining societies’, Landscape Ecology, 24(8), 1029–1035. Becker, F.D. and Mayo, C. (1971) ‘Delineating personal distance and territoriality’, Environment and Behavior, 3, 375–381. Berger, J. (1972) Ways of Seeing, London: British Broadcasting Association and Penguin. Blackburn, S. (2005) Truth: A Guide for the Perplexed, London: Allen Lane. Böhme, G. (1993) ‘Atmosphere as the fundamental concept of a new aesthetics’, Thesis Eleven, 36(1), 113–126. Böhme, G. (2005) ‘Atmosphere as the subject matter of architecture’, in Ursprung, P., ed. Herzog & DeMeuron: Natural History, Montreal: Lars Müller and Canadian Centre for Architecture, 398–406. Bowring, J. (1997) ‘Institutionalising the picturesque: The discourse of the New Zealand Institute of Landscape Architects’, PhD thesis, Lincoln University, New Zealand. Carter, P. (1987) The Road to Botany Bay, London: Faber & Faber. Cerwonka, A. (2004) Native to the Nation: Disciplining Landscapes and Bodies in Australia, Minneapolis, MN: University of Minnesota Press. Corner, J. (1990) ‘Origins of theory’, in Swaffield, S., ed. (2002) Theory in Landscape Architecture: A Reader, Philadelphia, PA: University of Pennsylvania Press. Cosgrove, D. (1984) Symbolic Formation and Symbolic Landscape, London: Croom Helm. Cosgrove, D. and Daniels, S. (1988) The Iconography of Landscape: Essays on the Symbolic Representation, Design and Use of Past Environments, Cambridge: Cambridge University Press. Crotty, M. (1998) The Foundations of Social Research: Meaning and Perspective in the Research Process, St. Leonards, Australia: Allen & Unwin. de Roo, G. (2011) ‘The ecosystem approach: Complexity, uncertainty, and managing for sustainability’, Planning Theory, 10(1), 92–95. de Roo, G. and Silva, E.A. (2010) A Planners’ Encounter with Complexity, Farnham: Ashgate Publishing. Deming, M.E. and Swaffield, S. (2011) Landscape Architecture Research: Inquiry, Strategy and Design, Hoboken, NJ: John Wiley & Sons. Dorling, D. (2014) All That is Solid: How the Great Housing Disaster Defines Our Times, and What We Can Do About It, London: Penguin. Failing, L., Gregory, R. and Harstone, M. (2007) ‘Integrating science and local knowledge in environmental risk management: A decision-focused approach’, Ecological Economics, 64(1), 47–60. Forman, R.T.T. (1995) Land Mosaics: The Ecology of Landscapes and Regions, Cambridge: Cambridge University Press. Forman, R.T.T. and Godron, M. (1986) Landscape Ecology, New York: John Wiley & Sons. Gramsci, A. (2011) Prison Notebooks, Volumes 1–3, Columbia University Press. Hall, H. (1966) The Hidden Dimension, New York: Doubleday. Hartig, T., Mang, M. and Evans, G.W. (1991) ‘Restorative effects of natural environment experiences’, Environment and Behavior, 23(1), 3–26. Hartig, T., Evans, G.W., Jamner, L.D., Davis, D.S. and Gärling, T. (2003) ‘Tracking restoration in natural and urban field settings’, Journal of Environmental Psychology, 23(2), 109–123. Heidegger, M. (1971) ‘Building dwelling thinking’, in Poetry, Language, Thought, New York: Harper and Row, 143–159. Ingold, T. (1993) ‘The temporality of the landscape’, World Archaeology, 25(2), 152–174. Jackson, J.B. (1984) Discovering the Vernacular Landscape, New Haven, CT: Yale University Press. Kagan, J. (2009) The Three Cultures: Natural Sciences, Social Sciences, and the Humanities in the 21st Century, Cambridge: Cambridge University Press. Kaplan, R. and Kaplan, S. (1989) The Experience of Nature: A Psychological Perspective, Cambridge University Press Archive. Kaplan, S. (1995a) ‘The restorative benefits of nature: Toward an integrative framework’, Journal of Environmental Psychology, 15, 169–182.

52 …

The role of theory „

Kaplan, S. (1995b) ‘The urban forest as a source of psychological well-being’, in Bradley G.A., ed. Urban Forest Landscapes: Integrating Multidisciplinary Perspectives, Seattle, WA: University of Washington Press, 101–108. Lyotard, J.-F. (1984) The Postmodern Condition: A Report on Knowledge, Manchester: Manchester University Press. Macey, D. (2000) The Penguin Dictionary of Critical Theory, London: Penguin. McHarg, I.L. (1969) Design with Nature, Garden City, NY: Natural History Press. Merchant, C. (1980) The Death of Nature: Women, Ecology, and the Scientific Revolution, New York: Harper. Merchant, C. (2005) Radical Ecology: The Search for a Liveable World, London: Routledge. Merleau-Ponty, M. (2012 [1945]) Phenomenology of Perception, London: Routledge. Meyer, E. (2002) ‘The expanded field of landscape architecture’, in Swaffield, S., ed. Theory in Landscape Architecture: A Reader, Philadelphia, PA: University of Pennsylvania Press, 167–172. Mitchell, D. (1996) The Lie of the Land: Migrant Workers and the California Landscape, Minneapolis, MN: University of Minnesota Press. Mitchell, D. (2003) ‘Cultural landscapes: Just landscapes or landscapes of justice?’, Progress in Human Geography, 27(6), 787–796. Mitchell, D. (2007) ‘Work, struggle, death, and geographies of justice: The transformation of landscape in and beyond California’s Imperial Valley’, Landscape Research, 32(5), 559–577. Murphy, M.D. (2005) Landscape Architecture Theory: An Evolving Body of Thought, Long Grove, IL: Waveland Press. Naveh, Z. and Lieberman, Y.A. (1984) Landscape Ecology: Theory and Applications, New York: Springer Verlag. Ndubisi, F. (1997) ‘Landscape ecological planning’, in Thompson, G.F. and Steiner, F.R., eds. Ecological Design and Planning, New York: John Wiley & Sons, 9–45. Olwig,K. (2002) Landscape, Nature and the Body Politic, Madison, WI: University of Wisconsin Press. Plumwood, V. (1993) Feminism and the Mastery of Nature, London: Routledge. Popper, K. (2002 [1934]) The Logic of Scientific Discovery, 2nd ed., London: Routledge. Raymond, C.M., Fazey, I., Reed, M.S., Stringer, L.C., Robinson, G.M. and Evely, A.C. (2010) ‘Integrating local and scientific knowledge for environmental management’, Journal of Environmental Management, 91(8), 1766–1777. Reiling, M. and Dolders, T. (2015) ‘Running Amsterdam: Designing a running friendly city’, MSc thesis, Wageningen University, the Netherlands. Relph, E. (1976) Place and Placelessness, London: Pion. Riley, R. (1990) ‘Editorial commentary: Some thoughts on scholarship and publication’, Landscape Journal, 9(1), 47–50. Rose, G. (1993) Feminism and Geography. The Limits of Geographical Knowledge, London: Polity Press. Rose, G. (1996) ‘Teaching visualised geographies: Towards a methodology for the interpretation of visual materials’, Journal of Geography in Higher Education, 20(3), 281–294. Sommer, R. (1969) Personal Space, Englewood Cliffs, NJ: Prentice-Hall. Snow, C.P. (2012 [1959]) The Two Cultures and the Scientific Revolution, Cambridge: Cambridge University Press. Steinitz, C. (1994) ‘A framework for theory and practice in landscape planning’, Ekistics, 61(364–365), 4–9. Steinitz, C. (2013) A Framework for Geodesign: Changing Geography by Design, Redlands, CA: Environmental Systems Research Institute. Swaffield, S., ed. (2002) Theory in Landscape Architecture: A Reader, Philadelphia, PA: University of Pennsylvania Press. Thompson, I.H. (2000) Ecology, Community and Delight: Sources of Values in Landscape Architecture, London: Spon. Ward Thompson, C. and Aspinall, P.A. (2011) ‘Natural environments and their impact on activity, health, and quality of life’, Applied Psychology: Health and Well Being, 3(3), 230–260. Wilson, E.O. (1975) Sociobiology: The New Synthesis, Cambridge, MA: Harvard University Press. Wilson, E.O. (2001 [1992]) The Diversity of Life, 2nd ed., London: Penguin. Williams, R. (1959) Culture and Society, London: Chatto & Windus. Wittgenstein, L. (1922) Tractatus Logico-Philosophicus, London: Kegan-Paul. Wylie, J. (2007) Landscape, London: Routledge.

53 …

Chapter 4: The relationship between research and design Sanda Lenzholzer, Ingrid Duchhart and Adri van den Brink

CONTEXT In a societal context of ‘scientification’ (Bayazit 2004; Cross 2006), landscape architecture research faces new perspectives: scope and scale of research studies is growing (Cushing and Renata 2015; Hewitt 2014; Weber 2015) and so is the involvement of other disciplines and citizens (Findeli 2001; Jonas 2007; Nowotny et al. 2001). The complexities arising from these developments, coupled with an increasing necessity to deal with uncertainty (Prominski 2005), call for a sound relationship of landscape architectural design and academic research. In this chapter we discuss the nature of this ‘sound relationship’. We focus on three relationships between research and design – that is research on design, research for design and research through design/ing – within landscape architecture. It is from these perspectives that landscape architecture can make significant contributions to scholarly design research in general and more specifically to the 21st century’s intra-scalar landscape transitions, for example from the design of outdoor objects to designing within regional development, as addressed by, among others Kempenaar et al. (2016). Our aim is to reflect on these different relationships, while referring to literature from both landscape architecture and neighbouring design disciplines. We pay particular attention to ‘research through designing’, its evolution to date, its current state and future potential. The subsequent methods may perhaps represent the most intimate relationship between research and design – the act of designing as an intrinsic part of research.

SOME DEFINITIONS AND BASIC CONCEPTS Although there are many different definitions of both research and design, we will use the terms as described within the general design disciplines’ discourse. Glanville (2015) concluded from this discourse that research in the academic sense means a rigorous and in-depth search for answers to research questions and to find new insights. He also makes a clear distinction between ‘design’ as a noun and as a verb (see also Steinitz 1995). ‘Design’ as a noun is described as the outcome of the design process in which a product, that is the design, has been drawn and given shape, and, in the case of a positive outcome of decision-making, may be implemented. In landscape architecture we talk about ‘design’ in the sense of giving three-dimensional form and function to, for example, the direct external living environment, be it urban, peri-urban or rural. Limiting the definition of ‘design’ to spatial dimensions is important in this context because many other definitions of ‘design’ exist

54 …

The relationship between research and design „

that have no relation to shaping space in any sense (e.g. IT, policy sciences, mechanical engineering, bioprocess technology and many other disciplines). Drawing, mapping, visualising, representing, and giving shape are the unique activities that constitute the act of designing. Lenzholzer et al. (2013) proposed to use the gerund form – designing – to more clearly distinguish the verb from the noun. Both – verb and noun – can be the subject of research. Designs (as a noun) can be studied after their implementation (post hoc), whereas design/ing (as a verb) can be studied during the design process. Important for this chapter is the role of designing as a part of a research process. Basically, all design disciplines connect design and research and there has been a growing body of literature since major figures such as Herbert Simon sparked the discussion in the 1970s (Simon 1975). Various architecture research theorists who built upon his insights had a strong focus on the engineering disciplines (Eder 1995; Hubka and Eder 1987; Simon 1996), but were also concerned with the artistic aspect of design. Subsequently, other design disciplines that represent artistic approaches also entered the discussion. In 1993 the art and design theorist Christopher Frayling published a seminal paper suggesting that there are three models of design research: research into design, research for design and research by design (Frayling 1993). This categorisation presented a widely used framework to describe the relationship between research and design. Although later on extended with other pronouns (see Archer 1995; de Jong and van der Voordt 2002; Jonas 2014) that denominate similar relations between research and design, the categorisation essentially remained the same and was also recently used in landscape architecture publications on the topic (Deming and Swaffield 2011, p.37; Lenzholzer 2010, p.19). In the following section we give hints on the meaning of these categories within landscape architecture.

RESEARCH INTO, FOR AND THROUGH DESIGN(ING) The first category described by Frayling is research into design, sometimes also referred to as research on or about design. Here, design is used as a noun, so studies in this field concern the design product (post hoc). Very typical for such research is a reflection on design products. In landscape architecture this can concern studies on garden, urban and landscape designs that can be interpreted from different points of view, that is historical (Nijhuis 2014; Treib 1993) or technological (Salingaros 2005), but also philosophical or aesthetical (van Etteger et al. 2016). Other typical research of this category are comparative case studies, plan analyses or design criticism (e.g. Brinkhuijsen 2008; Francis 2001). Additionally, empirical methods from the natural and social sciences can be used to assess designs after realisation, such as Post Occupancy Evaluation of buildings or public spaces (Meir et al. 2009; Sherman et al. 2005). This kind of research has already been conducted for a long time – also for the design products of landscape architecture. A wide range of examples can be found in Deming and Swaffield (2011). Remarkably, however, much of the research presented there was not conducted by landscape architects but by historians, geographers, environmental psychologists and social scientists. The second category pertains to research for design. This category covers all types of research that support the design product or design process. Here, both product and process benefit from research activities in the sense that the research outcomes inform the design process. Since design in landscape architecture is a highly complex undertaking, this research for design consequently requires a very wide knowledge base, ranging from the natural sciences and social sciences to the arts and humanities (see Thompson – Chapter 3). Research for design can be conceived of as

55 …

Sanda Lenzholzer, Ingrid Duchhart and Adri van den Brink

„

the creation of substantive knowledge through the generation of scientific data for application in ‘evidence based’ design (Brown and Corry 2011) (examples in Deming and Swaffield 2011; Groat and Wang 2002). In this context we can think of examples such as material studies, for example for green roofs (Blanusa et al. 2013; Susca et al. 2011) or for soil bioengineering in nature development projects (Simon and Steinemann 2000). It can also concern the generation of design guidelines for a user-friendly environment by translating knowledge from social sciences or environmental psychology (e.g. Groenewegen et al. 2006; Herzog 1992; Teel et al. 2010). When ‘design’ is addressed as a verb, the creation of knowledge for design can also be procedural: knowledge that aids the structuring of design processes (e.g. Dorst 1997; Schön 1987). In this category, too, the actual research part is not always conducted by landscape architects. However, the translation of the data into meaningful design guidelines is usually done by designers. The third category is research through design(ing), covering all the research processes that actively employ designing. Research through designing (RTD in the remainder of this chapter) is at the heart of all design disciplines. Its methods have been widely discussed and amongst design theorists there is no doubt that designing can be a valid research method (e.g. Rodgers and Yee 2015). Within landscape architecture, however, such ideas were regarded with reservation (see e.g. Deming and Swaffield 2011; Milburn and Brown 2003). It was correctly stated that for design to qualify as research it would need to meet certain methodological criteria, such as a clear research question, a theoretical framework and appropriate methods. Nonetheless, how to apply such criteria was not self-evident. This might be a result of the novelty of the discourse on research methods within landscape architecture (one of the major reasons for creating this book). The lack of methodological criteria might also result from historical applications of design/ing as a so-called research method within landscape architecture and related fields that would not always qualify as an established academic research method. In the next section we give a brief overview and discussion of these historical applications.

THE EVOLUTION OF RESEARCH THROUGH DESIGNING There are various developments within landscape architecture and neighbouring disciplines that led to the evolution of designing as a research method. We can observe three major strands occurring in this evolution. The first strand consisted of the developments in architecture, urbanism and landscape architecture on ‘rationalising’ design and planning processes. The second strand was the evolution of ‘research by design’ that occurred in the Netherlands. The third strand highlights the renewed international attention to research in landscape architecture design processes. In this section we discuss these strands in more depth and how a slow development took place from practice-oriented design towards RTD as a sound research method that generates new design-relevant knowledge. The first strand in the evolution of design as research dates back to the 1930s. There had been a tradition of a rational, science-oriented approach in architecture since the advent of modernism. Many architects, amongst them prominent figures like Le Corbusier, proposed architecture as a science (Le Corbusier, in Cross et al. 1981) and, for instance, in his version of ‘The Athens Charter of the International Congress of Modern Architecture’, Le Corbusier described urbanism as a ‘threedimensional … science’ (CIAM 1943, section 82). Alexander, in the earlier years of his career, also pursued a strictly rational, mathematically inspired approach to design in his Notes on the Synthesis of Form (1964), whereas in the 1970s Hillier et al. (1972) suggested bringing a scientific empirical

56 …

The relationship between research and design „

method into design. However, many architects also denied the possibility that design can be scientific or ‘research’. Jones (1970, p.10) posited that ‘designing should not be confused with … science, or with mathematics’. He emphasised that ‘when they [designers] deal with the future itself, as opposed to the present, scientific doubt is of no use’ (p.11). Lang (1987, p.19) asserted that ‘by definition, design cannot be scientific’, even though design should be based on scientific knowledge. In 1984 John Zeisel took the debate further and suggested that an architectural design can be a hypothesis or conjecture and that these hypotheses or conjectures can be tested (Zeisel 2006, pp.19– 25). His ideas inspired many colleagues. For instance Ungers (1997, in Steenbergen et al. 2002, p.24) suggested ‘… first there is the hypothesis, the idea, the image. Then there will be strategies of refutation – in fact. The real job.’ Especially at Delft Technical University in the Netherlands these methods were taken further during a major conference in 2001. Research by design was understood ‘as the development of knowledge by designing, studying the effects of this design, changing the design itself or its context, and studying the effects of the transformations’ (de Jong and van der Voordt 2002, p.455). Other scholars in Delft came up with similar models and claimed that research by design is literally the same procedure as used in the ‘classical’ empirical sciences: From a general scientific point of view, there appears to be, on closer examination, no essential difference between the steps in the empirical research cycle (statement of problem – analysis – generation of possible testable answers – formulation of hypotheses) and those of design research (task – analysis – generation of schematic … models – design). (Steenbergen et al. 2002, p.25)

Since then, the debate in architecture has grown towards full acceptance of design(ing) as a sound scientific research method. In landscape architecture the debate about the relationship between design and research was sparked off during the 1960s by McHarg (1968) and Steinitz and Rogers (1970). They argued that landscape architecture had to respond to the increasing complexity and the scale of issues of unprecedented growth through the adoption of ecological and rational planning and design methodologies. In 1969, Ian McHarg published his seminal book Design with Nature in which he presented new analysis tools by depicting various abiotic, biotic and human characteristics of the landscape as separate layers. This approach found many followers across the globe, but also received hefty critique for (apparently) denying the artistic aspect of designing, its weak academic qualities and lack of inclusiveness of the design process (e.g. Krieger and Litton 1971). Steinitz and Rogers (1970) responded by proposing a design methodology that included an iterative process of designing spatial configurations and testing them for their impact on, for example, hydrology, biodiversity and land use. Although the iterative nature and the testing process pointed towards basic principles of RTD, it was not coined as such; rather it reinforced a perceived antagonism between research and design (Milburn et al. 2003; Steiner 2006). In spite of the many experiments that were conducted at universities around the world, most of these experiments were actually typical research for design approaches that ran aground not only because of methodological complexities but also because landscape analyses dominating the design process left little room for creative and normative prospective designing. Moreover, in those early days digital data were scarce and geographic information system (GIS) technology still had little capacity – the powerful analytical tools we now take for granted were still primitive.

57 …

Sanda Lenzholzer, Ingrid Duchhart and Adri van den Brink

„

In an attempt to solve the complexities outlined above a new design process evolved, one that brought designing to the forefront of the designing process, at least in the Netherlands. In this second strand of thinking about the relationship between design and research Meto Vroom, from 1966 to 1994 professor of landscape architecture at Wageningen University, the Netherlands, played a prominent role. As a former student of McHarg, he and his students experimented with McHargian design procedures, but soon they realised that for innovative solutions, designing had to be brought to play a central role in the research process. In the mid-1970s, this resulted in the tandem ‘researchbased-design / design-based-research’ (ontwerpend onderzoeken / onderzoekend ontwerpen), a method explicitly aiming at connecting scientific research and designing without prioritising the one over the other. Te Boekhorst (2006) narrates how foresters, ecologists, agricultural scientists and landscape architects came together in a design charrette and co-designed a first conceptual design for a pilot case study area. This conceptual design generated various questions that were rephrased as research questions for each discipline involved. The outcomes of the subsequent (disciplinary) research were brought into the design process in a second design charrette. Several cycles of co-design activities and disciplinary research took place, slowly generating new knowledge and applying it in designs. This approach, in which research and design are linked by co-design sessions, can be best described as taking intermediate steps to explore and test evolving design alternatives. Unfortunately this ‘research-based-design / design-based-research’ discourse was not taken further in the years to follow (te Boekhorst 2006). However, the idea of the research-by-design charrette is still widely practiced as a means of facilitating community participation (e.g. Bouwmeester et al. 2009) or in solving multi-disciplinary problems through design (e.g. in forests, see Bell and Apostol 2008). It should be noted, however, that the actual research component of these charrettes neither meets the criteria of proper academic research, nor is designing appropriately elucidated as (part of) sound academic landscape architecture research. Next to the developments in landscape architecture practice in the Netherlands, a renewed interest – the third strand of thinking – for research in landscape architecture design processes occurred internationally (e.g. Kapper and Chenoweth 2000; Milburn and Brown 2003; Milburn et al. 2003; Milburn et al. 2001). Milburn and Brown, for example, defined five different model relationships between research and design: artistic, intuitive, adaptive, analytical and systematic. Each model is inclusive of both planning and design, but with a greater or lesser focus on aesthetics (2003, pp.47– 49). Nassauer and Opdam (2008) encouraged interdisciplinary collaboration with landscape designers to increase the usability of the results of academic research, arguing that jointly produced designs would enhance social acceptance and increase innovation. While acknowledging designing as part of doing research, they still suggested a sequential process, that is first research, then designing. Carl Steinitz, however, in his recent book A Framework for Geodesign (2012), gives designing in research and in inter- and trans-disciplinary interactions a much more prominent role. Nonetheless, the role of designing and its intrinsic values for the research process were still left obscure. Or as Deming and Swaffield (2011, p.205) observed, ‘design as an investigative strategy remains poorly understood and inconsistently applied, even if frequently invoked’. They used the term ‘projective design’ to emphasise ‘the unique quality of design process for research outcomes’, and explain how ‘[d]esign only becomes an autonomous research strategy when it produces new generalizable knowledge about the world through its purposes, protocols, and outcome’ (p.206). In their view, research-by-design is neither inductive (something must be: insights emerge from the design setting or context) nor deductive (something is actually operative: insights emerge from the testing and

58 …

The relationship between research and design „

challenging of established concepts), but rather an approach in which ‘researchers move back and forth between inductive and deductive perspectives, modifying their theoretical propositions in the light of the evidence, revising their understanding of the evidence (its categories, and its meaning and significance) in light of theoretical concepts and exploring new possibilities of understanding and new ways of knowing’. This ‘reflexive approach’ is called abduction, that is ‘an investigation of what might be’ (pp.8, 209) and is a hitherto overlooked approach to research in landscape architecture. This overview of the evolution of designing as a research method in landscape architecture shows that design(ing) as a research approach has gradually developed over more than four decades. We can observe that design(ing) as research was described using different and often multi-interpretable names. In all approaches, landscape and urban design remained the predominant focus of attention. However, the notion of designing as a part of soundly based research, following acknowledged criteria for research quality (e.g. Deming and Swaffield 2011, p.206; Milburn et al. 2003), was still not well addressed.

CURRENT MEANING OF RESEARCH THROUGH DESIGNING Lenzholzer et al. (2013) proposed that designing can be research provided it complies with the procedures, protocols and values of academic research. They suggested a framework of research criteria for design based on the widely used criteria set by Creswell (2014). Creswell described three different types of research strategies: qualitative, quantitative and mixed. He further distinguished four substantially different worldviews within which these approaches can be applied: (post) positivist, constructivist, transformative and pragmatic. Regardless of the worldview, the choice of the appropriate research strategy and subsequent research methods is always guided by the research question(s). Based on this categorisation and on Lenzholzer et al. (2013) we propose four different types of RTD in landscape architecture (see Table 4.1). For each type of RTD the overview addresses the kind of questions that can be answered, the kind of (new) knowledge that can be produced, the methods that can be employed and appropriate types of evaluation. The output of RTD should be new knowledge that is applicable in design practice or in further research. Typical outcomes can be generic design prototypes or design guidelines, but also ‘thick descriptions’ of lessons learnt in transformative RTD. The outcomes thus fundamentally differ from design products arising from ‘non-academic’, practical design processes that are site-specific and not created with the aim of broadening the disciplinary knowledge base. RTD methods within a (post)positivist approach are applied in many engineering-oriented landscape architectural projects. If, for instance, dike systems have to be improved, designs will usually be tested with computer simulations or mock-ups, evaluated, tested again in a range of predominantly quantitative feedback loops until an optimised solution is found. Typical constructivist RTD methods entail the generation of designs that are assessed according to their cultural, aesthetic, ethical or other socially ‘constructed’ values – either by expert groups or through citizens and are also brought to a satisficing (as opposed to optimal) result through predominantly qualitative feedback loops. The inclusion of citizens can lead to transformative RTD. Such research is oriented on co-creation and co-ownership of knowledge and process. The researcher uses a combination of participatory and ‘designerly’ research techniques. This RTD often concerns the development of a commonly shared vision on the endeavoured (future) quality of the living environment and an action plan to achieving and maintaining it.

59 …

Sanda Lenzholzer, Ingrid Duchhart and Adri van den Brink

„

Table 4.1 Overview of types of RTD (source: based on Lenzholzer et al. 2013)

However, we would like to stress that the fourth category – the ‘pragmatist’ RTD – might have the highest significance amongst the RTD methods. The pragmatic worldview is concerned with the research problem and uses pluralistic approaches to derive new knowledge about the problem and its solution. Landscape architectural design also deals with complex tasks and the smart integration of many interests to solve and/or identify a design problem and, accordingly, the new generated knowledge has to address such complexities too. Therefore, when research methods only follow one of the approaches, for instance, a positivist one, the new knowledge that is generated can only partially address the problem due to the reductionism inherent in this approach. In the case where

60 …

The relationship between research and design „

transformative methods are used exclusively, the use of certain research techniques or questions that are culturally unacceptable may also render the new knowledge inadequate. Design guidelines, spatial prototypes or other generalisable design knowledge will have to include various aspects in order to make it usable, design-relevant knowledge. For instance, aesthetics or social aspects need to be considered and these would have to be studied with a constructivist RTD approach. The pragmatic researchers ‘look to the what and how to research … where they want to go with it. Mixed methods researchers need to establish a purpose for their mixing, a rationale for the reasons why quantitative and qualitative data need to be mixed in the first place’ (Creswell 2014, p.11). These characteristics for pragmatic research show strong similarities with legitimate design processes. Thus, we suggest to use the strength of the landscape architect’s integrative and ‘designerly’ thinking and employ a pragmatic RTD, mixing approaches and methods when complex design-relevant knowledge is needed. The categorisation presented in Table 4.1 should, therefore, not only be understood as a dissection of RTD approaches but rather as an overview of different types of RTD and their potential to address the research questions at hand in the most suitable way. We assume that in most cases this will involve a combination of the different RTD methods we portrayed.

FUTURE POTENTIALS The discussion so far has indicated that RTD, as long as it is conducted according to academic standards, can yield reliable and valuable design-relevant knowledge. We hope that the RTD approaches we described above will be employed and developed further in future landscape architecture research. Where the pioneers of RTD had technological problems in handling large amounts of data, we believe that new perspectives may be induced by rapid developments in the IT sector, such as big data and virtual reality (e.g. Portman et al. 2015). For (post)positivist RTD methods new possibilities arise due to the growing capacity of computers because simulations of design hypotheses and proposals can be done much quicker and tested by prospective users. The software that enables such simulations is becoming increasingly sophisticated. Coupled with more computing capacity and more sophisticated simulation methods is ‘parametric design’. Parametric design is based on algorithms that express parameters and rules that define the eventual design responses (e.g. Gerber and Lin 2014). So actually, parts of the evaluation of the design to be tested in a (post)positivist RTD can be built into the algorithm. This can reduce feedback loops for certain parameters that would otherwise have to be tested by other means. Constructivist and transformative RTD will possibly profit from new representation techniques, such as virtual reality, that enable immersion of people into new spatial environments that do not yet exist. Virtual reality offers increasingly realistic simulations of visual, auditory and sensory environments and even the olfactory realm can be included. The potential for people to undergo ‘total immersion’ can be used to give their subjective evaluations of (aesthetic) experiences. Transformative RTD methods will also profit from these new possibilities. People will have easier access to new design proposals that need to be assessed, for instance via smartphone apps or other digital platforms. They have exciting opportunities to communicate more efficiently through new media and to vote for design proposals or to upload their own research data. The great amount of data supports the validity of the collected data. This will further advance community empowerment through transformative RTD processes. Generally speaking, we see many possibilities for the existing RTD methods, and even more for the future RTD methods, to generate new knowledge relevant for shaping the spatial environment.

61 …

Sanda Lenzholzer, Ingrid Duchhart and Adri van den Brink

„

Since our world is changing ever faster, the ‘transformative’ sciences that deal with the potential future states of our environment are likely to become increasingly important. RTD is expected to play an increasingly significant role in landscape architecture and in shaping landscape architecture as one of the ‘transformative’ sciences of the future.

SUGGESTED FURTHER READING Creswell, J.W. (2014) Research Design, Qualitative, Quantitative, and Mixed Methods Approaches, 4th ed., Thousand Oaks, CA: Sage. This book gives a broad overview of different approaches to academic research and is widely used in various academic disciplines. Duchhart, I. (2011) Annotated Bibliography on ‘Research by design’ (Ontwerpend Onderzoek), Landscape Architecture Group, Wageningen University, The Netherlands. This annotated bibliography provides insights into the backgrounds of RTD and related discussions in landscape architecture and neighbouring disciplines. Jonas, W. (2007) ‘Research through DESIGN through research: A cybernetic model of designing design foundations’, Kybernetes, 36(9/10), 1362–1380. Wolfgang Jonas has addressed RTD in different design fields. He has a rigorous approach to RTD as a valid research method, embedding his ideas in the current discourse on academic methods. Lenzholzer, S., Duchhart, I. and Koh, J. (2013). ‘”Research through designing“ in landscape architecture‘, Landscape and Urban Planning, 113, 120–127. This paper introduced RTD in landscape architecture. It is based on Creswell’s commonly shared academic values and shows that RTD plays a central role in landscape architectural research. Nowotny, H., Scott, P. and Gibbons, M. (2001) Re-Thinking Science: Knowledge and the Public in an Age of Uncertainty, Cambridge: Polity. This book introduces new approaches in science that incorporate the role of society as a generator of new knowledge. Rodgers, P. and Yee, J. (2015) The Routledge Companion to Design Research, Abingdon: Routledge. This recent book gives a broad overview of different relations between research and design in different disciplines.

REFERENCES Alexander, C. (1964) Notes on the Synthesis of Form, Cambridge, MA: Harvard University Press. Archer, B. (1995) ‘The nature of research’, Co-Design Journal, 2(11), 6–13. Bayazit, N. (2004) ‘Investigating design: A review of forty years of design research’, Design Issues, 20(1), 16–29. Bell, S. and Apostol, D. (2008) Designing Sustainable Forest Landscapes, London: Taylor & Francis. Blanusa, T., Monteiro, M.M.V., Fantozzi, F., Vysini, E., Li, Y. and Cameron, R.W. (2013) ‘Alternatives to Sedum on green roofs: Can broad leaf perennial plants offer better “cooling service”?’, Building and Environment, 59, 99–106. Bouwmeester, H., Elsinga, J., Hendrich, M., Lagendijk, O. and der Nederlanden, H., eds. (2009) Ontwerpen op het raakvlak van water en ruimte: Handreiking en voorbeelden, Actieprogramma Ruimte en Cultuur, Den Haag: Ministerie van VROM. Brinkhuijsen, M. (2008) ‘Landscape 1:1: A study of designs for leisure in the Dutch countryside’, PhD. thesis, Wageningen University, the Netherlands. Brown, R.D. and Corry, R.C. (2011) ‘Evidence-based landscape architecture: The maturing of a profession’, Landscape and Urban Planning, 100(4), 327–329. CIAM (1943) ‘The Athens Charter’, in Watson, D., Plattus, A.J. and Shibley R.G., eds. (2003) Time-Saver Standards for Urban Design, New York: McGraw-Hill. Creswell, J.W. (2014) Research Design, Qualitative, Quantitative and Mixed Methods Approaches, 4th ed., Thousand Oaks, CA: Sage. Cross, N. (2006) ‘Forty years of design research’, Design Research Quarterly, 1(2), 3–5. Cross, N., Naughton, J. and Walker, D. (1981) ‘Design method and scientific method’, Design Studies, 2(4), 195–201.

62 …

The relationship between research and design „

Cushing, D.F. and Renata, A. (2015) ‘Themes in landscape architecture publishing: Past trends, future needs’, Landscape Journal, 34(1), 15–36. de Jong, T.M. and van der Voordt, D.J.M. (2002) ‘Criteria for scientific study and design’, in de Jong, T.M. and van der Voordt, D.M.J., eds. Ways to Study and Research Urban, Architectural and Technical Design, Delft: Delft University Press, 455–457. Deming, E.M. and Swaffield, S. (2011) Landscape Architecture Research: Inquiry, Strategy, Design, Hoboken, NJ: John Wiley & Sons. Dorst, C.H. (1997) ‘Describing design: A comparison of paradigms’, PhD thesis, Eindhoven Technical University, the Netherlands. Eder, W.E. (1995) ‘Viewpoint engineering design: Art, science and relationships’, Design Studies, 16(1), 117–127. Findeli, A. (2001) ‘Rethinking design education for the 21st century: Theoretical, methodological, and ethical discussion’, Design Issues, 17(1), 5–17. Francis, M. (2001) ‘A case study method for landscape architecture’, Landscape Journal, 20(1), 15–29. Frayling, C. (1993) ‘Research in art and design’, Royal College of Art Research Papers, 1(1), 1–5. Gerber, D.J. and Lin, S.H.E. (2014) ‘Designing in complexity: Simulation, integration, and multidisciplinary design optimization for architecture’, Simulation, 90(8), 936–959. Glanville, R. (2015) ‘The sometimes uncomfortable marriages of design and research’, in Rodgers, P. and Yee, J., eds. The Routledge Companion to Design Research, Abingdon: Routledge, 9–22. Groat, L. and Wang, D. (2002) Architectural Research Methods, New York: Wiley & Sons. Groenewegen, P.P., van den Berg, A.E., de Vries, S. and Verheij, R.A. (2006) ‘Vitamin G: Effects of green space on health, well-being, and social safety’, BMC Public Health, 6(1), 149. Herzog, T.R. (1992) ‘A cognitive analysis of preference for urban spaces’, Journal of Environmental Psychology, 12(3), 237–248. Hewitt, R. (2014) ‘Globalization and landscape architecture: A review of the literature’, SAGE Open, January– March, 1–25. Hillier, B., Musgrove, J. and O’Sullivan, P. (1972) ‘Knowledge and design’, paper presented at the EDRA 3/AR 8, Los Angeles, CA. Available at: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.468.9559& rep=rep1&type=pdf. Hubka, V. and Eder, E.W. (1987) ‘A scientific approach to engineering design’, Design Studies, 8(3), 123–137. Jonas, W. (2007) ‘Research through DESIGN through research: A cybernetic model of designing design foundations’, Kybernetes, 36(9/10), 1362–1380. Jonas, W. (2014) ‘The strengths/limits of Systems Thinking denote the strengths/limits of Practice-Based Design Research’, FORMakademisk, 7(4), 1–11. Jones, J. C. (1970) Design Methods: Seeds of Human Futures, London: Wiley. Kapper, T. and Chenoweth, R. (2000) ‘Landscape architecture and societal values: Evidence from the literature’, Landscape Journal, 19(1–2), 149–155. Kempenaar, A., Westerink, J., van Lierop, M., Brinkhuijsen, M. and van den Brink, A. (2016) ‘“Design makes you understand” – Mapping the contributions of designing to regional planning and development’, Landscape and Urban Planning, 149, 20–30. Krieger, M. and Litton, B. (1971) ‘”Design with Nature by Ian McHarg” (book review)’, Journal of the American Institute of Planners, 37, 50–52. Lang, J. (1987) Creating Architectural Theory: The Role of the Behavioral Sciences in Environmental Design, New York: Van Nostrand Reinhold. Lenzholzer, S. (2010) ‘Designing atmospheres: Research and design for thermal comfort in Dutch urban squares’, PhD thesis, Wageningen University, the Netherlands. Available at: http://edepot.wur.nl/139053. Lenzholzer, S., Duchhart, I. and Koh, J. (2013) ‘”Research through designing” in landscape architecture’, Landscape and Urban Planning, 113, 120–127. McHarg, I. (1968) ‘Values, process and form’ in Steiner, F.R., ed. (2006) The Essential Ian McHarg: Writings on Design and Nature, Washington, DC: Island Press, 47–61. McHarg, I. (1969) Design with Nature, New York: John Wiley. Meir, I.A., Garb, Y., Jiao, D. and Cicelsky, A. (2009) ‘Post-occupancy evaluation: An inevitable step toward sustainability’, Advances in Building Energy Research, 3(1), 189–219.

63 …

Sanda Lenzholzer, Ingrid Duchhart and Adri van den Brink

„

Milburn, L.A.S. and Brown, R.D. (2003) ‘The relationship between research and design in landscape architecture’, Landscape and Urban Planning, 64(1), 47–66. Milburn, L.A.S., Brown, R.D., Mulley, S.J. and Hilts, S.G. (2003) ‘Assessing academic contributions in landscape architecture’, Landscape and Urban Planning, 64(3), 119–129. Milburn, L.A.S., Brown, R.D. and Paine, C. (2001) ‘”...Research on research”: Research attitudes and behaviors of landscape architecture faculty in North America’, Landscape and Urban Planning, 57(2), 57–67. Nassauer, J.I. and Opdam, P. (2008) ‘Design in science: Extending the landscape ecology paradigm’, Landscape Ecology, 23, 633–644. Nijhuis, S. (2014) ‘GIS-based landscape design research: Exploring aspects of visibility in landscape architectonic compositions’, in Lees, D.J., Dias, E. and Scholten, H.J., eds. Geodesign by Integrating Design and Geospatial Sciences, Cham: Springer, 193–217. Nowotny, H., Scott, P. and Gibbons, M. (2001) Re-Thinking Science: Knowledge and the Public in an Age of Uncertainty, Cambridge: Polity. Portman, M., Natapov, A. and Fisher-Gewirtzman, D. (2015) ‘To go where no man has gone before: Virtual reality in architecture, landscape architecture and environmental planning’, Computers, Environment and Urban Systems, 54, 376–384. Prominski, M. (2005) ‘Designing landscapes as evolutionary systems’, The Design Journal, 8(3), 25–34. Rodgers, P.A. and Yee, J. (2015) The Routledge Companion to Design Research, Abingdon: Routledge. Salingaros, N. (2005) Principles of Urban Structure, Amsterdam: Techne Press. Schön, D.A. (1987) Educating the Reflective Practitioner: Toward a New Design for Teaching and Learning in the Professions, San Francisco, CA: Jossey-Bass. Sherman, S.A., Varni, J.W., Ulrich, R.S. and Malcarne, V.L. (2005) ‘Post-occupancy evaluation of healing gardens in a pediatric cancer center’, Landscape and Urban Planning, 73(2), 167–183. Simon, H.A. (1975) ‘Style in design’, in Eastman, C.N., ed. Spatial Synthesis in Computer-Aided Building Design, Chichester: John Wiley, 287–309. Simon, H.A. (1996) The Sciences of the Artificial, 3rd ed., Cambridge, MA: The MIT Press. Simon, K. and Steinemann, A. (2000) ‘Soil bioengineering: Challenges for planning and engineering’, Journal of Urban Planning and Development, 126(2), 89–102. Steenbergen, C., Mihl, H. and Reh, W. (2002) ‘Introduction: Design research, research by design’, in Steenbergen, C., Mihl, H., Reh, W. and Aerts, F.A.M.F., eds. Architectural Design and Composition, Bussum: THOTH, 12–25. Steiner, F.R., ed. (2006) The Essential Ian McHarg: Writings on Design and Nature, Washington, DC: Island Press. Steinitz, C. (1995) ‘Design is a verb; design is a noun’, Landscape Journal, 14(2), 188–200. Steinitz, C. (2012) A Framework for Geodesign: Changing Geography by Design, Redlands, CA: Esri Press. Steinitz, C. and Rogers, P. (1970) A Systems Analysis Model of Urbanization and Change: An Experiment in Interdisciplinary Education, Cambridge, MA: MIT Press. Susca, T., Gaffin, S.R. and Dell’Osso, G.R. (2011) ‘Positive effects of vegetation: Urban heat island and green roofs’, Environmental Pollution, 159(8), 2119–2126. te Boekhorst, J. (2006) Landschapsarchitectuur en onderzoek: Een korte geschiedenis van de landschapsarchitectuur binnen DLO, Wageningen: ESG-Alterra. Teel, T.L., Manfredo, M.J., Jensen, F.S., Buijs, A.E., Fischer, A., Riepe, C., Arlinghaus, R. and Jacobs, M.H. (2010) ‘Understanding the cognitive basis for human-wildlife relationships as a key to successful protectedarea management’, International Journal of Sociology, 40(3), 104–123. Treib, M., ed. (1993) Modern Landscape Architecture: A Critical Review, Cambridge, MA: MIT Press. van Etteger, R., Thompson, I.H. and Vicenzotti, V. (2016) ‘Aesthetic creation theory and landscape architecture’, Journal of Landscape Architecture, 11(1), 80–91. Weber, G. (2015) ‘Auxiliary specialization opportunities in landscape architecture: Nature of profession, current view, allied relationships, skills & knowledge, and future directions’, MSc thesis, Kansas State University. Available at: https://krex.k-state.edu/dspace/bitstream/handle/2097/19044/GabrielaWeber2015.pdf? sequence=1. Zeisel, J. (2006) Inquiry by Design: Environment/Behavior/Neuroscience in Architecture, Interiors, Landscape, and Planning, New York: W.W. Norton & Company.

64 …

Chapter 5: The challenge of publication Maggie Roe

Editors are craftsmen, ghosts, psychiatrists, bullies, sparring partners, experts, enablers, ignoramuses, translators, writers, goalies, friends, foremen, wimps, ditch diggers, mind readers, coaches, bomb throwers, muses and spittoons – sometimes all while working on the same piece … But an editor’s primary responsibility is not to the writer but to the reader. He or she must be ruthlessly dedicated to making the piece stronger. (Gary Kamiya 2007)

INTRODUCTION This chapter aims to provide an ‘editor’s view’ of landscape architecture research in order to help understand if or where enhancement is necessary. This leads to three strands of thought: first, to consider the notion that landscape architecture should be regarded as a specific academic discipline in terms of research outputs; second, to provide a view on how to improve the quality of research papers in the field; and third, to provide some observations on the identification of possible useful future relevant research areas. In this chapter ‘landscape architecture’ is used in the broad sense defined by the UK Landscape Institute to include landscape planning, design, management and landscape science. The basis for the reflection and comment in this chapter is a review of a number of papers including research-based, viewpoint and comment-based editorials which examine the content of journals relevant to research published by landscape academics. In reflecting on these reviews I have also used my experience of editing the international peer-review journal Landscape Research for the last 12 years, and surveys I have carried out during that period as well as various discussions with researchers and colleagues over the last 20 years.

A VIEW OF ‘LANDSCAPE ARCHITECTURE RESEARCH’ In 1998 I took part in a symposium that aimed to review the state of landscape architecture research in the UK; this led to a number of useful papers that questioned what landscape architecture research actually is or should be (e.g. Benson 1998; Selman 1998; Thwaites 1998). Now, 17 years later, this debate has moved slowly forward. Most recently Milburn and Brown (2016) suggest that landscape architecture practitioners in the USA do not regard research being produced as particularly useful to practice. While a gap between academics’ views of what is ‘useful’ in terms of research and that of practitioners is not particularly surprising, what is more interesting to consider is whether it is really important for there to be an identifiable ‘landscape architecture’ body of research.

65 …

www.allitebooks.com

Maggie Roe

„

Landscape architecture can be seen as a form of ‘mode 2’ or ‘post-academic science’ (see Gibbons et al. 1994), that is where the acquisition of cross-disciplinary knowledge is through a collaborative approach in specialist areas and where the applicability of the findings is important rather than research for scientific knowledge alone, as is generally seen in fundamental (or ‘mode 1’) research. While the landscape architecture profession can be seen as a discipline with a particular training, there is a strong case to say that landscape architecture research is not ‘a’ discipline, rather a disciplinary field which uses many different methodologies and methods and examines many areas of interest that lie outside what may be relevant to the (generally applied) research interests of the landscape architecture profession. In most countries where landscape architecture is recognised as a profession such as the UK there is a professional body with specific codes of conduct to ensure that ‘practice, knowledge, skills and techniques are up to date’ (LI 2012, p.5). Professions ‘profess’ particular knowledge about and expertise concerning their subject, in this case the landscape. But as many commentators suggest, this knowledge is pulled from a wide variety of research areas and fields of study including natural sciences, arts and humanities, with research methods primarily developed and adapted from human, physical and cultural geography, psychology, aesthetics and philosophy, history and archaeology, agriculture and horticulture, fine arts and design. Academics who teach landscape architecture may be researching in fields very much on the edge of the areas of interest of those in practice (hence perhaps the results of Milburn and Brown), and may be involved in teaching other disciplines, and/or may not be professionally qualified in landscape architecture. The character of landscape architecture research is similar to planning and architecture fields of study that derive from ‘a broad interdisciplinary knowledge-base which can be distinguished from its professional application’ (Dyck 1994, p.143). Planning and architecture, like landscape architecture, are identified as ‘professionally oriented’ and not primarily focused on developing knowledge for its own sake. Cushing and Renata (2015) suggest that landscape architecture as a discipline is relatively new, then go on to cite the establishment of ASLA (American Society of Landscape Architects, which is the professional body in USA) as being over 100 years old. In the UK the professional body, the Landscape Institute, which was founded in 1929, is to date 86 years old. What Cushing and Renata are actually talking about is the profession. There are many early texts which contain relevant knowledge in terms of the disciplinary field of landscape architecture research. The first theoretical text that could be said to be based on ideas of landscape architecture was ‘A Treatise on the Theory and Practice of Landscape Gardening, Adapted to North America’ by Andrew Jackson Downing (1841). The research methods used by Downing to develop his theory seem to have been observation, cogitation and discussion. Since then there have been many kinds of publications that include relevant research emanating from both practice and academia. Academic journals that publish relevant work include Landscape Research which has been published in one form or another since 1968 (47 years) (Roe 2009). Another important outlet is Landscape and Urban Planning which has been published since 1974 (41 years) with Arnold Weddle as the first editor. The journal was then called Landscape Planning & Urban Ecology (Gobster 2011). Weddle was a landscape architect who was also a practitioner; he became the first professor of landscape architecture in the UK in 1967 at Sheffield University, UK. Weddle himself suggested that ‘landscape design (or landscape architecture...) is an activity rather than a discipline’ (Woudstra 2010, p.255). This brief outline of key events suggests that the field is not in its infancy. There is, however, a perceived need to develop a research profile within a multi-disciplinary field. Certainly in the UK the actual numbers of landscape

66 …

The challenge of publication „

architects who are in academia are small; it is also generally a small profession. An important issue is perhaps, do landscape architects in practice use research of any kind to inform their practice? Is it necessary for there to be identifiable ‘landscape architecture research’ that generates new knowledge, new methodologies and is then assumed to ‘strengthen the academic position of landscape architecture’ (Lenzholzer et al. 2013, p.120)? How can a small body of academics help raise the profile of the disciplinary field? The distinguishing activity for landscape and architecture professionals can be identified as problem-solving and the development of design methods including visioning. Thwaites (1998) critiques the lack of a cohesive research agenda grounded in design itself. Therefore perhaps a specific area where landscape architecture might make a unique contribution to the development of research methods is in relation to design(ing). The problem is that, as has already been suggested, this is only one potential area of research that may be relevant to landscape architecture, and many landscape architects would consider themselves to be other than primarily designers, for example they are landscape planners and assessors, landscape managers, landscape scientists and so on who are all brought together under the banner of ‘landscape architecture’. Another issue that is often suggested as important is that of the number of doctoral students identifiable as landscape architects. As discussed in Milburn and Brown (2003), there is a long-held view that there is a lack of research training within landscape architecture studies programmes and this is a key reason for the lack of profile for landscape architecture research. An analysis of landscape architecture research and the numbers of doctoral candidates that are registered as ‘landscape architecture’ candidates is reportedly relative low (van den Brink and Bruns 2014). Such assessments are problematic to use as an indicator of the state of landscape architecture research in institutions, primarily because in some academic institutions doctoral candidates receive a PhD or DPhil without an identifiable discipline ‘label’. Another problem is that candidates who may be qualified landscape architects may be studying under supervisors who do not identify themselves as landscape architects, and are situated within schools which are also not labelled as such either. Those supervisors that are in landscape architecture schools may or may not have landscape architecture as a first degree. Does all this matter? If trying to identify specific ‘landscape architecture research’ is difficult, how can the profile of landscape architecture research be raised and how can landscape architects make a greater contribution to the growing and important body of research related to landscape, mostly not done by trained landscape architects? The boundaries of the wider field of landscape research are constantly changing and under discussion (e.g. Jorgensen 2014). Landscapes are the focus for many different disciplines and disciplinary fields of study using various approaches, epistemologies, strategies, methods, discourses, framings, constructions of knowledge, ideologies, values and research objectives, some of which overlap with those of the disciplinary field of landscape architecture, and some of which are very different. Robust research is becoming more relevant in policy-making. If landscape architecture is to be seen by both the public and policy-makers as more than a way to ‘design out’ problems, it would seem important that those primarily concerned with landscape change (landscape architects) have awareness of these different viewpoints and are able to justify their practice on an authoritative basis of good research which indicates an in-depth understanding of the how and why of materials used as well as the wider consequences of design decisions. Both academics and practitioners are engaged in research in terms of being commissioned by a variety of funders or through various funding mechanisms to carry out research projects. Many funders consider landscape research as important in relation to cross-disciplinary problems where

67 …

Maggie Roe

„

overlapping discipline areas are relevant and multi-disciplinary approaches are encouraged. In this research environment it is increasingly unlikely for ‘pure’ landscape architecture research to be widely funded. Funders seem to be interested in integrated and interdisciplinary research and in encouraging larger projects. There is movement in both academic funding and practice to encourage venture into the ‘boundary areas between disciplines’ (see Roe 2012). While disciplinary training is useful to address specific disciplinary issues and there is still emphasis in academic institutions for single-disciplinary research strength, researchers who wish to engage in multi-disciplinary teams need to develop collaborative skills, often the ability to see an overview – or see ‘outside the box’ of the discipline – good communication skills and be creative in problem-solving. A landscape architectural training would seem to provide a good basis for all these necessary attributes for collaboration. However, training in research methods within landscape architecture programmes seems still to be an issue of concern. There are two critical questions that the rest of this chapter discusses in relation to research in the disciplinary field of landscape architecture: s s

How can the field be strengthened? What emerging and possible future research themes and research areas can be identified?

IS LANDSCAPE ARCHITECTURE RESEARCH ‘WISHY-WASHY’? One of the common cries is that landscape architecture does not have a good quantitative research basis and therefore is generally regarded by ‘hard’ scientists as ‘wishy-washy’. If landscape architecture is regarded as a crossover field then the expectation would be to see both quantitative and qualitative research as well as predominantly mixed methods, including the use of creative practice methods which respond to the arts and design aspects of the profession. It has been suggested that qualitative research in particular is less valued generally in academia although it is seen as having a particular ability in introducing new ideas into a field (Gilgun 2005). In academia generally, the focus is on the quality and novelty of the research, therefore a focus on how to improve quality should be of primarily importance. In addition, a key issue in crossover research areas is that there are often limited outlets for publication because many journals still tend to focus upon disciplinary lines. This indicates a mismatch between the emphasis of research funders and academic publishing. In particular there is a problem of finding good reviewers for multi-disciplinary papers and in the case of landscape architecture this is particularly a problem where papers not only cross disciplines, but may fall between the two stools in terms of paper structure – between design expression and more classic IMRAD (introduction, methods, results and discussion) type papers. A criticism often levelled is that there is no distinct theoretical base for landscape architecture research or practice. This is hard to address if it is seen as necessary to have a unique theory rather than base research on theory gathered from the many source disciplines evident in landscape architecture research. There are few real attempts to tackle this issue in a clear and evidence-based manner. Thompson carried out a survey of landscape professionals which resulted in his book Ecology, Community, Delight (2000). His proposition was that three overlapping value systems provide the theoretical basis for landscape architects (Figure 5.1). This book has been highly influential around the world, partly because it was the first real in-depth attempt to examine the issue of landscape architecture practice values through research. Since then a number of guidance texts have been

68 …

The challenge of publication „

Figure 5.1 Overlapping fields of value and sources of theory in landscape architecture (source: based on Thompson 2000)

published which further clarify useful theoretical approaches and set out the range of methods and methodologies for landscape architects and landscape studies. Wylie’s (2007) Landscape is a particularly helpful summary of theoretical approaches. Wylie is a cultural geographer, but this book is one that is constantly on academic reading lists in landscape architecture programmes and is commonly found on research papers focusing on landscape. Other relevant texts are Swaffield’s (2002) Theory in Landscape Architecture: A Reader, Howard et al.’s (2013) The Routledge Companion to Landscape Studies edited by a cross-disciplinary team, and Bell et al.’s (2012) Exploring the Boundaries of Landscape Architecture. The first two of these contain contributions from a wide variety of disciplines, not just from landscape architecture. Howard et al.’s (2013) text very specifically takes an eclectic approach to the subject of landscape. Based on Howard’s experience of editing the journal Landscape Research he uses four key themes: experiencing landscape; landscape culture and heritage; landscape, society and justice; and design and planning for landscape. In a slightly older text James Corner (2000) used writings of various contributors and disciplines to support the notion of the importance of landscape. There are other examples of notable contributions to relevant conceptual and theoretical ideas and emerging evidence from landscape architecture researchers, but on the whole, these do not call themselves ‘theories’, but rather ‘approaches’. These include: landscape urbanism (Waldheim 2006; Thompson 2012), landscape planning (Potschin et al. 2010), resilience and ‘reconnection’ (Selman 2012), green infrastructure (Benedict and McMahon 2006) and social constructivist approaches (e.g. Gailing and Leibenath 2015) to name a few. In terms of research methods recommended as particularly useful for landscape architecture, Mark Francis (2001) suggested that

69 …

Maggie Roe

„

the case study method – as used in many other professions – could provide both a ‘primary form of education, innovation and testing for the profession’ and a ‘collective record of the advancement and development of new knowledge in landscape architecture’ (p.15). Landscape ecology texts also often provide useful pickings for landscape architects and indicate that landscape ecology also demonstrates multi-dimensional, multi-disciplinary approaches to landscape issues (e.g. see Dramstad et al. 1994; Dupont and Jacobs 2008). Of course there are many other relevant texts based on landscape research issues, a number of which are based on multi-disciplinary contributions; this supports the view that the field of landscape research is broad and cross-disciplinary. In particular, Simon Swaffield has made a considerable contribution to the development of theoretical and methodological thinking in landscape architecture. He identifies six key themes in theoretical development that are particularly relevant to landscape architecture: ‘the central importance of the aesthetic and symbolic configuration of geological, hydrological, and biological forms and processes, and their ecological interrelationships. This has always underpinned questions of space, form and meaning in landscape architecture’ (Swaffield 2002, p.5). The focus in this text is primarily on helping to build a theoretical basis for landscape architecture practice (rather than wider landscape studies). Landscape architecture is referred to as ‘a discipline’. There is recognition of the wide range and potential for ‘fragmentation’ of the profession into ‘discrete subdisciplines’ (ibid., p.227). Meyer (1997 in Swaffield 2013) gets close to suggesting how an identifiable landscape architectural theory might develop. She suggests that theoretical development for landscape architecture should be contingent, particular, and situational. Grounding in the immediate, the particular, and the circumstantial – the attributes of situated criticism – is an essential characteristic of landscape architectural design and theory. Landscape theory must rely on the specific, not the general … It is not about idealist or absolute universals. It finds meaning, form, and structure in the site as it is … The site – and land – speaks prior to the act of design. (Meyer (1997) in Swaffield 2013, pp.167–168)

She remarks upon a potential ‘reconstructive’ approach that ‘assumes a multi-layered fabric that weaves together threads from primary sources and documents written by landscape architects and about landscape architecture with the concurrent history of ecological ideas, cultural and historical geography, design and planning criticism, and site interpretation’ (p.168). Herrington (2013) recognises how conceptual thinking has been ‘borrowed from a range of disciplines’ (p.356) and identifies three types of theory relevant to landscape design approaches: explanatory, normative and resistant. She identifies a very wide range of sources as inspiration in landscape design, describing the development of the profession as one of the reasons for this diversity. Academics such as Swaffield, Meyer and Herrington publish widely in many different outlets including books, professional publications and international peer-review academic journals such as Landscape Research. In 1,543 research papers published and available on the web in November 2015 in Landscape Research, only 24 had a keyword of ‘landscape architecture’ and 202 had ‘landscape architecture’ in the abstract. However, it would be difficult to dismiss any of the papers published in this journal as being irrelevant to landscape architecture practice, so perhaps the problem is in the understanding of relevance, in the identification by authors of where their work lies and in using keywords and abstracts to indicate allegiance to landscape architecture? Another view, of course, would be that authors in the journal are primarily not landscape architects or do not see their work as relevant to landscape architecture practice.

70 …

The challenge of publication „

RESEARCH AND RESEARCH PAPERS: HOW TO IMPROVE THE QUALITY? If the breadth of research is potentially not a problem, but can rather be seen as a strength, then the focus should be on research quality. Brown and Corry (2011) suggest that the key issue for making landscape architecture a ‘more scholarly profession’ is that it should ‘become a discipline of evidence-based landscape architecture’ based on an analysis of the development of the relationship between research and the medical professions. In the UK, the government has become completely obsessed with evidence-based policy-making and practice and there is a concentration in many academic disciplines on ‘evidence’ gained from traditional laboratory-based controlled experiments. Such methods are only relevant to a small area of landscape architecture practice. Landscape architecture research does not develop in an academic vacuum: there are many guidelines and standards for judging the quality of research that emanate from different disciplines and discipline fields around the world. While approaches and methods may differ enormously as suggested above, many of the standards for judging research quality are very similar and can be useful in developing guidelines for landscape architecture. In the UK there is a common framework established by the UK Government for assessing research quality (Research Excellence Framework) in all disciplines and fields. Medical research in particular is continually under the spotlight: there is a large research infrastructure and, in the UK, a legal requirement to promote research. Practitioners have to use evidence obtained through research in decision-making. Many of the research organisations linked to the medical profession have responsibilities for developing innovative research and links to practice (AMRC 2014). Medicine can be used as a model for improving the rigour of landscape architecture research in relation to practice needs (Brown and Corry 2011; Yang et al. 2016). It is therefore useful to look at some of the means by which medical research examines its outputs. Table 5.1 summarises the guidelines for research standards produced by the National Center for the Dissemination of Disability Research (USA) as a technical note following a review of primarily North American and UK literature (e.g. see Table 5.1) (NCDDR 2005).

Table 5.1 Standards for assessing the quality of research (source: adapted from NCDDR 2005)

71 …

Maggie Roe

„

As with most other disciplinary areas, this shows a focus on the quality of peer-review publication (research output) as an indicator of quality resulting from the research rather than an examination of the research itself. What could or should the possible products (or outputs) of research be in relation to landscape architecture? Products and outputs can be seen in terms of knowledge exchange methods with a wide range of possibilities. Traditional outputs of academic research are peer-review papers and books, reports, verbal/oral communications, discussions and presentations. But increasingly other forms of outputs are both possible and recognised in cross-disciplinary design, arts and humanities subjects such as performances, exhibitions, exhibition catalogues, paintings, installations, films, poetry, compositions, photography, stories, novels, pieces of music, artefacts, buildings, sculpture, curatorships, tweets, blogs, websites, screen writing, dance and actual landscape designs. Evidence of this in the UK can be seen in the submissions to the UK government Research Excellence Framework assessment (REF2014 2014). There is also a growth in non-traditional PhD routes such as through creative practice, which support new forms of research outputs as well as theoretical approaches. Creative practice approaches would seem interesting and pertinent to a design-based professionally focused subject like landscape architecture that is examining new possible ideas in research. Creative practice approaches aim to look at new ways to link research and practice by drawing on ‘subjective, interdisciplinary and emergent methodologies that have the potential to extend the frontiers of research’ (Barrett 2009, p.1). Also relevant is research that aims to develop methods to assess the benefits of creative outputs (designs). Yang et al. (2016, p.315) suggest the need for ‘performance research’ to determine the level of efficiency by which ‘designed landscape solutions fulfil their intended purpose and contribute to sustainability’. The discussion of how research or post-design analysis can evaluate designed landscapes is not new (see for example Manning 1995), but the approach by Yang et al. (2016) employs methods that are clearly aimed to attract the attention of policy-makers in focusing on the development of methods responding to the perceived need for evidence-based designs using quantitative performance measures relating to social benefits. In REF2014 the landscape architecture research outputs were assessed within several peer-review ‘panels’, but the majority came within the sub-panel 16 Architecture, Built Environment and Planning which accepted a range of research outputs. The main outputs were traditional research papers published in various peer-review journals (not just those specifically labelled as ‘landscape’ journals). In this panel there was a significant volume of work in the form of authored and edited books and monographs and design outputs; however, nearly 80 per cent were research papers (REF2014). The results also showed an increase in outputs that reported collaborative interdisciplinary research using mixed methodologies and methods funded by a range of sponsors with the objective of addressing major and global challenges, for example sustainability, carbon reduction and resilience to climate change. In the UK, the assessment of the impact of research has become increasingly important along with the development of the identification of indicators and measures to assess such impact. This is particularly significant for landscape architecture research in considering how research feeds into the development of policy and practice. Measurements of the recognition by scholars and research sponsors of outputs are extremely complex and increasingly used worldwide (Kelly and Burrows 2012). There is a plethora of metrics, league tables and attempts to measure quality and value of research outputs by academic institutions. The REF2014 peer-review panels examined various aspects of institutional research output including perceived ‘impact’ based on impact statements provided by those institutions and the amount of research funding gained per academic. The use of citation

72 …

The challenge of publication „

indices was not used by sub-panel 16 in this assessment because of the unreliability of the metrics for the subjects covered. Indices (e.g. Google Scholar) are hotly debated as signifiers of quality. Citation metrics have been used to develop journal rankings, but it is important to remember that metrics have many influences and, as suggested by Kelly and Burrows (2012, p.150), the impact factors of journals are shaped to a considerable extent by the ‘public life of data… and/or perhaps these very metrics are themselves the outcome of prior academic reputation (people read journals with a good historical reputation so they are the ones that get the citations and thus high impact factors)’. In the UK, post-REF2014 indications from government are that quantitative indicators (citation metrics etc.) will become more important in research quality assessment because they are seen as a ‘shorthand’ for assessing outputs and using such indices costs less than peer-review methods carried out nationally (DBIS 2015). The problem is that although good citation metrics may indicate significant impact in a particular field, weak metrics may not indicate lack of impact in a field, particularly if the field within which the academic is working is small, the publications are predominantly not in the English language and/or if much of the research is published in books (Harzing and van der Wal 2008). All these caveats would seem important for landscape architecture research. In improving the quality of research outputs, perhaps the first question to ask is: What makes what you are doing ‘research’? The objective of research, broadly speaking, can be seen to be knowledge formation so then the focus is on the form of the production and how the work sits within the disciplinary field. It is important to clarify what else is already out there. Basic issues need to be addressed such as a clear statement of the research question and hypothesis (if appropriate), and demonstration of the appropriateness of the approach/methodology (individual/sole researcher, collaborative, engagement, co-production, observational, insider/outsider, community action etc.). It is important to define the character of the research, for example whether it is experimental/testing, observational, deconstructive, critical, creative, and it is generally suggested that methods need to be stated in enough detail to make the approach to the research question clear, to explain the design of the study, to provide enough information to understand, evaluate and potentially replicate the study, and because to communicate an understanding of research results it is important that you communicate how the research was carried out. It is important that methods are appropriate to the answering of the research question; they can be innovative or well-used. Christensen (2006, p.2) as co-editor of the Journal of Planning Education and Research (JPER) suggests: Explain what evidence and research methods are appropriate for your research question and why. Then explain what data you collected, sources, how you sampled, how you collected it, how you coded it, and how you analysed it. Demonstrate that your methods are ‘explicit, sound and appropriate’. If you believe you need to explain highly technical methods, present them briefly in the text and provide more information and references to further detailed explanations in an appendix. Discuss any limitations, drawbacks or possible biases in your methodology and what you did to correct or compensate for the problem.

Robust methods are often used by assessors as an indicator of quality of research (but not necessarily quality of paper). If there is any doubt about the robustness or appropriateness of the methodology or methods it often means immediate rejection by assessors of the research, and methods quality may be used as an indication of clarity of thinking about the whole process or project. There is commonly confusion in landscape studies as to how long methods reporting should

73 …

Maggie Roe

„

Table 5.2 Methods reporting: survey of papers in 2009 published in Landscape Research

be within a paper. Again this should be appropriate to the type and character of the paper. In a survey I carried out of papers published in Landscape Research in 2009 methods reporting varied from 10 lines to 3.5 pages, or sometimes a general description of less than a page, or a critique throughout the paper (in relation to a ‘methodology’ paper) (see Table 5.2). Perhaps the most important question for any researcher to consider is: ‘What is new?’ In the UK REF2014, the highest quality research was based on the assessment that the research quality was world-leading in terms of originality, significance and rigour. This was applied to landscape architecture research in whatever form of output; whether traditional research paper or submission of a design-as-research, the criteria for assessment were the same. The impact of research was regarded as high quality if it was outstanding in terms of the reach and significance. These descriptions are on the face of it quite helpful, but the detailed assessment of such terminology spawns a host of sub-categories and description and the need for indicators and evaluations. It is perhaps useful to return to what knowledge formation is in the field of landscape architecture as a fundamental issue. If a key potential and defining characteristic of landscape architecture is the design and creative component, then it is helpful to examine the ideas around the growing area of creative practice research. In creative practice research knowledge formation is potentially through making, performing and inventing, often with communities or individuals in a ‘shared’ or co-produced methodological framework. The methodologies for creative practice research equate to the ‘how’ of the research. Key issues are how to engage critically with the theories/literature and provide the contextual framework, thinking about how to ask questions and seek answers, making use and sense of data and ideas, reflection on material practices and actions and the importance of the framework which helps the creation in a critical and reflexive manner (Grierson and Brearley 2009). In the standard research output, the research paper can be defined as a report that is based on an original idea, argument or thesis backed up by evidence of some kind. Depending on the discipline there are various forms of research paper so that it may be an exploration or thoughtful reading on a particular subject which allows for the development of an argument. The reading material may come from several sources and may be some kind of systematic review of past research which locates the context of the research, but also the use of ‘facts’ from surveys, experiments

74 …

The challenge of publication „

and so on with interpretive analysis, theoretical development (often) and conclusions. There are many sources which provide guidance on the detail of writing research papers which are relevant to landscape architecture questions (see for example Deming and Swaffield 2011; Azlan 2013). In relation to writing high quality papers, some further reflection on this is perhaps useful. In 2013 a survey I carried out of editors’ views provided information concerning what makes a high quality research paper. Simin Davoudi (associate editor, Journal of Environmental Planning and Management (JEPM) (pers. comm. 2013) suggests four key issues: 1 2 3 4

A clear and specific argument A clear and logical structure Something that is well-written and well-crafted Something that leaves the reader with a feeling that they have actually learned something

Clarity is obviously important here – not just clarity of writing, but also clarity of the argument; there is also another point to be gained from what she says: there must be something to say – that is there must be some new knowledge to be imparted by the paper, and the new knowledge must be understandable. This is critical: you must have something worth saying first; second you must consider your audience and construct your argument throughout the paper that is understandable to that audience. Identification of the audience can be particularly difficult in a cross-disciplinary paper. In their advice the editors identified some basic errors that are often committed by authors of papers. Neil Powe (managing editor, JEPM) suggests that more care needs to be taken over the abstract, the introduction and the conclusion: ‘Don’t rush the writing of the abstract – often this is written last and in a rush but it is the first thing that editors and potential reviewers often see’ (pers. comm. 2013). Paul Selman (former managing editor, Landscape Research) suggests that a good paper tells ‘an interesting story. There is a clear and overarching purpose to the paper, and that purpose is so clear in the author’s mind that they tell it with lucidity and conclusiveness’ (pers. comm. 2013). A key issue is to provide the context for the research through the literature review and as Klosterman (in Christensen 2006, p.2) suggests: Tying the research to the relevant literature helps to construct and show your conceptual framework. ‘This is not just a matter of recognizing one’s intellectual debts. More importantly, this literature can provide a wealth of intellectual and practical guidance in conducting the research. … [R]esearch deals with a host of important, complex and difficult topics and needs to draw on as much of the relevant literature as possible to insure that the research is as rigorous and intellectually sound as it can be.’

The literature for landscape architecture research is potentially very large, which is a considerable problem; this indicates the need to carry out a scoping exercise to map the relevant literature in response to the research questions. Ken Taylor (associate editor, Landscape Research) suggests that it is important to ask ‘what new or different perspectives or critique does the paper address if it’s a well-trodden topic?’ (pers. comm. 2013). He goes on to say that ‘nothing is more irritating than an author who claims this is a new approach and has not been done before, when rigorous review of literature and work in the area would reveal it is not new’. Although it is always interesting to have new robust methods demonstrated in research papers, it is also useful to consider that it is not necessary to ‘reinvent the wheel’ and part of the job of a good literature review is to review methods

75 …

Maggie Roe

„

used, and use this review to develop appropriate methods for the research as well as understand the locus of the ideas of the research. Clarity is something that all the editors consulted kept repeating in relation to various aspects of papers. For example, the nature or character of the paper should be clear. As Ken Taylor (pers. comm. 2013) suggests: ‘Is the paper focused on theoretical matters or practice and if the latter what theoretical underpinnings are included?’ Paul Selman (pers. comm. 2013) suggests that the paper should make a clear and original contribution, either by new evidence or independent critical insight. Too often I find myself reviewing a paper and thinking ‘but what did they actually do’? Sometimes authors will then re-write it in a way which much more clearly sets out their aims, methods, results, and original findings. A pity they couldn’t have done that first time round!

Most papers submitted to high quality peer-review journals get rejected; so if you get rejected it is the norm and the important thing is to try and learn from the process, resubmit after revision or submit the paper elsewhere. As Neil Powe (pers. comm. 2013) suggests: ‘Papers first have to get past the editor so they have to meet basic editorial standards.’ In particular the issue of language is one that frustrates editors. Although editors have sympathy with non-native English speakers, the English language journals are increasingly pressed because of the large number of papers submitted by non-native speakers. The demand for high quality language skills is important. This is not just about good language, but appropriate language, so disciplinary jargon is discouraged. As Peter Howard (former managing editor, Landscape Research) suggests: ‘Try and write English which is clear and concise; you are writing to be understood, not to prove how clever you are’ (pers. comm. 2013). Heather Campbell (senior editor, Planning Theory & Practice) suggests: ‘Plain speaking should not … be confused with over-simplification’ (2003, p.390), and Gary Fry (2001, p.162) wrote: I react against the trend [to use jargon] for several reasons, but mainly because it is fuzzy, it is sometimes patronising, and because this jargon externalises and objectifies landscapes and the people living in them. We talk about actors, stakeholders, customers, cross-compliance payments, sectoral targets, and we engage in optionalising, foresighting, outreaching, rightsizing, clustering and consensual reporting. Oh for simple, direct language and a drastic reduction of euphemisms!

Basic advice such as considering whether ‘paragraphs link together logically to give sense of flow’ (Ken Taylor, pers. comm. 2013) is surprisingly often ignored by research paper authors. An increasing issue with transdisciplinary or participatory methods is how to ensure that the ‘voices’ of the research emerge and/or are portrayed in the paper. This is linked to ethical concerns in research. Peter Howard (pers. comm. 2013) suggests: ‘When dealing with qualitative data (for example interviews) there should be a balance between your analysis of the respondents’ views, and allowing their own voices to be heard.’ Direct quotations used judiciously within the reporting of the data are often important and also some analysis of how things are being said, not just what is being said can be critical to understanding the data. The voice of the researcher should also be considered and should be clearly linked to methods used; so use ‘I’ (rather than the more traditional third person style) only when the ‘I’ matters in the research. Most research papers are not conversations, but you want to ‘engage’ the reader in the research because readers, in particular reviewers, have

76 …

The challenge of publication „

lots of reading to do, so interesting the reader is important. Some disciplines may prefer to see the researcher reflected in the research through the style of reporting, but most readers do not care about your impressions or feelings; they want to be stimulated to think or simply gain knowledge/ understanding. However, indicating a little passion about your subject is engaging, and writing in the active rather than passive tense is generally recommended. It is not easy to write a good research paper; it takes a lot of drafting and redrafting and one of the best pieces of advice is to find a critical reader who will examine drafts of the text carefully. In particular as Strunk and White (1979, p.12) suggest: Vigorous writing is concise. A sentence should contain no unnecessary words, a paragraph no unnecessary sentences, for the same reason that a drawing should have no unnecessary lines and a machine no unnecessary parts. This requires not that the writer make all his sentences short, or that he avoid all detail and treat his subjects only in outline, but that every word tell.

Originally written in 1918, this advice is still valid for today’s researchers.

KEY AREAS OF RESEARCH: PRESENT AND FUTURE POTENTIALS The issue of whether trends in published landscape research respond to perceived needs of landscape architecture practitioners is a difficult one to determine because the demands for academic publishing and research are not the same as the needs of practitioners as identified above. In addition different journals provide particular emphasis often according to the interests of the editors and in response to various other issues. Practitioners on the whole probably wish to read papers relating to applied science rather than theoretical approaches, while the remit of academia is to promote the acquisition of knowledge and understanding in a broader form. The question then is, should landscape architecture research respond solely to the needs of practice? Cushing and Renata (2015) suggest there is a mismatch between research and practice, but there are all kinds of difficult variables at work in this kind of assessment. Perhaps landscape architecture practitioners need to extend the breadth of their reading and perhaps landscape researchers need to provide research knowledge in a format that is more useful to practitioners. From a practitioner’s point of view there is much relevant research that is not published in any ‘landscape’ journal; the problem is more likely to be accessibility to papers and knowledge than that such research exists. This situation may change fairly quickly as more research publication becomes open access and is therefore more generally available. In addition journals such as the Journal of Landscape Architecture (JoLA), established in 2006 by the European Council of Landscape Architecture Schools (ECLAS), have pioneered a publishing approach that contains rigorous peer-review research content with high quality and lavish illustrative material and a variety of paper formats with the potential to attract a wide academic and practitioner audience. In looking at possible themes that may be of particular interest to landscape architects a number of issues are discussed by Meijering et al. in Chapter 6, this volume. Cushing and Renata (2015) identify climate change, active living, energy and health as areas of particular importance presently poorly represented from a landscape architecture research framing. In 2011 Howard identified six key future trend areas in landscape management which provide an additional guide to important potential areas of landscape research: access, wider protection, participation, intangible elements,

77 …

Maggie Roe

„

mobility and climate change (Howard 2011). In Europe, there is still much to be gained from a close examination of the European Landscape Convention (ELC) text in relation to developing relevant landscape research topics. The ELC has provided and continues to provide the inspiration for new research questions, particularly in relation to how ordinary landscapes can be protected, enhanced, restored and created. Quality and quality objectives are particularly important in the interpretation of the ELC text and the value of all landscapes need to be reflected and landscape values better understood. We need better ways that are more inclusive for evaluating landscapes which build on landscape character work and emphasise the role of landscapes in local distinctiveness. In order to engage with policy-makers the links between quality of landscape and economic and social success, health and wellbeing and so on need to be made clearly and with supporting evidence of the role landscapes can play in addressing these issues. It seems that an integrated and innovative or creative landscape approach may be useful as the basis for forward (visionary) planning to protect, manage and plan landscapes, but the ways in which landscape design can incorporate understandings that provide more just livelihoods are also important. The role landscape architecture can play in issues of social and ecological justice (e.g. climate change adaptation) is often ignored or forgotten as discussed above. As Sheppard (2015) suggests, there is a key role for landscape professionals in making climate change ‘visible’. Evidence that can be mobilised by practitioners and policy-makers is important and assessments of past landscape projects through, for example, landscape performance research (Yang et al. 2016), which focus on quantitative measurements of social benefits, may help to strengthen the justification for landscape change by design. Jorgensen (2014) identifies various potential research topics, particularly the threats to habitats and landscapes through increasing world population, urban densification, economic growth, and social and economic inequities. She throws out a considerable challenge to landscape architects to respond to such issues particularly those resulting from conditions where global populations live in unplanned settlements with basic facilities and infrastructure and little or no access to non-human environments. In identifying the frontiers in urban ecological design and planning research, Steiner (2014) identifies three key challenges: population growth, increasing habitation of urban areas and human influence on ecosystem change. The four interrelated fields of research he then identifies as a result of this as being of primary importance are: ecosystem services, the impact of natural disasters, the role of green infrastructure, the renewal of degraded urban places and the capacity of people to adapt to knowledge about their surroundings. Steiner suggests that these build on the expertise of landscape architects and planners and have the potential ‘to advance sustainable design and regenerative design’ (p.305). In March 2015 a one-day cross-disciplinary symposium was held at Newcastle University, UK, that aimed specifically to bring together practitioners, policy-makers and academics for discussion and to identify the key issues for policy, practice and research, primarily in the context of England and the wider UK. A number of emerging themes were identified which were articulated as questions (see Table 5.3). These provide useful indicators as to the kind of issues that are concerning landscape practitioners (planners, managers and designers) and policy-makers in particular. Many of these suggest the need to develop improved methods of practice and communication; many could be developed into research questions that would be of immediate relevance to practice and policy-making. In the UK, new, more collaborative and exploratory ways of working in both research and practice are emerging in spite of or perhaps in response to the demand for better evidence. For example, there is considerable interest by research funders in co-production and co-creation of knowledge.

78 …

The challenge of publication „

Table 5.3 Emerging themes from the Landscape Forward Symposium, 18 March 2015 (source: based on Roe 2015)

This is related to action research and participatory methods. It acknowledges approaches that have emerged from planning, cultural geography, ethnography and heritage studies. Methods now being used are often associated with participatory approaches and include oral histories, artistic expression and direct contact explorations of landscape using performance, walk-and-talk and so on. Many methods being developed owe some inspiration to experiential approaches with particular relevance to the perceptual, sensory and intangible aspects of landscape. Researching landscape perception through walking the landscape has long been used by educators and researchers based on the understanding that ‘the body is central to our understanding of landscape’ (Macpherson 2006), but there are some new and interesting methodological developments and much interest from cognate disciplines in landscape studies (see also Schultz and van Etteger – Chapter 11). Concepts based on sensory perceptions such a ‘smellscapes’, ‘soundscapes’, ‘touchscapes’ and ‘tastescapes’ are encouraging evaluative methods in landscape research, and also encouraging new ideas of what ‘landscape’ might be. But as Macpherson (2006) suggests, perceptions are complex and integrated; a whole body approach that understands people’s perceptions, embodiment and the dynamic nature of landscape is being encouraged through references to philosophers and cultural geographers, and being taken up in many different ways by those interested in the wider field of landscape studies. An example is the idea of the ‘underwater landscape’ explored by Musard et al. (2014); another is the examination of how people’s sensory perception of birdsong contributes to positive values in urban green spaces (Hedblom et al. 2014). On the natural sciences angle of landscape, empirical methods are expanding to potentially include DNA analysis of species, satellite imagery and other digital methods. Data gathering methods are also developing using web-based platforms and apps to gather crowd-based information and utilise citizen science methods in understanding attitudes to landscape as well as gather day-to-day

79 …

Maggie Roe

„

data (Natural England 2012; UKEOF 2016). Research in cognate discipline fields such as planning provides pertinent examples of potential future use of such methods. One example has indicated the use of ‘mobile participation’ through a variety of apps, particularly in relation to location-based data and local knowledge changing the role of ordinary people from information ‘receivers’, to ‘sensors’ and ‘partners’ in data observation and collection (Ertiö 2015). Another example uses a map-based app to gather spatial and character detail of a neighbourhood and for use as a visioning tool for future change (Jones et al. 2015). An example of a citizen-instigated project that has been used in policy discussions to provide evidence of the attachment to landscapes is the ‘Placebook Scotland’ (2016) which ‘is a community of people who love to share what Scotland’s landscapes and places mean to them’ through a website, blog and forum. Such tools would seem to have considerable potential for landscape research; however, they require critical use, and creativity in both the purpose and attitude of researchers. Through such methods exciting new worlds are opening up to landscape architecture researchers if only they are willing to investigate them (see also van Lammeren et al. – Chapter 9).

REFLECTIONS In 2013 Marc Antrop reflected that academic and applied landscape research was diverging as a result of a shifting focus in research goals, and funding sources, the academic merit system and the focus on pure research encouraged in PhD research. He also identified a shift in landscape research topics including, in Europe, a move away from quantitative landscape metrics to more holistic approaches that reference various disciplinary areas. This shifting and fragmenting picture provides many difficulties for landscape architecture researchers, but it also provides many opportunities to look outward and embrace change. Bell et al.’s (2012) brave attempt to explore the boundary areas of landscape architecture as a way of helping to develop the potential for cross- and interdisciplinary working and to ‘re-evaluate its old assumptions and above all to engage more closely and creatively with the many other disciplines who also have an interest in understanding and shaping our common landscape’ (p.2) is a call not just to landscape professionals, but also to researchers to ‘beef up’ their efforts towards and openness to transdisciplinary approaches. It was clear to me in reviewing the contributions to the book in order to write the concluding chapter, that intellectual agility is a key attribute that many landscape architects have and need to mobilise in order to respond to such challenges, as well as embracing and celebrating the wide range of influences on landscape architecture, rather than be bowed down by it (Roe 2012). My own approach to the issue of landscape architecture research is a pragmatic one (see Lenzholzer et al. 2013). So, for example multiple research methods that cross-reference and support understandings would seem sensible, and careful integration of new knowledge from various research and knowledge sources into existing knowledge (including indigenous and local knowledge for example) can be important. An outward gaze that embraces boundary crossing by other disciplines and fields of study is helpful in a world where environmental and social issues are increasingly complex. A key question is, ‘What can those with a landscape architecture training and outlook offer in terms of research insights or approaches that others cannot?’ What questions should we be asking? In an anniversary edition of the journal, Landscape and Urban Planning, Paul Gobster (2011) suggests that the major challenge for future research and practice is how we conceive of landscape itself; quoting Peter Jacobs, he suggests the importance of moving

80 …

The challenge of publication „

beyond questions of landscape that ask: What is it? or: How can it be measured? and toward questions of: What does it mean? ‘Landscape embodies the memory of natural process and human endeavor; the expression of who we are and what we value; it provides critical support for what we wish to become and how we wish to live within nature.’ (p.316)

As an editor of an international peer-review journal I see myriad potentials for landscape architecture research as indicated in Gobster’s suggestions, but as Kamiya indicates in the quotation used at the preface to this chapter, editors are ruthless and dedicated to raising the quality of research, so while there are many potentials for researchers, the quality of the outputs needs to be high to have the kind of impact needed to raise the profile of the disciplinary field. The exploration in this chapter suggests that landscape architecture research is based on composite knowledge and theoretical and methodological approaches drawn from many disciplinary fields. The key task for landscape architecture research is to define clearly what the important research questions are and what theories or methods will be appropriate and helpful in answering these questions. There is considerable potential for multi-disciplinary and interdisciplinary if not transdisciplinary research relating to landscape and increasing funding available for research in these areas. In particular the ability to see the overview of landscape and use creativity, that also allows landscape architects to solve problems on the ground, may be employed in the research context. This is particularly useful within research teams to identify innovative approaches to answering difficult research questions. Also important is an outward-looking approach that embraces the diversity and complexities of wider landscape research.

SUGGESTED FURTHER READING Benson, J.F. (2001) ‘Inside the Editor’s black box: 10 years of the Journal of Environmental Planning & Management’, Journal of Environmental Planning and Management, 44(1), 3–19. Cassatella, C. (2009) Landscape to be (Paesaggio al futuro), Torino: Marsilio. Council of Europe (2000) The European Landscape Convention text. Available at: http://conventions.coe.int/ Treaty/en/Treaties/Html/176.htm (accessed 30.06.2016). Council of Europe (2009) Guidelines for the Implementation of the European Landscape Convention, CEPCDPATEP (2009) 2E, Strasbourg: Council of Europe. Available at: http://www.coe.int/en/web/landscape/ guidelines-for-the-implementation-on-the-european-landscape-convention (accessed 30.06.2016). Dupont, A. and Jacobs, H., eds. (2008) Landscape Ecology Research Trends, New York: Nova Science. Grierson E. and Brearley, L., eds. (2009) Creative Arts Research, Rotterdam: Sense. Kienast, F., Wildi, O. and Ghosh, S., eds. (2007) A Changing World: Challenges for Landscape Research, Dordrecht: Springer. Roe, M.H. and Taylor, K., eds. (2014) New Cultural Landscapes, London: Routledge. Smith, H. and Dean, R.T., eds. (2009) Practice-Led Research, Research-Led Practice in the Creative Arts, Edinburgh: Edinburgh University Press.

REFERENCES AMRC (Association of Medical Research Charities) (2014) Research in the NHS. Available at: http://www.amrc. org.uk/our-work/research-nhs#sthash.TXEjuZZx.dpuf (accessed 07.04.2016). Antrop, M. (2013) ‘A brief history of landscape research’, in Howard, P., Thompson, I. and Waterton, E., eds. The Routledge Companion to Landscape Studies, Abingdon: Routledge, 12–22.

81 …

Maggie Roe

„

Azlan, N.I. (2013) Research Methodology in Landscape Architecture, Singapore: Trafford. Barrett, E. (2009) ‘Introduction’, in Barrett, E. and Bolt, B., eds. Practice as Research Approaches to Creative Arts Enquiry, London: Tauris, 1–13. Bell, S., Sarlöv Herlin, I. and Stiles, R., eds. (2012) Exploring the Boundaries of Landscape Architecture, Abingdon: Routledge. Benedict, M.A. and McMahon, E.T. (2006) Green Infrastructure, Linking Landscapes and Communities, Washington, DC: Island Press. Benson, J.F. (1998) ‘On research, scholarship and design in landscape architecture’, Landscape Research, 23(2), 198–204. Brown, R. and Corry, R. (2011) ‘Evidence-based landscape architecture: The maturing of a profession’, Landscape and Urban Planning, 100, 327–329. Campbell, H. (2003) ‘Talking the same words by speaking different languages: The need for more meaningful dialogue’, Planning Theory & Practice, 4(4), 389–392. Christensen, K.S. (2006) ‘The scholarly paper’, Journal of Planning Education and Research. Available at: http:// www.sagepub.com/sites/default/files/upm-binaries/12946_Manuscript_Tips.pdf (accessed 10.11.2015). Corner, J. (2000) Recovering Landscape: Essays in Contemporary Landscape Theory, New York: Princeton Architectural Press. Cushing, D.F. and Renata, A. (2015) ‘Themes in landscape architecture publishing: Past trends, future needs’, Landscape Journal, 34(1), 15–36. DBIS (Department for Business, Innovation & Skills) (2015) Review of Research Excellence Framework (REF): Terms of Reference. Available at: https://www.gov.uk/government/publications/research-excellenceframework-review-terms-of-reference (accessed 11.04.2016). Deming, M.E. and Swaffield, S. (2011) Landscape Architectural Research: Inquiry, Strategy, Design, Hoboken, NJ: Wiley. Downing, A.J. (1841) A Treatise on the Theory and Practice of Landscape Gardening, Adapted to North America, New York & London: Wiley and Putnam; Boston: C.C. Little & Co. Dramstad, W., Olson, J.D. and Forman, R.T.T. (1994) Landscape Ecology Principles in Landscape Architecture and Land-Use Planning, Washington DC: Island Press. Dupont, A. and Jacobs, H., eds. (2008) Landscape Ecology Research Trends, New York: Nova Science Publishers. Dyck, R.G. (1994) ‘Discipline vs profession: A comment on the meaning of design’, Journal of Planning Education and Research, 13(2), 143–146. Ertiö, T.-P. (2015) ‘Participatory apps for urban planning: Space for improvement’, Planning Practice & Research, 30(3), 303–321. Francis, M. (2001) ‘A case study method for landscape architecture’, Landscape Journal, 20, 15–29. Fry, G.L.A. (2001) ‘Multifunctional landscapes: Towards transdisciplinary research’, Landscape and Urban Planning, 57(3–4), 159–168. Gailing, L. and Leibenath, M. (2015) ‘The social construction of landscapes: Two theoretical lenses and their empirical applications’, Landscape Research, 40(2), 123–138. Gibbons, M., Limoges, C., Nowotny, H., Schwartzman, S., Scott, P. and Trow, M. (1994) The New Production of Knowledge: The Dynamics of Science and Research in Contemporary Societies, London: Sage. Gilgun, J.F. (2005) ‘The four cornerstones of evidence-based practice in social work’, Research on Social Work Practice, 15(1), 52–61. Grierson, E. and Brearly, L., eds. (2009) Creative Arts Research: Narratives of Methodologies and Practices, Rotterdam/Boston/Taipei: Sense Publishers. Gobster, P.H. (2011) ‘Editorial: Landscape and urban planning at 100: Looking back moving forward’, Landscape and Urban Planning, 100, 315–317. Harzing, A.W. and van der Wal, R. (2008) ‘Google Scholar as a new source for citation analysis?’, Ethics in Science and Environmental Politics, 8(1), 61–73. Hedblom, M., Heyman, E., Antonsson, H. and Gunnarsson, B. (2014) ‘Bird song diversity influences young people’s appreciation of urban landscapes’, Urban Forestry & Urban Greening, 13(3), 469–474. Herrington, S. (2013) ‘An ontology of landscape design’, in Howard, P., Thompson, I. and Waterton, E., eds. The Routledge Companion to Landscape Studies, Abingdon: Routledge, 355–365.

82 …

The challenge of publication „

Howard, P.J. (2011) An Introduction to Landscape, Farnham: Ashgate. Howard, P.J., Thompson, I.H. and Waterton, E., eds. (2013) The Routledge Companion to Landscape Studies, Abingdon: Routledge. Jones, P., Layard, A., Speed, C. and Lorne, C. (2015) ‘MapLocal: Use of smartphones for crowdsourced planning’, Planning Practice & Research, 30(3), 322–336. Jorgensen, A. (2014) ‘Looking backwards, looking forwards’, Landscape Research, 39(1), 1–6. Kamiya, G. (2007) ‘Let us now praise editors’, Salon. Available at: http://www.salon.com/2007/07/24/editing/ (accessed 12.11.2015). Kelly, A. and Burrows, R. (2012) ‘Measuring the value of sociology? Some notes on the performative metricisation of the contemporary academy’, Sociological Review, 59(s2), 130–150. Lenzholzer, S., Duchhart, I. and Koh, J. (2013) ‘“Research through designing” in landscape architecture’, Landscape and Urban Planning, 113, 120–127. LI (Landscape Institute) (2012) The Landscape Institute Code of Standards of Conduct and Practice for Landscape Professionals. Available at: http://landscapeinstitute.org/PDF/Contribute/CodeofConduct-May2012. pdf (accessed 15.04.2016). Macpherson, H. (2006) ‘Landscape’s ocular-centrism: and beyond?’, in Tress, B., Tress, G., Fry, G. and Opdam., P., eds. From Landscape Research to Landscape Planning: Aspects of Integration, Education and Application, Dordrecht: Springer, 95–106. Manning, O. (1995) ‘Landscape revisited: A note on the methodology of criticism’, Landscape Research, 20(2), 77–86. Meijering, J.V., Tobi, H., van den Brink, A., Morris, F. and Bruns, D. (2015) ‘Exploring research priorities in landscape architecture: An international Delphi study’, Landscape and Urban Planning, 137, 85–94. Meyer, E. (2002 [1997]) ‘The expanded field of landscape architecture’, in Swaffield, S., ed. Theory in Landscape Architecture: A Reader, Philadelphia: University of Pennsylvania Press, 167–170. Milburn, L.-A.S. and Brown, R. (2003) ‘The relationship between research and design in landscape architecture’, Landscape and Urban Planning, 64, 47–66. Milburn, L.-A.S. and Brown, R. (2016) ‘Research productivity and utilization in landscape architecture’, Landscape and Urban Planning, 147, 71–77. Musard, O., Le Dû-Blayo, L., Francour, P., Beurier, J.-P., Feunteun, E. and Talassinos, L., eds. (2014) Underwater Seascapes: From Geographical to Ecological Perspectives, Dordrecht: Springer. Natural England (2012) Evidence in Natural England: Our evidence strategy. Available at: http://publications. naturalengland.org.uk/file/2550314 (accessed 07.04.2016). NCDDR (National Center for the Dissemination of Disability Research) (2005) What Are the Standards for Quality Research? Technical Brief No. 9. Available at: http://ktdrr.org/ktlibrary/articles_pubs/ncddrwork/focus/ focus9/Focus9.pdf (accessed 10.11.2015). Placebook Scotland (2016) Sharing Your Places and Landscapes. Available at: http://www.placebookscotland. co.uk/ (accessed 07.04.2016). Potschin, M.B., Klug, H. and Haines-Young, R.H. (2010) ‘From vision to action: Framing the Leitbild concept in the context of landscape planning’, Futures, 42, 656–667. REF2014 (2014) Research Excellence Framework Reports. Available at: http://www.ref.ac.uk (accessed 30.06.2016). Roe, M.H. (2009) ‘Editorial: The times they are a-changin’’, Landscape Research, 34(1), 1–6. Roe, M.H. (2012) ‘Crossing the boundaries?’, in Bell, S., Sarlöv Herlin, I. and Stiles, R., eds. Exploring the Boundaries of Landscape Architecture, Abingdon: Routledge, 299–315. Roe, M.H. (2015) Landscape Strategies and Strategic Thinking in England. Presentation to the National Landscape Forum, Dublin 25 June 2015. Selman, P. (1998) ‘Landscape design as research: An emerging debate’, Landscape Research, 23(2), 195–196. Selman, P. (2012) Sustainable Landscape Planning: The Reconnection Agenda, Abingdon: Routledge. Sheppard, S.R.J. (2015) ‘Making climate change visible: A critical role for landscape professionals’, Landscape and Urban Planning, 142, 95–105. Steiner, F. (2014) ‘Frontiers in urban ecological design and planning research’, Landscape and Urban Planning, 125, 304–311.

83 …

Maggie Roe

„

Strunk, W. and White, E.B. (1979) The Elements of Style: How to Speak and Write Correctly, 3rd edn, New York/ London: Macmillan. Swaffield, S., ed. (2002) Theory in Landscape Architecture: A Reader, Philadelphia, PA: University of Pennsylvania Press. Swaffield, S. (2013) ‘Sustainability practices in New Zealand agricultural landscapes under an open market policy regime’, Landscape Research, 39(2), 190–204. Thompson I. (2012) ‘Ten tenets and six questions for landscape urbanism’, Landscape Research, 37(1), 7–26. Thompson, I.H. (2000) Ecology, Community, Delight: Sources of Value in Landscape Architecture, London: Spon Press. Thwaites, K. (1998) ‘Landscape design is research: An exploration’, Landscape Research, 23(2), 196–198. UKEOF (UK Environmental Observation Framework) (2016) Citizen Science. Available at: http://www.ukeof.org. uk/our-work/citizen-science (accessed 07.04.2016). van den Brink, A. and Bruns, D. (2014) ‘Strategies for enhancing landscape architecture research’, Landscape Research, 39(1), 7–20. Waldheim, C. (2006) The Landscape Urbanism Reader, New Haven, CT: Princeton Architectural Press. Woudstra, J. (2010) ‘The “Sheffield Method” and the first department of landscape architecture in Great Britain’, Journal of Garden History, 38(2), 242–268. Wylie, J. (2007) Landscape, London: Routledge. Yang, B., Lia, S. and Binder, C. (2016) ‘A research frontier in landscape architecture: Landscape performance and assessment of social benefits’, Landscape Research, 41(3), 314–329.

84 …

Chapter 6: Assessing research priorities and quality Jurian Meijering, Hilde Tobi, Adri van den Brink and Diedrich Bruns

INTRODUCTION In its commitment to serve people and society the field of landscape architecture has generated substantial knowledge through practice and research, especially in the last twenty or so years. The field has become very diverse in the topics it addresses, the areas and regions it develops, and the scales it covers. Moreover, the type of contributions that landscape architects make to planning, design and management processes are changing and today landscape architects fulfil a number of different roles. These changes to the substance of landscape architecture and the broadening of the role of landscape architects in society call for the generation of new knowledge. As a consequence, the scope of landscape architecture research includes a wide array of different research domains, many of which considerably overlap with each other as well as with the domains of a number of ’neighbouring disciplines’ such as architecture, landscape ecology, geography and environmental psychology. This calls for a discourse on specifying and prioritizing research that would, without sacrificing the field’s diversity, contribute to the promotion of landscape architecture as a discipline with a specific identity, defined by its own body of knowledge (van den Brink and Bruns 2014). To do so, landscape architecture must strive to focus and sharpen its research agenda, and to address the grand landscape challenges of our time successfully, it must continue to develop its body of knowledge. The purpose of this chapter is to give an indication of how to focus the landscape architecture research agenda and how to enhance the reporting of the resulting research. These insights may also help to set the stage for a discourse about the nature of landscape architecture research. They may also serve as a kind of baseline for the remaining chapters of this book. So, in this chapter, we report on and synthesize the results of two recent studies. The first is a Delphi study with an international sample of landscape architecture experts from academia and practice. The study aimed to answer two research questions s s

Which research domains do these experts prioritize as the most important for landscape architecture as a scientific discipline? Which research domains do these experts prioritize as the most useful for landscape architecture practice?

The second study is a concise systematic review of a sample of papers published in 2013 in a number of international peer-reviewed journals which publish landscape architecture research

85 …

Jurian Meijering, Hilde Tobi, Adri van den Brink and Diedrich Bruns

„

and/or research relevant to landscape architecture. This systematic review aimed to answer the following questions: s s s

What percentage of papers reported empirical studies and their findings? What is the quality of empirical papers in terms of methodological transparency? To what extent are the most important and useful research domains, as identified in the Delphi study, represented in the empirical papers?

This chapter is organized as follows: the next section presents and discusses the Delphi study. A description of the Delphi method as a research method is given, and the methods and results of the study are reported. The systematic review is then addressed. This section begins with a description of the systematic review as a research method, followed by a report on the methods and the results of the study. Several recommendations for improving the methodological quality of empirical papers are also provided. By synthesizing and discussing the results of the Delphi study and the systematic review, the final section sets the stage for a wider discourse on landscape architecture research.

MAPPING THE TERRAIN: RESEARCH PRIORITIES FOR LANDSCAPE ARCHITECTURE1 Understanding the Delphi method as a research method The Delphi method was developed in the 1950s (Dalkey and Helmer 1963) and has, since the 1960s, been applied in the context of many different disciplines such as health care and education as a means of allowing a group of experts to achieve agreement on various aspects of a particular topic. The Delphi method has several defining characteristics (see among others Hung et al. 2008; Linstone and Turoff 1975). Participation of experts on the topic of interest is required. It is generally recommended to select experts based on specific criteria (Keeney et al. 2006) and to assemble a heterogeneous sample so as to reflect a wide range of views (Powell 2003). A Delphi study consists of at least two rounds. In the first round experts are questioned about their opinion on the topic of interest, usually by means of a standardized questionnaire. In this questionnaire experts express their opinion by rating a number of items that were presented to them by the study team (see below). In some instances the experts are also asked to provide a rationale for their ratings. To prevent undue influence from dominant people and group pressure, commonly present in everyday meetings, the experts remain anonymous throughout the course of the study and are not expected to communicate with one another. Instead, the study manager (i.e. the researcher) provides experts with so called ‘controlled opinion feedback’ based on the answers given in the previous round. In most Delphi studies this feedback includes summary statistics showing majority opinions (Boulkedid et al. 2011). Feedback of the rationales indicating why experts hold certain opinions is also strongly advocated (Bolger and Wright 2011). Starting from the second round, the experts receive feedback and are presented with a more-or-less adapted second version of the questionnaire. Based on the feedback the experts are allowed to reconsider and possibly change their ratings. This procedure continues until a desired level of agreement has been achieved or until a certain stability in experts’ responses has been reached. Figure 6.1 illustrates the general procedure of a Delphi study.

86 …

Assessing research priorities and quality „

Figure 6.1 The general procedure of a Delphi study

Methodology of the landscape architecture Delphi study We conducted the Delphi study to achieve agreement among international landscape architecture experts from academia and professional practice on two issues: (1) the most important research domains for landscape architecture as a scientific discipline, and (2) the most useful research domains for landscape architecture practice. The methodology of this study involved the sampling of landscape architecture experts, the development of a list of research domains in landscape architecture, the design of the Delphi questionnaire and the analysis of the collected data. We assembled a sample of landscape architecture academics and professionals from around the world. Names and contact details of landscape architects were acquired from various sources, including websites of (inter)national landscape architecture organizations affiliated with the International Federation of Landscape Architects (IFLA) and conference proceedings of the European Council of Landscape Architecture Schools (ECLAS) and the Council of Educators in Landscape Architecture (CELA). Several criteria were used to select the landscape architects to include as ‘experts’ in the sample. We decided only to include academics who hold a position at an academic institution and who were known to be actively engaged in research, having published at least one landscape architecture research paper in an international peer-reviewed journal. We also decided only to include professionals engaged in professional practice (i.e. with a position in a private company or public institution involved in landscape architecture) and, to ensure the inclusion of high quality professionals, who were also either jurors or winners of competitions administered or promoted by IFLA. Eventually, 279 landscape architecture academics and professionals from Africa, Asia, Australia, Europe, North America and South America were selected and invited to enrol in the Delphi study. We then compiled a list of 12 landscape architecture research domains to be presented to the experts who took part in the Delphi study. The main sources of these domains were the list of research areas presented by Deming and Swaffield (2011, p.25), inventories of contents of scholarly journals (i.e. Gobster et al. 2010) and the authors’ own knowledge of the field. Participating experts were asked to give their opinion on this selection of research domains and also invited to suggest additional ones. Four groups of research domains were defined: ‘academic perspectives’, ‘subjects of study’, ‘competencies of the landscape architect’ and ‘other areas of knowledge and expertise’. The first group, academic perspectives, included the domains of historical, human and biophysical dimensions of planning and design. Examples of corresponding research topics include landscape architecture history, cultural heritage preservation, landscape perception, place attachment, green infrastructure and agro-ecology. The second group, subjects of study, included the domains of rural and natural environments, built environment and infrastructure, and aquatic environments. Examples of corresponding research topics include agriculture, nature

87 …

Jurian Meijering, Hilde Tobi, Adri van den Brink and Diedrich Bruns

„

reserves, public squares, motorways, coastal management and urban water fronts. The third group, competencies of the landscape architect, included the domains of theories, tools and technologies, and artistic creativity. Examples of corresponding research topics include design theories, geo-information technology and modelling tools. Finally, the fourth group included the following other areas of knowledge and expertise: values and ethics (e.g. landscape values, multiculturalism), global landscape issues (e.g. climate change, energy transition), and policy and governance (e.g. landscape governance, policies affecting landscapes). For round one, a questionnaire was designed in which experts were asked to rate these 12 research domains according to how they saw their importance for landscape architecture as a scientific research discipline and also according to their usefulness for landscape architecture practice. The experts were also invited to provide their rationale for some of the ratings and to make suggestions for domains that should be added to the list. For round two a questionnaire was designed that was very similar to that of the first round. The experts were asked to rate the importance and usefulness of the 12 original research domains as well as several new domains that were added following suggestions from the first round. Each of the 12 original domains was accompanied by controlled opinion feedback. This feedback consisted of a summary of the rationale that experts had provided, sometimes complemented by summary statistics. The questionnaire that was used for the third and final round had a different set-up. The experts were first given the opportunity to receive feedback before selecting their three most important and three most useful research domains. Because of the many different nationalities included in the study, all three questionnaires were written in English and programmed as web surveys. Data were collected from April to June 2013. Data collected in rounds one and two were analysed to develop the summaries that were fed back to the experts in rounds two and three respectively. Per research domain, the median importance and usefulness were calculated, as well as the percentage of ‘very important’ and ‘very useful’ ratings. Experts’ rationales regarding the importance and usefulness of each domain were analysed by means of content analysis. Based on the data that were collected in round three the most important and useful domains were identified. For each domain the percentage of experts that selected it as most important and as most useful was calculated. The strict agreement index (Meijering et al. 2013) was used to estimate the level of agreement among experts.

Results of the Delphi study Of the 279 landscape architecture experts who were invited to participate in the Delphi study, 86 completed the first round. Of these, 55 and 46 also completed the second and third rounds respectively. More experts from academia than from professional practice participated as well as more experts from Europe than from other continents. Nonetheless, the final round included experts from academia and professional practice as well as experts from all six continents. After the first and second rounds of the Delphi study, experts rated the research domains ‘human dimensions of planning and design’, ‘green urban development’, ‘built environments and infrastructure’ and ‘global landscape issues’ most often as ‘very important’ for landscape architecture research. The domains ‘human dimensions of planning and design’ and ‘built environments and infrastructure’ were also rated most often as ‘very useful’ for landscape architecture practice. Table 6.1 shows that consistent results were found in the third and final round of the Delphi study.

88 …

Assessing research priorities and quality „

Table 6.1 Percentage of experts that selected each research domain as ‘most important’ or ‘most useful’ in round 3 of the Delphi study

Experts gave various reasons for why they selected the domains as most important or most useful. With regard to the domain ‘human dimensions of planning and design’, experts explained that landscape planning and design is done for people and that research into this domain is needed to understand how people perceive and respond to landscapes. With regard to the domain ‘built environments and infrastructure’, experts explained that the world is urbanizing and that research is needed to improve the sustainability of cities and to provide practice with knowledge, ideas and solutions necessary to meet future challenges. While the domains ‘human dimensions of planning and design’ and ‘built environments and infrastructure’ were most frequently selected as most important and most useful, even these two domains were not selected by a majority of experts. This indicates that there was considerable disagreement among the experts about which domains are the most important and most useful. The experts seemed to agree, however, that the domains ‘landscape architecture education’, ‘tools and technologies’, ‘aquatic environments’ and ‘policy and governance’ are not the most important for landscape architecture research. Likewise, they also seemed to agree that research into ‘values and ethics’, ‘theories’ and ‘historic dimensions of planning and design’ would probably not be the most useful for landscape architecture practice. Some substantial differences regarding the importance and usefulness of domains were identified between experts from academia and professional practice as well as between experts from different parts of the world. Most notably, in the final round of the Delphi study, experts from Europe selected the domain ‘human dimensions of planning and design’ more often as most useful for practice (70%) than experts from outside Europe (26%).

89 …

Jurian Meijering, Hilde Tobi, Adri van den Brink and Diedrich Bruns

„

THE SYSTEMATIC REVIEW OF LANDSCAPE ARCHITECTURE PAPERS Understanding the systematic review as a research method Fink (2009, p.3) defines a systematic review as a ‘systematic, explicit, and reproducible method for identifying, evaluating, and synthesizing the existing body of work produced by researchers, scholars, and practitioners’. This means that a systematic review should be conducted and reported in such a way that another researcher can repeat the work and obtain similar results. Every systematic review starts with one or more explicit research questions, followed by the sampling of research units from which data can be collected. In a systematic review the research units are publications in the form of papers, books chapters and reports. Publications are usually acquired from one or more bibliographic databases. Popular online databases for scientific papers are Scopus and the Web of Science. Based on the research question(s) a search query is developed and entered into the selected database(s) to acquire appropriate publications. Usually these publications are screened by the researcher who, based on certain criteria, excludes those that are considered irrelevant or which have a poor methodological quality. The remaining publications are read in detail. Data are extracted from the text of the publications by one or more researchers using a standardized form or coding scheme to identify and code (i.e. mark) relevant text fragments. The coded fragments are then evaluated and synthesized descriptively.

The methodology of the landscape architecture systematic review We conducted a systematic review in order to explore how many landscape architecture papers (i.e. papers that report landscape architecture research but are not necessarily written by landscape architects), published in international peer-reviewed journals, report empirical studies and their findings. We also evaluated the quality of the empirical papers in terms of their methodological transparency. This means that we read the empirical papers to find out to what extent they report various aspects of the methods they used, such as the study design, the sampling of research units and the data analysis. Finally, we determined to what extent the most important and useful research domains, as identified in the Delphi study, were represented in the sample. The methodology of this study involved the sampling of landscape architecture papers, the extraction of data from those papers by means of a coding scheme and the analysis of the data. The sampling of papers followed a stepwise and iterative procedure. The first step involved selecting the appropriate international peer-reviewed journals from which papers would be sampled. To make sure that the selection included journals that at least partly focus on landscape architecture, the selection was limited to journals with the word ‘landscape’ in their title (as indexed in Scopus, the world’s largest database of scientific literature; see http://www.scopus.com/). This resulted in the selection of 12 journals: Journal of Environmental Engineering and Landscape Management, Journal of Landscape Architecture, Journal of Landscape Ecology, Landscape and Ecological Engineering, Landscape and Urban Planning, Landscape Ecology, Landscape History, Landscape Journal, Landscape Online, Landscape Research, Living Reviews in Landscape Research, Studies in the History of Gardens & Designed Landscapes. The next step involved the sampling of papers. In 2013, the last fully completed year at the time this study was conducted, a total of 591 papers were published in the 12 selected journals. To make sure that the sample only included papers reporting relevant landscape architecture research, a search query was developed and applied in Scopus (see Table 6.2). This search query

90 …

Assessing research priorities and quality „

Table 6.2 Search query applied in SCOPUSa

contained terms about important study objects (e.g. landscapes) and activities (e.g. designing) in landscape architecture that were searched for in the title, abstract and keywords of the papers. Search terms were derived from definitions of landscape architecture according to the American Society of Landscape Architects (ASLA), the International Federation of Landscape Architects, the European Council of Landscape Architecture Schools, the Landscape Institute and the Oxford English Dictionary. The search query also defined the publication year (2013) and the type of publications to exclude from the search (such as conference reviews, editorials and business articles). After the application of the search query, 379 papers remained to be considered further in the study. Next, we randomly selected, within each of the 12 journals, 25 per cent of the remaining papers, with a minimum of five and a maximum of 30 papers per journal. Within the journals Landscape History and Landscape Online fewer than five papers matched the search query and hence all of these papers were selected. The final sample comprised 99 papers. The abstracts of these papers were read by two of the authors of this chapter (van den Brink and Bruns) after which they decided to remove a further 36 papers from the sample because they did not report landscape architecture research. Five more papers were also removed from the sample as they were not written in English. The remaining 58 papers were then checked to find out if they reported empirical research. Papers were defined as empirical if they described a study in which one of the following types of data were collected and analysed: primary data (i.e. data collected within the researcher’s own study), secondary data (i.e. data collected in a previous study) or simulated data (i.e. data generated by the researcher by means of computer simulations). As 16 papers did not report empirical research we were left with 42 empirical papers. Figure 6.2 gives an overview of the entire sampling process. The final sample of 42 papers was read in full. Data were extracted from the papers by means of a coding scheme which comprised several steps, as explained over the page (also described in

91 …

Jurian Meijering, Hilde Tobi, Adri van den Brink and Diedrich Bruns

„

Figure 6.2 Overview of the systematic review sampling process

the Introduction to this book and further elaborated by Tobi and van den Brink in Chapter 2). For each step several codes were created and applied to relevant data (i.e. text fragments) within the papers. For example, a code ‘research question’ was created to code one or more explicitly reported research questions within each paper. If, within a paper, a particular code could not be applied, this meant that no relevant text fragments about that aspect of the research could be found. For the step formulation of the research question codes were created (e.g. ‘research question’ and ‘hypothesis’) to code the research questions as well as any explicitly reported research objectives and hypotheses. For the step study design a code (‘study design’) was created to code text fragments that explicitly stated the applied study design. Furthermore, codes were created (e.g. ‘level of control researcher’) to code text fragments about the researchers level of control (i.e. concerning a possible intervention or manipulation and the random assignment of research units to groups), the number of

92 …

Assessing research priorities and quality „

data collection waves, and the timeframe of the study (i.e. retrospective, present, prospective). For the step sampling and data collection separate codes were created (e.g. ‘landscapes: sampling method used’ and ‘other research units: sampling method used’) to code text fragments about the sampling of landscapes and about the sampling of other research units such as humans and animals. This was done because in landscape architecture research the research units are usually landscapes. Codes were also created (e.g. ‘sampling frame’) to code text fragments about the applied sampling frame (i.e. the list of research units from which a random sample was drawn) and selection scheme (i.e. the criteria or logic on the basis of which research units were included in a non-random sample). Additionally, codes were created (e.g. ‘data collection methods used’) to code text fragments about the applied data collection methods and the year(s) in which data were collected. For the step data analysis separate codes were created (‘description quantitative data analysis’ and ‘description qualitative data analysis’) to code text fragments about the quantitative and qualitative analysis of data. Finally, the overall structure of the paper was scanned to determine whether the information was properly organized in the following sections: introduction, methods, results, discussion. Sections about the applied theory or conceptual framework were disregarded as they were not important for the purpose of the current study. Within each paper the coded text fragments were examined and evaluated to gain an insight into their methodological transparency. For example, text fragments with the code ‘data collection methods used’ were examined to find out exactly which methods were applied (e.g. standardized questionnaire, interview) and text fragments with the code ‘description quantitative data analysis’ were evaluated to find out whether the reporting of the quantitative data analysis was complete or incomplete. Finally, two of the authors of this chapter (van den Brink and Bruns) read the abstracts of the papers and independently decided for each paper to what extent it was related to any of the research domains arising from the Delphi study. Differences of opinion were discussed and resolved.

Results of the systematic review Seventy-two per cent of the sampled papers described the collection of primary, secondary or simulated data and were thus considered to be empirical. Table 6.3 shows that most (83%) of these papers explicitly reported a research objective that clearly stated which knowledge was to be acquired. However, research questions and hypotheses were far less commonly found: only in 12 per cent and 14 per cent of the papers respectively. In some papers (12%) no research question, objective or hypothesis could be found at all. Table 6.4 shows that the research objectives and questions reported in most (76%) empirical papers were focused on acquiring knowledge on a current issue, while far fewer objectives and questions were focused on investigating an issue of the past or in the future (in 2% and 14% of the papers respectively). As stated earlier, a number of papers (12%) did not state a research question, objective or hypothesis and thus their timeframe could not be established. In other papers (7%) the research objectives were unclear regarding the timeframe, for example when a research objective focused on the measurement of landscape change but did not indicate whether it concerned only change that occurred in the past or change that was tracked into the future. Furthermore, the timeframe as suggested by the research objective or question did not always match the timeframe as suggested by the applied study design. For example, an objective could be focused on a present timeframe, while the study design as described in the methods section made clear that the research had a retrospective timeframe.

93 …

Jurian Meijering, Hilde Tobi, Adri van den Brink and Diedrich Bruns

„

Table 6.3 Percentage of empirical papers reporting a research objective, question or hypothesis

Table 6.4 Percentage of empirical papers reporting a research objective or question focused on a retrospective, prospective or present timeframe

Table 6.5 Percentage of empirical papers explicitly reporting the use of a case study, experiment or other study design

Table 6.5 shows that a small majority (55%) of the papers did not explicitly report their study design. If a study design was explicitly reported, it was mostly a case study (33%). Only two papers reported an experiment as the applied study design. These were also the only two papers in which the level of control of the researchers, regarding the introduction of a manipulation or intervention and the assignment of research units across groups, could actually be regarded as truly experimental

94 …

Assessing research priorities and quality „

Table 6.6 Percentage of empirical papers reporting the use of a random or non-random sampling method

(i.e. the studies included a manipulation or intervention and the researchers randomly assigned research units to a control and experimental group). In all other papers the researchers had neither experimental nor quasi-experimental control. Furthermore, in the great majority of papers (88%, not presented in the table) only one data collection wave could be identified. This demonstrates a lack of longitudinal research in which the change of a phenomenon across time is studied. In all but two of the papers, landscapes (e.g. natural parks, cities, gardens) were selected as research units. Some papers contained a nested sampling design in which, for example, several cities were purposively selected as study cases and a large number of smaller research units (e.g. neighbourhoods) were randomly selected within these cases. For each step in such a nested sampling design the sampling method should normally be reported. This includes the selection of one or more cases as explained in more detail by Swaffield in Chapter 7. Table 6.6 shows that two thirds of the papers were wholly or partly unclear about the method(s) applied for selecting landscapes. In many of these papers one or more cases or study areas were described but it remained unclear how or why these particular ones were selected. Although it is highly probable that a non-random (e.g. purposive) sampling method was applied, no indication of this could be found anywhere. Almost half of the papers clearly reported the use of a non-random sampling method for the selection of landscapes, while only a few clearly reported the use of a random sampling method (e.g. the random selection of landscape grids from a map). Roughly half of the papers that clearly reported a random or nonrandom sampling method did not report a corresponding sampling frame or selection scheme. Table 6.6 shows that a minority (10%) of papers reported the selection of research units other than landscapes, such as people or animals. In all these papers the sampling method applied was clearly described. The use of non-random sampling methods (in 31% of the papers) seemed to be more common than the use of random sampling methods. Remarkably, as with the sampling of landscapes, a description of the corresponding sampling frame or selection scheme was often missing. Table 6.7 shows that the retrieval of secondary data and observation were most often reported as data collection methods (45% and 38% respectively). In this regard it is worth noting that in 36 per cent

95 …

Jurian Meijering, Hilde Tobi, Adri van den Brink and Diedrich Bruns

„

Table 6.7 Percentage of empirical papers reporting the use of various data collection methods

of the papers (not shown in the table) landscape imagery was retrieved and processed for data analysis. In some papers the imagery consisted of actual hardcopy or digital images (e.g. maps, satellite images, photographs), whether or not equipped with some form of data, for example concerning land use or land cover types, the height of the landscape, or temperature. In other papers the imagery took the form of a spatial data set which could be converted into a map using geographic information system (GIS) software. Depending on how the imagery was processed, it was either considered to be secondary data or an observation. In this systematic review (digitized) images and spatial data sets were considered to be secondary data when they were automatically processed for the purpose of (further) data analysis. In some papers, however, images or spatial data sets (converted into images) were processed manually based on visual inspection by human actors, for example when researchers divided a satellite image of a city into grids and decided for each grid which type of land use is most prevalent. In these papers images or spatial data sets were used in place of field observations during which researchers actually visit a place to observe the landscape (or the actors/artefacts within the landscape). Therefore, the manual processing of images or spatial data sets based on visual inspection by human actors was considered to be an observation. Generally, the papers that described the retrieval of landscape imagery did not clearly describe how it was processed. Nevertheless, for most of these papers, some information could be found indicating the automatic or manual processing of the imagery and thus a decision whether to consider it as secondary data or as an observation could be made. Other data collection methods which were quite frequently reported within the papers are data simulation (29%) and the standardized questionnaire (21%). Methods that are mainly focused on the collection of qualitative data, such as the interview and the focus group, were less common. Remarkably, in almost one quarter of the papers it was partly unclear how data were collected. Furthermore, only 36 per cent of the papers (not shown in the table) clearly and completely stated the year(s) in which data were collected. In particular, often no information could be found stating how old the retrieved secondary data were and in which year an observation of imagery was performed (based on visual inspection by human actors). Table 6.8 shows that almost all papers (90%) reported research in which quantitative data were analysed. Most of these (66%) also offered a complete description of how this analysis was done by

96 …

Assessing research priorities and quality „

Table 6.8 Percentage of empirical papers reporting the quantitative and qualitative analysis of data

clearly describing all the calculations and statistical analyses which were performed to arrive at the reported results. Nevertheless, in a number of papers the description was either incomplete (21%) or even wholly missing (13%). A minority of papers (29%) reported research in which qualitative data were (also) analysed. However, in most of these papers (58%) a description of the qualitative analysis was wholly missing. Only two papers provided a complete description by clearly describing the type of qualitative analysis performed (e.g. content analysis), the coding procedure, and the strategy by which main themes or constructs were identified. Finally, a few papers (10%) were at least partly unclear about what kinds of data were collected and how these data were analysed. The majority of the empirical papers (62%) had a basic scientific structure in the sense that at least the following four sections were clearly distinguished: introduction, methods, results and discussion. A paper was considered to have an incomplete scientific structure when one or more of these four sections was missing, when large amounts of information that ought to be reported in one of the four sections was reported in one or more additional sections, or when the results and discussion sections were combined. Combining the results and discussion sections is undesirable as a clear distinction between the actual results of the study and the interpretation of those results by the researcher is lost. Table 6.9 shows that the papers were spread across a wide range of the research domains identified from the Delphi study. A considerable proportion were classified as fitting the domains ‘built environments and infrastructure’ (45%) and ‘human dimensions of planning and design’ (33%), both of which were identified in the Delphi study as most important for landscape architecture research and most useful for landscape architecture practice. However, a number of papers were (also) classified as fitting the domains ‘biophysical dimensions of planning and design’ (45%), ‘development of applied methods and techniques’ (43%), and ‘tools and technologies’ (36%), a domain that experts considered to be less important for landscape architecture research. Remarkably, few papers fitted the domains ‘green urban development’ (17%) and ‘global landscape issues’ (2%), both of which were identified in the Delphi study as important for landscape architecture research.

97 …

Jurian Meijering, Hilde Tobi, Adri van den Brink and Diedrich Bruns

„

Table 6.9 Percentage of empirical papers related to a particular research domain

Recommendations Based on the results of the systematic review we have several recommendations for researchers who report their empirical research in the field of landscape architecture. First of all, the introduction of a paper should explicitly state a clear research objective, question and/or hypothesis. Without a research objective or question it is impossible for readers to judge whether an appropriate study design was used and whether the results properly filled the knowledge gap identified at the outset of the research. Simply put: a paper without a research objective or question does not contribute anything to the landscape architecture body of knowledge. If landscapes were sampled as research units (e.g. as case studies), a paper should also report how they were selected. In most of the empirical papers that we analysed, landscapes such as parks, cities and gardens were selected as research units. Often, however, it was not entirely clear whether these landscapes were selected by means of a random or non-random sampling method. This is important, because the applied sampling method determines the extent to which the results of a study may be generalized to other landscapes. With the use of a random sampling method results may potentially be (statistically) generalized to all other research units listed in the corresponding sampling frame. With the use of a non-random sampling method, generalization of results is always doubtful and should at best be limited to research units that fit the criteria or logic of the applied selection scheme. It is therefore important that the reporting of the sampling method is accompanied by a clear description of the corresponding sampling frame or selection scheme, both of which were missing in many empirical papers. If a nested sampling design is used, researchers should report the sampling method for each step, including the initial (and usually non-random)

98 …

Assessing research priorities and quality „

selection of one or more cases or study areas. After all, as explained by Swaffield in Chapter 7, it matters quite a lot whether the results of a study are based on a paradigmatic or extreme case. If imagery was used as a data source, a paper should describe how it was processed. A number of papers reported the use of imagery but it was often not entirely clear whether this imagery was processed automatically by some kind of GIS software or processed manually based on visual inspection by human actors. If imagery was processed manually this needs to be clearly reported, for example by describing how many human actors were involved and which criteria were used in the categorization of image polygons. In this way readers are able to judge the quality of the observational data. Furthermore, authors should not forget to report the year in which the imagery was obtained and manually processed. A paper should offer a complete description of how the data collected were analysed. In a substantial number of papers the description of the quantitative and especially the qualitative data analysis was incomplete or even entirely missing. Without a complete description of how the data were analysed readers cannot judge the appropriateness of the data analysis techniques and thus the validity of the subsequent results. Quite often papers did not follow a basic scientific reporting structure. Every empirical research paper should consist of at least four sections: introduction, methods, results, discussion. By organizing a research paper in such a way, readers are easily able to find the information they are looking for, regardless of the journal in which the paper was published. A final important result of the systematic review concerns the study designs used: in most papers researchers were unable to exercise experimental control and only one data collection wave could be identified. The results of the systematic review thereby seem to confirm the suggestion made by Tobi and van den Brink in Chapter 2 that landscape architecture research tends to be more about exploring and describing phenomena and less about explaining why certain phenomena occur and predicting their future occurrence. Both explanatory and predictive research may require the application of a longitudinal study design, in which a phenomenon is monitored over time by collecting data in several waves, or an experimental study design, in which researchers test a specific hypothesis about the future occurrence of a phenomenon by introducing an intervention in a controlled environment. To develop further the landscape architecture body of knowledge it is important that more explanatory and predictive research is conducted.

SYNTHESIS AND DISCUSSION In this chapter a summary of a Delphi study among landscape architecture experts was presented as well as a fuller report on a systematic review of landscape architecture journal papers. The results of the studies should help to (1) focus landscape architecture research agendas towards the issues which are important, and (2) enhance the reporting of research and thereby improve the way in which research is approached and conducted – with improved rigour, clear objectives, well-defined methods and improved analysis. By synthesizing results of both studies insight is gained that should also help to guide discourses about the future of landscape architecture research. The results of the Delphi study suggest that the research domains ‘human dimensions of planning and design’, ‘built environments and infrastructure’, ‘green urban development’ and ‘global landscape issues’ were perceived as being the most important for landscape architecture as a field of academic research. Research into the domains ‘human dimensions of planning and

99 …

Jurian Meijering, Hilde Tobi, Adri van den Brink and Diedrich Bruns

„

design’ and ‘built environments and infrastructure’ was also seen as most useful for landscape architecture practice. Hence, both of these domains in particular may be put at the top of landscape architecture research agendas. These results confirm what we have observed, for example at international conferences, namely that social and human studies in relation to urban environments and global issues have become central to landscape architecture research. The relevance of research into ‘human dimensions of planning and design’ also corresponds with studies in which comparable domains were identified as important within the academic journals, Landscape and Urban Planning (Gobster 2014) and Landscape Journal (Powers and Walker 2009). The Delphi study also showed that some research domains were clearly not regarded as most important for landscape architecture research (landscape architecture education, tools and technologies, aquatic environments, policy and governance) or most useful for landscape architecture practice (values and ethics, theories, historic dimensions of planning and design). This does not mean, however, that research into these domains is irrelevant. It merely indicates that these domains should perhaps not form the core of landscape architecture research agendas. The research domains as used in the Delphi study are rather broad and may seem to be overlapping with one another. ‘Built environments and infrastructure’ may, for example, also include parts of ‘green urban development’, while ‘global landscape issues’ may encompass several parts of other separate domains. The list of research domains was inspired by existing ones; in fact, we discovered that in the field of landscape architecture several different lists exist that contain many different research domains (see among others: Deming and Swaffield 2011, p.25; Weber 2015, p.84). Some of them are rather narrowly defined, while others are broader. This abundance of different research domains represents, to a large degree, the lack of clarity that has developed over the years within the field of landscape architecture. This lack of clarity may be related, at least partly, to the interdisciplinary character of landscape architecture, both as an academic discipline and as professional practice. Other assumptions may also be reasonable: for example, the large number of research domains may hint at – already existing or envisaged – opportunities for specialization (cf. Weber 2015). Furthermore, the fact that none of the domains was selected by a majority of experts may hint at several ‘schools of thought’ and/or regionally different priorities. Obviously, within an applied discipline such as landscape architecture, such prioritizations are context-dependent, and, as a consequence, different issues have emerged as important in different parts of the world. Therefore, to continue developing landscape architecture as a discipline that relies on and builds its own body of knowledge, it would be advisable to become more specific about the domains that, together, constitute its intellectual core. In this light, the results of the Delphi study mark a first step on a longer path of specifying research needs within landscape architecture. A necessary next step would be, for example, to further delineate the research domains and also to clarify the knowledge gaps that need bridging. Taking such steps would provide a basis for a continuously focused discourse on research agendas. The results of the systematic review of journal papers show that a considerable portion of the sample addresses the two research domains that experts considered important and useful (i.e. human dimensions of planning and design, built environments and infrastructure). More work may also be needed to collect evidence on the domains ‘global landscape issues’ and ‘green urban development’. Both were identified, in the Delphi study, as important for landscape architecture research, while the systematic review showed that only a few papers addressed these domains. The systematic review also clearly shows that, generally speaking, future empirical research papers

100 …

Assessing research priorities and quality „

ought to be more transparent about the methodology used. Most papers did not describe one or more crucial aspects of research design. Methodological clarity is a prerequisite for all research so that anyone is able to understand how the research was conducted, is able to replicate it and can judge the validity of the results. Therefore, the lack of methodological transparency in the sample must be of concern for the pursuit of further developing the body of knowledge on which landscape architecture may rely as an evidence-based discipline. Results clearly indicate that researchers in or related to landscape architecture need to pay more attention to research methodology. In summary, for landscape architecture to continue developing its body of knowledge, researchers need to conduct more empirical research that is based on observed and measured phenomena. Researchers also ought to report their findings in such a way that anyone can understand how the research was done and, therefore, is able to assess the methodological rigour that was applied. A reliable body of knowledge is one upon which future research agendas may be developed proactively and, eventually, without reliance on a few experts’ opinions. Where evidence has been established, knowledge gaps may be identified more safely. Review papers might be written that would support the field discussing how research agendas might be established regionally or around specific themes. Monitoring research activity might also reveal how legacies are evolving and how specific sub-disciplines are establishing.

NOTE 1 This section is largely a summary of an empirical study that has been published in the journal Landscape and Urban Planning. For more details on the methodology and results of the study we kindly refer the reader to Meijering et al. (2015).

SUGGESTED FURTHER READING The following may be consulted to obtain a general understanding of the Delphi method: Keeney, S., Hasson, F. and McKenna, H. (2006) ‘Consulting the oracle: Ten lessons from using the Delphi technique in nursing research’, Journal of Advanced Nursing, 53(2), 205–212. Linstone, H.A. and Turoff, M., eds. (1975) The Delphi Method: Techniques and Applications, Boston, MA: Addison-Wesley. Powell, C. (2003) ‘The Delphi technique: Myths and realities’, Journal of Advanced Nursing, 41(4), 376–382.

The following offer insight into several specific methodological issues of the Delphi method, such as the measurement of agreement and the provision of controlled opinion feedback: Bolger, F., Stranieri, A., Wright, G. and Yearwood, J. (2011) ‘Does the Delphi process lead to increased accuracy in group-based judgmental forecasts or does it simply induce consensus amongst judgmental forecasters?’, Technological Forecasting and Social Change, 78(9), 1671–1680. Meijering, J.V., Kampen, J.K. and Tobi, H. (2013) ‘Quantifying the development of agreement among experts in Delphi studies’, Technological Forecasting and Social Change, 80(8), 1607–1614. Rowe, G., Wright, G. and McColl, A. (2005) ‘Judgment change during Delphi-like procedures: The role of majority influence, expertise, and confidence’, Technological Forecasting and Social Change, 72(4), 377–399.

Some well-known and critical reviews of the Delphi method are the following: Sackman, H. (1974) Delphi Assessment: Expert Opinion, Forecasting, and Group Process, No. RAND-R-1283-PR, Santa Monica: Rand Corporation. Woudenberg, F. (1991) ‘An evaluation of Delphi’, Technological Forecasting and Social Change, 40(2), 131–150.

101 …

Jurian Meijering, Hilde Tobi, Adri van den Brink and Diedrich Bruns

„

The following may be consulted to learn more about how to conduct a systematic review: Gough, D., Oliver, S. and Thomas, J., eds. (2012) An Introduction to Systematic Reviews, London: Sage. Petticrew, M. and Roberts, H. (2006) Systematic Reviews in the Social Sciences: A Practical Guide, Oxford: Blackwell.

Online training videos and guides on how to conduct a systematic review can be found here: The Campbell Collaboration Research Center: http://www.campbellcollaboration.org/resources/training/The_ Introductory_Methods.php The Georgia State University: http://research.library.gsu.edu/systematicreview

REFERENCES Bolger, F. and Wright, G. (2011) ‘Improving the Delphi process: Lessons from social psychological research’, Technological Forecasting and Social Change, 78(9), 1500–1513. Boulkedid, R., Abdoul, H., Loustau, M., Sibony, O. and Alberti, C. (2011) ‘Using and reporting the Delphi method for selecting healthcare quality indicators: A systematic review’, PLoS ONE, 6(6), e20476, available: doi: 10.1371/journal.pone.0020476. Dalkey, N. and Helmer, O. (1963) ‘An experimental application of the Delphi method to the use of experts’, Management Science, 9(3), 458–467. Deming, E. and Swaffield, S. (2011) Landscape Architectural Research: Inquiry, Strategy, Design, Hoboken, NJ: Wiley. Fink, A. (2009) Conducting Research Literature Reviews: From the Internet to Paper, 3rd ed., Thousand Oaks, CA: Sage. Gobster, P.H. (2014) ‘(Text) Mining the LANDscape: Themes and trends over 40 years of landscape and urban planning’, Landscape and Urban Planning, 126, 21–30. Gobster, P.H., Nassauer, J.I. and Nadenicek, D.J. (2010) ‘Landscape Journal and scholarship in landscape architecture: The next 25 years’, Landscape Journal, 29(1), 52–70. Hung, H.L., Altschuld, J.W. and Lee, Y.F. (2008) ‘Methodological and conceptual issues confronting a crosscountry Delphi study of educational program evaluation‘, Evaluation and Program Planning, 31(2), 191–198. Keeney, S., Hasson, F. and McKenna, H. (2006) ‘Consulting the oracle: Ten lessons from using the Delphi technique in nursing research’, Journal of Advanced Nursing, 53(2), 205–212. Linstone, H.A. and Turoff, M., eds. (1975) The Delphi Method: Techniques and Applications, Boston, MA: Addison-Wesley. Meijering, J.V., Kampen, J.K. and Tobi, H. (2013) ‘Quantifying the development of agreement among experts in Delphi studies’, Technological Forecasting and Social Change, 80(8), 1607–1614. Meijering, J.V., Tobi, H., van den Brink, A., Morris, F. and Bruns, D. (2015) ‘Exploring research priorities in landscape architecture: An international Delphi study’, Landscape and Urban Planning, 137, 85–94. Powell, C. (2003) ‘The Delphi technique: Myths and realities’, Journal of Advanced Nursing, 41(4), 376–382. Powers, M.N. and Walker, J.B. (2009) ‘Twenty-five years of Landscape Journal: An analysis of authorship and article content’, Landscape Journal, 28(1), 96–110. van den Brink, A. and Bruns, D. (2014) ‘Strategies for enhancing landscape architecture research’, Landscape Research, 39(1), 7–20. Weber, G. (2015) Auxiliary Specialization Opportunities in Landscape Architecture: Nature of Profession, Current View, Allied Relationships, Skills & Knowledge, and Future Directions (MSc), Kansas State University, available: https://krex.k-state.edu/dspace/handle/2097/19044.

102 …

PART III SELECTED APPROACHES AND METHODS

This page intentionally left blank

Chapter 7: Case studies Simon Swaffield

INTRODUCTION Case studies are widely used in landscape architectural research. They are ideally suited to the investigation of complex phenomena such as designed landscapes, which are the focus of the discipline. However, the way landscape architecture researchers typically select and undertake case studies is also a limitation to achieving greater practice, policy and scientific relevance for the outcomes of our research. The objective of this chapter is to explain the methodological rationale for case studies and the implications of this for the selection and treatment of cases in landscape architecture. The central question that I ask is how can researchers in landscape architecture use case studies in order to enhance our discipline’s knowledge base? In particular, I focus upon the use of cases as research tools, rather than educational or professional exemplars. I aim to demonstrate that a more theoretically informed selection of case studies could sharpen the way landscape architecture researchers shape and answer their research questions, strengthen their contribution to the theory of the discipline, and hence improve the quality of their design and planning practice and contribution to wider debates over policy. These are all fundamental to ensuring the future relevance of landscape architecture. The chapter is based upon several analyses of published research in landscape architecture. The first was a review of articles from leading landscape architecture journals undertaken during the preparation of a textbook on research strategies (Deming and Swaffield 2011); the second was a review of articles published over the past 40 years of the journal Landscape Research (Vicenzotti et al. 2016); and the third has been a summary analysis of the use of cases in articles in the Journal of Landscape Architecture 2006–2014. The chapter has also drawn upon a number of research projects in which I have been involved, including student dissertations at PhD level. This chapter is focused on case study research as applied to landscape architecture. Other forms of case-based research, such as case series and (nested) case-control studies, are not discussed in this book because they are mainly used in medicine and public health studies and uncommon in landscape architecture research. First, the chapter describes the current status of case study research in landscape architecture and identifies potential shortfalls in its approach. Second, it reviews the concept of a case study. Third, it examines in more detail the theoretical and methodological rationale for case studies, with particular focus upon case study selection and how that task can be best undertaken to achieve research objectives. Finally it identifies directions for improving the way that case study research practice in landscape architecture can contribute to the theoretical development

105 …

Simon Swaffield

„

of the discipline. Examples are used to illustrate the key points. The focus of the chapter is the role of case studies in landscape architectural research, rather than in landscape research more generally. In drawing upon the various sources noted above I have therefore placed most attention upon research examples that generate knowledge upon the way landscape architecture practice transforms landscapes.

CASE STUDIES IN LANDSCAPE ARCHITECTURE Case studies are emerging as a popular mode of investigation in the discipline in response to a range of commercial, public policy and institutional pressures for enhanced practice and research performance. In North America for example, the Landscape Education Foundation has a website and publication series that is focused upon case studies of exemplary design practice (https:// lafoundation.org/research/landscape-performance-series/). In Europe, the European Union (EU) funded LE:NOTRE programme raised awareness of case study process through workshops and seminars, and prepared case studies which are now held by the LE:NOTRE Institute (http://www. le-notre.org/). Case studies also feature widely in academic landscape architecture publications. For example, over the period 2011–2014, cases were cited in 78 per cent of published peer reviewed articles in the journal Landscape Research. This figure contrasts with only 11 per cent of articles in a similar period twenty years ago (1991–1995). The Journal of Landscape Architecture (JoLA) includes a section in each issue called ‘Under the Sky’ dedicated to case studies, and case studies featured in 32 per cent of all peer reviewed articles in JoLA from 2006 to 2014. Case studies also appear prominently in conference presentations by both landscape architecture professionals and academics. On the face of it therefore case study research is well established in landscape architecture. The critical question is whether it is being used to best effect. For the most part, landscape architecture researchers take a pragmatic approach to case studies. Francis (1999) established a template for case study reporting in landscape architecture which provides a guide to the data that could and should be collected, and how it might be presented. He defined case studies as ‘a well-documented and systematic examination of the process, decision making and outcomes of a project that is undertaken for the purpose of informing future practice, policy, theory and/or education’ (Francis 1999, p.9). In a subsequent article Francis used an exemplar of integrated and sustainable community design – the Village Homes development in Davis, California – to show how a case study method might be applied (Francis 2001). The approach is primarily descriptive, and this is reflected in the projects that have been recorded and lodged on the Landscape Architecture Foundation website (http://landscapeperformance.org/case-study-briefs). This collection of case studies establishes a record of projects undertaken by the profession in North America, which is useful for both practitioners and students, who also contribute many of the cases. The availability of a template means there is the basis for some measure of consistency in reporting, so that basic comparisons can be made. The crucial decision in using case studies for research rather than professional education, however, is the rationale for the selection of the cases. Most professionally reported cases are opportunist – they are those which are to hand and that appear interesting – and are frequently chosen for their value in promoting the designer, a practice or the wider profession. The methodological problem with opportunist studies is that even if a template is followed, it is a slow process accumulating knowledge that can be generalised, as the cases reported are hard to compare in any meaningful

106 …

Case studies „

way (Ostrom 2007). In particular, case studies that seek to add to the theoretical knowledge of the discipline need to provide insight into relationships that have relevance beyond the individual events or situations being investigated. Most research using multiple cases recognises this need for a framework of comparison. However, the same logic applies when interpreting the significance of individual case studies – they need to be able to be compared with other cases in a systematic way that enables something to be learnt from similarities and differences seen in different contexts. Cases selected by opportunity alone may offer some comparative insights, but more often than not the value of the comparison is limited. Analysis of articles reporting landscape architecture research that include case studies reveals that even in a peer reviewed publication the rationale for case study selection typically receives little or sometimes no explanation (see Meijering et al. – Chapter 6). In the sample of JoLA case studies noted above, the rationale for case selection was mentioned or could reasonably be inferred in 79 per cent (26/33) of the articles using cases, but the explanation seldom warranted more than a sentence. Half of these cases were exemplars chosen to illustrate a proposition or argument, rather than to investigate a relationship. In short, case study applications in landscape architecture frequently lack sufficient rationale for the knowledge claims they make, and hence the validity of their findings frequently remains undetermined. This is a significant limitation upon the progress of the discipline both in developing its own systematic knowledge and in contributing to wider debates in science and policy. In the next section I therefore consider the nature of case study research from a theoretical and methodological perspective.

THE NATURE OF CASE STUDY RESEARCH A case study may be broadly defined as a study of a specific event, situation or complex phenomenon investigated in their real world context (Yin 2014). Frequently, multiple data or information sources are used for analysis, and often the analysis spans a period of time. In many research projects only one case study is undertaken, although multiple comparative cases are increasingly investigated. The focus upon a single or small number of situations or ‘cases’ examined in all their complexity and in their wider context contrasts with some other types of scientific investigation, such as largescale surveys, that seek understanding of the world by narrowing down attention upon a limited number of relationships between defined variables, and that use statistical techniques to analyse large numbers of events chosen from a population of cases. When seen like this, case study research is not in opposition to other types of research. Rather, it expresses a different research design chosen to best suit the purpose of the investigation (Flyvbjerg 2006). Gerring (2004) emphasises that case study research is ‘a particular way of defining cases, not a way of analyzing cases’ (Gerring 2004, p.341; see also Gerring 2007). Case study research is thus a general approach to research, and not a set of specific methods. A key consideration in undertaking case study research is to understand how the concept of a case study relates to the decisions on methods of sampling, data collection methods and methods for data analysis that shape every research project. In the next sections I will focus upon four main steps – understanding how the findings may be generalised, how to select appropriate cases, choosing a framework for comparative analysis, and matching the case(s) to the question being asked. These sections will help you to decide whether or not you want to adopt a case study approach for answering your research question. Please note that once you have decided upon a case study approach, many other

107 …

Simon Swaffield

„

decisions on research methods remain to be made, such as how you sample informants, collect data and analyse data (see also Tobi and van den Brink – Chapter 2). Obviously, this chapter cannot cover all of these decisions, but instead focuses on the selection of the case or cases.

GENERALISATION OF CASE STUDY RESEARCH All research makes some underlying methodological assumptions about how knowledge can be formalised, justified and generalised. In deductive scientific methodology, theory is built by developing hypotheses. These are conjectures about causal relationships in the world based upon the best current knowledge which are then tested empirically in some way, usually using statistical analyses. Case study research on the other hand seldom includes sufficient numbers of cases to enable any form of cross-case statistical analysis. Researchers therefore use a different logic to determine the significance of the findings, known as analytical generalisation (Yin 2014). This uses existing theory as a template against which to compare the results of a new study. Hence theory develops inductively, as evidence accumulates. The comparison of evidence with theory may usefully be guided by a working proposition – that is, a provisional statement about the relationships being investigated. However, there is an important choice to make in how the comparisons are made. Landscape architecture researchers often seek to build theory through positive proof – providing cases as exemplars of a proposed theory – usually indicated in the explanation by the phrase ‘I will demonstrate (my) theory with an example.’ The problem with this approach is that positive proof is biased to a tendency to confirm the researchers’ preconceived ideas as we all tend to favour that which is familiar (Kahneman 2011). In contrast, the philosopher Karl Popper (1959) showed that the most robust logical test of a theory is falsification – that is, trying to disprove a hypothesis or proposition, rather than trying to demonstrate its truth. Hence all good research designs, whether they involve an experiment, a large-scale survey or a case study, contain a critical challenge to the starting hypothesis or proposition. It is typically assumed that case-based research is unsuited to falsification. However, Flyvbjerg (2006) points out that cases can be used to test theory by falsification if a case is selected that is least likely to be consistent with a particular proposition. This may be because it is an extreme case, or because it has been critically chosen for this purpose. If the case still confirms the proposition under investigation, then the theory has been tested through falsification and found to be robust – until the next test. Case selection aimed at falsification rather than positive exemplars has the potential to significantly strengthen the power of landscape architecture theory in a way that has wider scientific credibility. However, it does require a reorientation in the way cases are selected.

CHOOSING THE CASE(S) Cases are normally described as purposive samples, which means they have been individually selected because their characteristics enable conclusions and hence generalisations to be drawn based upon their type, rather than upon their statistical occurrence. Flyvbjerg (2006) identifies several types of purposive case: paradigmatic, extreme and critical. Paradigmatic cases are exemplars that have prototypical or metaphorical value. They are selected because the researcher believes they represent a situation that has general relevance – for example it may be a new approach to a difficult problem in landscape planning. Hence when Spicer et al. (2014) investigated how management of non-point

108 …

Case studies „

source pollution from animals and fertiliser influences the way an agricultural landscape changes, they selected a prototypical situation in which a new way of managing pollution is being trialled. The case investigated was the Lake Taupo project in New Zealand which is reportedly the first attempt globally to manage nutrient discharge using a ‘cap and trade’ system for a defined water catchment. This means that the regulatory authority set a maximum for total discharge of nitrates in the catchment. Landowners were then each allocated a permit (a right) to discharge a specified level of nitrates based upon historical patterns of use, and also given the power to buy or sell these rights to discharge to other landowners in the basin. The idea was that this would allow land uses to change to become more efficient while remaining within the overall environmental capacity of the catchment, and it would also encourage landowners to innovate to reduce discharge. In addition, a trust was established to purchase discharge rights from landowners, and hence bring down the overall nutrient loading in the catchment. Landscape biography (Roymans et al. 2009) provided a way to trace what has happened since the new regime was introduced. By selecting this prototypical case the conclusions will be relevant to the design of environmental management regimes in other landscapes using a cap and trade approach. Extreme cases on the other hand are selected because they test an idea to its limits. The rationale is that they are particularly effective in testing hypotheses or falsifying propositions, as the outcome is likely to be reasonably clear-cut. Hoversten (2013) investigated decision making in alternative futures planning, which involves developing scenarios of change and then projecting the possible landscape outcomes. He identified a series of decision points (or discursive moments as he called them) which appeared to be particularly important in shaping the way that the different futures were chosen. However, it was possible that the discursive moments vary depending upon how the decision making responsibilities are allocated. In particular the investigator wanted to better understand the influence of experts in shaping the decisions. He therefore chose two contrasting cases – one in which outside experts took a leading and directive role in shaping the futures, and another which involved a highly collaborative and bottom up approach. He found that in both extreme cases there were a series of similar discursive moments. This gave him confidence that the moments he identified were features that are likely to be found in alternative futures planning irrespective of the way the process was implemented. Another example of an extreme case is provided in Yu et al. (2008). Their research was investigating how rapidly urbanising city regions could draw upon adaptive strategies from the past to manage flood risk. The research focused upon the Yellow River Basin in China as ‘probably the most difficult river to regulate in the world’ due to its high sediment load and flood regime. Lessons that could be identified in such an extreme case would have relevance for other extreme situations facing enhanced flood risk due to climate change. Critical cases are chosen because they have features that are central to the theoretical purpose of the investigation. Zonneveld (2011) asked the question of how small towns in New Zealand could manage the tensions between the globalising influence of tourism and community desires to retain a sense of local identity. He developed a theoretical model of how these different influences might be expressed within a ‘tourism circuit’ (Figure 7.1). He then selected case studies that expressed the different types of global/local interface within his model – transit towns, gateway towns and destination towns – and used these cases to examine the way in which small towns are already changing and to undertake design experiments about how the tensions might be managed. The use of a typology enabled selection of several contrasting cases that expressed the critical relationships

109 …

Simon Swaffield

„

Figure 7.1 Theoretical cases – different types of small town on a tourism circuit (source: after Zonneveld 2011)

he was investigating in different contexts, and by placing them within a broader theoretical model he was able to draw more generally relevant conclusions. Another example of a critical case is provided by Bolleter (2009) in an examination of landscape architecture practice in Dubai. The focus of the investigation is the emerging typology of designed landscape in this rapidly developing city state, which is widely regarded as a possible model for urbanism across significant parts of Asia. The research addresses a question with wider ramifications, which is what role should a professional charter have in shaping the actions of practitioners? The particular case is critical because of the apparent discrepancy between the types of design emerging from contemporary landscape practice and the ideals of the International Federation of Landscape Architects, as expressed in their charter. In a theoretically informed discipline, such critical cases should arguably be the most dominant form of case study.

COMPARING CASES These examples of purposive cases all share a common feature, which is that they each allow a particular type of relationship to be investigated and compared. Gerring (2004) argues that the best case study design is that which provides the most powerful understanding of variation – why one situation differs from or is similar to another. Case study comparisons can be undertaken at two levels (Figure 7.2). Within-case comparisons are typically used when there is a single complex case being studied, and may be undertaken in several ways: by applying multiple analyses to a complex case; by adopting multiple theoretical perspectives across the case; or by undertaking several smaller case studies embedded within a larger complex case. Cross-case comparison is undertaken when the research design seeks to answer its questions by comparing different cases. I will now examine these in turn. Multiple analyses are a well established approach both within and beyond landscape architecture case studies, and involve using different data collection methods such as documentary research,

110 …

Case studies „

Figure 7.2 Within-case and cross-case comparisons

participant observation and interviews all focused upon the ‘same’ case. The advantage of using multiple data collection methods is that each method is best suited to different types of data, and so applying multiple methods can generate a greater diversity of information and insight than use of a single method (Creswell 2014). However, the combination of different types of data requires care. Researchers who seek to maximise the objectivity of their work may use multiple methods as a form of triangulation to achieve greater precision and certainty. Hence identification of a relationship between variables in a situation that is revealed with one method would be regarded as being more reliable if the same relationship is also revealed through other methods. However, researchers who regard research findings as dependent in part upon the researcher’s involvement as an active agent may see multiple analyses as providing a richer but not necessarily a more precise outcome, as each technique may reveal a different aspect of reality – depending on the assumptions. Hence the value in multiple methods may not be in their combination (as in triangulation) but rather in the scope they offer to raise questions of why there are differences between the findings from different perspectives. This opens the possibility of multiple perspective analysis. Multiple perspective comparisons are research approaches which apply more than one conceptual or theoretical frame to a complex case. The key feature and advantage of multiple perspective comparisons is that the different conceptual perspectives or lenses may reveal a different interpretation of what is happening. Comparison between the interpretations can have significant explanatory power by allowing the researcher to ask which of these perspectives provides the most compelling explanation, and how do the different explanations interrelate? A classic textbook example of using different perspectives was Allison’s political science investigation of the Cuban Missile Crisis (1969), where he interpreted the decisions that were taken during the military and political confrontation between the USA and USSR in the 1960s over the location of nuclear missiles in Cuba. Allison applied three analytical perspectives in an attempt to understand the complexities of the situation (Figure 7.3). Multiple perspective analyses of single complex cases are not common in landscape architecture. One example is Abbott’s (2008) investigation of ways to understand wilderness through design. The case he used was Fiordland – an extensive wilderness area in south-west New Zealand, mostly

111 …

Simon Swaffield

„

Figure 7.3 Multiple perspective analyses of a single complex case – two examples (source: after Allison 1969 and Abbott 2008)

National Park and recognised as a United Nations Educational, Scientific and Cultural Organization (UNESCO) World Heritage Area. Abbott undertook a series of investigations using different design based techniques – including mapping, waymarking, equipment design – all intended to reveal and compare different ways in which design can construct and deconstruct wilderness (Figure 7.3). A third and more common approach to multiple analyses in landscape architecture case study research is to undertake several smaller embedded case studies within a single overall study, for example to investigate geographically different parts of a complex landscape. In the prototypical case study of Lake Taupo described above, the responses by landowners appeared to vary according to which part of the catchment they were located within. It seemed that the social legacy and land use history was a significant factor. Spicer therefore undertook three micro-studies within the overall Lake Taupo case, focused upon different parts of the catchment, embedded within the overall case, from which she was able to compare and hence distil a more nuanced understanding of the overall trends and patterns (Spicer et al. 2014) (Figure 7.4). Evidence from multiple cases is more compelling than single cases, as it allows more comprehensive and robust comparisons (George and Bennett 2005; Gerring 2007; Yin 2014). Multiple cases chosen purposively help us to adopt ‘an outside view’ (Kahneman 2011). A recently published example of multiple cases is Lenzholzer (2008), which compared eleven Dutch urban squares. Lenzholzer was investigating the implications of a dominant design approach in postWorld War II design for urban microclimate, and selected eleven squares frequently mentioned in the media for their lack of appeal. These cases were all chosen for their shared characteristics, in a search for regular patterns. Multiple cases are more often chosen to explore different dimensions. Gerring (2007) identifies temporal and spatial variation as two fundamental descriptive dimensions, which are also frequently used in landscape architecture for selection of multiple cases. Butenschön and Säumel (2011), for example, selected five parks in Berlin for their study of the relationship between cultural and ecological processes in planting choice and patterns. The parks are from two different time periods (19th century and 20th century) and two different parts of Berlin (Former East and West Berlin).

112 …

Case studies „

Figure 7.4 Embedded geographical cases (source: Spicer et al. 2014)

In another example, Sinha (2010) selected five types of park in Lucknow, India, to investigate the shifting ideologies and changing aesthetics of memorial parks. She emphasised variation in time – with two types that were developed in a colonial period, and three in a post-colonial period. However, while space and time are useful points of reference, they do not in themselves provide much information about the functional characteristics of a situation. Steenbergen (2008) offers an alternative framework for stratified sampling focused on design investigations, emphasising the importance of context and the nature of the design object. George and Bennett (2005) recommend use of typologies – the classification of phenomena into types – as a way to conceptualise comparisons in a theoretically informed way, and typology can be useful when designing both within-case and cross-case comparisons. Petrow (2011), for example, investigated the way landscape architecture can express the self-conception and image of a city, and selected three cases of redevelopment in Germany based upon a common design type – the waterfront. Hence the design type was constant, and the variation was in the self-conception. In an

113 …

Simon Swaffield

„

Figure 7.5 Cases based on types (source: after Tavares et al. 2013)

investigation of cultural adaption to microclimate in Christchurch, New Zealand, Tavares et al. (2013) selected four cases according to the type of urban setting, which was defined as a combination of two characteristics: place dynamics (established and emerging) and spatial configuration (active social space and passive retreat space) (Figure 7.5). In each of the examples above, the selection of cases was the critical factor that determined the value of the findings.

MATCHING THE CASE(S) TO THE RESEARCH QUESTION – AND HOW UNDERSTANDING CASE STUDIES CAN HELP REFINE YOUR RESEARCH QUESTION How can these insights into the nature of case study methodology help in developing a research proposal? As the preceding discussion illustrates, case studies express a particular way of shaping an investigation. The choice of case determines what data is collected, influences how findings can be analysed and reported, and most significantly from a research perspective, determines how the findings may be related to other cases and how they can contribute to the wider theory of the discipline. At the same time, choosing to undertake a case-based investigation helps to shape and sharpen your research question(s). In this section I therefore examine the relationship between case study methodology and the research question being asked. There are several types of research question in landscape architecture for which a case study approach is particularly well suited. These include: 1) questions that relate to particular landscape settings; 2) questions of an exploratory nature; 3) questions that seek in-depth understanding about particular types of situation; and 4) questions that test or challenge theoretical claims. Landscape practice is (or has been hitherto) almost always focused on specific landscapes – it deals with design and planning interventions and actions in particular locations, by particular people, at particular times – both contemporary and historical. These are by definition unique cases. As Francis (2001) has shown, following a case study protocol provides a powerful organisational framework for investigating questions focused upon such individual cases. Given the nature of our discipline, it could therefore be argued that a case study approach can be seen as a default methodological position for landscape architectural research, in the absence of another more compelling approach. This is, however, a starting point, not a conclusion, as it raises a further question for the researcher. As Gerring (2007) points out, once a case is defined, it implicitly raises the question, ‘what is this a case of?’ This in turn provides the potential for comparison with other similar types of case, and hence the beginnings of theoretical generalisation. The second type of research question for which case studies are well suited are questions of an exploratory nature, that seek to discover patterns and potential relationships in a little understood

114 …

Case studies „

aspect of the discipline. The reason that cases are a preferred approach is that they enable the scope, content and analytical approach to evolve during the investigation, in response to emerging patterns and insights. Unlike many other research designs, the validity of case study findings are not compromised by shifts in the methods or types of data collected during the course of the study, and data collected at any stage can contribute to the shaping of an overall interpretation of the situation. Hence if your initial question is open rather than narrowly focused, a case study can provide a robust design. Zonneveld’s (2011) investigation of small town responses to global tourism addressed an exploratory question. However, just as with a question that starts from a particular landscape, an exploratory case will be more theoretically fruitful if the research question specifies the type of case being investigated. In Zonneveld’s example, the use of a conceptual model of a tourism circuit enabled the identification of different types of small town setting, which could be compared. Hence exploratory research is not a-theoretical. To be most conceptually useful, the choice of case should be purposive rather than opportunist. The third type of relevant question is well recognised in case study literature; it highlights that case studies are particularly well suited to investigations that add in-depth understanding to an emerging body of theory. The particular contribution of the case study approach in this situation is the way it enables multiple analyses of a situation, or multiple perspective analysis, or multiple embedded cases, all giving both greater richness to the investigation, and enabling within-case comparisons. For example, Spicer et al.’s (2014) investigation of environmental cap and trade at Lake Taupo was able to drill down into the complexity of the landscape setting and its social and cultural legacy using multiple embedded case studies. As with exploratory cases, however, the choice of case is critical in determining the value of an in-depth investigation. Devoting extensive resources and time to a study that cannot be compared or generalised is an extravagant use of the limited resources of the discipline. And as with exploratory cases, the decision to adopt a case-based design for in-depth analysis reflects attention back to the nature of the initial research questions, encouraging the researcher to sharpen their focus and clarify their rationale. The fourth category of research questions well suited to case study are those that aim to test or challenge theoretical claims or assumptions about complex situations which cannot easily be investigated using experimental or quasi-experimental designs. Hoversten’s (2013) investigation of discursive moments in alternative futures was this type of ‘testing’ question, as he was challenging the assumption that the choices made in alternative futures planning are technical rather than political. The topic is complex, and there are only limited instances that can be investigated. His question was framed as a proposition and the research design involved selection of contrasting expert and community-led cases to test the proposition in two extreme situations. In each of these examples, it is evident that selecting a case study design to match a research question is a reflexive process – that is, while the first step may be to say, ‘would a case study design be a good way to address my question, and if so what type of case study?’, there is inevitably a second phase which asks ‘have I framed my question so that a case-based investigation will generate useful new knowledge?’ Hence the type and choice of case once again comes to the foreground in developing the research project. A second common feature of all these types of question is highlighted by Gerring (2007) who makes the point that case studies are better suited in general to investigation of situations where the causal relationships are strong rather than weak – in other words, where it is possible to draw conclusions based upon an holistic ‘whole of case’ view.

115 …

Simon Swaffield

„

DISCUSSION AND CONCLUSION The chapter objective has been to explain the methodological rationale for case-based investigation and the implications of this for selection of cases used in landscape architectural research. My intention has been to demonstrate to researchers in landscape architecture how they might be more effective in the way they incorporate case studies into their research strategies. I have briefly reviewed the concept of a case study and described the current status of case-based research in landscape architecture. Analysis of published research indicates that case studies are emerging as a popular mode of research, but that landscape architecture researchers take a largely pragmatic approach to case studies, and frequently provide insufficient justification for the knowledge claims they make. This highlights the need to pay increased attention to the selection of cases. The research design is the framework that determines what data to collect, why and how. In case study research, the choice of case or cases is the critical part of the design and is the key to creating generalisable knowledge of strategic design value. Several ways to select cases have been reviewed – paradigmatic (prototypical), extreme and critical – and these purposive categories each provide a different theoretical basis for interpreting the findings. However, many published landscape architecture cases are exemplars, chosen to demonstrate how a theoretical approach may be applied, rather than to challenge a theoretical proposition, and as such are biased to a tendency to confirm the researchers’ preconceived ideas. Shifting the emphasis in case study selection away from attempts to demonstrate the veracity of theoretical statements towards research designs that use purposive cases to challenge theoretical propositions has potential to significantly strengthen the way the discipline develops, and to enhance its credibility in the wider science and policy community. The best case study design is that which provides the most powerful understanding of variation, and this requires comparisons. Within-case comparisons are typically used when there is a single complex case being studied, and may involve multiple analyses, multiple theoretical perspectives, or multiple embedded case studies within a larger complex case. Many cross-case comparisons use geographical or temporal variation. Given the focus of landscape architecture upon planned and designed landscapes, typology is a particularly powerful way to frame multiple cases in such a way that theoretical propositions about how to make effective design choices can be systematically tested and challenged. The identification of the type of case under investigation is also central to matching a case study design to the research question, and in reviewing four types of research question well suited to a case study approach I have highlighted the way that developing an appropriate design reflects questions back upon how the research question itself is framed. This reminds us that case study design for research purposes is not a routine matter but an integral part of the creative and critical process of developing a productive and insightful research project. In addition to emphasising the importance of case selection, the review has identified several opportunities to enhance the use of case study research in landscape architecture: first, there is unrealised opportunity for systematic comparative review of the already accumulative case-based knowledge in landscape architecture: that is, to use established cases to provide evidence for the long term outcomes of landscape architecture inspired actions. The growing record of case studies in repositories such as the Landscape Architecture Foundation, and LE:NOTRE Institute, and the cumulative record of peer reviewed cases in the main journals of the discipline offer a major resource for further comparative investigation. These cases offer a way for landscape architecture researchers to build understanding of what types of design decision have produced the best outcomes and to identify

116 …

Case studies „

effective planning and design strategies. The major limitation in seeking to draw new knowledge from already completed case studies is their variable assumptions and opportunist coverage (Ostrom 2007). However, the process may at least identify critical questions for new case investigations. A second significant opportunity for the discipline is to undertake new case studies – with cases chosen purposively – in order to identify the types of design that have proven to be effective in steering change down sustainable pathways in different contexts and different policy regimes. Working in an applied discipline – political science – with useful lessons for landscape architecture, George and Bennett (2005) suggest paying particular analytical attention to the aspects of a situation which offer significant choices to decision makers. Choices in a complex situation and their relationship to outcomes can be identified through a method of process tracing – tracking the decisions made and their consequences for the project trajectory. A major challenge in analysing the process biography of completed projects through process tracing will be to backfill knowledge on critical decisions that are not in the public record. This will require interviews with key decision makers, and careful documentary investigation. Fortunately, historical research is one of the strongest areas of scholarship in landscape architecture. One of the most widely recognised but least acknowledged truths about expertise is that we learn most from our mistakes – yet researchers seldom systematically investigate landscape architecture projects that failed, and the reasons for failure. One notable exception is Kirkwood (1999) who used analysis of failures in landscape construction as a powerful tool to improve detailed design. Just as process tracing could help identify effective strategies, so it could also identify typical mistakes, and thus help focus on new research questions, as well as informing education and practice. A particularly strong argument for decision makers is the counterfactual. What might have happened, or might yet happen, if no decision is taken, or if a different course of action is taken? Frequently the projection of the ‘do nothing’ option provides the strongest argument for action, by highlighting the wider public policy goals that might be compromised by inaction. One of the major strengths of the alternative futures planning approach (Shearer 2005) is the ability to model and project outcomes depending upon a range of assumptions, and hence show what difference might be made by different types of design and planning intervention. If alternative futures projects were framed as cases with wider theoretical significance they could contribute directly to the knowledge base of the discipline, as well as informing particular communities. In a similar way, considering cases as theoretically informed explorations of possibilities can also inform and be integrated with the growing use of research through design (Lenzholzer et al. 2013), and there is a need and opportunity for research being undertaken by and through design to explore situations selected as purposive cases that contribute directly to answering the critical questions of our discipline. A critical approach to case study selection therefore has potential to open up rich lines of casebased research involving research through design, cases as possibilities, learning from mistakes, and identifying the most effective design strategies. There is also scope to undertake systematic analysis of the growing archive of cases already published in peer reviewed journals or available through organisations such as the Landscape Education Foundation and LE:NOTRE Institute. Overall, however, the single most effective improvement that can be made in the use of case studies in landscape architecture is to shift away from seeing them as exemplars, by which a preconceived normative position can be ‘demonstrated’, towards theoretically informed selection of cases to build and test theory about important design and planning relationships.

117 …

Simon Swaffield

„

SUGGESTED FURTHER READING Creswell, J.W. (2014) Research Design: Qualitative, Quantitative and Mixed Method Applications, 4th ed., Thousand Oaks, CA: Sage. Flyvbjerg, B. (2006) ‘Five misunderstandings about case study research’, Qualitative Inquiry, 12(2), 219–245. Francis, M. (2001) ‘A case study method for landscape architecture’, Landscape Journal, 20(1), 15–29. Ragin, C.C. and Becker, H.S., eds. (1992) What is a Case? Exploring the Foundations of Social Inquiry, New York: Cambridge University Press. Swanborn, P. (2010) Case Study Research: What, Why and How?, Thousand Oaks, CA: Sage. Thomas, G. (2011) How to Do Your Case Study: A Guide for Students and Researchers, Thousand Oaks, CA: Sage. Yin, R.K. (2014) Case Study Research: Design and Methods, 5th ed., Thousand Oaks, CA: Sage.

REFERENCES Abbott, M. (2008) Designing wilderness as a phenomenological landscape: design-directed research within the context of the New Zealand conservation estate (PhD), Lincoln University, available: http:// researcharchive.lincoln.ac.nz/handle/10182/1026?show=full (accessed 1 July 2016). Allison, G.T. (1969) ‘Conceptual models and the Cuban missile crisis’, American Political Science Review, 63(3), 689–718. Bolleter, J. (2009) ‘Para-scape: landscape architecture in Dubai’, Journal of Landscape Architecture, 4(1), 28–41. Butenschön, S. and Säumel, I. (2011) ‘Between cultural and ecological processes: Historical plant use in communal parks in Berlin, Germany’, Journal of Landscape Architecture, 6(1), 54–67. Creswell, J.W. (2014) Research Design: Qualitative, Quantitative and Mixed Method Applications, 4th ed., Thousand Oaks, CA: Sage. Deming, M.E. and Swaffield, S. (2011) Landscape Architectural Research: Inquiry, Strategy, Design, Hoboken, NJ: John Wiley. Flyvbjerg, B. (2006) ‘Five misunderstandings about case study research’, Qualitative Inquiry, 12(2), 219–245. Francis, M. (1999) A case study method for landscape architecture, final report to the Landscape Architecture Foundation, Washington, DC, available: https://lafoundation.org/myos/my-uploads/2010/08/19/ casestudymethod.pdf (accessed 15 April 2015). Francis, M. (2001) ‘A case study method for landscape architecture’, Landscape Journal, 20(1), 15–29. Gerring, J. (2004) ‘What is a case study and what is it good for?’, American Political Science Review, 98(2), 341–354. George, A.L. and Bennett, A. (2005) Case Studies and Theory Development in the Social Sciences, Cambridge, MA: The MIT Press. Gerring, J. (2007) Case Study Research: Principles and Practice, New York: Cambridge University Press. Hoversten, M.E. (2013) Decision-making, values, and discursive moments in alternative futures landscape planning (PhD), Lincoln University, available: https://researcharchive.lincoln.ac.nz/bitstream/ handle/10182/5631/Hoversten_PhD.pdf;jsessionid=799AB236614AC271B04D2FFED4106C70?sequ ence=3 (accessed 30 September 2015). Kahneman, D. (2011) Thinking: Fast and Slow, London: Penguin. Kirkwood, N. (1999) The Art of Landscape Detail: Fundamentals, Practices, and Case Studies, New York: John Wiley. Lenzholzer, S. (2008) ‘A city is not a building: Architectural concepts for public square design in Dutch urban climate contexts’, Journal of Landscape Architecture, 3(1), 44–55. Lenzholzer, S., Duchhart, I. and Koh, J. (2013) ‘“Research through designing” in landscape architecture’, Landscape and Urban Planning, 113, 120–127. Ostrom, E. (2007) ‘A diagnostic approach for going beyond panaceas’, Proceedings of the National Academy of Sciences, 104(39), 15181–15187.

118 …

Case studies „

Petrow, C.A. (2011) ‘Hidden meanings, obvious messages: Landscape architecture as a reflection of a city’s selfconception and image strategy’, Journal of Landscape Architecture, 6(1), 6–19. Popper, K.R. (1959) The Logic of Scientific Discovery, London: Hutchinson. Roymans, N., Gerritsen, F., van der Heijden, C., Bosma, K. and Kolen, J. (2009) ‘Landscape biography as a research strategy: The case of the South Netherlands Project’, Landscape Research, 34(3), 337–359. Shearer, A.W. (2005) ‘Approaching scenario-based studies: Three perceptions about the future and considerations for landscape planning’, Environment and Planning B: Planning and Design, 32(1), 67–87. Sinha, A. (2010) ‘Colonial and post-colonial memorial parks in Lucknow, India: Shifting ideologies and changing aesthetics’, Journal of Landscape Architecture, 5(2), 60–71. Spicer, A., Swaffield, S.R., Fairweather, J.R. and Moore, K. (2014) ‘Environmental regulation, decision networks and land change: A Taupo, New Zealand case’, in Global Land Project, eds. Proceedings of the Global Land Project, 2nd Open Science Meeting, Berlin, 19–21 March, 78–79. Steenbergen, C. (2008) ‘Composing Landscapes: Analysis, Typology, and Experiments for Design, Basel: Birkhäuser. Tavares, S.G., Swaffield, S.R. and Stewart, E. (2013) ‘Sustainability, microclimate and culture in post-earthquake Christchurch’, Land Environment and People Research Paper no. 19, Lincoln University, available: https://researcharchive.lincoln.ac.nz/bitstream/handle/10182/5422/LEaP_rp_19.pdf?sequence=3 (accessed 30 September 2015). Vicenzotti, V., Jorgensen, A., Qvistrom, M. and Swaffield, S. (2016) ‘Forty years of Landscape Research’, Landscape Research, doi.org/10.1080/01426397.2016.1156070. Yin, R.K. (2014) Case Study Research: Design and Methods, 5th ed., Thousand Oaks, CA: Sage. Yu, K., Lei, Z. and Dihua, L. (2008) ‘Living with water: Flood adaptive landscapes in the Yellow River Basin of China’, Journal of Landscape Architecture, 3(2), 6–17. Zonneveld, R. (2011) Lost in transitions: Staging global tourism in local small towns (PhD), Lincoln University, available: https://researcharchive.lincoln.ac.nz/handle/10182/4543?show=full (accessed 30 September 2015).

119 …

Chapter 8: Landscape biography Jan Kolen, Hans Renes and Koos Bosma

INTRODUCTION Landscape biography is an approach that is often used both in practice and in research for the purpose of analysing and exploring the history of a landscape. In this chapter we present the background, conceptual principles and sources of landscape biography as a means aimed at informing landscape design and planning which researchers have developed for understanding landscapes (Roymans et al. 2009; Bosma and Kolen 2010; Kolen et al. 2015a). Landscape biography is particularly useful in cases where, for the purpose of considering future landscape transformations, knowledge is needed about long-term landscape changes, about regional and local heritage, and about historical narratives of space and place. This chapter starts by explaining the rationale for using landscape biography in landscape research. It then provides a conceptual framework, followed by a description of principles and of methods, including sources of information, data collection techniques and so on. Examples of landscape biography application are discussed in more detail, and its potential for the study of urban landscapes is explored. Ideas about a ‘biography of landscape’ can be traced back to a variety of different sources. Some of the first ideas appeared more than three decades ago. In an inspiring essay, Samuels (1979), a geographer, introduced his ideas on ‘The biography of landscape’. In it he used the term ‘authored landscapes’ to emphasise that all landscapes bear the imprint of people’s personal and collective authorship. He argued that landscapes should not be seen as the anonymous by-products of social and economic development but that landscape research should treat each landscape as having its own specific biography. Later, around the middle of the 1990s, and initially without reference to the earlier ideas discussed in geography, the biography concept arose within the discourses of cultural anthropology and landscape archaeology (Kopytoff 1986). Starting around the turn of the twenty-first century, and with several ground-breaking projects (some of which are presented below), landscape biography developed into the interdisciplinary approach that it is today. A landmark application of landscape biography took place in 2001 when it was defined as the basic concept underpinning a large research programme of the Netherlands Organisation for Scientific Research (NWO), under the heading ‘Protection and Development of the Dutch ArchaeologicalHistorical Landscape’. This programme was interdisciplinary and, at the same time, aimed at a strong integration of landscape research, heritage management and spatial planning.

120 …

Landscape biography „

RATIONALE FOR USING LANDSCAPE BIOGRAPHY IN LANDSCAPE RESEARCH The last three decades have been an exciting period for landscape and heritage studies in many ways. First, research has generated a wealth of new insights about landscape history. Second, the history of landscape has become more ’democratised’, with popular narratives, place-bound social memories and academic interpretations of past landscapes being combined and frequently included in public debates about the values of space and place. Third, awareness is growing about the histories of landscapes and how these are best conceptualised in complex ways, expanding beyond the simplistic and monolithic reconstructions of linear chronologies that used to be common practice. Instead, landscape researchers such as landscape archaeologists, historical geographers, landscape ecologists, sociologists, cultural anthropologists and others, have begun to reflect on landscape change in transdisciplinary ways and in terms of non-linear developments where past and present (and perceptions of the future) mingle (see the chapters covering these themes in Bell et al. 2011). In addition, landscape researchers need to handle growing amounts of data generated by the ever wider number of disciplines becoming involved in studies about landscape history and, at the same time, employing an ever-growing arsenal of techniques. As landscape research and design research have become multi-, inter- and trans-disciplinary (van der Valk 2010), landscape biography has developed as an analytical, explorative and interpretative approach that works like an interface where different disciplines can meet and combine their efforts and their methods in joint studies of landscape history. New developments are also happening at the interface of landscape and heritage. Landscapes are increasingly valued as heritage while, at the same time, the concept of heritage has also evolved. Formerly denoting a collection of objects that had to be kept intact and shielded from the forces of modernisation, ’heritage’ has now developed into a concept that incorporates a continuously evolving environment, thereby referring to traces of change, dynamic reservoirs of memory and place-bound stories. As a result, the old experts’ emphasis on authenticity of material remains has given way to the inclusion of diverse opinions and attitudes towards adaptations of landscapes and architecture to facilitate new functions and to incorporate new memories and historical narratives. For landscape designers and urban planners heritage changed from a precondition that limits or frames the designer’s creativity into a source of inspiration, from a ’ready-made’ chronicle of life and dwelling to a constantly transforming palimpsest that carries the past with it in ever changing ways (Fairclough 2003; Auclair and Fairclough 2015; see also Janssen et al. 2014a; 2014b). These new developments are calling for new concepts, and for integrative approaches and narratives with which the appreciation for old landscapes can be connected to stakeholder participation, sustainable development and considerations of social identity and quality of life. The landscape biography approach seems well-attuned to meet such ambitions, since its aim is to bring histories and memories of landscape and place to the attention of relevant societal actors in an appealing way and to make good use of existing landscape values in spatial plans and landscape designs for future development (Kolen and Witte 2006).

CONCEPTUAL FRAMEWORK A number of paradigms existed and shifted during the periods of landscape biography development mentioned above. When the approach was first mentioned, during the 1970s, landscapes were thought of as absorbing parts of the life histories of numerous persons and groups that build

121 …

Jan Kolen, Hans Renes and Koos Bosma

„

their livelihood in a particular area. They were conceived of as living and dynamic documents that include a biographical dimension for which the term ’authored landscape’ was coined. According to Samuels (1979), this authorship takes shape in the interdependence between what he described as ‘landscapes of impression’ and ‘landscapes of expression’. He defined the first as the internalised, imaginary and utopian landscapes of, for example, personal experience or collective planning doctrines, the second as the world of physical structures and forms that are the concrete expression of those ideas and doctrines in the environment. Physical landscapes of expression, such as the canonical gardens of the past or the designed landscapes of the avant-garde may in turn be important inducement and sources of inspiration for landscape experiences and appreciation. In this way there is a continuing dialectic movement in which the environmental experience and spatial structures constitute each other continuously. With his new ’phenomenological’ perspective, emphasising the constructive interaction between people and their environment, Samuels distanced his thoughts from those of the former quantitative approaches hitherto dominant in human geography. During the 1990s, shortly after landscape archaeologists had conceived their version of a biographical approach, for example when describing the life histories of objects, land and landed property and so on, they started to think of such ‘objects’ as having a long history of social transmission. In due course, the ‘objects’, including monuments and landscapes, not only ’gather’ stories and information about their (temporary) owners and users, but in this way they also were thought of as building a life history of their own. In the case of landscapes such history manifests itself on a long timescale and, if lengthy and/or impressive, in turn often influences the social and economic values of the landscapes concerned or of specific parts thereof. We can think, for example, of the memory value of places and landscapes for individuals, but also of the social and economic values (and the brand identity) attached to landscapes with a United Nations Educational, Scientific and Cultural Organization (UNESCO) World Heritage status. Focusing on the temporal dimension of material culture, publications by Appadurai (1988) and Kopytoff (1986) gained almost immediate popularity among landscape archaeologists. Their studies increasingly concentrated on the long-term changes in use, layout and perception of striking (pre) historic monuments. Examples are the biographical reconstructions of archaeological sites such as Vindolanda, the Roman fort on Hadrian’s Wall (Birley 2012), Avebury (Pollard and Reynolds 2002), Evora in Portugal (Holtorf 2008), a deserted medieval settlement in the German Rhineland (Kolen 1995) and the late-prehistoric urn fields in the southern part of the Netherlands (Roymans 1995). In all these cases, the life histories of the monuments are followed up to the present and attention is paid to their value and meanings for present-day science, heritage management and society. These archaeological examples illustrate how successive communities and generations have appropriated, empowered and transformed places and monuments through time, and how they ever and again gave them a functional and meaningful place in their contemporary living environments, economies and memories. From 2001, the landscape biography approach has been applied to interdisciplinary regional studies. Paradigmatic innovations took place, this time in a number of studies that were part of a large Dutch research programme (Bloemers et al. 2010). In these studies the landscape biography approach integrated, for the first time, research methods from different disciplines in a way such that they were combined, forming a synthesised vision of the historical development of, on present-day heritage values of, and also on the future life and use of the landscapes in the regions concerned. In developing the approach these research studies also exemplify a paradigmatic shift towards an

122 …

Landscape biography „

increased emphasis on human–land relationships and in also interpreting landscape biography in terms of the impact that these relationships have on ecosystems and rural land use (for which the term ‘historical ecology’ was coined; see Spek at al. 2015). A fascinating aspect is the intensive use of field names, which were seen as a clue to the environmental perceptions of the local population in the past and which also provided a theme for the intensive involvement of the present local population (Elerie and Spek 2009).

PRINCIPLES FOR USING LANDSCAPE BIOGRAPHY AS A RESEARCH APPROACH During the development of landscape biography as a research approach, five operational principles have been established. Firstly, landscape biography essentially is a ‘historicising’ approach, meaning that it aims at improving our understanding of the making of the present-day landscape. This understanding may be combined with more ‘constructivist’ research that highlights the multiple ways in which the landscape is experienced, and also how its dimensions of the past are constructed, for example by and in discourses within the different groups that make up present-day societies. Landscapes are always contested to some extent and historical studies reveal that contestation can be identified in the present as much as in the past. Secondly, landscape biography does not necessarily start from a conventional chronological arrangement of the history of landscapes in discrete phases or periods of use. Rather, the development of the human-made landscape is understood as a continuum or succession of transformations, in which the past is – consciously as well as subconsciously – experienced and processed in ever changing ways. Processes of remembering and forgetting have constantly been important formative principles in the design of our living space. To the extent that landscape archaeologists and historians can reconstruct past processes, societies have always appeared to be fascinated by places and landmarks in their environment which they knew, or felt intuitively, had survived for many generations. In some cases this fascination has even led to the deliberate development of large-scale memorial landscapes, such as the burial landscapes (barrow complexes and urn fields) of the Neolithic, Bronze and Iron Ages (such as Salisbury Plain in England) and the memorial landscapes of the First and Second World War (Ypres, the Somme Valley, the Normandy beaches, etc.). The ways in which prehistoric and historic communities have built their existence in landscapes that were full of reminders of earlier generations, and the multiple ways in which they (re)used those traces from the past, provide new insights in the historical layering of landscapes, both past and present (Meinig 1979). Thirdly, landscape biography employs no discipline-specific or sectoral method alone, but aims at the combined use of archaeological, geographical, historical, social, architectural and other methods and sources. Landscape biographers also seek collaboration with landscape designers and urban planners. Fourthly, landscape biography starts from a ‘dwelling perspective’ on the landscape (Ingold 2000). Therefore, an important field of interest is how local communities have, through time, expressed something of themselves, their ideas and their identities in the landscapes they occupied, used and reshaped, and how they (conversely) derived their identity and existence from the landscape as ’formative agent’. This shows how the ‘hardware’ and ‘software’ of landscapes and regions have mutually influenced and shaped each other through time. Not only does the community with its corporate identity, but also the individual with his or her idiosyncratic personality take part in this

123 …

Jan Kolen, Hans Renes and Koos Bosma

„

process of landscape making. An important place within landscape biography is therefore given to the ‘authorship’ of landscapes, elaborating on Samuels’ concept of ‘authored landscapes’ (see above). Fifthly, landscape biography aims at making historical research and reflection relevant and productive for current issues of designing, planning and heritage management. As landscape biography does not define a sharp boundary between past and present, contemporary heritage practices are also studied from or incorporated into landscape biography projects. The interaction of interest groups with the past in the landscape is an integral part of the spatial conditions of communities and, hence, of spatial transformations. Heritage is always human-made and value-laden and processes of cultural transmission in the landscape are inseparable from the construction of memories and identities.

SOURCES AND METHODS As landscape biography puts the focus on an integrative, long-term perspective of landscape changes, it relies on a large and varied set of historical, environmental and other sources as data to inform studies about the diverse ways in which communities have interacted with their natural and cultural environments through time. Researchers may want to start by looking into ’conventional’ sources, such as data on soils and soil properties, on geomorphology, and also might include elevation models, data on vegetation and land use and data on infrastructure (e.g. ancient road and water systems), on settlement dynamics and on the emergence and development of cities and urban networks, of industrial and post-industrial transformations, etc. (for more information on sources and data, see e.g. Kolen et al. 2015a; Spek et al. 2015). The next step includes linking geographical, ecological and other physical-material data to region-specific knowledge on land use practices, on forms of social organisation and cultural values, and other social and cultural knowledge. For this reason, landscape biography also makes use of less conventional sources that have been explored only sporadically in the study of landscape change so far, such as place-bound narratives (oral history), field names (Elerie and Spek 2010; Kolen 2015), local memories (van Veldhoven 2015; Sooväli-Sepping 2015), ego-documents – such as diaries and photo albums (Elings 2015), maps, landscape paintings and prints (Ronnes 2015), sources that inform about the uses and perceptions of sound, smell and (artificial) light in past landscapes (see e.g. Ekirch 2006), and – in the case of twentieth-century landscapes – novels, films and documentaries (cf. Koren 2015). This information can be found in private collections, libraries and museum collections, but also in governmental archives as official institutes have often been involved in knowledge development, policy research and decision-making on these themes in the past. By synthesising physical, social and cultural data it becomes possible to try to reconstruct the often idiosyncratic social and environmental contexts that set the agenda for human-nature exchanges and for the emergence of characteristic biotopes, such as human-influenced wetlands, fenlands, forests and meadow systems, as well as the knowledge and exploitation systems that sustained these biotopes (Huijbens and Pálsson 2015; Kolen 2015; Purmer 2015). In order to gain a deeper understanding of these and other aspects of change, landscape biographers focus on, for example, the impact of religious beliefs and religious transformations on the built environment and on the ordering, use and perception of space and place (Kolen in press; Roymans et al. 2009), the political and social histories of landed property and the changing cultural appreciation of the

124 …

Landscape biography „

landscape heritage by people through time (Roymans et al. 2009; Papmehl-Dufay 2015; Gillings and Pollard 2015; Holtorf 2015; van der Laarse 2015). Methodologically, therefore, landscape biography does not primarily entail the morphogenetic or retrospective analysis of present-day landscapes, as done in traditional (and still useful) historical landscape assessments and applied historical geography projects. Instead, it focuses on the multidimensional (cultural, social and economic) aspects of landscape change within a certain period or the ‘layerdness’ of landscape at specific moments in time, understanding ’layerdness’ as much as the outcome of memory-informed reuse and experience as physical and patterned overlays (Renes 2015). As in other approaches to landscape assessment and historical reconstruction, subsequent generations of maps and other cartographic documents are considered important sources in order to fulfil this exercise, although maps in landscape biography are seen as much as representations of ancient world views and spatial perceptions as they are treated as documents of past physical realities (e.g. Huijbens and Pálsson 2015). Although landscape biography emphasises the importance of an integrative approach to landscape, it does not strive, of course, to reach completeness. As a rule, Landscape Biographies select specific sets of places or focus on particular ecosystems, buildings, landscape sections or types of transformation, such the religious or political ordering and use of a specific area (for examples see Kolen et al. 2015a). More detailed and project-specific information on methods and sources are discussed in the examples below.

EXAMPLES OF RESEARCH USING THE LANDSCAPE BIOGRAPHY APPROACH The Boerenverstand project in the Vecht region near Utrecht, the Netherlands Assignment and research question In the project Boerenverstand (‘common sense’, the Dutch term could be translated as ‘farmers’ wit’), landscape designers were commissioned by local and regional governments to enhance the recognisability of a former rural area by designing gates and fences that evoke images of the past and a sense of regional identity. The traditional fences that are found in the rural landscape along the river Vecht near Utrecht, the Netherlands were studied (Christiaansen and Montens 2009). Wooden fences and gates, constructed by local carpenters or the farmers themselves, are characteristic features of the Western Netherlands fenlands. In recent decades many of these have been replaced by mass-produced metal versions that last longer and need less maintenance. The aim of the study was to answer the question of how to reconstruct the partly forgotten practice of fence making. The reason for this study was that the disappearance of the wooden fences and gates makes the region appear more monotonous and – as the commissioners of the project stated – the landscape loses some of its charm. Information was needed on how to design a present-day version of the wooden gates, in order to re-invest the landscape with a nostalgic value and in this way to make the region more recognisable for the many visitors who cross this countryside on foot or bicycle.

Methods and materials The landscape biography approach was used by landscape designers Krijn Christiaansen and Cathelijne Montens to study the practice of fence making which was once characteristic of a farmer’s

125 …

Jan Kolen, Hans Renes and Koos Bosma

„

daily work in the pastureland. Starting with field reconnaissance and using a recording system developed for the task, the researchers first described and photographed surviving old fences. Then local farmers were interviewed and asked to describe their memories about these landscape elements. In addition, timber planks that were stored in sheds and elsewhere on the farm were described and documented with museum-like precision, as these apparently were also the ’relics’ of the old practice of fence construction – stored to be used some time in the future for maintenance activities. Interviews were recorded and filmed; information about fences and planks (photographs, descriptions, memories) was included in a separate catalogue. Landscape biography was thus used as an exploratory tool to highlight a practice from the recent past that was typical for both the landscape and the farmer’s work and to shed light on the memories and stories ’surrounding’ this practice in relation to the dynamic life histories of the fences themselves.

Results The interview contents revealed how a rich corpus of memories and anecdotes exists, not only about the fences, gates and planks themselves, but also about the way these were built and repaired. The fences turned out to be much more than functional pivots in the agrarian landscape and the planks to be much more than simple building materials. In a nutshell, people told the story of life and work on farmyard and farmland over the years. Plank ‘KS P1’, for example, was a 250 year old former schoftbalk, a timber cross beam that was used to tie cows to, and ‘KB P19’ a’plank of a sun blind that was later used as banner on a carrier cycle in a Queen’s Birthday procession, with the text: auxiliary service of the future, helps earth, air and water’. In some cases, the gates or planks helped farmers to remember occasions that happened almost half a century earlier in and around the farm. These memories were, moreover, filled with colourful people that had lived and worked in this landscape many decades ago: [this plank was] part of the lower end of a freight car that was bought somewhere during the 1950s or 1960s from people who went door-to-door selling wood. These people later started firms in second-hand wood and metal, such as Blauwe Bertus on the Horstermeer, Nelis van den Berg (you recognise his house by the wooden shoe beside the door) in Muiden and Kroon in Kortenhoef (plank ‘CD P87’).

These results showed that ‘biographies’ could be written of the fences and planks, comparable with those presented by the anthropologists Kopytoff and Appadurai in their seminal book already referred to (Appadurai 1988). The histories of the gates and fences were apparently connected to the life histories of the people who made the farmland into what it is today and in this sense were considered repositories of beloved landscape memories. In addition, the memories collected by the researchers also clearly illustrated the daily management practices in farmyard and landscape. These practices are reminiscent of a concept described by Lévi-Strauss (1962) as bricolage. In the distinction he made between different ways of production, bricolage is far away from mass production (think of the modern steel gates). The bricoleur improvises and uses materials that are readily available. In the process, craftsmanship and inventiveness convert old elements into something new and functional. Process and materials are as important as the final product. The wooden fences and gates of the Vecht region mirrored, par

126 …

Landscape biography „

excellence, the farmers’ talent to improvise, even when they, pragmatic as they are, don’t reject the convenient steel mass-produced gates.

Outcome The knowledge obtained by applying the landscape biography approach helped to decide against developing a new concept for imposing wooden fences and gates on the landscape. Instead, from ‘story-rich’ planks from the storage barns of the farms, the designers selected materials for a series of new fences and gates that were made in the old way. The wooden gates were coated with black polyurethane, which not only conserves the old wood, but also accentuates the silhouette of the gates in the landscape. Thus, the new gates are the result of a slightly paradoxical design and production process that we can best describe as ‘imitated bricolage’. As well as the designs, the research resulted in the revitalisation of this rural practice which was taken up again by enthusiastic farmers.

’The Limes nowadays – A biography’ (Central Netherlands) Assignment and research question ‘The Limes nowadays – A biography’ is a study that focused on the remains of the former Roman frontier along the river Rhine (Berkers and van Stiphout 2009). Since its construction the long linear border zone (the Limes) has undergone a process of urbanisation. It runs through the urban networks of the Randstad and the Arnhem-Nijmegen metropolitan regions, is locally intersected by busy transport routes and alternates with rivers and fragments of agrarian landscapes and nature reserves. The research question was: how might it be possible to reconstruct the numerous transformations of the Limes landscape through time and to visualise the layered historical dimension of that landscape to be appreciated in the present?

Methods and materials For the aim of assessing the historical change of the Limes landscape, landscape biography was used to map the historical layeredness of this landscape up to the present day. Quantification of temporal change and spatial discontinuities formed part of this, as well as the analysis of historical and archaeological maps and data. The quantitative mapping procedure was combined with field investigations. The whole Dutch stretch of the Limes was divided into sections (or legs) surveyed in a single day and was covered on foot and partly by boat without concessions to obstacles in the actual landscape. High fences around industrial estates, watercourses, houses and private gardens were crossed. In this way, the contrasts and discontinuities in the modern Limes landscape could be confronted with the continuity shown on the maps and the (romanticised) reconstructions of the archaeologists. Contrasts are not only found at a regional level, but also at the micro level in local textures and stratigraphy. Where the existence of gravel in the soil still betrays the old course of the Limes road, the surface is now covered with concrete, shell rubble, stone pavement and traces of ploughing. The daily legs were covered by teams of changing composition, consisting of designers, administrators, archaeologists and heritage specialists who, in the field, searched for dialogues with residents and other users of

127 …

Jan Kolen, Hans Renes and Koos Bosma

„

space. The aim of the daily legs was always twofold. They served to collect and exchange knowledge about the Limes landscape, and to search for opportunities for designing the Limes landscape of today. Returning to some of the spots visited, in order to check any local impact of the exchange of knowledge and stories, was also part of the research.

Results It was impressive to realise that the Dutch part of the Limes road was 152.99 km long and in Roman times took a normal person six to eight days to travel the whole length. For the researchers it was not only important to learn how the Limes appeared in the river landscape of 2000 years ago, but also how it has been transformed intensively since then. The road, for example, is nowadays divided into over 5665 fields, 3722 landowners (most of them private) and 39 villages and towns. It now takes us, on foot, much longer (eighteen days) than during the Roman period to cover the distance. And, not to forget, 34.18 km of the Limes road now lies in areas with planned developments. As a result of conducting the study it could be seen how the Limes had, in some instances, been ‘woken up’ by the collecting and sharing of knowledge, stories and memories during the study walks, as they led to more-or-less spontaneous and apocryphal design actions. The final study publication mentioned some examples, such as a farmer in the village of Pannerden who created a footpath over his land exactly on the site of the Limes road and a plan developed by the owner of a brick factory to make stones with Roman inscriptions. In this sense, landscape biography generated a (spatial) process by evoking the past and initiating new and experimental design practices on a local scale. The main strength of the Limes road, however, according to the designers, remains its mysterious invisibility (van Stiphout 2009).

Outcome The designers who conducted the study decided, on the basis of their research results, not to make concrete design proposals. Rather, based on the landscape biography study, the suggestion was made to delay developing definitive designs.

LANDSCAPE BIOGRAPHY OF URBAN LANDSCAPES In his essay cited earlier, Samuels (1979) mainly used examples of urban landscapes to illustrate his ideas. For Samuels, the principle of ‘authorship’ can be read perfectly from the urban space of large metropolitan landscapes such as New York or Shanghai. In The Practice of Everyday Life (1984), the anthropologist Michel de Certeau explains how the authors of the urban landscape are not simply planners or architects or the urban elite of entrepreneurs. For the real authors of the urban space we must look at the social practice of people’s everyday life in the street. There we meet ‘the ordinary practitioners’ who really move through the town and thereby embody an elementary spatial experience: ‘Wandersmänner … whose bodies follow the thicks and thins of an urban “text” they write without being able to read it’ (de Certeau 1984, p.93). Here the town presents itself not as a space that is planned and carefully designed in advance, but as a wrinkled world that is ‘remade’ over and over again by the routines and experiments of residents and visitors.

128 …

Landscape biography „

The core question to be answered by applying the landscape biography approach to studying urban landscapes is how day-to-day social practices, interventions in the physical environment, spatial perceptions and official plans might influence each other and ultimately bestow form and meaning to urban areas. One example addressing this question is a study investigating successive transformations of urban landscapes in the south-eastern part of the Dutch province of Brabant since 1920 (Janssen 2006). Brabant-city (Brabantstad in Dutch) is the unofficial name for the urban network of which the line of cities in the region (Breda, Tilburg, Eindhoven, Helmond and Den Bosch) forms the backbone. The main research question was: how could this network conurbation develop so quickly in a region where the population traditionally identifies itself with the values and ideals typical of the Catholic countryside? According to the study, the answer lies – paradoxically – in a strongly developed fear of the metropolis and a forced contact with the advancing urban culture felt by the locals at the time; in other words, the answer lies in a collectively shaped common mentality. During the 1920s and 1930s, among regional administrators, a fear arose that the unplanned growth of towns would lead to an unworkable spatial structure and one that would become a threat to harmonious Catholic life in all the villages surrounding the cities. Active planning was therefore seen as a necessity, but the evidence-based approach of official town planners collided with what politicians and administrators had in mind. The urbanists preferred the town over the countryside, whereas the administrators kept clinging on to the principle of subsidiarity. The regionalist movement Brabantia Nostra, consisting of prominent intellectuals and writers of undisputed Roman Catholic background, that presented the cities as places of doom and as a source of devaluations and immorality was very influential in spreading this view among the regional population. The Catholic rural ideal remained dominant after the Second World War, even when it became clear that the more than 1.2 million inhabitants of Brabant in 1950 (as against slightly less than 600,000 in 1900) could not possibly be distributed amongst small villages. A policy of dispersal was chosen. New urban areas should be the size of a parish of 6000 people and the church would be the undisputable religious and social focal point. The result of this policy, together with other measures such as the well-reasoned spreading of industry over the region, was exactly the opposite of what the administrators originally aimed at: a rapid and large-scale urbanisation of the entire Brabant countryside. This example illustrates what it takes for the landscape biography approach to be successful, when applied to help understanding of the history of urban landscapes, because ’authorship’ in urban landscapes is historically complex. Urban transformations tend to take shape in complex interactions between personal motivations and shared cultural values, and between existing physical structures and the practices and dynamics of daily life, in which each actor could be considered a true author of urban space. How these interactions actually evolved in particular cases is an important issue for urban planners and landscape designers operating in these urban settings today, and landscape biography has proven to be a suitable exploratory method for research into those cases.

LANDSCAPE BIOGRAPHY: THE NEXT DEVELOPMENTS The landscape biography approach will, with every application, continue to help researchers integrate methods from a variety of fields. What is more, landscape biography has been and will continue to be part of the way heritage and historic landscapes are managed.

129 …

Jan Kolen, Hans Renes and Koos Bosma

„

Digital facilities and geospatial technologies have developed recently that are well suited to support researchers wishing to apply the landscape biography approach and to use landscape biography for synthesising findings obtained by employing methods from different fields. Where diverse quantitative and qualitative data sets are used, as part of a landscape biography these might best be organised by means of so-called spatial data infrastructures (SDI). An SDI is an infrastructural facility for combining, exploring and exchanging large amounts of digital data and multidisciplinary information about the history and heritage of landscapes and for making new connections and comparisons by cross-cutting existing boundaries between different disciplines, time periods and geographical areas. SDIs function as coherent systems of digital data and information, agreements, standards, technology (hardware, software and electronic communication) and knowledge, providing the different users with the combinations of information they need within their specific research and design frameworks. At the core of an SDI lies the technical infrastructure of services, varying from data viewing to download and more complex processing services. On top of these services applications can be built with which users, with different research questions, objectives and ’geo-information literacy’ levels can perform their research tasks. SDIs have been used in landscape biography projects within the framework of the European Union (EU) FP7 HERCULES project, more particularly for the Dutch river landscape, the Uppland area in Sweden and Vooremaa in Estonia (Kolen et al. 2015b). SDIs that have been constructed in landscape biography projects can be used as a sound empirical and interpretative basis for geographic information system (GIS)-based dynamic models (DMs) and agent-based models (ABMs). These models reconstruct and visualise the dynamic landscape changes in the past in a specific area, taking into account both region-specific changes and more general trends in land use, and both economic and social/cultural transformations. It has been proposed that SDIs and models for landscape biography be used more intensively in digitised versions of geodesign (Steinitz 2012). Following a series of steps, the first step entails a description of the selected area in the form of representation models which contain the geographical and landscape data upon which the regional study relies. This is followed by an analysis of the processes that currently operate in the area (urbanisation, erosion, increasing tourism, etc.), and, subsequently, by evaluation models, change models, impact models and decision models. The landscape biography approach, highlighting the historical background and long-term character of landscape change, as well as the often prolonged impact of decisions taken in the distant past, has an obvious place in the geodesign framework. The same holds for detailed information on heritage values and how these (may) influence current social values, spatial developments and land use. Therefore, landscape biography and geodesign would, if applied in concert, mutually reinforce each other. An important step in this direction is the recent development of digital cultural biography or DCB. This is a digital tool that acts as a foundation for the application of knowledge exchange tools in landscape biography projects, such as SDI, as well as for assessing the value of such tools in specific design contexts. Exchange tools are used to improve the exchange of multidisciplinary information and to stimulate dialogue between different stakeholders in landscape biography projects, including future-oriented urban planners and landscape designers, past-oriented archaeologists and historians, as well as local citizens and private enterprises. Empowering the dialogue between these various stakeholders allows for better-informed decisions leading to more sustainable landscape plans. The value of DCB as a digital exchange and decision making tool has been explored in a couple of studies, for example by a multidisciplinary team of archaeologists, architects and urban planners in the historical neighbourhood of Testaccio in Rome, Italy, and, by experimenting with digital landscape

130 …

Landscape biography „

biographies in a study investigating the eastern part of Brabant in the Netherlands. In the latter case, after testing the usefulness and workability of the resulting Biografie van de Zandstad website it became apparent that designers generally prefer digital biographies that allow ample possibilities for making creative associations, all the while cross-cutting periods, areas and disciplines. At the same time historians and heritage students generally seem to prefer well-structured databases with predefined chronologies, geographical locations and source-classifications. New versions of the DCB, such as the recent Testaccio version, may help to bridge this divide. Landscape biography has contributed a lot to help heritage management in changing to adopt a culture of profit (van der Laarse 2005; Frijhoff 2007; Eerden et al. 2009; van der Zande and During 2009; Janssen et al. 2014b). During the nineteenth and twentieth centuries, a whole corpus of laws, treaties and formal institutions was established, designed for the protection and survival of antiquities, historic buildings and landscapes. The care for this collection (a term that sometimes can be taken literally, see Thurley 2013) of ‘monuments of art and history, nature and culture’ was seen as a necessary counterweight for modernisation and industrialisation. In this context the heritage sector cultivated a ‘culture of loss’. The primary driving force was the fear of losing valuable buildings, historic landscapes and archaeological sites. Heritage management meant a constant fight against damage and deterioration which, during the post-war period, was particularly caused by urbanisation of the countryside, modernisation of the city centres and the large-scale reconstruction of rural landscape for modern mechanised agriculture. All this changed during the 1990s. Although legal protection and preservation are still primary aims of heritage management, society has new demands from the past that require more dynamic solutions. Heritage plays a constructive role in the creative industry (for example the redevelopment of industrial heritage), the rise of tourism and recreation (leisure landscapes) and the reshaping of urban regions. In this context, ‘heritage’ no longer refers to isolated historic buildings and archaeological sites, but includes complete landscapes and urban structures. The heritage sector is increasingly characterised by a ‘culture of profit’, one that leaves behind (perhaps even too much) the old feelings of threats and losses, as can be seen, for example, in the Kassel Bergpark Wilhelmshöhe, particularly after its designation as a UNESCO World Heritage Site. In a number of European countries plans were already developed before 1990 to integrate historic buildings and old industrial landscapes into new spatial transformations and regional plans. A benchmark project was the Internationale Bauaustellung (International Building Exhibition) Emscher Park in the German Ruhr Area. The area now attracts more than half a million visitors annually. Other examples include derelict mining and industrial areas such as Nord-Pas-de-Calais in France, the Walloon region in Belgium and the South Wales mining region, all of which also chose their heritage as the core of a new regional identity. Further examples come from England, Denmark and Italy and other European countries. Everywhere in Europe, lieux de mémoire, industrial monuments, ancient buildings and historic landscapes were deployed for the revitalisation of rural and urban landscapes over the past twenty years (Nota Belvedere 1999; Saris et al. 2008; Kolen 2008). All these examples show that modern heritage management is mainly about managing change, in which aspects of economic importance, cultural value production and social vitality are inseparable. That of course also demands responses from the scientific disciplines that claim landscape and heritage as their fields of research. While (traditional) morphological approaches succeeded in generating a large corpus of information and data on regional landscapes, it is the landscape biography approach that helps researchers to understand better the larger spatial, historic

131 …

Jan Kolen, Hans Renes and Koos Bosma

„

and social contexts. What landscape designers are looking for is inspiration such as can be derived from stories about the role of people in their environment, about long-term processes, about the mechanisms behind stability and change and so on. In landscape biography the landscape is not simply considered as the backdrop to human action or a pre-given environment for the people living in it, but as the historically grown and meaningfully constituted habitat of people. In this habitat, the landscape is ‘a chronicle of life and dwelling’, as Ingold (2000) put it. According to him the landscape ‘unfolds the lives and times of predecessors who, over the generations, have moved around in it and played their part in its formation. To perceive the landscape is therefore to carry out an act of remembrance’ (p.189). In this chapter we have seen that the chronicle of life and dwelling has always been and still is very dynamic, being constantly rewritten, not least by landscape designers and urban planners. For them to do their task in a better prepared way, landscape biography is designed as a research approach for exploring the memories of place as well as for localising unexpected and deep histories that may contribute to the understanding of the ways in which human space and place has come about. By following this route, opportunities are created to present an area, such as a region or a city quarter, by means of varied and inspirational images of the past. These images are more informative than simply taking account of the historical icons and canons that appear (too) often as advertised in place promotion and regional marketing and that used to be the sole content of research on landscape history in the past.

ACKNOWLEDGEMENT An earlier version of this chapter was published in Kolen et al. (2010). Recently, the ideas were extended in Kolen et al. (2015a).

SUGGESTED FURTHER READING The term ‘landscape biography’ was introduced by the American geographer Marwyn Samuels to express the fact that the landscape cannot be understood properly in isolation from the life histories of all those people who – over the centuries – have shaped and reshaped it as a meaningful and dynamic palimpsest: Samuels, M.S. (1979) ‘The biography of landscape: Cause and culpability’, in Meinig, D.W., ed. The Interpretation of Ordinary Landscapes, Oxford: Oxford University Press, 51–88. In the early 1990s the concept was rediscovered, partly autonomously, by archaeologists who applied the anthropological concept of the ‘biography of things’ to the history of place, land and landscape: Holtorf, C. (1998) ‘The life-history of megaliths in Mecklenburg-Vorpommern (Germany)’, World Archaeology, 30, 23–38. Kolen, J. (1995) ‘Recreating (in) nature, visiting history: Second thoughts on landscape reserves and their role in the preservation and experience of the historic environment’, Archaeological Dialogues, 2(2), 127– 159. Kopytoff, I. (1986) ‘The cultural biography of things: Commoditization as process’, in Appadurai, A., ed. The Social Life of Things: Commodities in Cultural Perspective, Cambridge: Cambridge University Press, 64–91.

132 …

Landscape biography „

Roymans, N. (1995) ‘The cultural biography of urnfields and the long-term history of a mythical landscape’, Archaeological Dialogues, 2(1), 2–24.

After several critical reformulations of the concept and through a large number of regional case studies, landscape biography has now developed into a broadly applied tool for landscape research, connecting landscape history with current issues of interpretation, experience and design, both in Europe and in other parts of the world, see: Kolen, J., Renes, J. and Hermans, R., eds. (2015) Landscape Biographies: Geographical, Historical and Archaeological Perspectives on the Production and Transmission of Landscapes, Amsterdam: Amsterdam University Press.

REFERENCES Appadurai, A., ed. (1986) The Social Life of Things: Commodities in Cultural Perspective, Cambridge: Cambridge University Press. Auclair, E. and Fairclough, G., eds. (2015) Theory and Practice in Heritage and Sustainability: Between Past and Future, Abingdon: Routledge. Bell, S., Sarlöv Herlin, I. and Stiles, R., eds. (2011) Exploring the Boundaries of Landscape Architecture, London: Routledge. Berkers, M. and van Stiphout, M., eds. (2009) Limesweg, Amsterdam: Architectura & Natura. Birley, R. (2012) Vindolanda: A Roman Frontier Fort on Hadrian’s Wall, Stroud: Amberley Publishing. Bloemers, T., Alders, G., Van Heeringen, R., Kok, M., van Londen, H., Theunissen, L., Vos, P. and Pálsson, G. (2010) ‘From Oer-IJ estuary to metropolitan coastal landscape: Assessing and preserving archaeological-historical resources from 4000 years of living between land and water’, in Bloemers T., Kars, H., van der Valk, A. and Wijnen, M., eds. The Cultural Landscape & Heritage Paradox: Protection and Development of the Dutch Archaeological-Historical Landscape and Its European Dimension, Amsterdam: Amsterdam University Press, 203–238. Bosma, K. and Kolen, J., eds. (2010) Geschiedenis en ontwerp: Handboek voor de omgang met cultureel erfgoed, Nijmegen: Uitgeverij Vantilt. Christiaansen, K. and Montens, C. (2009) Boerenverstand, Utrecht: VBK Uitgevers. de Certeau, M. (1984) The Practice of Everyday Life, Los Angeles, CA: University of California Press. Eerden, M., Kapelle, M., Labuhn, B., Linssen, M. and Strolenberg, F., eds. (2009) Belvedere.nu: Praktijkboek cultuurhistorie en ruimtelijke ordening, Utrecht: Matrijs. Ekirch, A.R. (2006) At Day’s Close: Night in Times Past, New York: W.W. Norton. Elerie, H. and Spek, T., eds. (2009) Van Jeruzalem tot Ezelakker: Veldnamen als levend erfgoed in het Nationaal landschap Drentsche Aa, Utrecht: Matrijs. Elerie, H. and Spek, T. (2010) ‘The cultural biography of landscape as a tool for action research in the Drentsche Aa National Landscape (Northern Netherlands)’, in Bloemers, T., Kars, H., van der Valk, A. and Wijnen, M., eds. The Cultural Landscape & Heritage Paradox: Protection and Development of the Dutch ArchaeologicalHistorical Landscape and Its European Dimension, Amsterdam: Amsterdam University Press, 83–114. Elings, A. (2015) ‘De Loowaard: Een uiterwaard als ideaalbeeld van het verleden’, in Kolen, J., Ronnes, H. and Hermans, R., eds. Door de lens van de landschapsbiografie: Een nieuwe kijk op de geschiedenis en het erfgoed van landschappen, Leiden: Sidestone Press, 221–246. Fairclough, G. (2003) ‘Cultural landscape, sustainability and living with change?’, in Teutonico, J.M. and Matero, F., eds. Managing Change: Sustainable Approaches to the Conservation of the Built Environment, Proceedings of the 4th annual US/ICOMOS International Symposium, Philadelphia, April 2001, Los Angeles, CA: Getty Publications, 23–46, available: http://www.getty.edu/conservation/publications_ resources/pdf_publications/pdf/managing_change_vl_opt.pdf. Frijhoff, W. (2007) Dynamisch erfgoed: Heeft de cultuurgeschiedenis toekomst?, Amsterdam: SUN. Gillings, M. and Pollard, J. (2015) ‘Authenticity, artifice, and the druidical temple of Avebury’, in Kolen, J., Renes, J. and Hermans, R., eds. Landscape Biographies: Geographical, Historical and Archaeological

133 …

Jan Kolen, Hans Renes and Koos Bosma

„

Perspectives on the Production and Transmission of Landscapes, Amsterdam: Amsterdam University Press, 117–142. Holtorf, C. (2008) ‘The life-history approach to monuments: An obituary?’, in Goldhahn, J., ed. Gropar & monument: En vänbok till Dag Widholm, Kalmar: University of Kalmar, 411–427. Holtorf, C. (2015) ‘What future for the life-history approach to prehistoric monuments in the landscape?’, in Kolen, J., Renes, J. and Hermans, R., eds. Landscape Biographies: Geographical, Historical and Archaeological Perspectives on the Production and Transmission of Landscapes, Amsterdam: Amsterdam University Press, 167–181. Huijbens, E.H. and Pálsson, G. (2015) ‘The marsh of modernity: The bog in our brains and bowels’, in Kolen, J., Renes, J. and Hermans, R., eds. Landscape Biographies: Geographical, Historical and Archaeological Perspectives on the Production and Transmission of Landscapes, Amsterdam: Amsterdam University Press, 49–70. Ingold, T. (2000) The Perception of the Environment: Essays on Livelihood, Dwelling and Skill, London: Routledge. Janssen, J. (2006) Vooruit denken en verwijlen: De (re)constructie van het plattelandschap in Zuidoost-Brabant, 1920–2000, Tilburg: Stichting Zuidelijk Historisch Contact Tilburg. Janssen, J., Luiten, E., Renes, H. and Rouwendal, J. (2014a) ‘Heritage planning and spatial development in the Netherlands: Changing policies and perspectives’, International Journal of Heritage Studies, 20(1), 1–21. Janssen, J., Luiten, E., Renes, H., Rouwendal, J., Faber, O., Pen C.-J. and Stegmeijer, E. (2014b) Character Sketches: National Heritage and Spatial Development Research Agenda: Part 1 – Research Agenda, Netwerk Erfgoed & Ruimte, Amersfoort: Netwerk Erfgoed & Ruimte, onderzoek & onderwijs, available: http:// dspace.library.uu.nl/bitstream/handle/1874/306449/Charactersketsches_060214.pdf (accessed 18-82015). Kolen, J. (1995) ‘Recreating (in) nature, visiting history: Second thoughts on landscape reserves and their role in the preservation and experience of the historic environment’, Archaeological Dialogues, 2(2), 127–159. Kolen, J. (2008) ‘Een functionele geschiedenis’, in Eerden, M., Luiten, E., van der Zande, A., Kolen, J. and During, R., eds. Op historische gronden: Erfgoed in een context van ruimtelijk ontwerp, planning en democratie, Utrecht: Belvedere, 83–102. Kolen, J. (2015) ‘Biographies of biotopes’, in Kolen, J., Renes, J. and Hermans, R., eds. Landscape Biographies: Geographical, Historical and Archaeological Perspectives on the Production and Transmission of Landscapes, Amsterdam: Amsterdam University Press, 71–98. Kolen, J. (in press) ‘The religious transformation of a landscape: Drenthe (the Netherlands)’, in Ó Carragáin, T. and Turner, S., Making Christian Landscapes in Atlantic Europe: Conversion and Consolidation in the Early Middle Ages, Cork: Cork University Press. Kolen, J. and Witte, M. (2006) ‘A biographical approach to regions, and its value for spatial planning’, in van der Knaap, W. and van der Valk, A., eds. Multiple Landscape: Merging Past and Present, selected papers from the Fifth International Workshop on Sustainable Land Use Planning, 7–9 June 2004, Wageningen: Wageningen University, 125–145. Kolen, J., Bosma, K. and Renes, H. (2010) ‘De landschapsbiografie: Instrument voor onderzoek, planning en ontwerp’, in Bosma, K. and Kolen, J., eds. Geschiedenis en ontwerp: Handboek voor de omgang met cultureel erfgoed, Nijmegen: Uitgeverij Vantilt, 212–237. Kolen, J., Renes, J. and Hermans, R., eds. (2015a) Landscape Biographies. Geographical, Historical and Archaeological Perspectives on the Production and Transmission of Landscapes, Amsterdam: Amsterdam University Press. Kolen, J., Crumley, C., Burgers, G.-J., von Hackwitz, K., Howard, P., Karro, K., de Kleijn, M., Löwenborg, D., van Manen, N., Palang, H., Plieninger, T., Printsmann, A., Renes, H., Scholten, H., Sinclair, P., Veldi, M. and Verhagen, P. (2015b) ‘HERCULES: Studying long-term changes in Europe’s landscapes’, Analecta Praehistorica Leidensia, 45, 209–219. Kopytoff, I. (1986) ‘The cultural biography of things: Commoditization as process’, in Appadurai, A., ed. The Social Life of Things: Commodities in Cultural Perspective, Cambridge: Cambridge University Press, 64–91. Koren, D. (2015) ‘Shanghai: The biography of a city’, in Kolen, J., Renes, J. and Hermans, R., eds. Landscape Biographies: Geographical, Historical and Archaeological Perspectives on the Production and Transmission of Landscapes, Amsterdam: Amsterdam University Press, 253–282.

134 …

Landscape biography „

Lévi-Strauss, C. (1962) La Pensée sauvage, Paris: Presses Universitaires de France. Meinig, D.W. (1979) ‘The beholding eye: Ten versions of the same scene’, in Meinig D.W., ed. The Interpretation of Ordinary Landscapes: Geographical Essays, Oxford: Oxford University Press, 33–47. Nota Belvedere (1999) Nota Belvedere: Beleidsnota over de Relatie Cultuurhistorie en Ruimtelijke Inrichting, Den Haag: VNG Uitgeverij. Papmehl-Dufay, L. (2015) ‘Places that matter’, in Kolen, J., Renes, J. and Hermans, R., eds. Landscape Biographies: Geographical, Historical and Archaeological Perspectives on the Production and Transmission of Landscapes, Amsterdam: Amsterdam University Press, 143–166. Pollard, J. and Reynolds, A. (2002) Avebury: The Biography of a Landscape, London: Duckworth. Purmer, M. (2015) ‘Authorship and the agrarian landscape of Eerde’, in Kolen, J., Renes, J. and Hermans, R., eds. Landscape Biographies: Geographical, Historical and Archaeological Perspectives on the Production and Transmission of Landscapes, Amsterdam: Amsterdam University Press, 183–204. Renes, J. (2015) ‘Layered landscapes’, in Kolen, J., Renes, J. and Hermans, R., eds. Landscape Biographies: Geographical, Historical and Archaeological Perspectives on the Production and Transmission of Landscapes, Amsterdam: Amsterdam University Press, 403–422. Ronnes, H. (2015) ‘The quiet authors of an early modern Palatial Landscape’, in Kolen, J., Renes, J. and Hermans, R., eds. Landscape Biographies: Geographical, Historical and Archaeological Perspectives on the Production and Transmission of Landscapes, Amsterdam: Amsterdam University Press, 205–234. Roymans, N. (1995) ‘The cultural biography of urnfields and the long-term history of a mythical landscape’, Archaeological Dialogues, 2(1), 2–24. Roymans, N., Gerritsen, F., van der Heijden, C., Bosma, K. and Kolen, J. (2009) ‘Landscape biography as research strategy: The case of the South Netherlands Project’, Landscape Research, 34(3), 337–359. Samuels, M.S. (1979) ‘The biography of landscape: Cause and culpability’, in Meinig, D.W., ed. The Interpretation of Ordinary Landscapes, Oxford: Oxford University Press, 51–88. Saris, J., van Dommelen, S. and Metze, T., eds. (2008) Nieuwe Ideeën voor Oude Gebouwen: Creatieve Economie en Stedelijke Herontwikkeling, Rotterdam: NAi Publishers. Sooväli-Sepping, H. (2015) ‘Biographies of landscape: Rebala Heritage Reserve, Estonia’, in Kolen, J., Renes, J. and Hermans, R., eds. Landscape Biographies: Geographical, Historical and Archaeological Perspectives on the Production and Transmission of Landscapes, Amsterdam: Amsterdam University Press, 423–435. Spek, T., Elerie, H., Bakker, J.P. and Noordhoff, I., eds. (2015) Landschapsbiografie van de Drentsche Aa, Assen: Koninklijke Van Gorcum. Steinitz, C. (2012) A Framework for Geodesign: Changing Geography by Design, Redlands, CA: Esri Press. Thurley, S. (2013) Men from the Ministry: How Britain Saved Its Heritage, New Haven, CT: Yale University Press. van der Laarse, R., ed. (2005) Bezeten van vroeger: Erfgoed, identiteit en musealisering, Amsterdam: Het Spinhuis. van der Laarse, R. (2015) ‘Fatal attraction’, in Kolen, J., Renes, J. and Hermans, R., eds. Landscape Biographies: Geographical, Historical and Archaeological Perspectives on the Production and Transmission of Landscapes, Amsterdam: Amsterdam University Press, 345–376. van der Valk, A. (2010) ‘Planning the past: Lessons to be learned from “Protecting and Developing the Dutch Archaeological-Historical Landscape” (PDL/BBO)’, in Bloemers, T., Kars, H., van der Valk, A. and Wijnen, M., eds. The Cultural Landscape & Heritage Paradox: Protection and Development of the Dutch Archaeological-Historical Landscape and its European Dimension, Amsterdam: Amsterdam University Press, 21–53. van der Zande, A. and During, R., eds. (2009) Erfgoed en ruimtelijke planning, Den Haag: Sdu uitgevers. van Stiphout, M. (2009) ‘Ontwerp van een proces’, in Berkers, M. and van Stiphout, M., eds. Limesweg, Amsterdam: Architectura & Natura, 16–21. van Veldhoven, F. (2015) ‘Post-industrial coal-mining landscapes and the evolution on mining memory’, in Kolen, J., Renes, J. and Hermans, R., eds. Landscape Biographies: Geographical, Historical and Archaeological Perspectives on the Production and Transmission of Landscapes, Amsterdam: Amsterdam University Press, 327–344.

135 …

Chapter 9: Social media Ron van Lammeren, Simone Theile, Boris Stemmer and Diedrich Bruns

INTRODUCTION Social media have serious impact on societal processes. Successful start-ups like Airbnb (https:// www.airbnb.com) to rent lodging facilities and Uber (https://www.uber.com) to arrange taxi transport heavily rely on social networking services. These start-ups show the power of contacting by internet and set processes in motion, known as ‘disruptive innovation’, where high technology successfully replaced low technology. The objective of this chapter is to examine how internetbased social media might be used to insert some ‘disruptive innovations’ into landscape architecture research. To explore innovative ways of enhancing the field’s knowledge base researchers might, for example, be experimenting with ‘crowdsourcing’ and geo-social media. Social media might be useful as a source of data for project research (such as discovering what happens around a design or planning area) and as a data collection tool for academic research (for example, setting up a platform and inviting members of the public to make entries which become qualitative data sets). Findings from the recent Delphi study summarised in Chapter 6 and in Meijering et al. (2015) point to ‘human dimensions of planning and design’ as one of landscape architecture’s most important and useful research challenges. Our starting claim is that research into this domain could benefit from the use of social media as a new resource. This chapter discusses social media as vehicles for knowledge acquisition through collaborative processes and it also presents methods to analyse data shared on social media sites.

Social media Most of the current definitions of social media combine aspects of use and technology. Kaplan and Haenlein (2010, p.61), for example, define social media as ‘a group of Internet-based applications that build on the ideological and technological foundations of Web 2.0, and that allow the creation and exchange of User Generated Content’. In line with this definition, other authors emphasise the fact that social media usually depend on mobile and web-based technologies to create highly interactive platforms through which individuals form communities to interact and create, share, discuss and modify user-generated content, such as information and ideas (e.g. Kietzmann et al. 2011). Social media are applications and take advantage of social networking services. They are platforms upon which to build virtual social networks among people, including but not limited to those who share similar interests, lifestyles and so on. In general, network services support the

136 …

Social media „

creation of a user profile (representation of the user), a contact persons list (representation of the social network), a sharing profile (interaction type with contact persons), and tools for information (text, graphics, photo, audio, video formats) and communication (view, edit, read, evaluate, modify, send, share, delete). By and large, social networking services are individual-centred services which can be used to create group-centred services in favour of virtual communities. Classifications of social media abound and not all of them are consistent. For example, Kaplan and Haenlein (2010) created a classification scheme that includes different types of social media: s s s s s s

Social networking sites (Facebook, LinkedIn) Blogs and microblogs (Twitter, Tumblr) Social news networking sites (Digg and Leakernet) Content communities (YouTube, DailyMotion, Panoramio, Flickr) Collaborative projects (Wikipedia, Open Street Map) Virtual social worlds (Second Life) and virtual game-worlds (World of Warcraft)

These classifications illustrate the great variety of social media which may range from those for general use and with an extremely high number of users, like Facebook, to small, dedicated virtual communities with a specific purpose, like Wiki groups (also see Web 2.0 applications, Haklay et al. 2008). With the growing number of location-aware mobile devices such as the current generation of smartphones, many social media types also tag uploaded information with location coordinates (Stefanidis et al. 2013). Geo-social information differs from Volunteered Geographic Information as defined by Goodchild (2007) because people posting geo-social information are not geotagging deliberately and for a purpose. For that reason the term ‘ambient geo-information’ has been coined (Stefanidis et al. 2013). The geographic component of the social media data is a by-product of the content and included coordinates, address or other locational reference. These tags offer data sets with great potential for various kinds of landscape research since they combine aspects of image (photos taken of specific places to represent some aspect), action (what people may be doing in a specific place) and structure (the characteristics of the geo-referenced location).

Big data and crowdsourcing Social media users generate content that can be understood as the output of social processes supported by internet facilities. Such content can be defined as data or, owing to the amount and diversity of data produced and flowing, as big data. Researchers using materials generated by social media have to deal with large volumes and varieties of data that needs to be accessed while it is flowing and which has uncertain quality and value. The challenge of using big data effectively is dealing with volume, velocity and variety, and also validity and veracity (de Mauro et al. 2015). Volume refers to the amount of data. Velocity refers to how fast data is processed and, in research, to watching and collecting data flows that never stop (‘streams’). Variety refers to the various types of data and to variation in forms of data, including content and format. Validity and veracity refer to uncertainties about the quality and usability of data. With the objective of gaining insights about data-related processes and systems, data science aims to discover and extract actionable knowledge from such data, including knowledge that can be used to make decisions and predictions reliably, not just to explain what is going on.

137 …

Ron van Lammeren, Simone Theile, Boris Stemmer and Diedrich Bruns

„

Computational neurobiology, the science concerned with study of brain function in terms of its information processing properties, relates functions of human senses to the different sensors used in robots. Expanding on this analogy, users of social media can be understood as ‘human sensors’ that are responsive to their environment. Data is collected by human sensors, and these data collections are supplemented by the use of mobile sensors (such as laser scanners, car mounted sensors, etc.) and static sensors (such as the Internet of Things, weather stations, fixed surveillance cameras, etc.). Some devices are also able to measure impacts of our own behaviour on ourselves and our immediate environment (such as Global Positioning System (GPS) trackers, heart-rate monitors, etc.). Together, all of these sensors, human and device based, create increasingly large masses of heterogeneous and unstructured data sets including numerical data, text, images, audio and video. In this seemingly nebulous mass, these data sets might conceal relationships that are potentially interesting for systematic study (Dhar 2013; Dhar et al. 2014). For example, geo-social media such as Geo-Spatial Web, public participatory geographic information systems (PPGIS), Urban Geo-Wiki, geo-tracking and other tools based on geo-positioning applications (e.g. of smartphones) may be applied to support research, including research involving volunteers who, together, generate large amounts of data. Big data and crowdsourcing are related insofar as one specific data set might be generated via crowdsourcing. However, not all sets called big data are equal to crowdsourcing and not all crowdsourcing lead to big data. Crowdsourcing in general terms means web participation. Public participation in research and co-production of knowledge is also called citizen science. Citizen science may be understood as an approach where the conduct of research involves a ‘crowd’ of non-scientists and their ‘wisdom’ (Surowiecki 2005). Before enthusiastically deciding on a citizen science and crowdsourcing approach researchers should, for ethical reasons and scholarly soundness, seriously reflect on expected knowledge gains (Wechsler 2014). In addition, researchers will need to consider providing and managing proper quality control and allow for the rigorous evaluation necessary for making any sort of generalisation from expected study findings. Of course, the use of this approach and these data sets must also be relevant to answering a research question and not selected out of enthusiasm before a problem has been identified and framed (see Tobi and van den Brink – Chapter 2).

CONCEPTUAL FRAMEWORK TO STUDY THE ROLE OF SOCIAL MEDIA Marking the rising importance of communication in planning, mainly during the 1980s, some pertinent theoretical foundations were laid during the 1990s (Fischer and Forester 1993). Planning for and designing of space and landscape is highly dependent on information exchange and discourse, both of which are occurring ‘offline’ and, increasingly, also ‘online’. Research on the role and impact of social media in landscape planning and design must therefore try to understand the communication (and negotiation) that is going on online, in the internet in general, and in social media in particular. The conceptual framework offered in this chapter has two components: one is a transdisciplinary analytical framework with communication at the centre and the other includes a number of theoretical approaches that help to conceptualise social media, social discourse and the construction of reality. The two components are meant to help understand social media as a ‘place’ where different ‘actors’ are engaging in communication, such as in discourse about and through landscape knowledge, and where reality is ‘constructed’ and ‘mediated’ in ways that are particular to social media.

138 …

Social media „

Figure 9.1 Transdisciplinary framework with social media interface (source: based on Enengel et al. 2012)

To position the role of social media in research we build upon the concept of transdisciplinary knowledge generation as elaborated by Enengel et al. (2012) (see Figure 9.1). This model is extended as illustrated in Figure 9.2. We include a typology of the actors involved and their knowledge dimensions, forms of actor interaction, study phases and other elements (Louwsma et al. in prep). The precision and order of this typology in each phase and the duration of these phases depend on the purpose of the phase. Relationships between study phases and actors can be mapped, as well the type and frequency of interactions between actors during study phases. Interaction between the different actors may vary in terms of intensity, duration and so on. Building on the international literature, de Waal et al. (2013) identified five intensity levels, namely: informing, consulting, advising, co-creating and co-deciding as the main forms of interaction supporting the generation of knowledge. In general, informing refers to one-way communication in which no feedback is involved, while consulting implies two-way communication in which feedback is given. More interactive levels and formats may be related to individual study phases.

Social media as ‘place’ and interface of communication In order to include social media into the analytical framework, an interaction interface must be created. In the model used here, this social media interface is the virtual space where communication and interactions take ‘place’ (Figure 9.2). The social network that researchers decide to include in their study greatly affects the look-and-feel of this ‘place’, that is the functionality, the data use

139 …

Ron van Lammeren, Simone Theile, Boris Stemmer and Diedrich Bruns

„

Figure 9.2 Extended transdisciplinary framework with interaction typology (source: based on Louwsma et al. in prep.)

and the presentation of the interface. Specific tools are offered to users of social media that help to customise the look and feel and also, for example, to improve the media user presence in the network. The interface may have an impact on the role of the different actors in processes in which communication and interaction occur. The interface is not simply the technology that supports interactions between actors within, say, a research process. The interface is also where all the data generated are collected and stored. In

140 …

Social media „

other words, the social media is the ‘place’ that acts as big data supplier. These data can be used for many research purposes. It has been found that by employing so-called harvesting techniques, for example via application programming interfaces (APIs), researchers are able to collect user-generated data for the purposes of their study. Media communication is moulded by whatever media are used. The term ‘mediation’ has been adopted to describe media-based communication. Mediation refers to impacts of the ‘logic and form of any medium involved in the communication process’ (Altheide and Snow 1988, p.195). While personal communication is transferred, at least partly, to using media based services, processes of communication are changing people’s construction of reality.

Social media, social discourse and the ‘construction of reality’ To conceptualise how media users perceive and understand a ‘place’ that is positioned on, or ‘mediated’ by, a social media site, we can refer to social constructionist theory (Berger and Luckmann 1966; Gailing and Leibenath 2015). Before we discuss how space (and landscape) are ‘constructed’ we need to try to conceptualise how people communicate and interact in the ‘place’ that is a social media site. We can refer to discourse theory here (Richardson and Jenson 2003). Discourses are conceived of in at least two different ways. First, in social and cultural studies, a discourse is understood to be people using signs, words and other utterances that they relate to objects and practices in a contingent manner (Androutsopoulos 2013). A specific discourse may be referred to, in social and communication studies when meanings are socially, thematically and semantically contextualised. Examples of social contexts are ‘real’ communities such a city council but also ‘virtual’ communities such any social media network. The advance of information and communication technology is constantly adding new forms of utterance and, possibly, also new meaning. According to social constructionists (Berger and Luckmann 1966; Gailing and Leibenath 2015), all human experience is socially constructed. Place, space and landscape in particular (but also nature, heritage, wellbeing, etc.) are understood as ‘social constructions of reality’. As persons and groups interact in social systems, they create, over time, concepts or mental representations of their environment and of each other’s actions. These concepts eventually become, by the way of reciprocal interactions, ‘institutionalized’ (Berger and Luckmann 1966). ‘Institutionalism’ is a research approach that operates on the assumption that the analysis of social rules and regulations is vital for the understanding of social agency. Differences are made between formal and informal institutions. Formal institutions include sets of rules and regulations (e.g. laws or statutes). Informal institutions include traditions, customs, shared values, interrelated practices and routines (Gailing and Leibenath 2015) such as those found to occur in internet-based social, and thereby open, ‘collectives of dispersed individuals’ (Faraj et al. 2011, p.1224). Geo-social media can be understood as augmenting reality whereby perceptions of our surroundings are supplemented by computer-generated sensory input (such as Global Positioning System (GPS) data and visual graphics). Geo-social media thus comprise computer-aided constructions of space and landscape (Shelton et al. 2015). Social networks, social media and geo-social media in particular have advanced in ways that allow people to share ideas and images in ‘unparalleled scale and scope’ (Faraj et al. 2011, p.1224). Little is known, however, about what exactly the influence of using geo-social media and of social networking might be on the ‘social construction of reality’ in general, and of place, space and landscape in particular. Social and spatial sciences merge to develop a dialectical perspective in which the spatiality of social life is conceived of as simultaneously being a field of action and a basis for action (Lefebvre

141 …

Ron van Lammeren, Simone Theile, Boris Stemmer and Diedrich Bruns

„

1991). Treating society and space as a duality, rather than as a dualism (Giddens 1986), acknowledges that space is produced through human action (including, for example, using maps or GPS), and through social discourse (driving, for example, is a different experience whether navigating by paper maps or by GPS): ‘social life is both space-forming and space-contingent’ (Soja 1985, p.98). Findings from research investigating how relevant construction processes might be conceptualised and analysed suggest that ‘space’ is constructed in both non-institutionalised and institutionalised contexts (Richardson and Jensen 2003). Shared understandings and usages of such concepts are components of what Giddens (1986) has termed ‘practical consciousness’. Online technology is said to be pervasive among certain user groups (Evans-Cowley 2010), and internet-based knowledgesharing appears to follow particular rules (Faraj et al. 2011).

RESEARCH METHODS: THREE EXAMPLES OF APPLICATION Ostwürttemberg, Germany: landscape assessment using WebGIS technologies Assessing the definition of landscape quality by using social media was the objective of this study. The two questions ‘Do geo-social media help elucidating local knowledge that informs decisions on landscape quality at regional scale and if yes, how?’ and ‘Is data from geo-social media reliable for the definition of landscape quality?’ addressed this objective. The hypothesis was made that a database can be created which provides the basis for establishing which areas people perceive as landscape (local scale) and what they give value to in their surroundings (Bruns and Stemmer in prep.).

Study design With the purpose of including members of the public, a case of real-world landscape policy making was identified. The statutory landscape planning process for the Ostwürttemberg region in southwestern Germany was selected mainly because the competent planning authorities were interested in and prepared to generate the resources needed for taking part in a study on employing geosocial media. A WebGIS platform was established that fulfilled all the technical needs of the study, including a minimum level of interactivity. This platform was hosted and made publically available through the website of the regional planning authority with the aim to reach out to as many people as possible. The study was partly funded by federal and state agencies.

Data collection and sampling This ran from 21 July to 20 August 2014 while the website was online. Publicity was generated through a press conference and press releases, postcards, email and other kinds of announcements. A public event was held at the State Garden Exhibition from 17 September to 28 September 2014. Exhibition visitors were invited to make website entries using computers available on the exhibition grounds. When making an entry, people would draw one or more objects (a polygon, a line, a point or a vista) and they would add written comments such as ‘this is where I go walking the dog after work’. Using a categorised system, people could choose one or more locations and enter a site or an area into the website, and describe them in a narrative way. Geographic information includes, for example, bathing spot, hiking trail and so on. People could also choose from categories such as

142 …

Social media „

Figure 9.3 Detail of a map of public input to landscape assessment with descriptive texts

‘favourite places / areas’, ‘historic / heritage sites, areas’, ‘leisure and sport sites / areas’, ‘sightseeing locations’ and several others. The descriptions that people made with each entry included personal perceptions and values, as well as activities, things that people usually do when visiting a specific site or area (Figure 9.3). A total of 280 entries were registered in which people had posted data and information about places that they feel give value to their surroundings. In particular, they had identified and marked their most favourite places. While entries had been made from local perspectives, they were collected for the entire Ostwürttemberg region.

Data analysis People gave a short description in written form and a name identifying the drawn object. Both were stored in the GIS together with the geometry of geographic locations (point, line or area). For data analysis purposes, geometric data were first grouped and clustered according to location. In the next step, by following the geographic distribution of clustered entries, regionally relevant hotspots of public interest were identified. Third, people’s explanations (narratives) were subjected to systematic and qualitative content analysis. Analysis of location-specific comments provided information about the reasons why participants chose one or other spot or area as being interesting (‘a spectacular view from up there!’), and also some understanding and explanation of how and why preference and value judgments were made (‘all of the village meets here after work’, ‘this is the spot for our Easter Fire’). The areas identified as special places and as highly valued landscapes were not equally distributed across the region. The kind of evidence needed to respond to the two research questions above were successfully obtained. Pieces of local knowledge were provided in sufficiently large numbers to paste together

143 …

Ron van Lammeren, Simone Theile, Boris Stemmer and Diedrich Bruns

„

a mosaic in which regional hotspots became visible. In addition, it is important to note how information was generated, also for the first time at regional scale, about intangible landscape preferences. Narratives included descriptions and names, reasons for choosing a particular place as important and so on. Together, all the website entries pointed at areas that were particularly cherished, making them relevant beyond local interest, something that would have been impossible without the use of the online platform. Similarly, as any object or entry could be sorted according to one of the predefined categories mentioned above, and, by clustering entries made by individuals without having had personal contact with others, researchers could learn which activities many people commonly engage in and whether or not people share common interests and values. By referring to the analytical framework above, we note that the website called mitmachenostwürttemberg.de (mitmachen means to participate) served as an ‘interface’ between ‘actors’ from public agencies, research institutions and local communities. We discovered that, for any engaging of non-experts in a landscape study to be successful, it is critical that attractive, interesting and userfriendly invitations are extended. For ethical reasons it was important to assure participants that website entries would be kept anonymous. People knowing that they would not personally be exposed through the entries they made on a public website somewhat limited the extent by which entries could be analysed. Comparing entry data with social data might have revealed more valuable information. Another factor limiting analytical ambitions were predefined categories, such as landscape types and catalogues of landscape features, that a participant could choose from while making entries. These categories, selected in order to be able eventually to make public assessment results compatible with official planning documents, include ‘site for swimming’, ‘hiking trail’, ‘cultural heritage site’ and so on. Having to choose from a set of fixed categories may narrow down options for describing landscapes. Participants complied in making selections from the predefined categories. However, in their descriptions of sites and areas people mixed content that pertained to different categories. For producing maps the use of categories was very helpful but for conducting content analysis they were only partly helpful. Constructivist landscape theory (Gailing and Leibenath 2015) provided the conceptual framework for analysing content that members of the public posted on the website and, by using evidence collected via social media, this theory helped in understanding shared landscape values. For example, one entry explains how ‘Keuerstadt is very important for me; it is a very special landscape. I like its remoteness. I like to walk along the creek and see rare plants and old trees. I also know how the chapel is important for the history of our region.’ From a constructivist perspective this narration includes content pertaining to the following categories: (a) physical (material) features, such as natural (water, plants) and cultural (building) landscape elements; (b) activities (walking, watching) that people engage in, taken as expressions of how people take ownership of specific sites and areas; (c) emotional linkage (‘very special’, ‘important’), taken as expressions for symbolic landscapes, and of identity and place making. Maps and information that result from public involvement using geo-social media in a way explained above are considered, in this chapter, as expressions of which locations and areas are mentally constructed, by local people, as attractive and valuable landscape. The results obtained from the Ostwürttemberg study also provide evidence about how sites and areas are perceived, for example whether natural or anthropogenic (man-made). At the beginning of the study it was hypothesised that not only preferences for certain places but also landscape perceptions and values would be shared among several study participants, while some site-specific preferences would remain highly individual. Geo-social media helped our understanding of how and to what extent these hypothesis are true.

144 …

Social media „

Zuid-Holland, the Netherlands: zonal scheme development using social media as text interface With the aim to create zoning scheme sketches (ruimtelijke structuurvisies in Dutch) two municipalities in the province of Zuid-Holland in the south-west of the Netherlands took a participatory approach. From November 2010 until April 2011 they ran pilot projects, called ‘Qwiek’ and ‘City of Tomorrow’ respectively, using the ‘text options’ of social media. The social mediagenerated results of both projects led to some interesting research questions, including ‘What geo-information is present in texts exchanged by social media during a participatory planning process?’ and ‘Does this geo-information show successive stages of planning, even co-creation’? These questions were addressed by Eikelenboom (2012). The two projects used existing social media platforms, such as Twitter, Facebook and LinkedIn. Residents were invited to share their opinions and ideas on the future of their municipalities through one or more channels: participation evenings, workshops, a dedicated website and the social media. In April and May 2012 the results were discussed by the municipal council. During this meeting, the public could take part, also by using social media. During the full process the organisers of the social media channel enabled the process by posing statements.

Study design It was hypothesised that the shared messages of LinkedIn, Twitter and Facebook, which consist of all tweets, comments and posts, supported the participatory planning process and were able to ‘harvest’ and store the messages in databases for further analysis. The messages expressed the ‘knowledge’ of the participants in a restricted text format, for example by a maximum of 140 characters in a Tweet. These expressions showed abbreviations, letter–number combinations and emoticons which complicated the understanding and interpretation of the content. In addition, messages conveyed many different semantics such as trusts, beliefs and preferences, forwarding facts, reasons, opinions, emotions and decisions. For that reason rules were developed to classify and to geocode the messages. It was expected that the classified and geocoded messages would be related to a timeline and to related stages of the planning process. This was expected to provide analysis of the development of location-based semantic meanings during the planning process. It was also expected that this approach might give an insight into how the knowledge was constructed. Given these assumptions, the study aimed to find three key performance indicators of the planning process: the number and frequency of distinctive toponyms and their spatial scale, the shared geo-located semantics and the characteristics and number of social media participants. Toponym in this study is loosely defined as place name or a word coined in association with the name of a place.

Methods and data analysis In total 4588 messages from Twitter archives, relevant LinkedIn and Facebook discussion lists were harvested, stored, semantically classified and geocoded by toponyms. The count of toponyms was done in monthly intervals due to the frequency of messages contributed by the participants. This count showed the relative importance of locations. Time graphs of seven and twelve intervals gave

145 …

Ron van Lammeren, Simone Theile, Boris Stemmer and Diedrich Bruns

„

an indication of location-based semantics development. In addition, developments in importance as a result of the evolution of the discussion were analysed by usage count per interval. The semantics of the messages were extracted using six categories: fact, question, opinion, idea, event or report. These categories were partly based on Pang and Lee (2008) who made a distinction between objective information, providing verifiable facts about a spatial plan, and more subjective information, giving opinions or sentiments. Subjective information regarding spatial plans can also include proposals for alternative developments. Events and reports were related to the planning logistics and the other categories related to the spatial planning phases. Post-hoc interviews with the organisers of the social media intended to validate the found results.

Results The study showed that 26 per cent of all messages included a toponym. Detected toponyms in the messages of the Twitter archives of the two projects varied between 16 per cent and 26 per cent. In the LinkedIn data it varied between 67 per cent and 71 per cent. The Twitter archives presented 30 per cent (89) and 26 per cent (180) unique toponyms respectively and LinkedIn 96 per cent (56) and 132 per cent (94). The names of the municipalities and their populated places were the most used toponyms. However, per project differences were detected. In the Qwiek project messages named the main cities nearby and the main infrastructure that connects the municipality with these cities. In the City of Tomorrow project more detailed locations in the area were mentioned, including the intended section of a new by-pass road. In general, populated places, infrastructures and major areas showed toponyms at a regional scale. In both Twitter databases one or more buildings were often named. These were actually the venues of the face-to-face participatory planning meetings and not the subjects of zoning discussions. Messages mostly discussed relatively large geographic areas, which could be explained by the character of the planning objective being to sketch zoning schemes for the total area of the municipalities. The semantic analysis of the messages (Table 9.1) showed that 53 per cent supported the planning process. The other 47 per cent of referred to events (38 per cent) and reports (9 per cent) which were only beneficial to the logistics of the process. The number and characteristics of information sharers showed that only a small number of people joined in the process. The top ten users shared between 47 per cent (City of Tomorrow tweets) and Table 9.1 Semantic analysis of the messages

146 …

Social media „

91 per cent (Qwiek posts) of the messages. A much larger share of users contributed only once. The top ten users consisted mainly of the actors involved in the organisation of the project. The number of stakeholders including the general public was limited. The ones who did contribute were seriously committed to making this approach a success. In the Qwiek project participants mainly shared facts and opinions about the current situation of their surroundings. In the City of Tomorrow project a number of people launched ideas about the allocation of future land use options which seemed more on the level of co-production.

Discussion and conclusion Deriving toponyms from messages and especially geotagging them posed challenges because of the ambiguous nature of place names. Abbreviations and the use of the same toponym for different locations showed to be an important source of error. This can be both coincidental, for example a place named after a common landscape feature or an important person, and intentional because, for example, during colonisation places get named after each other. Automated procedures to derive toponyms by using dedicated software and databases such as GeoNames were available. In this study, and for the sake of evaluating toponyms, they were derived manually. The geographic scale was explored by matching the dominating toponyms to the simple geographic ontology of GeoNames (http://www.geonames.org/). Alternative occurrences of locational descriptions were found by hashtags (#), telephone codes and uncommon abbreviations. Timeline graphs in this study showed that almost all messages were found close to, before and during organised events. For that reason toponym development seemed strongly influenced by the face-to-face meetings and turned out to be of very limited value. New toponyms that gave evidence for co-creation were rare, which is of course an issue when using automated procedures. Participants seemed to prefer the traditional written and/or colloquial toponyms rather than abbreviations and informal alternatives such as telephone codes. However, due to the informal nature of social media, and in Twitter’s case its limited text space, it was expected that more people would use shorter text expressions. According to the post-hoc interviews, the lack of public involvement could be explained by the short time span for preparing the social media role and the way it had been announced. The study showed that messages conveyed geo-located semantics in a participatory planning process. Analysing such messages as stored in social media archives gave insight in the ongoing planning process including the development of the intended design. Activities and geo-located semantics of each participant at a certain phase of the planning process could be tracked and studied by the timeline graph. The semantic analysis of geolocated messages into facts, questions, opinions and ideas linked to a phase of the timeline helped to give an impression of the level of participation and the roles of participants. In this respect more recent studies have showed the power of social media data as a result of the significant increase in smartphone use and the extension of semantics, for example by analysing emotions (e.g. Dunkel 2015).

Research challenges It seems that it is a considerable challenge to create strategies that increase awareness and interest to the level at which people really like to participate in planning endeavours. Participants need

147 …

Ron van Lammeren, Simone Theile, Boris Stemmer and Diedrich Bruns

„

to be made aware of how much their input will be a serious part of the discussion, for example during the analysis and presentations of time graphs and web maps which show toponyms and semantics. Similar ideas are mentioned by, for example, Sakaki et al. (2010) and Dunkel (2015) who also refer to the experiences of people’s immediate surroundings. Web maps may stimulate participation (Dunkel 2015; Kahila-Tani et al. 2016). Participants must agree upon the way in which social media are harvested and used. Related to that agreement is the mode by which the data are exchanged and shared during the planning process that can also be used for analysis. The following questions are still of interest. How can public involvement be increased, such as in the number of entries? How can the link between location and semantic load be improved in order to support co-creation through the use of messages? What semantic classes may benefit from a better text understanding? How can a toponym or semantic cloud map be combined (Ahern et al. 2007; Hahmann and Burghardt 2011) with a timeline graph to give a more real-time and better understanding of the participatory planning progress?

Amsterdam, the Netherlands: analysing tourist behaviour by social media posted photos Cities struggle with the number of visitors, which may cause overcrowding, high noise levels, litter and dangerous traffic situations. To tackle these problems a study analysing tourists’ whereabouts was carried out (van der Drift 2015). Inspired by studies on geospatial webs (Haklay et al. 2008) and on location-based social networks (Roick and Heuser 2013) new ways of exploring tourism dynamics in a city were developed. This study started with the question of which parts of the city are the most visited and whether uploaded photos on social media could reveal these tourist preferences. It was assumed that the metadata (tags) of photos in the cloud are important sources for analysis which may help to find spatio-temporal patterns. This metadata presents among other things camera settings, the date and time when the photo was taken and geotags. The number of photos on or near a road is an indicator of the popularity of a route (Sun et al. 2015). Knowing that a route can be constructed out of a chronological sequence of locations, such geotagged data would make it possible to create routes and to analyse route densities in order to locate the most crowded streets and squares of a city. This way of analysing densities favours the notion that tourist routes in a city are not the shortest routes. Due to their daily behaviour, given different interests like dwelling, shopping and visiting cultural places, the distance and location of routes is based on the daily combination of such interests.

Study design The method selected was the harvesting and subsequent analysis of geotagged photos and metadata of tourists visiting Amsterdam, obtained from the social media platform Flickr. Harvested photos were classified by unique user accounts in two classes, respectively photos taken by tourists and those taken by residents. To validate the outcomes of this classification method a confusion matrix on location tags of fifty tourists and fifty residents randomly picked from the set with classified users was used. Afterwards the two classes of photos were compared by location and time. Validation of classified photos was done by finding photos that gave evidence for the correctness of time and location tags. To detect major tourist hotspots in time and place the photos were clustered by using the density-based clustering algorithm DBSCAN (Scikit-learn 2015). Routes

148 …

Social media „

were created by network construction based on photo geotags per user (unique Flickr account) in order to detect the individual tourist’s movement. All routes were linked to the public spaces with the highest densities of routes. Hotspots and route densities gave an overview of tourist pressure on the urban fabric. The seventeen top tourist locations of Amsterdam ranked by Lonely Planet were used to validate the identified hotspots and an expert validation was performed to validate the calculated routes.

Data collection and analysis During five weeks in 2015, the metadata of 2,849,261 geotagged photos was harvested from Flickr and stored in a spatial database. From this data set 393,828 photos were located in Amsterdam. Photo locations in Flickr are either automatically captured by the camera or manually specified by placing the image on a map. Flickr automatically adds an accuracy value for all photos with a geographical location. The platform bases this value on the zoom level of the map when the image is geotagged. Values range between world (1), country (3), region (6), city (11) and street level (16). Many users marked the photos with various other forms of information including a title, description and a wide variety of textual tags. Visual analytics (Andrienko et al. 2010) were applied by analysing the tagged photos in space (spatial hotspots) and time (temporal hotspots). The analysis included distribution and clustering techniques. The temporal distribution of photographers (tourist and locals) was analysed for three granularities: the days of the week, months of the year and days of the year. For the detection of hotspots and landmarks spatial clustering methods were applied. The most probable routes of tourists between subsequent photo locations were calculated by an enhanced network construction approach which included a Euclidean distance calculation. Out of the labelled segments of the networks a route density map was derived. Finally, a qualitative approach was used to validate the study outcomes by interviewing eight tourism experts.

Results The classification showed an average of 24.6 photos per user, unique Flickr account, for the harvested photos (Table 9.2). A manual check of the Flickr profile pages which contain albums, a photo stream, textual expressions and other indicators gave evidence of the 98 per cent correct classification of tourists. The clustering techniques identified thirteen hotspots. The identified hotspots and tourist route densities are illustrated in Figure 9.4. Table 9.2 Distribution of the harvested photos

149 …

Ron van Lammeren, Simone Theile, Boris Stemmer and Diedrich Bruns

„

Figure 9.4 Identified hotspots (above) and tourist route densities (below) (source: van der Drift 2015)

150 …

Social media „

Table 9.3 Overview of invalid Flickr photos

Discussion Not all photos were selected because of certain criteria (see Table 9.3). The spatial accuracy of street level (so-called level 16) is an especially interesting issue. In this research only data from Flickr users who shared their photographs with the whole Flickr community were used. Most of these were tourists from North America. It seems advisable, for future studies, to enlarge the photo data set with photos from social media which include users from a wider area in order to gain a better representation of all tourists. The timestamps of photos in relation to time zone settings are critical, especially in case of temporal studies. The use of different time intervals (year, month, day, hour) in the analysis supported different spatial interests, for example to find seasonal differences in behaviours. Spatial precision given by the tags could have a serious impact on the understanding of the spatial distribution and density of locations. Figure 9.5 shows the effect of spatial granularity on the level of detail. The calculated routes were validated by experts. These experts, working in the tourism department of the city of Amsterdam, agreed upon these routes. They suggested that including the number of tourists that use routes in a certain direction would support further interpretation of spatial density.

Conclusions The approach presented here might help to gain a better understanding of the daily use of the urban landscape based on large data sets of in-situ photographs. In this study the example of tourist photos was chosen. The semi-automatic classification method classified 39.1 per cent of the users as tourists to a very high precision (98 per cent). During the research the thirteen hotspots identified were compared with the seventeen top tourist locations ranked by Lonely Planet. Four hotspots, mainly nodes for travelling and food, were not mentioned by Lonely Planet. The validation by the tourism experts showed a high agreement of the detected spatial clusters and route density map with expert knowledge. Despite several imperfections in the geo-social data, meaningful insights into the spatial and temporal patterns of tourists (spatial densities of main locations and routes) could be derived from the geo-social media metadata. Such insights may offer a valuable addition to planning and design research tools for understanding better the dynamics of human behaviour on a local urban scale. The approach of the example ‘what photo – what route’ might support understanding the communication format of the photo as used in social media platforms. It also shows the potential power of geo-social media metadata for studying spatial patterns. Although it is very attractive to link human preferences to densities of photo locations, the data as such do not offer such possibilities

151 …

Ron van Lammeren, Simone Theile, Boris Stemmer and Diedrich Bruns

„

Figure 9.5 Effect of spatial granularity on level of detai (source: van der Drift 2015)

yet. From this study we can see that photos on social media may have potential for spatial and temporal pattern analysis of certain target groups by studying the metadata of geotagged photos. Most of the analytical tools to do this already exist. Validation of the time stamps of photos needs further attention.

152 …

Social media „

Research challenges Harvested data tends to be demographically skewed and many records will have spatial and temporal inaccuracies. At the moment, validation strategies are needed to address this issue. In the long run larger amounts of data may solve this problem. The metadata of the geotagged photos from the main data source alone will not give evidence as to why the photos were taken. One question remaining is: if a photo shows a location of personal preference is this proof of a place being a touristic highlight as advertised by the tourist office and travel books? If so, has the highlight actually been visited, or has the entry been made into a pictorial diary during a holiday trip? In general an additional survey is needed to find the normative load of this geo-social data. In contrast to social media as the environment of a participatory planning process, these types of social media gained a lot of data especially in relation to locations of great touristic interest. However, harvesting a sufficient amount of geotagged photos of locations that present issues like ‘unpleasant place’ are harder to come by using social media. To gather context-dedicated photos is a challenge on its own.

METHODS FOR STUDYING SOCIAL MEDIA IMPACTS This section gives an overview of analytical approaches and methods that researchers might use better to understand and explain online activities, including impacts that social media have on planning and design and on relevant decision making. Referring to the examples above, types of analysis and underlying concepts are addressed first and the workflow of data tracking, monitoring and so on is then explained. As forms of data analysis social media analysis, social network analysis (SNA), semantic network analysis and spatial-temporal pattern analysis are important in various ways. Social media analytics (SMA) helps ‘collecting, monitoring, analysing, summarizing and visualizing information from Social Media’ (Stieglitz and Dang-Xuan 2012, p.1282), including a large amount of data from within social media (see ‘Big data’ above). Who (people, organisations, institutions) are linked via social media and who are opinion leaders? To find answers to this question it would be appropriate to analyse media and network structures and to identify key users using SNA that would help to identify how central some users are to specific communication processes, and the extent to which individual users reach out into communities (Stieglitz and Dang-Xuan 2012). Text analysis and trend analysis may be linked to SMA. One important part of text analysis is called ‘sentiment analyses’, or ‘opinion mining’. Running a SNA would be useful, for example, to identify opinion leaders and relevant user communities (Stieglitz et al. 2014). Conceptually, the computer sciences have borrowed heavily from a number of different disciplines. For example, graph theory (applied in network analysis) and network theory have been adapted to help to understand content and activities occurring in social media (Stieglitz and Dang-Xuan 2012). Data analysis may include clustering correlations and regressions for data on different scales; such tools help to make use of data mined from social media. As illustrated in the examples above, the researcher first decides which ‘object’ to study, for example Twitter, Facebook, (public) weblog and so on, or a combination of these. Social media sites and platforms offer a range of tracking sources that might be used. With Twitter these sources include public tweets; with Facebook the content of the ‘wall’ may be mined, including commentaries. Access rights are different with every media and these rights must be observed. In order to track data the researcher usually needs to access

153 …

Ron van Lammeren, Simone Theile, Boris Stemmer and Diedrich Bruns

„

interfaces, and not all services offer such access. Most blogs do not offer interfaces but their RSS feeds might be tracked instead. Twitter and Facebook offer APIs but, due to special configurations, researchers might be challenged by restrictions and constant changes. One may encounter temporal restrictions and the amount of downloadable data might be limited. If one wanted, for example, to identify the moment of the highest tide in public outcry about a bridge proposed to be built in a designated heritage landscape, it might be necessary to jump on the train while it is running, so to speak. If searching for data at a later date the researcher might find that social media services are not run like libraries where sources are reliably stored. If, for example, services like Twitter and Facebook were to be accessed, it would be advisable to download data regularly, as was done in the Amsterdam tourism study. Data tracking and data collecting takes place either by monitoring, when researchers are in control of the social media application, or by harvesting, if researchers are tracking data of interest (web search). Systematic sampling depends on the research question. When using APIs and RSS feeds, predefined search words or key words pertaining to specific themes, organisations, projects and so on related to the research question would be necessary. Stored data may then be analysed according to structure, to thematic or action related interest, according to opinion and sentiment and so on (Stieglitz and Dang-Xuan 2012). In those cases where large amounts of data are available, automated quantitative methods of content analysis or text mining would be most suitable. Advanced forms of learning programs are also being developed which are based on ‘machine learning’ and include tools used for semantic network analysis. Based on ‘pre-processed’ data, further processing may include text analysis, content analysis, sentiment analysis and so on. A number of text analysis and SNA tools are commercially available that non-experts with little or no programming experience may use. Stieglitz and Dang-Xuan (2012) have compiled a list of text mining and automated content analysis tools including, for example, WordStat (add-on for QDA Miner) and Linguistic Inquiry and Word Count (LIWC). Tools for text mining and manual content analysis include Atlas.ti and QDA Miner. By applying certain algorithms, automated tools might be used to identify specific opinions and sentiments that are being uttered in social media about individual projects and topics. For sentiment analysis and opinion mining, sentiment attributes are linked to words that people are using. For mining purposes words are identified and subjected to automated analysis. Sentiment attributes are dictionary-based classifications of sentiment orientation including polarity (positive, negative, neutral) and strength. Online communications still offer challenges including the use of emoticons and acronyms. Sarcasm and irony cannot be easily understood by machines. Currently, manual analysis is recommended, as illustrated in the Amsterdam example, as is the inclusion of background information for interpreting sentiment from social media data. The volume and velocity of data will define the limits as to which manual analysis is still possible. Examples for opinion mining and sentiment analysis include tools like SentiStrength and SentiWordNet. Finally, for SNA, tools such as Gephi, NodeXL (Excel plugin), UCINET, Pajek, Netvizz and others are available. By using these methods and tools it would be possible to understand online specifics, for example while a stakeholder analysis is being conducted in preparation for a participatory process. An open-access tool such as NodeXL or Gephi might be used to study how some of the main stakeholder groups are behaving, mainly online, regarding individual projects.

154 …

Social media „

DISCUSSION Expectations, efforts, gains and satisfaction are all related to trust, and trust is important for researchers to assess and learn how their study will yield reliable and solid results. Unlike controlled experiments, studies employing social media may yield inconsistent and unreliable results. In addition, there are challenges from the large amount of data that social media help to collect in different formats. Data originate from individuals who intend to exchange and share this data. The semantics of these data range from beliefs to facts and are mainly based on perception. In a study using social media one needs to be able to check how reliably one finds the same outcome every time one uses specific sites. We might find particular social media to be reliable, meaning that they give a good performance for extracting data time after time, while others may turn out to be less reliable. What can be regarded as accurately describing the virtual or the real world can also be regarded as valid. Validity defines the strength of the study results. The challenge is how to extract knowledge from sampling and analysing data and, by trying to explain what is being observed on social media, to use such findings to start understanding what is going on in any ‘data storm’ that might be brewing. Before discussing the validity and value of social media data we must first consider the challenges of the other big ‘v’s of big data: volume, velocity and variety. Volume refers to the challenge of dealing with large sizes. When purposefully employing social media we expect to generate more data and thus extract more information compared with doing research without social media. Regarding social media as data source, the amount of shared information expressed on social media grows. Almost all the data represents private perceptions which are time and space tagged. Thus geo-social media providing data in very large amounts offer new dimensions for doing research. Handling such large volumes may be demanding and calls for more efficient tools for collecting, storing and retrieving data, updating it when required as well as sampling and analysing it. Velocity refers to the challenge of watching and analysing data flows (sometimes in near real time). Researchers need to deal with the pace at which data flows and be able to make use of voluminous data streams which never stop. Looking again at process first, it can be seen how involvement and participation via social media is closely related to specific events (see the Ostwürttemberg, ZuidHolland and Amsterdam examples). Events appear to trigger a social media activity momentum with, most of the time, a short lifespan of exchange and sharing, given the number of active participants. To be able to explain process peculiarities, turning to the social sciences and the humanities might be helpful, because analysing different process stages might yield important results: who is engaging in interactions, what opinion development may be observed and, with a view to practice, what planning and design development (e.g. in the example of Zuid-Holland) might find the greatest support? In order to understand what has been termed ‘knowledge collaboration’ in online communities (OCs) we must remember that OC-based communication ‘takes place despite the absence of existing social relationships’ and ‘without the structural mechanism traditionally associated with knowledge collaboration in organizational teams: stable membership, convergence after divergence, repeated people-to-people interactions, goal sharing, and feelings of interdependence among group members. There can even be total absence of existing social relationships’ (Faraj et al. 2011, p.1225). Due to the velocity and ongoing activities on social media platforms, data might be obtained over periods of time that would otherwise require considerable resources to be generated in order to conduct longitudinal studies, as can be seen from the examples above. Crucial for this

155 …

Ron van Lammeren, Simone Theile, Boris Stemmer and Diedrich Bruns

„

is the speed at which flows can be accessed, particularly since streams may offer nearly real time information. Technically this is not easy to do. Variety refers to the challenge of dealing with many sources and types of data, both structured and unstructured. There is great variation in data content and format. It is necessary to be able to discern amongst many different configurations of technology and to assess the value. These potentially include millions of wearable wireless health monitors, billions of hours of videos, unimaginable numbers of different pieces of content generated by billions of users of social media network providers and so forth. During any research the variety of data can be purposely predefined according to a certain format (such as text, video, etc.). By tapping into social networks one could find how the communication formats of social media and micro-blogging styles differ from one another. The development of planning content may be limited to a certain level (see the Ostwürttemberg and Zuid-Holland examples), and different phases of the process may deliver different formats and differences in content. Knowing that data formats may vary greatly, finding associations between different data ensembles is one of the greatest challenges of big data exploration (see the Amsterdam example). There are social media data (based on experiences and perception) and measured data (collected by sensors and other calibrated measurement technologies). Combinations of both might enable researchers to relate data types to each other and to identify social–temporal patterns. Returning now to validity and veracity we should consider the criteria applied to ascertain the quality of research. All researchers need to be able to check data for quality and for their usability to generate plausible results. The question is, how much of the data collected and sampled is correct, accurate and trustworthy for the intended use? During the process, participants’ responses may be monitored and a ‘lineage’ of arguments found as well as a certain consistency in content development through SNA. The main actors and relations between them can also be analysed. The ability to ‘read, see, or listen’ (Evans-Cowley and Griffin 2011) to what, say, citizens are contributing about their community, offers an opportunity to engage the public into a research project. Another point worth considering regarding validity is bias. In the Ostwürttemberg and ZuidHolland examples ongoing land-use changes and projects might have influenced people’s landscape awareness, and attention may, at least with some entries made on websites and social media platforms, be directed more towards some areas and less to others. When observing social network activities one may need to watch for demographic or other social bias. Friendship networks and other user groups tend to be demographically highly segmented (Evans-Cowley 2010) as people link with others of similar demographic format, status and class, belief and value system and so on. This principle, described as ‘homophily in social networks’, has as much an implication for the information people receive and the interactions they experience (McPherson et al. 2001) as the so called ‘filter bubble’ (Pariser 2011), the ‘echo chamber’ effect (Pfeffer and Zorbach 2015) and ‘power-law distribution’ (Barabasi et al. 2000), that is people will link with those who are well known, and those who are not or less well known receive, even when trying hard, few or no links in return. Considering how friends mainly communicate with friends, and how people seldom break out of their groups, one question is, what it would it take to engage with groups of people with whom they have never interacted before? Trying to find answers to such questions means exploring new territory and developing innovative study designs. Validation of data is possible by comparing social media-derived data with measured data (for example by sensors and other calibrated measurement technologies). In the Zuid-Holland examples the words and codes used to describe locations could be controlled by an online gazetteer. However,

156 …

Social media „

the gazetteer will not check the geocoding. Validating metadata (including geo- and time-tags) is supported by finding the origin of the social media data, which may help to ascertain accuracy. In the Amsterdam example the time-tags have been validated by making use of time references (e.g. regional time settings). It is well known that the geotag is often inaccurate or even missing due to the information creator’s habits such as alternating being on or offline, related data uploading strategies in relation to the exact location of the device (Agichtein et al. 2008) and the expertise of the creator (Elwood 2008). In studies where local people provide knowledge about intangible landscape values, such as their place attachment, memory and identity, how and to what extent citizen science-produced data sets are subject to error is still poorly understood (Dickinson et al. 2010). One issue that needs careful consideration is how the use of ‘collective intelligence’ and the ‘wisdom of crowds’ (Surowiecki 2005) might reliably and consistently lead to generalisable results. Another issue is how data and information might be checked for being reliable, that is accurate, true and honestly provided. The spread of rumours and false information, whether this is done on purpose or not, can also be an issue (Mendoza et al. 2010). When certain tasks are assigned to a large volume of people, as happens in citizen science projects, it is important to consider how the inclusion of non-scholars in academic research might be explored for their knowledge-building potential and also assessed for their scientific reliability, consistency and validity (Ulrich 2000). At the same time, validation might, again, be organised and enriched by crowdsourcing processes (Wechsler 2014). Finally, we discuss the last of the big ‘v’s, which is value. We also return to the disruptive character of social media for landscape research. Regarding social media and big data, having many small and distinct parts, granularity in time and space becomes a serious subject to consider in pattern analysis. Granularity may strongly influence the discussion about scaling and scale levels and it may also lend a new dimension to the idea of incrementalism used to model planning processes. In addition, the impact of large numbers may show, in the long run, a normal distribution of data, which means that statistical confidence will improve. Thirdly, locations of expression become a commodity. The precision and accuracy of such locational expression still requires more quality control.

CONCLUSIONS AND OUTLOOK Expectations are high regarding the knowledge that may be gained through research using social media. Researchers hope, for example, to generate and collect valuable data by purposefully establishing online platforms. We also hope to find information and knowledge by mining the internet. It is advisable to look for the many pitfalls hidden below the seemingly shiny surface of social media studies, some of which we have pointed out in this chapter. Mainly, we encourage researchers to use social media in their studies. Advantages include running internet-based surveys that allow researchers to collect landscape data, to undertake locality-based studies of human experiences and everyday behaviour or to investigate web participation in planning and design processes (Eräranta et al. 2015). Much would also be gained if we understood how public opinion develops in social media: how are tides of emotions rising and ebbing? Studying the content of social media sites and the interfaces between different actors adds new dimensions to understanding how opinion changes occur. Since special fields of study are developing around social media and around the approaches and methods we have been discussing here, we conclude by making some observations and presenting an

157 …

Ron van Lammeren, Simone Theile, Boris Stemmer and Diedrich Bruns

„

outlook into how we see these fields developing. Traditionally, landscape architecture research often focuses on the case study. As more and more cases are being studied and the demand for evidence is growing, temporal and human dimensions are making inroads into landscape architecture research. We have arrived at a situation where we are studying, in concert, the who, the what, the where and the when. At this point the role of social media may be seen as disruptive for data collection, analysis and visualisation. A large proportion of our communication and transactions are internet and social media based. All relevant data is stored at this ‘interface’. New research domains like data sciences are developing frameworks and technologies to interrelate and associate these data in great varieties of formats. By linked-data strategies it becomes possible to explore data patterns, and to enhance information about spatial–temporal behaviour, preferences and beliefs. Developing such frameworks cannot be done without expertise from different scientific domains coming together, including landscape architecture. The availability and accessibility of these frameworks will become subject to political debates, including those on open access versus proprietary nexus, and those regarding knowledge representation through certain data and tools. For any field and discipline concerned with a common good, such as landscape, these are important debates. There are many questions that need addressing. While employing social media as a means of data collection does not guarantee better results, it may enable researchers to derive new insights by delving deeply into unexpected areas where valuable data are to be found. Like it once happened through advances in cartography, for example, current social media developments will lead towards new insights and discourses including those on citizen science, expanded notions of trans-disciplinarity and the question of how democracy gains are observed through using social media in research.

SUGGESTED FURTHER READING de Waal, R., Kempenaar, A., van Lammeren, R. and Stremke, S. (2013) ‘Application of social media in a regional design competition: A case study in the Netherlands’, Proceedings Digital Landscape Architecture 2013, 186–199. Dunkel, A. (2015) ‘Visualizing the perceived environment using crowdsourced photo geodata’, Landscape and Urban Planning, 142, 173–186. Kahila-Tani, M., Broberg, A., Kyttä, M. and Tyger, T. (2016) ‘Let the Citizens Map—Public Participation GIS as a Planning Support System in the Helsinki Master Plan Process’, Planning Practice & Research, 31(2), 195–214. Roick, O. and Heuser, S. (2013) ‘Location based social networks – Definition, current state of the art and research agenda’, Transactions in GIS, 17(5), 763–784. Stefanidis, A., Crooks, A. and Radzikowski, J. (2013) ‘Harvesting ambient geospatial information from social media feeds’, GeoJournal, 78(2), 319–338.

REFERENCES Agichtein, E., Castillo, C., Donato, D., Gionis, A. and Mishne, G. (2008) ‘Finding high-quality content in social media’, Proceedings of the International Conference on Web Search and Web Data Mining (WSDM’08), 183–194. Ahern, S., Naaman, M., Nair, R. and Hui-I Yang, J. (2007) ‘World Explorer: Visualizing aggregate data from unstructured text in geo-referenced collections’, Proceedings of the 7th ACM/IEEE-CS Joint Conference on Digital Libraries, 1–10. Altheide, D. and Snow, R.P (1988) ‘Toward a theory of mediatization’, in Anderson, J.A., ed. Communication Yearbook (Vol. 11), New York: Sage, 194–223.

158 …

Social media „

Andrienko, G., Andrienko, A., Demsar, U., Dransch, D., Dykes, J., Fabrikant, S., Jern, M., Kraak, M., Schumannh, H. and Tominski, C. (2010) ‘Space, time and visual analytics’, International Journal of Geographical Information Science, 24(10), 1577–1600. Androutsopoulos, J. (2013) ‘Participatory culture and metalinguistic discourse: Performing and negotiating German dialects on YouTube’, in Tannen, D. and Trester, A.M., eds. Discourse 2.0: Language and New Media, Washington, DC: Georgetown University Press, 47–71. Barabasi, A-L., Albert, R. and Jeong, H. (2000) ‘Scale-free characteristics of random networks: The topology of the world-wide web’, Physica A: Statistical Mechanics and its Applications, 281(1–4), 69–77. Berger, P.L. and Luckmann, T. (1966) The Social Construction of Reality: A Treatise in the Sociology of Knowledge, Garden City, NY: Anchor Books. Bruns, D. and Stemmer, B. (in prep.) ‘Landscape assessment in Germany’, in Fairclough, G., Sarlöv-Herlin, I. and Swanwick, C., eds. Routledge Handbook of Landscape Character Assessment, Abingdon: Routledge. de Mauro, A., Greco, M. and Grimaldi, M. (2015) ‘What is big data? A consensual definition and a review of key research topics’, AIP Conference Proceedings, 1644, 97–104. de Waal, R., Kempenaar, A., van Lammeren, R. and Stremke, S. (2013) ‘Application of social media in a regional design competition: A case study in the Netherlands’, Proceedings Digital Landscape Architecture 2013, 186–199. Dhar, V. (2013) ‘Data science and prediction’, Communications of the ACM, 56(12), 64–73. Dhar, V., Jarke, M. and Laartz, J. (2014) ‘Big data’, Business & Information Systems Engineering, 6(5), 257–259. Dickinson, J.L., Zuckerberg, B. and Bonter, D.N. (2010) ‘Citizen science as an ecological research tool: Challenges and benefits’, Annual Review of Ecology, Evolution, and Systematics, 41, 149–172. Dunkel, A. (2015) ‘Visualizing the perceived environment using crowdsourced photo geodata’, Landscape and Urban Planning, 142, 173–186. Eikelenboom, J. (2012) Sharing Views on Spatial Plans: Scales of Geographical Information and Public Participation in Social Media (MSc), Wageningen University, Centre for Geo-Information, the Netherlands. Elwood, S. (2008) ‘Volunteered geographic information: Future research directions motivated by critical, participatory, and feminist GIS’, GeoJournal, 72(3), 173–183. Enengel, B., Muhar, A., Penker, M., Freyer, B., Drlik, S. and Ritter, F. (2012) ‘Co-production of knowledge in transdisciplinary doctoral theses on landscape development – An analysis of actor roles and knowledge types in different research phases’, Landscape and Urban Planning, 105(1–2), 106–117. Eräranta, S., Kahili-Tani, M. and Nummi-Sund, P. (2015) ‘Web-based public participation in urban planning competitions’, International Journal of E-Planning Research, 4(1), 1–18. Evans-Cowley, J.S. (2010) ‘Planning in the age of Facebook: The role of social networking in planning processes’, GeoJournal, 75(5), 407–420. Evans-Cowley, J.S. and Griffin, G. (2011) Micro-Participation: The Role of Microblogging in Planning, available: http://ssrn.com/abstract=1760522 (accessed 13 April 2016). Faraj, S., Jarvenpaa, S.L. and Majchrzak, A. (2011) ‘Knowledge collaboration in online communities’, Organization Science, 22(5), 1224–1239. Fischer, F. and Forester, J., eds. (1993) The Argumentative Turn in Policy Analysis and Planning, Durham, NC and London: Duke University Press. Gailing, L. and Leibenath, M. (2015) ‘The social construction of landscapes: Two theoretical lenses and their empirical applications’, Landscape Research, 40(2), 123–138. Giddens, A. (1986) ‘Action, subjectivity, and the constitution of meaning’, Social Research, 53(3), 529–545. Goodchild, M.F. (2007) ‘Citizens as sensors: The world of volunteered geography’, GeoJournal, 69(4), 211–221. Hahmann, S. and Burghardt, D. (2011) ‘Maple – A Web Map Service for verbal visualisation using tag clouds generated from map feature frequencies’, in Ruas, A., ed. Advances in Cartography and GIScience, Vol 1, Selection from ICC 2011, Paris. Heidelberg: Springer, 3–12. Haklay, M., Singleton, A. and Parker, C. (2008) ‘Web mapping 2.0: The neogeography of the GeoWeb’, Geography Compass, 2(6), 2011–2039. Kahila-Tani, M., Broberg, A., Kyttä, M. and Tyger, T. (2016) ‘Let the Citizens Map—Public Participation GIS as a Planning Support System in the Helsinki Master Plan Process’, Planning Practice & Research, 31(2), 195–214.

159 …

Ron van Lammeren, Simone Theile, Boris Stemmer and Diedrich Bruns

„

Kaplan, A.M. and Haenlein, M. (2010) ‘Users of the world, unite! The challenges and opportunities of social media’, Business Horizons, 53(1), 59–68. Kietzmann, J.H., Hermkens, K., McCarthy, I.P. and Silvestre, B.S. (2011) ‘Social media? Get serious! Understanding the functional building blocks of social media’, Business Horizons, 54(3), 241–251. Lefebvre, H. (1991) The Production of Space, Oxford: Blackwell. Louwsma, M., Bregt, A., Edelenbos, J. and van Lammeren, R. (in prep.) Analysing citizens’ adoption of e-government from the situational context of spatial planning. McPherson, M., Smith-Lovin, L. and Cook, J.M. (2001) ‘Birds of a feather: Homophily in social networks’, Annual Review of Sociology, 27, 415–444. Meijering, J.V., Tobi, H., van den Brink, A., Morris, F. and Bruns, D. (2015) ‘Exploring research priorities in landscape architecture: An international Delphi study’, Landscape and Urban Planning, 137, 85–94. Mendoza, M., Poblete, B. and Castillo, C. (2010) ‘Twitter under crisis: Can we trust what we RT?’, Proceedings of 1st Workshop on Social Media Analytics (SOMA 2010), New York: ACM, 71–79. Pang, B. and Lee, L. (2008) ‘Opinion mining and sentiment analysis’, Foundations and Trends in Information Retrieval, 2(1–2), 1–135. Pariser, E. (2011) The Filter Bubble: What the Internet is Hiding from You, London: Viking, Penguin Books. Pfeffer, J. and Zorbach, T. (2015) ‘Shitstorms: Social Media und die Veränderungen der digitalen Diskussionskultur’, in Stiegler, C., Breitenbach, P. and Zorbach, T., eds. New Media Culture: Mediale Phänomene der Netzkultur, Bielefeld: transcript, 125–141. Richardson, T. and Jenson, O.B. (2003) ‘Linking discourse and space: Towards a cultural sociology of space in analysing spatial policy discourses’, Urban Studies, 40(1), 7–22. Roick, O. and Heuser, S. (2013) ‘Location based social networks – Definition, current state of the art and research agenda’, Transactions in GIS, 17(5), 763–784. Sakaki, T., Okazaki, M. and Matsuo, Y. (2010) ‘Earthquake shakes Twitter users: Real-time event detection by social sensors’, Proceedings of the 19th International Conference on World Wide Web, April 16–30, Raleigh, NC, 851–860. Scikit-learn (2015) ‘Overview of clustering methods’, http://scikit-learn.org/stable/modules/clustering.html (accessed 18 April 2016). Shelton, T., Poorthuis, A. and Zook, M. (2015) ‘Social media and the city: Rethinking urban socio-spatial inequality using user-generated geographic information’, Landscape and Urban Planning, 142, 198–211. Soja, E. (1985) ‘The spatiality of social life: Towards a transformative retheorisation’, in Gregory, D. and Urry, J., eds. Social Relations and Spatial Structures, London: Macmillan, 90–127. Stefanidis, A., Crooks, A. and Radzikowski, J. (2013) ‘Harvesting ambient geospatial information from social media feeds’, GeoJournal, 78(2), 319–338. Stieglitz, S. and Dang-Xuan, L. (2012) ‘Social media and political communication: A social media analytics framework’, Social Network Analysis and Mining, 3, 1277–1291. Stieglitz, S., Dang-Xuan, L., Bruns, A. and Neuberger, C. (2014) ‘Social media analytics: An interdisciplinary approach and its implication for information systems’, Business & Information Systems Engineering, 6(2), 89–96. Sun, Y., Fan, H., Bakillah, M. and Zipf, A. (2015) ‘Road-based travel recommendation using geo-tagged images’, Computers, Environment and Urban Systems, 53 (special issue), 110–122. Surowiecki, J. (2005) The Wisdom of Crowds, New York: Anchor Books. Ulrich, W. (2000) ‘Reflective practice in the civil society: The contribution of critically systemic thinking’, Reflective Practice, 1(2), 247–268. van der Drift, S. (2015) Revealing Spatial and Temporal Patterns from Flickr Photography: A Case Study with Tourists in Amsterdam (MSc), Wageningen University, Centre for Geo-Information, the Netherlands. Wechsler, D. (2014) ‘Crowdsourcing as a method of transdisciplinary research – Tapping the full potential of participants’, Futures, 60, 14–22.

160 …

Chapter 10: Virtual environments Sigrid Hehl-Lange and Eckart Lange

INTRODUCTION Virtual environment technology has great promise to be used for purposes of basic and applied research in the wider field of the planning and design disciplines in general and in landscape planning and landscape architecture in particular. Virtual environments can be broadly characterised as computer-generated, three-dimensional environments providing interactivity and immersion (cf. Gaggioli 2001). Virtual environments are used for static or dynamic landscape representations; they are in essence descriptive and synthetic models of real environments, and they typically focus on external representation (Ervin 2001; Deming and Swaffield 2011, p.89). In order to represent possible future landscape developments, as required in all forms of landscape planning and design, virtual environments need to be linked to a simulation model, or they might follow a normative scenario approach with a particular target concerning, for example, environmental, social, economic and cultural factors (cf. Börjeson et al. 2006). In experimental settings, such as presented in this chapter, virtual environments are modelled to give participants the experience of computer-synthesised landscapes. In landscape planning and design static scenes are commonly used (e.g. Downes and Lange 2015). In addition, animated walks or sequences are increasingly utilised in order to enhance communication with stakeholders. In the first experiment presented here, people are invited to go on a virtual walk that follows a path leading through an agricultural landscape; alternative scenarios of future landscape development are presented and rated using preference scores. In the second experiment a mix of quantitative and qualitative approaches is pursued to investigate how real-time virtual landscape models might be used in a participatory setting, such as a stakeholder workshop (see also Stokols 2011; Schroth et al. 2011). Sophisticated visualisation has become increasingly important in the context of participatory planning and designing of landscapes. At policy level the foundations for democratic and bottomup decision making include the Rio Declaration on Environment and Development (1992), the Aarhus Convention on Access to Information, Public Participation in Decision-Making and Access to Justice on Environmental Matters (1998) and, most prominently for the field of landscape, the European Landscape Convention (2000). Much is still waiting for public involvement to have greater impact on planning and design practice (Lane 2005, p.283). Resulting from increasing demands for experts to transparently communicate with members of the public, we need to have more research in three dimensional (3D) landscape visualisation to support planning and design processes.

161 …

Sigrid Hehl-Lange and Eckart Lange

„

In the remainder of this chapter, a short history of visualisation techniques in landscape architecture is given, concluding with the current status of research in landscape architecture that employs virtual environment technology. Then, a conceptual framework is provided upon which virtual environment research might be based. Two detailed examples of research using virtual environment technology in an experimental setup are presented and discussed. Finally, the chapter identifies directions for improving the way that virtual environment supported research in landscape architecture can contribute to knowledge building of the discipline.

HISTORY OF VIRTUAL ENVIRONMENTS IN LANDSCAPE ARCHITECTURE Research in landscape visualisation and related fields has advanced considerably in the previous decades. The turning point in terms of methods and approaches came during the late 1980s and early 1990s. It was then when the shift from analogue to digital techniques started to take effect. Before that time, landscape architects and planners communicated their ideas mainly by using hand-drawn sketches and perspectives or physical models (e.g. Markelin and Fahle 1979). During the early ages of digital 3D landscape imagery several major issues had to be tackled. First, hardware that could deal with complex graphics was very expensive (Danahy and Wright 1988) and such graphics still were far from real time. Second, software specifically dedicated to landscape visualisation purposes was largely missing (as e.g. in Orland et al. 2001). Third, the relatively small number of trained experts in this domain was a limiting factor (see Lange 2002, p.14). As a result, digitally rendered visualisation was, at best, used as a tool for a crude visual representation of landscape and landscape change (e.g. Nickerson 1979), or purely for visualising the results of a planning process. At the same time, digital 3D visualisation was not yet appreciated as a methodological opportunity that could potentially enhance landscape planning and designing together. However, landscape architecture soon made great progress in 3D modelling and visualisation. Within a relatively short period of time the field has come a long way: starting with the laborious production of static images, either through digital photomontages (Vining and Orland 1989) or, more sophisticated, through digital modelling of 3D environments (Hehl-Lange and Lange 1993) visualisation has, particularly with the advancement of real-time visualisation software, become more and more interactive (Danahy 2001). As a result, new opportunities arose for landscape architects to be better prepared to engage with the public in planning and design. Currently, research in virtual environments includes investigating multiple use of virtual reality models for on-site display on mobile devices (Gill and Lange 2015), novel approaches in augmenting the real world with digital 3D data (Haynes and Lange 2016) and studying the effects of multi-sensory environments on user perception (e.g. Lindquist et al. 2016).

CONCEPTUAL FRAMEWORK Landscape architects are challenged by both the need to involve different stakeholders in planning and design, as well as wishing to establish linkages between participatory methods with 3D landscape visualisation. Provided that stakeholders have been properly identified (through systematic stakeholder analysis) and invited, the main research question that needs answering in this context is: ‘How should visualisations be prepared and presented for stakeholder involvement?’ (see e.g. Scottish Natural Heritage 2014).

162 …

Virtual environments „

The particular research gaps we are addressing in this chapter relate to the experience of dynamic movement through virtual environments. While static visualisations have a predetermined viewpoint, virtual environments offer opportunities for users to move around and explore their virtual surroundings. Gibson (1979), in his ‘Ecological Approach to Visual Perception’, underlines how important it is for people to be looking around and moving about. He argues that we normally perceive the world dynamically and that human perception is not confined to a particular frame of view (see e.g. Nassauer 1995). Any person exploring real environments is used to being able to move freely in and through the space and to view whatever attracts their attention. When it comes to virtual environments, there is a need to learn how ‘real’ people experience and explore them. We must, through research, also try to better understand how such experience can affect decision making, particularly from a planning and design perspective. Humans experience their daily environment while in motion (see e.g. Burckhardt 2006; Thwaites and Simkins 2007). Heft and Nasar (2000) point out three important aspects of our visual perception while in motion. The optical flow causes a streaming of features from a centre of expansion in the field of view accompanying forward movement. Motion parallax stands for the rates of ‘movement’ of static objects as a function of their relative distances from the perceiver. This means near objects seem to move faster than far objects. Through occlusion objects are gradually covered and uncovered while in motion. Considering an experience of space while in motion is typically not explicitly taken into account in planning and design. However, there are a number of famous Chinese examples, for example the Garden of the Master of Nets in Suzhou (see Henderson 2012) and Japanese parks including, for example, Katsura Imperial Villa and Shugaku-in Imperial villa in Kyoto, that are designed especially to be experienced in a sequential way (see Johnson 2003). Similarly, in the English landscape style there are purposefully designed vistas, which open up to visitors, in sequence, when driving or walking through a park. A more recent example where the notion of movement, in this case in architectural space (Samuel 2010), is taken explicitly into account is Le Corbusier’s concept of the ‘promenade architecturale’. In previous research studies Appleyard et al. (1966) analysed the perception of car drivers along a highway by using a set of photographs or perspective sketches combined with written descriptions of this sequential experience. In the Berkeley Environmental Simulation Laboratory (Appleyard and Craik 1978) a miniature endoscopic camera was used that was hung from overhead gantries. This setup gave users freedom of movement through the physical model of central San Francisco at eye level. Building on such research involving analogue technology in the planning and design disciplines the research described in the next section looks into the effects of static versus dynamic representation in virtual environments.

COMPARISON OF DYNAMIC MOVEMENT AND STATIC REPRESENTATION IN AN EXPERIMENTAL SETTING Research question and hypothesis The aim of the study is to better understand which effect different visualisation modes (i.e. static or dynamic) have on landscape evaluation and on people’s assessment of scenarios of future landscapes. The main research question in this experiment is: ‘Does the way of exposure to a landscape have an effect on how landscape is rated?’ The underlying hypothesis is that scores will differ when people are experiencing static images in comparison to dynamic sequences.

163 …

Sigrid Hehl-Lange and Eckart Lange

„

A virtual walk through different landscape scenarios: the Zürich case study For the purpose of this experiment we aimed at exposing participants to a virtual walk allowing navigation in real time. The virtual walk takes participants along a path leading through a rural landscape at the urban fringe near the university campus of ETH Zürich-Hönggerberg, Switzerland. The status quo is a farmed open space with a few fruit trees scattered about the fields and meadows. In addition to the status quo, three scenarios of possible future landscape change were developed: ‘agriculture’, ‘recreation’ and ‘nature conservation’ (see Lange et al. 2008). In the scenario ‘agriculture’ all of the fruit trees within the fields and meadows are removed, and instead fruit trees are newly planted at the edges of the plots and along the paths crossing the farmland. In the scenario ‘recreation’ additional single trees and new hedges of variable height and length have been added, compared to the status quo. The main feature of the scenario ‘nature conservation’ is a large new forest that could function as a wildlife corridor between the currently disconnected forests located east and west of the campus. Additionally, for the ‘nature conservation’ scenario, a new wetland with a buffer of shrubs is also created.

The visual stimuli The virtual model was constructed using Polytrim visualisation software from the Centre for Landscape Research at the University of Toronto (e.g. Danahy and Hoinkes 1995). The model consists of a digital terrain model with a draped orthophoto at a resolution of 0.5 metres (m). 3D objects such as ETH Zürich-Hönggerberg buildings and vegetation were visualised with geo-specific textures.

Static images The status quo and each of the three scenarios were represented with five still images (identical to the respective frames in the animations) spread evenly along a public path (Figures 10.1 and 10.2) including the start and endpoints. Thus, in total, the participants saw 20 images.

Dynamic walkthroughs For the status quo and each of the three scenarios, participants were presented with an animated sequence (Figure 10.3) of 60 seconds duration consisting of 856 single frames, leading viewers along an identical path of 131 m in length. This frame rate still provides fluid motion. The speed of the animation is equal to 7.8 kilometres/hour (km/h) which corresponds to the speed of a jogger or a slow cyclist (cf. Teknomo 2002).

Study participants The participants in the experiment (n=62) were recruited from the Department of Landscape at the University of Sheffield. The sample included staff as well as graduate and doctoral students. The participants were deliberately chosen for their unfamiliarity with the landscape they were going to be exposed to during the experiment, thus avoiding potential effects related to local knowledge.

164 …

Virtual environments „

Figure 10.1 The path in the virtual model. The numbers indicate the locations of the five viewpoints of the still images

Figure 10.2 Still images of the status quo and the three scenarios and five viewpoints

165 …

Sigrid Hehl-Lange and Eckart Lange

„

Figure 10.3 Filmstrip representing the animated walkthrough

Study design and protocol Before the experiment started all participants received a participant information sheet and a short questionnaire (four questions) to record their scores. The participants were briefly introduced to the task of rating a set of images or walkthroughs showing a virtual representation of an agricultural landscape at the urban–rural fringe of the City of Zürich, Switzerland. The actual purpose of the study, the comparison of static versus dynamic mode of representation, was not revealed. Neither was it mentioned to the participants that they would be looking at different scenarios for the same landscape. The participants were split up randomly into five separate groups (12–13 persons per group). One group saw the static images only. Each of the four other groups saw one of the walkthroughs only, that is one of the three scenarios or the status quo. No group saw more than one set of stimuli. As a warm-up for the group looking at the static scenes, four images from within the virtual model were shown. These four images were not considered for the results of the experiment. As a warm-up for the four groups looking at the sequences, an animation along a public footpath within the same landscape model was shown. This animation was not rated in the experiment. All images were shown with a data projector on a projection wall (approx. 3 m x 2 m). Each of the static images was shown for 30 seconds. During this time the participants ticked the boxes with their respective scores. To avoid an ordering effect the images were shown in random order. The walkthroughs were presented in real time using a laptop and a data projector. After seeing the walkthrough once, participants responded to all four questions. Participants were using a verbally anchored five-point rating scale, ranging from very little (1) to very much (5). For each of the 20 images the same questions (addressing preference) were asked: s s

How much do you like this landscape? Would you enjoy walking in this landscape?

After all participants rated all of the images, two further questions (addressing orientation) referring to the entire set of images were asked: s s

To what degree do the images help you to envisage the landscape? To what degree do the images help you to understand where you are in the landscape?

166 …

Virtual environments „

As the participants were not familiar with the site, assigning scores for orientation when just looking at single images is impossible. Thus, for the last two questions the whole set of images was scored. For each of the four animations the same four questions (as above) were asked, that is questions referring to walkthroughs instead of images.

Results The results are looked at in terms of representation type (walkthrough vs static) and in terms of the ratings of the status quo and the three scenarios. For the questions ‘How much do you like this landscape?’ and ‘Would you enjoy walking in this landscape?’ (Figure 10.4), the scores obtained for the walkthroughs are consistently higher than those given for the images. The difference in the scoring ranges from 0.14 (status quo, 2.46 vs 2.32) to a maximum of 0.62 (scenario ‘recreation’, 3.42 vs 2.8). As mentioned earlier, the questions ‘To what degree do the images / walkthroughs help you to envisage the landscape?’ and ‘To what degree do the images / walkthroughs help you to understand where you are in the landscape?’ (Figure 10.4) were asked after one group of participants saw all images. This is why the values for the images across the scenarios are the same. For the question ‘To what degree do the images / walkthroughs help you to envisage the landscape?’, with the

Figure 10.4 Participant’s ratings of four different landscape scenarios, dynamic walkthroughs versus static images

167 …

Sigrid Hehl-Lange and Eckart Lange

„

exception of the status quo, the walkthroughs scored lower in the other three scenarios than the images. For the question ‘To what degree do the images / walkthroughs help you to understand where you are in the landscape?’, the walkthroughs scored consistently higher than the images. The scores for the individual scenarios regarding the questions addressing preference, ‘How much do you like this landscape?’ and ‘Would you enjoy walking in this landscape?’, show highest ratings for the scenario ‘nature conservation’, followed by the scenario ‘agriculture’ and then the scenario ‘recreation’, with the status quo scoring the lowest. Animations of individual scenarios consistently receive higher scores than static images. Regarding the two questions addressing orientation there is no clear pattern of scores for animations versus static images.

Discussion The way visualisations are presented, that is static images or dynamic walkthroughs, has an effect on ratings for landscape preference and on people’s assessment of scenarios of future landscapes. Given the setup, the experiment could be conducted in a timeframe of approximately 15 minutes. Such a relatively short duration keeps the attention levels of the participants high and prevents fatigue, which is known to potentially have an effect on the results (Rathod and La Bruna 2005; Galesic and Bosnjak 2009). As the images were shown in random order and not identified as part of a particular scenario, it was not recognisable for the participants that the images were part of different scenarios. Also, because the static images were shown randomly and the sequences only once to one set of participants, it was possible to exclude an ordering effect. The disadvantage is that one needs a relatively large number of participants for the overall number of responses generated, that is there are rather low participant numbers (12–13 per test set), but this was accounted for by the rather homogenous composition of the participants. In terms of ranking the status quo and the three scenarios, the scoring pattern for the two landscape preference questions is fairly consistent (i.e. either as walkthroughs or as static images) with the scenario ‘nature conservation’ receiving the highest scores. Regarding the two questions addressing orientation in the model, there is a slight tendency that walkthroughs score higher, but the results do not support a conclusion that either of the two modi is better than the other. The reason for this could be a slight ambiguity in the phrasing of the question: that is ‘To what degree do the images / walkthroughs help you to envisage the landscape?’ is perhaps rather similar to ‘understand where you are’. On the other hand it may well be the case that a static image is better for focussing on particular features in order to envisage the landscape. Animations resemble walking. This might explain the comparatively large differences between high scores for walkthroughs and lower scores for images when asked ‘would you enjoy walking in this landscape?’ There are some important factors with a likely influence on the results that merit further research investigation. It would be interesting, for example, to repeat the same experiment with people from different backgrounds (e.g. laypeople only) and with larger numbers. There is a potential effect of comparing a predetermined path (as in this example) with a freely navigable environment. The predetermined movement itself, without free navigation, does not seem to be sufficient to support the assumptions used in the theory of ecological perception (Gibson 1979). Also, the length of the path or the duration of the experience could play a role. Another important factor could be connected to the complexity of the landscape that is studied, especially when it comes to the questions that relate to navigation and orientation.

168 …

Virtual environments „

Further research is also needed in comparing an image to a walkthrough in a truly immersive environment in 3D that is freely navigable. There have been a number of examples in which immersive and dynamic visualisation with a public participation approach is applied to real-world case studies (e.g. Orland and Uusitalo 2000; Stock and Bishop 2002; Danahy 2005). But, to date, there is still a considerable gap in empirical research in investigating whether people respond in a ‘natural’ way when moving or driving through a virtual model, considering sequential and/or immersive experiences in the landscape.

Participatory stakeholder workshop with freely navigable virtual environments: the Alport Valley case study In this study the Alport Valley forest landscape management project in the Peak District National Park (UK) is used as a real-world case in order to look into the potential use of immersive and freely navigable virtual environments, as opposed to predetermined animation paths, when engaging with stakeholders. A time series of future landscapes was developed from a forest management plan and translated into 3D visualisation models. These were explored and assessed in two stakeholder workshops in an immersive environment facility supporting 3D vision. The Alport Valley is a steep sided valley carved into the upland gritstone plateau of the Dark Peak area of the Peak District National Park, which was established in 1951 as the first national park in the UK. While the surrounding upland plateau is very exposed and without any trees, in contrast, the Alport Valley is a sheltered and forested landscape. It is one of the largest and essentially ‘trafficfree’ valleys in the Peak District and has therefore an important function of tranquil enjoyment for walkers and hikers. As a result of policies to provide the UK with a strategic timber reserve (see e.g. Essex 1990), mostly in the first half of the twentieth century dense coniferous forests dominated by non-native species including Sitka Spruce (Picea sitchensis (Bong.) Carrière), with smaller stands of Japanese Larch (Larix kaempferi (Lamb.) Carrière), Lodgepole Pine (Pinus contorta Dougl. Ex Loud.) as well as native Scots Pine (Pinus sylvestris L.) were planted for fast growth. Not only are large areas dominated by non-native trees in conflict with the notion of a national park in general, nowadays forest policies also take into account the social and landscape benefits as well as benefits for plant communities and wildlife. Because of the overall unique landscape character, initial proposals for large scale timber extraction, including the construction of suitable access roads for heavy timber trucks, caused major opposition. Subsequently, a joint planning approach integrating the key stakeholders and landowners as well as the views of the public was pursued. As a result of the collaboration among the stakeholders the Alport Valley management plan (Figure 10.5) was developed. This includes unconventional measures such as trees felled to rot on site as well as ring-barking of trees (see Lange and Hehl-Lange 2010a). Both measures had been introduced because of the predicted impact on the tranquil valley associated with the removal of the logged trees and construction of new access roads. In line with forests requiring adaptation to climate change, the overall aim is to establish native woodlands mostly by natural regeneration through the preservation of individual native seed trees and partly if necessary through active plantings and seeding with material of local provenance. This process will take place in a period spanning several decades.

169 …

Sigrid Hehl-Lange and Eckart Lange

„

Figure 10.5 Long-term forest management plan indicating the year when felling takes place

The visual stimuli Some decades ago Dame Sylvia Crowe used hand-drawn sketches to illustrate forest management options (Crowe 1978). In the Alport Valley example, a virtual landscape model of the valley is constructed that consists of a digital terrain model (DTM), an orthophoto and a range of object types such as trees, dry stonewalls, buildings, paths and the sky as a backdrop. The DTM is based on an original map scale of 1:10,000 and has a resolution of 10 m. The resolution of the orthophoto is 1 m. This orthophoto is draped over the DTM using the visualisation software Simmetry3d. The terrain along the path was edited manually to provide a smooth animation. The geometry of the buildings is constructed in Sketchup. In order to achieve a realistic representation of the built objects they were photographed in the field and their textures applied to the building geometry. Similarly, for the vegetation, a library of geo-specific textures that were acquired on site is used. In addition to the billboards, in prominent locations along the main access to the valley, trees with texture-mapped 3D-geometry (see Paar 2003) were also included. The visualisation of several thousand trees while still being able to move around in real time was a major challenge that required fine-tuning with several iterations of model improvement. The landscape is shown in several stages over time (Lange and Hehl-Lange 2010b): ‘2005’ before forest management activities began, ‘2020’ after harvesting most of the existing woodlands, ‘2030’ when new woodlands have started to be established and ‘2090’ the proposed ‘final state’ with oak–birch woodland.

Study participants A key issue for a success in conducting stakeholder workshops as in the example of the Alport Valley is to identify and involve the relevant stakeholders. Also, this ensures that management planning takes into consideration local experience. If the key decision-making parties are involved early there

170 …

Virtual environments „

is strong potential for the discussions in stakeholders’ workshops to lead to concrete action on the ground. The relevant stakeholders are the National Trust (the main landowner), the Forestry Commission, the Peak District National Park Authority (PDNPA), the British Mountaineering Council, the Campaign to Protect Rural England (CPRE), Friends of the Peak District and The Kinder and High Peak Advisory Committee.

Study design and protocol The workshops were held in the virtual reality studio of the University of Sheffield. To provide unobstructed views for the participants, chairs were placed at the back of the room approximately 3–4 m away from the screen (3 m x 2.5 m) in a semi-circular arrangement. Prior to the actual stakeholder workshops a test run was conducted in order to eradicate any potential issues and provide a smooth operation. An essential requirement is a skilled operator of the system and a moderator guiding and structuring the process (Figure 10.6). In order to keep a record of the workshops, they were recorded with photographs and video. Also, a researcher took written notes during the workshops. A practical hurdle was finding a time slot in which everyone would be available for a workshop. In order to arrange the date a preparatory meeting was held with the main stakeholders, that is the National Trust and the Forestry Commission. Because of the limited availability of the stakeholders it turned out to be impossible to find a joint single slot, and therefore the workshop was conducted twice. In total 11 persons representing different stakeholder groups participated in the workshops. The workshops gave the stakeholders the opportunity to see, for the first time, and in three dimensions, the management plan and the landscape as it would develop over time. Workshop participants were presented with 4 x 5 static images (Figure 10.7). Each of the four landscape models were represented through five images, with five seconds showing time per image and

Figure 10.6 Setup of the participatory workshop using 3D visualisation

171 …

Sigrid Hehl-Lange and Eckart Lange

„

Figure 10.7 Static images of the four landscape models represented through five images along the path

animations at 20 frames/s leading participants along a stretch of approximately 200 m of the sole access route in the Alport Valley. The speed of 8 km/h corresponds to the pace of a jogger. In addition, participants had the opportunity to explore the virtual environment on their own by using devices such as a joystick as used in computer games and stereo glasses for 3D immersion. After they had seen and explored the virtual environments on their own, participants were asked to fill in a short questionnaire and to answer four questions for which the answers could be given using a five-point rating scale with tick boxes (ranging from ‘not at all’ equalling 1 to ‘very much’ equalling 5). The four questions were: ‘To what degree do the following types of visualisations help you to envisage the landscape?’, ‘To what degree do the following types of visualisations help you to understand where you are in the landscape?’, ‘To what degree do the following types of visualisations help you to understand the visual transformation of the landscape in the future?’ and ‘How helpful are the different visualisations for you to participate in the forest management plan for the Alport valley?’

Results In the second experiment particular focus was put on gathering qualitative feedback (e.g. Lewis and Sheppard 2006; Schroth et al. 2011). When the stakeholders saw the images and the walkthroughs along the pre-recorded animation path they were sitting quietly, focussing and concentrating on the screen. Only a few comments were whispered. The stakeholders were then encouraged to get up from their seats and move towards the screen to try exploring the virtual environment at their own pace and along their own routes. They got a short introduction of how to use the joystick

172 …

Virtual environments „

for navigation. In particular, when they moved away from the path and explored the whole valley they engaged much more and spontaneously commented on the visualisations, for example ‘It’s fantastic, this is brilliant’, ‘The river there! That’s actually the view!’, ‘Look, the ridge there!’, that is they were immediately interacting more with the landscape and relating their own experiences and geographic knowledge to what they saw on the screen, or even deliberately going to particular locations that they wanted to explore. In comparison to the still images (‘the still images added very little, whereas the real time navigation was amazing and added real value’) the stakeholders made it clear that the navigation brought the landscape to life (‘captures the visual sensation of being in the valley from different vantage points’), because it gave them a sense of place and scale, expressing a feeling like being in that landscape, as well as a sense of ownership and control as they could explore the landscape freely (‘being able to navigate through the valley was a useful experience’). The visualisations were seen as providing a good representation of how the landscape might look in years to come. Some study participants wanted to see more foreground details including shrubs or bracken as well as boundaries such as fences (‘good to show some more boundaries’) to give more reality to the visualisation. For the majority of the stakeholders a high level of foreground detail was not considered to be important. For them the overall impression of the landscape as a whole is the key factor (‘at the scale of grand landscape the model is at its best’) as well as the ability to show how to manage large landscapes (‘overall it has given me great benefit and understanding on the impacts that the proposed felling and plantation will have’; ‘the visualisation will help to consider whether the landscapes we are working towards will meet our original objectives’; ‘it is an excellent tool to “get over” to people how you intend to manage large landscapes’). For a detailed quantitative analysis the sample with 11 stakeholders is rather small. However, the analysis of the responses shows clear patterns in terms of how the different representation media are rated by the individual stakeholders (Figure 10.8). Also, the results show that the scores overall tend to be between 3 and 5, that is clustering around 4. In general, there is a clear trend for all four questions that the real-time navigation scored higher than the animations, whereas the animations scored either equal or in most cases slightly higher than the images. Depending on the questions, in some cases the stakeholders treat the three representation methods as equal.

Discussion The results provide quantitative and also qualitative information of people’s perception regarding the use of static imagery, animations and interactive computer visualisations as a basis for making decisions about future landscapes. All stakeholders participating in the study were directly involved in the Alport Valley management plan. They know this landscape well. Because of their high level of local knowledge and their familiarity with the site they were able to target specific locations they wanted to explore in order to view the landscape from different perspectives. For them, the question about orientation in the landscape, whether it is related to imagery, animations or self-navigation, in statistical terms did not have a distinct discriminatory effect. For the question referring to the role of visualisations for participation in the management plan, for the majority of the stakeholders (6 out of 11) both visualisation approaches resulted in high scores averaging clearly above 4, whereas 5 out of 11 clearly favoured the real-time navigation over images and animations. In particular the open comments

173 …

Sigrid Hehl-Lange and Eckart Lange

„

Figure 10.8 Stakeholders’ ratings of different representation media

that were gathered beyond the rigidly structured questionnaires, especially when the stakeholders themselves took over and experienced the virtual landscape on their own, gave further insights that would not have been received by only relying on a typical questionnaire format. In terms of scheduling participatory events, a key effort is needed for organising them. This should not be underestimated, and thorough preparation and scripting needs to be done by the hosts. In addition, workshops and surveys need to be approved by the respective ethics committees. All this can take a considerable amount of time and needs flexibility in planning ahead and also in addressing alternatives in terms of timings. When digital landscape visualisations are used as an integral component of the process (e.g. Gill and Lange 2013) this complicates the setup even more. In such a case, because of the interactive nature of the involvement of the public this is very challenging and needs skilled operators and moderators. Detailed choreography is absolutely essential but can only account to a certain degree for the unpredictability of the focus of the aspects investigated when involving the public. While the level of complexity is relatively low when only static landscape visualisations are used, the challenge when using interactive landscape visualisations increases considerably. Also, as a virtual landscape

174 …

Virtual environments „

model can be freely explored, it needs a relatively high level of detail throughout (rather than providing detail only along certain routes) which increases rendering time. Compared to a workshop setting that does not rely on interactive 3D landscape models, this needs a skilled operator and a moderator who recognises the role of the visualisation. It also needs participants willing to engage. For digital natives the interactive nature of computer games is something they are used to and probably expect. For novices the moderator will have to be patient and encouraging in order for the workshop participants not to shy away. In the process of the preparation of the management plan, as part of the ongoing stakeholder involvement, the tree felling method was changed to ring-barking. As a result the dead tree trunks will be standing for some years in order to prevent erosion. As a consequence, the virtual model ‘2020’ (after harvesting most of the existing plantations) had to be adapted. Instead of visualising areas where harvested trees were lying on the ground to rot, these areas were populated with trunks of dead trees still standing. In this example dead tree trunks as seen from a distance are displayed as thin lines and when their positions change slightly from each pre-recorded image to the next image flickering appeared in parts of the animations. This was noticed by some of the stakeholders and is a common problem in computer graphics. It does not occur in still images. In forest management long-term planning decisions are made today but they will impact future generations. 3D visualisation gives us the opportunity to decide on the effects of forest management decisions today and to experience what future generations will potentially experience.

GENERAL DISCUSSION AND CONCLUSIONS The setup of both of the experimental studies reported on above was exploratory in nature, and with the experience gained from them we were able to transfer the approaches as role models to other research settings (see e.g. Hehl-Lange et al. 2012; Gill et al. 2013; Hehl-Lange et al. 2015). Decisions to be made in planning and design processes are often complex. This requires sophisticated decision support techniques and favours the use of visual communication techniques that can potentially simplify and explain complex settings and spatial information in order to improve design and decision making. Landscape visualisation not only has the potential to visually communicate spatial characteristics of possible future landscapes to stakeholders, it can also be used to explore conflicting interests by involving the relevant stakeholders early on, for example by adapting models for public involvement such as focus groups, public hearings, round tables, workshops, design charrettes and so on or, further, it could be the basis to integrate adaptive, analytical and systematic approaches (see Milburn and Brown 2003) in research into a design context. Regarding the presentation mode perceived by the user/stakeholder, it seems not so much a question of static (an image) versus dynamic representation (an animation along a predetermined route). The real benefit lies in the possibility to explore virtual environments freely. More than any other form of virtual reality exposure, the self-determined exploration most resembles the behaviour of a real person in a real landscape. Moreover, it may even go beyond real world exposure, as in virtual representations of real environments, a user can easily navigate to, perhaps, otherwise inaccessible locations. The visualisation approaches in the research presented rely on representing the real world through a virtual surrogate. In the future, other approaches are likely to play an important role as

175 …

Sigrid Hehl-Lange and Eckart Lange

„

well (e.g. Gill and Lange 2013, 2015). In particular, this includes developments for augmenting the real world with textual or graphical data. In combination with increasingly ubiquitous mobile devices such as tablets and smartphones, augmented reality for planning and designing our environments is likely to develop into a new field of research in the near future.

SUGGESTED FURTHER READING Appleyard, D. (1977) ‘Understanding professional media: Issues, theory and a research agenda’, Reprint No. 150 from Human Behavior and Environment 2, Berkeley, CA: Institute of Urban and Regional Development, University of California. Bishop, I. and Lange, E., eds. (2005) Visualization in Landscape and Environmental Planning: Technology and Applications, London, New York: Taylor & Francis. Buhmann, E., Paar, P., Bishop, I. and Lange, E., eds. (2005) Trends in Real-Time Landscape Visualization and Participation, Heidelberg: Wichmann. Conan, M., ed. (2003) Landscape Design and the Experience of Motion, Washington, DC: Dumbarton Oaks Trustees for Harvard University. Lange, E. (2011) ‘99 volumes later: We can visualise. Now what?’, Landscape and Urban Planning, 100, 403–406. Sheppard, S.R.J. (1989) Visual Simulation: A User’s Guide for Architects, Engineers, and Planners, New York: Van Nostrand Reinhold.

REFERENCES Appleyard, D. and Craik, K.H. (1978) ‘The Berkeley environmental simulation laboratory and its research program’, Applied Psychology: An International Review, 27, 53–55. Appleyard, D., Lynch, K. and Myer, J.R. (1966) The View from the Road, Cambridge, MA: The MIT Press. Börjeson, L., Höjer, M., Dreborg, K.-H., Ekvall, T. and Finnveden, G. (2006) ‘Scenario types and techniques: Towards a user’s guide’, Futures, 38(7), 723–739. Burckhardt, L. (2006 [1979]) Warum ist Landschaft schön? Die Spaziergangswissenschaft, Berlin: Martin Schmitz Verlag. Crowe, S. (1978) ‘The landscape of forests and woods’, Forestry Commission Booklet 44, London: HMSO. Danahy, J. (2001) ‘Technology for dynamic viewing and peripheral vision in landscape visualization’, Landscape and Urban Planning, 54, 125–137. Danahy, J.W. (2005) ‘Negotiating public view protection and high density in urban design’, in Bishop, I.D. and Lange, E., eds. Visualization in Landscape and Environmental Planning: Technology and Applications, London, New York: Taylor & Francis, 195–202. Danahy, J.W. and Wright, R. (1988) ‘Exploring design through 3-dimensional simulations’, Landscape Architecture, 78(4), 64–71. Danahy, J.W. and Hoinkes, R. (1995) ‘Polytrim: Collaborative setting for environmental design’, in Tan, M. and Teh, R., eds. The Global Design Studio. Proceedings CAAD Futures ‘95. CASA, National University of Singapore, 647–658. Deming, M.E. and Swaffield, S. (2011) Landscape Architectural Research: Inquiry, Strategy, Design, Hoboken, NJ: John Wiley & Sons. Downes, M. and Lange, E. (2015) ‘What you see is not always what you get: A qualitative, comparative analysis of ex ante visualizations with ex post photography of landscape and architectural projects‘, Landscape and Urban Planning, 142, 136–146. Ervin, S.M. (2001) ‘Digital landscape modeling and visualization: A research agenda’, Landscape and Urban Planning, 54, 49–62. Essex, S.J. (1990) ‘Woodland planning in the Peak District National Park, UK: Formulation and implementation of a land use policy’, Land Use Policy, 7(3), 243–256.

176 …

Virtual environments „

Gaggioli, A. (2001) ‘Using virtual reality in experimental psychology’, in Riva, G. and Galimberti, C., eds. Towards Cyberpsychology: Mind, Cognition and Society in the Internet Age, Amsterdam: IOS Press, 157–174. Galesic, M. and Bosnjak, M. (2009) ‘Effects of questionnaire length on participation and indicators of response quality in a web survey’, Public Opinion Quarterly, 73(2), 349–360. Gibson, J.J. (1979) The Ecological Approach to Visual Perception, Boston, MA: Houghton Mifflin. Gill, L. and Lange, E. (2013) ‘Visualizing landscapes’, in Howard, P., Thompson, I. and Waterton, E., eds. The Routledge Companion to Landscape Studies, London, New York: Routledge, 417–426. Gill, L. and Lange, E. (2015) ‘Getting virtual 3D landscapes out of the lab‘, Computers, Environment and Urban Systems, 54, 356–362. Gill, L., Lange, E., Morgan, E. and Romano, D. (2013) ‘An analysis of usage of different types of visualisation media within a collaborative planning workshop environment’, Environment & Planning B: Planning and Design, 40(4), 742–754. Haynes, P.S. and Lange, E. (2016) ‘In-situ Flood Visualisation Using Mobile AR‘, IEEE Symposium on 3D User Interfaces, 19–23 March 2016, Greenville, SC, 241–242. Heft, H. and Nasar, J.L. (2000) ‘Evaluating environmental scenes using dynamic versus static displays’, Environmental Behavior, 32(3), 301–322. Hehl-Lange, S. and Lange, E. (1993) ‘2D3D4D’, Anthos, 2, 12–16. Hehl-Lange, S., Gill, L., Henneberry, J., Keskin, B., Lange, E., Mell, I.C. and Morgan, E. (2012) ‘Using 3D virtual geodesigns for exploring the economic value of alternative green infrastructure options’, in Buhmann, E., Ervin, S. and Pietsch, M., eds. Digital Landscape Architecture 2012, Berlin-Offenbach: Wichmann, 273–280. Hehl-Lange, S., Lange, E. and Bilge, G. (2015) ‘Plan and design together – Just a vision?‘, Proceedings REAL CORP 2015, Ghent, 483–487. Henderson, R. (2012) The Gardens of Suzhou, Philadelphia, PA: University of Pennsylvania Press. Johnson, N.B. (2003) ‘Mountain, temple, and the design of movement: Thirteenth-century Japanese Zen Buddhist landscapes’, in Conan, M., ed. Landscape Design and the Experience of Motion, Washington, DC: Dumbarton Oaks Research Library and Collection, 158–186. Lane, M.B. (2005) ‘Public participation in planning: An intellectual history’, Australian Geographer, 36(3), 283–299. Lange, E. (2002) ‘Visualization in landscape architecture and planning: Where we have been, where we are now and where we might go from here’, in Buhmann, E., Nothhelfer, U. and Pietsch, P., eds. Trends in GIS and Virtualization in Environmental Planning and Design, Heidelberg: Wichmann, 8–18. Lange, E. and Hehl-Lange, S. (2010a) ‘Citizen participation in the conservation and use of rural landscapes in Britain: The Alport Valley case study’, Landscape and Ecological Engineering, 7(2), 223–230. Lange, E. and Hehl-Lange, S. (2010b) ‘Making visions visible for long-term landscape management’, Futures, 42(7), 693–699. Lange, E., Hehl-Lange, S. and Brewer, M.J. (2008) ‘Scenario-visualization for the assessment of perceived green space qualities at the urban-rural fringe’, Journal of Environmental Management, 89(3), 245–256. Lewis, J.L. and Sheppard, S.R.J. (2006) ‘Culture and communication: Can landscape visualization improve forest management consultation with indigenous communities?’, Landscape and Urban Planning, 77(3), 291–313. Lindquist, M., Lange, E. and Kang, J. (2016) ‘From 3D landscape visualization to environmental simulation: The contribution of sound to the perception of virtual environments’, Landscape and Urban Planning, 148, 216–231. Markelin, A. and Fahle, B. (1979) Umweltsimulation: Sensorische Simulation im Städtebau, Schriftenreihe 11, Städtebau-Institut Universität Stuttgart, Stuttgart: Krämer. Milburn, L.-A.S. and Brown, R.D. (2003) ‘The relationship between research and design in landscape architecture’, Landscape and Urban Planning, 64, 47–66. Nassauer, J.I. (1995) ‘Messy ecosystems, orderly frames’, Landscape Journal, 14(2), 161–170. Nickerson, D.B. (1979) ‘Sightline, perspective plot, scope: Three desktop computer programs for forest landscape design’, Journal of Forestry, 77(1), 14–17. Orland, B. and Uusitalo, J. (2000) ‘Immersion in a virtual forest – some implications’, in Sheppard, S.R.J. and Harshaw, H.W., eds. Forests and Landscapes: Linking Ecology, Sustainability and Aesthetics, Wallingford, Oxon: CABI, 205–224.

177 …

Sigrid Hehl-Lange and Eckart Lange

„

Orland, B., Budthimedhee, K. and Uusitalo, J. (2001) ‘Considering virtual worlds as representations of landscape realities’, Landscape and Urban Planning, 54, 139–148. Paar, P. (2003) ‘Lenné 3D – The making of a new landscape visualization system: From requirements analysis and feasibility survey towards prototyping’, in Buhmann, E. and Ervin, S.M., eds. Trends in Landscape Modeling, Proceedings at Anhalt University of Applied Sciences, 78–84. Rathod, S. and La Bruna, A. (2005) ‘Questionnaire length and fatigue effects – does size really matter?’, ESOMAR, Conference on Panel Research, April 17–19, Budapest. Samuel, F. (2010) Le Corbusier and the Architectural Promenade, Basel: Birkhäuser. Schroth, O., Wissen, U., Lange, E. and Schmid, W.A. (2011) ´Visulands – A multiple case study of landscape visualization as a tool for participation’, Landscape Journal, 30(1), 53–71. Scottish Natural Heritage (2014) Visual Representation of Wind Farms, http://www.snh.gov.uk/publicationsdata-and-research/, accessed 22.3.2016. Stock, C. and Bishop, I.D. (2002) ‘Immersive, interactive exploration of changing landscapes’, Proceedings Integrated Assessment and Decision Support. Lugano, Switzerland, International Environmental Modelling and Software Society 1, 30–35. Stokols, D. (2011) ‘Transdisciplinary action research in landscape architecture and design’, Landscape Journal, 30(1), 1–5. Teknomo, K. (2002) Microscopic Pedestrian Flow Characteristics: Development of an Image Processing Data Collection and Simulation Model, published PhD thesis, Tohoku University, Japan. Thwaites, K. and Simkins, I. (2007) Experiential Landscape: An Approach to People, Place and Space, London: Routledge. Vining, J. and Orland, B. (1989) ‘The video advantage: A comparison of two environmental representation techniques’, Journal of Environmental Management, 29, 275–283.

178 …

Chapter 11: Walking Henrik Schultz and Rudi van Etteger

INTRODUCTION Walking is one of the most common and revealing ways to explore a landscape. There are volumes of writing in which researchers report on the landscape knowledge gained by walking, from fields ranging as wide apart as geomorphology (Lukas and Bradwell 2010) and psycho-geography (Coverly 2010). Walking, in our view, holds great promise for obtaining knowledge to inform the design of complex landscapes, and walking can be profitably used in the context of research through designing. Walking is especially suitable for answering research questions dealing with complex and unfamiliar tasks that require engagement with the object of research in order to understand and frame the problem properly. For example, a research project on strategies to integrate new infrastructures for energy production would benefit from engaging through walking. The nature of walking itself can be viewed as experimental (Fischer 2011, p.289). In this context ‘experimental’ means a test, trial or tentative procedure; an act or operation for the purpose of discovering something unknown, in this case the characteristics and potentials of the terrain crossed by the walk. Using walking as a research method implies being aware of its experimental character. Conducting walking as experiment in landscape research means to intervene and change the object of research. In walking experiments researchers provide a framework for fostering creative engagement and for combining planned and unplanned elements (von Seggern 2000, p.316). The character of a walking experiment can best be described by quoting Bruno Latour: ‘A good experiment is not one that offers some definite knowledge, but one that has allowed the researcher to trace the critical path along which it will be necessary to pass so that the following iteration will not be carried out in vain’ (Latour 2004, p.196). The chapter begins with a short history of walking used as a method in research and we briefly describe research in landscape architecture that employs walking. Secondly, we provide a conceptual framework upon which research using walking might be based. Thirdly, we present and discuss two examples of research using walking. Finally, we identify directions for continuing to improve the way that walking-supported research can contribute to knowledge building in the field of landscape architecture research.

CONCEPTUAL FRAMEWORK Rebecca Solnit (2000; 2005) provides both a historical perspective on walking and an insight into why one should walk. Geoff Nicholson (2008) speaks – in the title of his book on the subject –

179 …

Henrik Schultz and Rudi van Etteger

„

about The Lost Art of Walking. Referring to, for example, Jean-Jacques Rousseau who claimed to be unable to work, or even think, when not walking, Frédéric Gros (2014, p.5) describes why going for a walk is the ‘best way to free your mind’. With his famous ‘strollology’ experiments Lucius Burckhardt influenced designers, including landscape architects, and pointed at the ‘invisible’ aspects of landscape (Fezer and Schmitz 2012). In urban design, Appleyard (1982), Jacks (2007) and Gehl (2010) have emphasised the importance of walking for the liveability of the urban environment, and Speck (2013) advertises the walkable city. Costa et al. (2015) have looked into the experiences of walkers in urban parks. Wunderlich (2008) has sketched walking in an urban context, while Wylie (2005) has described the experience of a walk along the coast. Outside academia, authors have also touched upon the experience of walking in relation to landscape, for example, in the stories by Ian Sinclair on the urban fringe in his book London Orbital (2003), and by Robert Macfarlane about the English landscape in The Wild Places (2007) and The Old Ways (2012). Bill Bryson (1998), in his book A Walk in the Woods, describes both intensive landscape perception and insight on fundamental questions as side effects on his walk on the Appalachian Trail. In this chapter we consider landscape from a constructivist point of view (Berger and Luckmann 1966, p.210; Creswell 2009, p.8; Crotty 1998, p.43). Landscape is constructed through the active interplay between observer and the observed. Landscape is considered as coming into being through a complex, non-linear process of transformation. The European Landscape Convention builds on this concept of landscape: according to article 1 A, ‘landscape’ means an area, as perceived by people, whose character is the result of the action and interaction of natural and/ or human factors. People play a constitutive part by perceiving, using and changing landscape both physically and in their minds. According to Hille von Seggern´s (2008) definition of landscape as Raumgeschehen, all space that surrounds us can be understood only by actively becoming part of and perceiving its ongoing process of transformation. Thus, landscape is not simply a physical entity that can be analysed by measuring and observing; studying landscapes needs to take the observer into account. This brings us to phenomenological theory. The founding father of phenomenology, Edmund Husserl (1970 [1936]), had serious concerns about the course of scientific thinking of his time, and also about the role that science plays in society. Though science had made great advances throughout the nineteenth century, the abstraction and mathematisation of science brought, to Husserl’s mind, science further away from people’s everyday life-world. For example, the measurable fact that wood and metal consist mainly of empty space that exists between nuclear cores and electron clouds does not stop anyone from sitting on a chair built of such material. The central theme for phenomenology was ‘back to the things themselves’. Phenomenology advocates studying the phenomena and the experiences of people and their actions in their life-world. Studying should not focus on an abstract world of atoms, but on a contingent specific place that surrounds people. With this life-world people bodily interact, as theorised by phenomenologist Merleau-Ponty (1968), not as subjects between objects, but by being a part of it. Phenomenological studies may best be described as accounts of ‘subjective experience’, rather than subjective accounts of experience (Gallagher and Zahavi 2008). The phenomenological thinking and approach has been further developed in recent decades by Moustakas (1994), Smith et al. (2009) and by Ihde (2012). From the original transcendental aspirations of Husserl an approach has been developed that is grounded in an existential position. The earlier theoretical understandings of phenomenology by Husserl in the 1930s and MerleauPonty in the 1960s on the situatedness of knowledge have been corroborated by findings from

180 …

Walking „

experimental research (McLeod and Dienes 1996) and neuroscientific research which show how interconnections exist between the body, the senses and the world we live in (e.g. Noë 2004). Knowledge about the world is actively constructed by engaging with the world. According to Barbaras (2006, p.90) it is only attention for what is already there in perception, so the landscape does not appear as a view – it is already there in the flow of experience and then singled out. McLeod and Dienes (1996) describe how baseball players do not know where the ball will go, but can move towards that spot. Similarly, scholarly knowledge is not just waiting to be found by scientists, but is actively constructed by scientists engaging with situations. Experiences of the landscape are not there in the landscape, but come into existence when an observer moves into a specific place. Design researchers need to get involved but also need to reflect on their role as active forces in the processes by which life-world phenomena become transformed and when developing ideas which help shape the environment and which anticipate certain experiences to be had there. This brings us to our underlying concept of research aimed at obtaining knowledge within the field of landscape architecture. As landscape is a complex, ever changing object, there is a choice either to simplify and isolate aspects of landscape which are then universally true or to accept the socially constructed nature of landscape and follow Martin Prominski´s classification of landscape research as part of ‘mode 2’ research (Prominski 2004). The term mode 2 research was established by Helga Nowotny who distinguished between ‘reliable knowledge’ of ‘mode 1’ research and ‘socially robust knowledge’ of mode 2 research (e.g. Nowotny et al. 2001, p.166). Nowotny describes mode 2 knowledge generated in contemporary research processes as ‘working knowledges’. Working knowledges ‘include the knowledge that works and knowledge with which one works. … Working knowledge therefore means working upon an object, which is created by knowledge’ (Nowotny 2008, p.13). According to Nowotny, understanding life means to change it (Nowotny and Testa 2009, p.15). This mode 2 knowledge is not developed to be universally applicable, but exists in and relies on a particular social setting. This is also in keeping with the Rittel’s description of planning and design problems as wicked problems: every formulation of the wicked problem corresponds to a statement of the solution and vice versa (Protzen and Harris 2010, p.156). Walking is a way of contributing to and of playing with processes of transformation in landscapes. The working knowledges one generates while walking relates to tangible elements of a landscape. For example, to get a sense of the comfort or discomfort of a set of stairs one has to walk up them. To get a feeling for the appropriate width of a path, one has to walk it and feel the comfort or discomfort of passing a stranger. To understand such things the designer has to become part of the landscape, its movement, its performance. Equally important is the fact that walking fosters a process of understanding. Von Seggern (2008, p.233) says that ‘as a transformative process, understanding is directly linked to design’. Walking is such a transformative process of understanding and creativity. While walking and bodily perceiving a landscape researchers become part of the landscape and, thus, no longer distinguish between themselves (subject) and the landscape (object). Walkers explore what is already there, immediately creating and thus changing this ‘reality’ by walking through it and by connecting elements in their minds and with their bodies and by reflecting on the insights gained. These insights reach beyond knowledge that primarily serves design activities such as making plans for an open space. In fact, it is a way to engage with the object of research – one precondition of understanding.

181 …

Henrik Schultz and Rudi van Etteger

„

ENGAGING WITH THE LANDSCAPE William Whyte (1980), Jane Jacobs (1961) and Jan Gehl (2010) are famous explorers of public life. Their insights were gained by sitting at one place and observing people’s behaviour, thus focussing on relatively small urban areas. For researchers to study the qualities and complexities of larger areas (von Haaren et al. 2014), moving through space would be more suitable than sitting at one spot. Apart from considering scale, three aspects are crucial for using walking as part of a research approach: ‘generating implicit knowledge’, ‘identifying problems, research questions and ideas’ and ‘reflection in action’. First, walking helps generate implicit knowledge through involvement with the object of research. Walking is a certain way of understanding landscapes that differs from the desk study of maps and other landscape data. What, at first glance, may seem like empty space on a map, might actually be filled with invisible walks (Careri 2005, p.42) and contain vast amounts of information relevant for both design and research. Walking can reveal such information, including, to begin with, the logic of logistics of the landscape. Certain shapes that seem clear on the map might appear much more eclectic when viewed from the route of a walk. Tim Ingold (2011, p.46) says that ‘cognition should not be set off from locomotion, along the lines of a division between head and heels, since walking is itself a form of circumambulatory knowing’. The knowledge Ingold refers to unfolds while experiencing and being part of a landscape in motion: animals, other people, the wind, moving clouds, sun, shade and seasons change landscapes constantly. The walking researcher understands dynamics, experience and atmospheres. The rhythm that characterises the act of walking enables a complex interplay of body and mind. Walking rhythmically merges the changing landscape, the motion of our bodies and our lines of thought. This process allows the walking researcher to feel particularities and ephemeral aspects that might play an important role in a research project. When doing research on site specifics in landscape architecture, for example, these dynamic elements have to be grasped (Diedrich 2013). Second, walking helps to identify problems, research questions and ideas. When dealing with complex objects of research, formulating relevant research questions is a creative act that benefits from ‘walking around the problem’ rather than sitting at a desk, reading books and extrapolating. As Jonas (2007, p.1365) puts it: ‘No information is available, if there is no idea of a solution, because the questions arising depend on a kind of solution, which one has in mind. One cannot fully understand and formulate the problem, before it is solved. Thus, in the end, the solution is the problem.’ The process of walking helps to get an idea of the solution. Ideas cannot be produced only by referring to rational strategies. Intuitional and bodily strategies are needed as well. Walking is a bodily activity and the creative processes described by so many different people are based on intuition (Schultz 2014b, p.129). Third, walking fosters reflection in action. The knowledge generated during a walk may be implicit but it can be shared among other researchers and members of the general public. Such sharing, when done right away and on site – for example when walking with other researchers or with people living in the area – can help make knowledge explicit. One may even develop a new concept for a research project. In other words, while walking, researchers can practise reflection in action (Schön 1984, pp.76ff). Researchers benefit from engaging in transdisciplinary dialogue. Through the mutual sharing of conceptions that people have of landscapes they gain insights into their dynamic and ever changing object of research. The fact that almost everybody is capable of walking makes the method a low-threshold activity. This is a crucial quality because it fosters the generation of ‘socially

182 …

Walking „

robust knowledge’ of mode 2 research (Nowotny et al. 2001, p.166). The methods presented below illustrate how walking used as a research method can support researchers in meeting their tasks of knowledge production within processes of complex landscape transformation.

The wandering-method The first example presents a method of walking research where, through participatory activity, we include two types of research roles, those who participate as full-time researcher, and those who participate as competent contributors. The latter participants follow certain guidelines provided by the researcher, engage with the landscape and make their findings explicit after the walk (Schultz 2014b). All participants should be considered as playing a role in doing research (Wechsler 2014, p.18). Through the participatory arrangement the classical distinctions are overcome that used to be made between ‘the researcher’ and others who were assumed to have no active role to play in the execution of a study. There is the researcher and the members of a participating group, together doing research. The participant-researchers are free to choose a path; deciding which path to take is an intuitive act leading them to relevant sites. In the first example one researcher and seven research participants took part (Schultz 2014b). Participants engaged in walking through the urban landscape of metropolitan Hamburg, Germany. The assumption made, in advance of conducting the experiment, was that walking fosters two things: first, the generation of implicit knowledge through engaging with the object of research and, second, the identification of relevant research questions and ideas. The study was designed in a way that would exemplify the benefits researchers would get from using the wandering-method. The study design included recruiting competent contributors as participants, framing and setting guidelines for conducting the experiment, instructing participants, carrying out the walking experiment, collecting information based on participants’ experiences, and analysing and synthesising their experiences (Schultz 2014b). In preparation for the experiment seven competent contributors were recruited as participants. These were people involved in planning and designing and who were based in the metropolitan region of Hamburg. The sample included practitioners from the disciplines of landscape architecture, urbanism and geography. This sample was made deliberately because it included people who were regularly working on large-scale landscape assignments. The guidelines for this walking experiment were designed based on previous experience from walking in different research and design contexts. The setting and the guidelines were subjected to a pre-test with students of the Leibniz University of Hannover who did the same walk. From previous experiences we learned that there must be enough time for participants to engage and discover the landscape and that it is beneficial for the walks to be strenuous because this intensifies bodily perception. Apart from that, the pre-test showed that it is important to express the first ideas and findings right after the walk and before engaging in group discussion. This ‘saves’ the implicit knowledge in words and images that otherwise vanishes quickly. The guidelines prepared in advance were there to inspire and enhance three modes of experiencing landscapes through walking: (1) the ‘discovery mode’, (2) the ‘flow mode’ and (3) the ‘reflective mode’ (Schultz 2014a; see Figure 11.1). In the ‘discovery mode’, participants focus their attention on the traversed space. They constantly orientate themselves, searching curiously without a specific objective. While moving and connecting views, feelings and places, they get new perspectives, see things from different angles or in a different light. The scenery becomes a

183 …

Henrik Schultz and Rudi van Etteger

„

Figure 11.1 Interplay of three walking modes (source: Schultz 2014a)

spectacle in which the continuously moving walkers play an active role. When walkers enter the ‘flow mode’, the space becomes diffuse scenery and walkers let their thoughts stray, following their intuition. Walking and awareness merge. Flow is a state where body and mind are aligned, says physiologist Csikszentmihalyi (1996). Elsewhere he explains how walkers are fully involved and open for holistic experiences (Csikszentmihalyi 1985, p.58). The ‘reflective mode’ re-establishes the distance, which means that walkers distinguish between themselves and the object of research. This is an analytical mode allowing them to understand what they experienced. They can reflect on and sharpen their findings, and transfer implicit and embodied knowledge into words and images.

184 …

Walking „ The three modes feed on each other and can appear in long or very short periods of time. During the strenuous climb of a hill, for example, walkers will focus on their breathing and might be able to switch off thinking. It is likely that they will experience flow. Once they arrive at a viewpoint on top of the hill their mind is open for new thoughts and ideas. And if they enter the discovery mode again, because something triggers their attention, there can be a creative moment. The experiences they had before their climb will influence the ideas and the knowledge they generate during the climb. The ‘reflective mode’ can be reached alone or in discussions. Walks taken together with people from different backgrounds and professions are opportunities to exchange views, perspectives and interpretations of the perceived landscapes right away, while still on site. Certainly, the perceptions and ideas will differ but still it is much easier to discuss, for example, the effects of a cluster of windmills while walking the landscape around them instead of sitting in a conference room with everybody having their set arguments ready. Walking together can help to find common points of reference. To acquaint participants with their task before conducting the experiment they received a short briefing by phone where the procedures and the way the experiment was embedded in the research project were explained to them. Participants were told that their task would be to walk for a whole day and that this walk was likely to be strenuous. Participants would not have to prepare for the experiment. Some days before the actual walk all participants received an email providing further information: they were asked to bring suitable shoes, a backpack, rain gear, a bottle of water, some provisions and a compass. They were reminded to be ready to use and document their imagination while engaging with and perceiving the space they would walk through. The experiment was conducted on 21 April 2012. The task was to walk for the day and to find ideas and research questions regarding the future development of the greater region around the city of Hamburg. In the morning, before commencing the walk, all participants met and were reminded how the experiment should be understood as a scientific adventure. After this briefing the participants were given three sheets of A4 paper: a walking map of the area, guidelines and a blank sheet. The walking map provided an overview but deliberately showed no detail, so that, for example, street names were not readable. Participants were encouraged not to use the map during their walk but rather simply as a rough orientation before starting out. On the blank paper participants were asked to draw a single sketch and to write a short description expressing how they imagined the landscape they were going to walk through. These exercises were not just for a warm-up before the walk but formed important documentation that would be compared with the impressions that participants would put down on paper after the walk. Each participant walked individually and covered around 20 kilometres in some seven hours. They were asked to explore particularities of the landscape and reflect on possible transformations. In particular they were asked: What are suitable research questions for transforming the landscape? On their search for these questions they applied a set of rules provided. One important rule was to avoid looking at their maps; instead they were to try navigating the landscape by following intuition and using their compass. Participants were encouraged to try talking to people they met on the way and to learn about the landscape through conversations with them. Another rule was that participants could freely take a maximum of ten photographs of situations that attracted their attention. They were also asked to find their own individual rhythm of walking. They had to try immersing themselves into the landscape, enjoying the walk. After seven hours of walking, all the participants reconvened at a previously determined location. They expressed what they had experienced on the way and immediately started documenting

185 …

Henrik Schultz and Rudi van Etteger

„

images and questions on two sheets of paper, relying on their professional competence as a standard for relevance. Then the participants exchanged experiences with each other. They were encouraged to engage in conversations by triggering a process of making implicit knowledge explicit. All discussions were recorded and – together with the sketches drawn on the sheets – analysed during the days following the walk.

Results What did the participants experience? They were able to perceive intensively the space they traversed while enjoying being part of the landscape, which included animals, other walkers, the changing weather and more. Their sketches and reports reflect how they created new mental ensembles out of spatial elements, which they connected by walking. Ownership structures, municipal borders and patterns of land use dominated the sketches drawn before the walk. After the walk, paths, fringes, passages, contrasts and landmarks became more important. Participants found new and even poetic words for the different parts of the landscape, such as ‘black sand forest’, and drew images of the whole area. These images are working knowledge that could inform both design research and practice. Through highlighting the particularities of a landscape, it provides starting points for designs. Through introducing research questions it also unveils themes that can only be found when engaging with space and its ephemeral qualities. One participant used newspaper clippings to connect different parts of the traversed space in order to create a walkable landscape. His collage expressed this idea and also represented his way of reading the space and of creating a new landscape. The collage was also an implicit way of expressing the research question that this participant had identified during the walk: ‘What roles do pathways in metropolitan peripheries play for landscape perception and transformation?’ The research question was a finding of the walk and the subsequent analysis and discussion. It could be used to deepen knowledge about the traversed landscape. That question is also relevant in other landscapes and can thus be decisive for landscape research beyond this particular site.

Discussion Results obtained by applying the wandering-method include working knowledge gained about the object of research. In addition, research questions were identified and their relevance confirmed by dialogue between experiment participants. It is crucial to make explicit the implicit knowledge gathered on a walk and to do so without losing its inventiveness and productive fuzziness. Walking researchers can describe the total experience of a landscape, which means that they are able to understand it intuitively in all its complexity. It is crucial to express such understanding immediately after the walk and to use images, such as producing a collage, to tell stories, or to give the perceived landscape a name. Only then, after expressing the totality of the experience, may factual knowledge be added, enhanced and refined.

Continuous/stop-motion walking The method in the second example aims specifically to allow for aesthetic evaluations of the landscape and evaluation of selected aspects of landscape experiences to be made, for instance the

186 …

Walking „

‘naturalness’ of the experienced landscape. Evaluations are facilitated by one researcher doing two walks, both following a pre-set route, making findings explicit after the first continuous walk and the second stop-motion walk (van Etteger 2013). Sensory perceptions are expressed by using words written after the first walk and during the second walk and by drawings and photographs produced during the second walk. The process of data collection structures the walk. In this second example the research question is: ‘What is the phenomenology of a designed landscape, or in other words how does a designed landscape present itself to my experience?’ (van Etteger 2013). Walking is used for research by applying what can be called the continuous/stop-motion (CSM) method. The researcher is the walker. A route is walked twice. First the walk is done by continuous walking and describing the experience from memory afterwards. This allows the researcher to follow the flow of the walk. Then, the walk is done again, the second time in stop-motion, stopping at regular time or space intervals, each time to record findings and to put a representation of the sensory perceptions into words, drawings and photographs. The data are descriptive, made by putting aside (‘bracketing’ in phenomenological jargon) existing ideas about the object of exploration. This is done to suspend the superficial ordinary way of looking at landscape (the natural attitude) and thus to reach a state of openness towards the experiences of the world (the epoché in phenomenological jargon). Following Ihde (2012) the walker is asked to stay close to sensory experiences and to stay away from referring to knowledge, past memories and so on. After completing both of the two walks a phase follows where the singular character of the walking experiences is reflected and findings are analysed. The elements of the CSM method are described in Figure 11.2. The CSM method was employed as part of research into the aesthetic evaluation of designed landscapes (van Etteger 2013). The phenomenological part of this research investigated the landscape of the Dutch island of Walcheren in the Province of Zeeland. This landscape was designed in 1945 by landscape architects Jan Bijhouwer, Roel Benthem and Nico de Jonge (Steenhuis and Hooimeijer 2009). The two walks by the researcher were done, as described above, to experience the designed characteristics of the landscape first-hand and in an open-minded fashion, whilst suspending preconceived thoughts about this landscape in an attempt to be as unbiased as possible. The phenomenological approach in general does not require that the sample chosen to reflect upon is representative: it just needs to be able to offer insights for reflection (Casey 2000, p.23). In this particular case, studying a large landscape through walking, special care was taken that the two routes cut through the different parts of the area in order to be able to reflect on most of the studied landscape. The two routes run along the edges of the island and also through open areas in the interior of the island. Walking the two routes in stop-motion took twice as long as the first continuous walk. The first walk took 2.5 hours for route 1 and 3 hours for route 2; the second walk took 5 hours for route 1 and 6 hours for route 2.

Results There was a significant difference in the data collected between the first and second manner of walking. The uninterrupted walking of both of the two routes and writing down findings afterwards provided about 3,000 words of description of experiences. The second time, after walking both routes in stop-motion, in this case stopping every 500 metres, delivered about 3,000 words of description of experiences and, in addition, ten hand-drawn sketches and about 1,000 photographs. In the first description, the one of the continuous walk, attention was paid to changes in direction.

187 …

Henrik Schultz and Rudi van Etteger

„

Figure 11.2 The schematic model of the CSM method (source: van Etteger 2013)

In this walk visual attention is forward-orientated. For example, walking away from a road the sound of cars rushing by is experienced progressively less intensively as walking towards a road. Also, by stopping and photographing, one steps out of the continuity of the walk. The descriptions show how things are seen in the second walk that lie behind the observer in the first walk. These things were not noticed during the first walk facing forward and moving in continuous motion. When stopping next to a field, for example, horses get curious and come up to the observer, making noises and smells, which are not registered while walkers are moving at a steady pace along the field. Stopping also puts an end to the sounds of being in motion and allows other sounds, even those of insects, to be heard. The experiences of the continuous walk are a closer emulation of a walker in progress, whilst the second stop-motion walk provides a wider range of possible experiences to be obtained from the landscape.

Discussion In keeping with phenomenological methods, the findings of the particular walks were subjected to imaginative variation afterwards. The walker reflected on the dependence between the experiences of this observer compared with those of another observer of a different gender, age or level of physical health. The mode of transport was reflected upon and the influence of seasonal variety on

188 …

Walking „

the experiences during the walk was considered. Reflecting on the variations, observer-identity was not considered to be of significant influence on the experience, but the mode of transport would have made a difference. Between walking and cycling things change gradually, but walking or driving in a car makes a large difference in terms of experiential data.

REFLECTION AND CONCLUSION According to Husserl walking is our normal way of experiencing the world, a kind of experiencing that is done in a natural attitude in which we assume all kinds of things about experience, rather than explore our experience in itself. When walking to the supermarket we rarely reflect on the nature of that walk. The walk is instrumental in getting us somewhere. However, in the examples presented above, walking is conceived as a research method and thus instrumental for gaining knowledge. Given a relevant research question, the two methods described are accessible and may be applied by all who are working on landscape design and research. While intensely engaging with and experiencing the landscape, walkers may call up visual memories, add new experiences to their implicit memory and create new images (Bauer 2004, p.71). In addition, when walking for research, researchers themselves may identify relevant research questions. However, walking as an academic research method needs to be practised like any other academic method. Practising the methods used in both examples is most certainly a process of learning to make conscious use of intuitional and bodily strategies, to link them with rational ones and to express implicit knowledge. The findings from the two methods presented above do not qualify as mode 1 knowledge that remains true independent of the location and the social context in which they were formed. It is knowledge about the subjective experience of landscape made by the walkers. The accounts of these experiences are open to intersubjective scrutiny. The knowledge developed is not of an exclusive and idiosyncratic nature. It is mode 2 knowledge, on landscape as a social construct. It is working knowledge that can help to guide the transformation of the explored landscapes. It is knowledge that can be used to support finding the answers to the wicked questions that are asked in planning and design contexts (Protzen and Harris 2010). What can be taken from the walks to other situations is the kind of questions that are asked. These questions can be asked at new locations in new social contexts without, however, the guarantee that they are always the right questions to ask. The two methods have much in common. Both deliver grounded, implicit, embodied knowledge that cannot always be explained in words. This is why visual means of communication are included, such as mappings, collages and drawings. One of the challenges of both methods is to represent the multitude of experiences of the complex interaction between body and landscape. Walks do not readily deliver objective ‘facts’ about a landscape, but help to generate ‘working knowledges’ (Nowotny 2008) for tackling the wicked problems of planning and designing landscapes for people. The wandering-method, as applied in the periphery of Hamburg, exemplifies the power of walking as a research method for understanding large-scale landscapes. Although the participants did walk and experienced only a section of the periphery, all were able to gain knowledge about and to find relevant questions for the whole area. For the study design researchers need to decide who is walking with whom and also how long the walk should be. A requirement of the wandering-method is that the walk should be at least of a length that allows for the flow mode to develop. When applying the continuous/stop-motion

189 …

Henrik Schultz and Rudi van Etteger

„

walking method the length needs to be linked to the topic of exploration and the amount of variation in the landscape that can be expected. In terms of data collection the two methods are very different. The wandering-method allows participants to engage freely with the physical surroundings and to meet with people they encounter along the way. Ideas can be picked up through discovery or can develop in the interplay of the three modes explained above. The CSM method is much more particular in what is to be collected. Mainly sensory impressions are collected first, ordered through the continuous walk, and then more detailed information is gathered during the stop-motion walk. Special attention can be paid to particular aspects such as the degree of public access or the experience of naturalness. Both methods also differ in terms of data sampling. The participants in Hamburg were left to improvise, within certain prescribed boundaries. The beginning and end were set, but not the alignment of the route. The CSM method is much more structured. The route was planned in advance, to allow researchers to visit all of the different parts of the landscape in question. The stop-motion walk was set to distance or time intervals that appeared to be relevant for the specific landscape studied, and also to the researched aspect of that landscape. In terms of data generation the wandering-method again is quite open. A map and a blank sheet for a representation of experiences before and after the walk is all that researchers provide to participants. Data analysis occurs during the discussions between participants. The data analysis plan for the CSM method, however, is set by the traditional rules of the phenomenological approach. The descriptive data generated from the fieldwork has to be processed through imaginative variation in order to get closer to the core of the experience. The data also has to be analysed for its content. Graphs can be drawn plotting, for instance, the degree to which space is perceived as open to the public or experienced as natural along the route. Such data can also be translated into maps. The fact that walking allows different groups of people to contribute to research in a transdisciplinary setting enables significant contributions to be made, through walking, to what has been called ‘transformative science’. As Schneidewind and Singer-Brodowski (2013) observe, transformative scientists are actively involved in ongoing transformation processes in order to learn about them. Creswell (2009, p.208), for example, uses this term in the context of an advocacy/ participatory approach in research for planning and design. These observations are fully applicable to walking research. Both walking methods described in this chapter offer opportunities to involve people in the creative process of exploring the object of research and of finding relevant research questions. First, the methods allow the researcher-designer to come into contact with people who are in the process of engaging with the object of research and thus open and informed about its characteristics. The walking researcher can hold discussions with users and co-producers of landscapes and use them as sources of information. Second, people can be invited to rediscover their own territory and walk together in groups with landscape design research experts. These organised walks can be part of the research design allowing stakeholders, landowners, decision makers and the interested public to contribute to generating working knowledge. Armed with the personal experience of a walk, people can contribute to an engaged dialogue about the characteristics of a landscape, its challenges and questions. They can generate working knowledges that are shared and discussed right away on site. Third, walks can be conducted as interventions. This can be a way to test first assumptions, specify questions and allow everybody to participate in experiencing processes of landscape transformation. Trespassing, for example, can open the discussion on accessibility of

190 …

Walking „

spaces. Walking silently as a group already turns a walk into a performance that changes the whole setting. When people make a walk they leave traces. These traces can be highlighted by specific signs that together form a kind of path, inscribing the walk into the landscape and thus initiate a debate on path-making as creative practice. When describing processes of design research, Christopher Frayling (1993), Henk Borgdorff (2006) and Wolfgang Jonas (2007) used slightly different terms, but they all emphasised the importance of embodied knowledge generated in the creative process of changing the object of research. As Jonas (2007, p.1368) puts it: ‘The active improvement of an unsatisfactory, problematic situation is the primary motivation for thinking, designing, and, finally – in a more refined, purified, quantitative manner – for scientific knowledge production.’ Walking as a method can foster research through designing, as defined by Jonas. Researchers are part of the object of research, in this case the landscape, and, at the same time, involved in a process of building up a network between elements of the landscape. The perceptions, stories and images found during the walk are a special form of working knowledges. The knowledge is generated while creating something and while changing the object of research. However, as noted before, problems change character during the course of finding solutions and are only fully understood and formulated in relation to a solution. This understanding of the problem is based on the engagement with the object of research while walking and on the complex interplay of rational, intuitive and bodily strategies that characterise the process of designing. In Jonas’ (2007, p.1369) words: ‘There are forms of knowledge special to the awareness and ability of a designer, independent of the different professional domains of design practise.’

SUGGESTED FURTHER READING Walking in general: Careri, F. (2005) Walkspaces: El Andar como Practica Estetica – Walking as an Aesthetic Practice, Barcelona: Editorial Gustavo Gili. Jacks, B. (2007) ‘Walking and reading in landscape’, Landscape Journal, 26(2), 270–286. Solnit, R. (2000) Wanderlust: A History of Walking, New York: Viking.

On the epistemological context of knowledge and walking: Ingold, T. (2011) Being Alive: Essays on Movement, Knowledge and Description, New York: Routledge. Nowotny, H., Scott, P. and Gibbons, M. (2001) Re-Thinking Science: Knowledge and the Public in an Age of Uncertainty, Cambridge: Polity Press. Schön, D.A. (1984) The Reflective Practitioner: How Professionals Think in Action, New York: Ashgate Publishing.

The landscape architectural perspective: Schultz, H. (2014) ‘Designing large-scale landscapes through walking’, Journal of Landscape Architecture, 9(2), 6–15. van Etteger, R. (2013) ‘Wish you were here walking with me: Walking as a tool for the aesthetic evaluation of designed landscapes’, in Yeung, H.H. and Collier, M., eds. Selected Essays from the On Walking Conference, 28–29 June, The University of Sunderland, Sunderland: Art Editions North, 322–332, available: http://issuu.com/stereographic/docs/walkonconference.

Walking in specific landscape types: Sinclair, I. (2003) London Orbital, London: Penguin. Macfarlane, R. (2012) The Old Ways: A Journey on Foot, London: Penguin Books.

191 …

Henrik Schultz and Rudi van Etteger

„

REFERENCES Appleyard, D. (1982) Liveable Streets, Berkeley, CA: University of California Press. Barbaras, R. (2006) Desire and Distance: Introduction to a Phenomenology of Perception, Stanford, CA: Stanford University Press. Bauer, J. (2004) Das Gedächtnis des Körpers: Wie Beziehungen und Lebensstile unsere Gene steuern, München: Piper. Berger, P.L. and Luckmann, T. (1966) The Social Construction of Reality: A Treatise in the Sociology of Knowledge, New York: Anchor Books. Borgdorff, H. (2006) The Debate on Research in the Arts (Sensuous Knowledge 02), Bergen: Bergen National Academy of the Arts. Bryson, B. (1998) A Walk in the Woods: Rediscovering America on the Appalachian Trail, London: Black Swan. Careri, F. (2005) Walkspaces: El Andar como Practica Estetica: Walking as an Aesthetic Practice, Barcelona: Editorial Gustavo Gili. Casey, E.S. (2000) Imagining: A Phenomenological Study, 2nd ed., Indianapolis, IN: Indiana University Press. Costa, S., Coles, R. and Boultwood, A. (2015) ‘Landscape experience and the speed of walking’, in Niin, G. and Mishra, H.S., eds. Landscapes in Flux: Peer-reviewed Proceedings ECLAS 2015 Conference, Tartu: Estonian University of Life Sciences, 123–127. Coverly, M. (2010) Psychogeography, Harpenden: Pocket Essentials. Creswell, J.W. (2009) Research Design: Qualitative, Quantitative and Mixed Methods Approaches, 3rd ed., London: Sage. Crotty, M. (1998) The Foundations of Social Research: Meaning and Perspective in the Research Process, London: Sage. Csikszentmihalyi, M. (1985) Das Flow-Erlebnis, Jenseits von Angst und Langeweile im Tun aufgehen, Stuttgart: Klett-Cotta. Csikszentmihalyi, M. (1996) Creativity, Flow and the Psychology of Discovery and Invention, New York: HarperCollins Publishers. Diedrich, L. (2013) ‘Translations – Radicant design for transforming harbour sites’, Portusplus, issue 2012, available: https://www.researchgate.net/publication/294090226_Translations-Radicant_Design_for_ Transforming_Harbour_Sites. Fezer, J. and Schmitz, M., eds. (2012) Lucius Burckhardt Writings, Rethinking Manmade Environments: Politics, Landscapes & Design, New York: Springer Verlag. Fischer, R. (2011) Walking Artists: Über die Entdeckung des Gehens in den performativen Künsten, Bielefeld: Transcript. Frayling, C. (1993) ‘Research in art and design’, Royal College of Art Research Papers, 1(1), 1–5. Gallagher, S. and Zahavi, D. (2008) The Phenomenological Mind: An Introduction to Philosophy of Mind and Cognitive Science, Abingdon: Routledge. Gehl, J. (2010) Cities for People, Washington, DC: Island Press. Gros, F. (2014) A Philosophy of Walking, London: Verso. Husserl, E. (1970 [1936]) The Crisis of European Sciences and Transcendental Phenomenology: An Introduction to Phenomenological Philosophy, Evanston, IL: Northwestern University Press. Ihde, D. (2012) Experimental Phenomenology: Multistabilities, 2nd ed., Albany, NY: SUNY Press. Ingold, T. (2011) Being Alive: Essays on Movement, Knowledge and Description, New York: Routledge. Jacks, B. (2007) ‘Walking and reading in landscape’, Landscape Journal, 26(2), 270–286. Jacobs, J. (1961) The Death and Life of Great American Cities, New York: Random House. Jonas, W. (2007) ‘Research through DESIGN through research: A cybernetic model of designing design foundation’, Kybernetes, 36(9/10), 1362–1380. Latour, B. (2004) Politics of Nature: How to Bring the Sciences into Democracy, Cambridge, MA: Harvard University Press. Lukas, S. and Bradwell, T., eds. (2010) The Quaternary of Western Sutherland and Adjacent Areas: Field Guide, London: Quaternary Research Association. Macfarlane, R. (2007) The Wild Places, London: Granta Books.

192 …

Walking „

Macfarlane, R. (2012) The Old Ways: A Journey on Foot, London: Penguin Books. McLeod, P. and Dienes, Z. (1996) ‘Do fielders know where to go to catch the ball or only know how to get there?’, Journal of Experimental Psychology: Human Perception and Performance, 22(3), 531–543. Merleau-Ponty, M. (1968) The Visible and the Invisible, Evanston, IL: Northwestern University Press. Moustakas, C. (1994) Phenomenological Research Methods, Thousand Oaks, CA: Sage Publications. Nicholson, G. (2008) The Lost Art of Walking: The History, Science, Philosophy, and Literature of Pedestrianism, New York: Riverhead Books. Noë, A. (2004) Action in Perception, Cambridge, MA: The MIT Press. Nowotny, H. (2008) ‘Designing as working knowledge’, in von Seggern, H., Werner, J. and Grosse-Bächle, L., eds. Creating Knowledge: Innovationsstrategien im Entwerfen urbaner Landschaften – innovation strategies for designing urban landscapes, Berlin: Jovis, 12–15. Nowotny, H. and Testa, G. (2009) Die gläsernen Gene: Die Erfindung des Individuums im molekularen Zeitalter, Frankfurt am Main: Suhrkamp. Nowotny, H., Scott, P. and Gibbons, M. (2001) Re-Thinking Science: Knowledge and the Public in an Age of Uncertainty, Cambridge: Polity Press. Prominski, M. (2004) Landschaft entwerfen: Zur Theorie aktueller Landschaftsarchitektur, Berlin: Reimer. Protzen, J.-P. and Harris, D.J. (2010) The Universe of Design: Horst Rittel’s Theories of Design and Planning, Abingdon: Routledge. Schneidewind, U. and Singer-Brodowski, M. (2013) Transformative Wissenschaft: Klimawandel im deutschen Wissenschafts- und Hochschulsystem, Marburg: Metropolis. Schön, D.A. (1984) The Reflective Practitioner: How Professionals Think in Action, New York: Ashgate. Schultz, H. (2014a) ‘Designing large-scale landscapes through walking’, Journal of Landscape Architecture, 9(2), 6–15. Schultz, H. (2014b) Landschaften auf den Grund gehen: Wandern als Erkenntnismethode beim Großräumigen Landschaftsentwerfen, Berlin: Jovis. Sinclair, I. (2003) London Orbital, London: Penguin. Smith, J.A., Flowers, P. and Larkin, M. (2009) Interpretative Phenomenological Analysis: Theory, Method and Research, London: Sage publications. Solnit, R. (2000) Wanderlust: A History of Walking, New York: Viking. Solnit, R. (2005) A Field Guide to Getting Lost, London: Penguin. Speck, J. (2013) Walkable City: How Downtown Can Save America, One Step at a Time, New York: North Point Press. Steenhuis, M. and Hooimeijer, F., eds. (2009) Maakbaar Landschap, Nederlandse Landschapsarchitectuur 1945– 1970, Rotterdam: NAI Uitgevers. van Etteger, R. (2013) ‘Wish you were here walking with me: Walking as a tool for the aesthetic evaluation of designed landscapes’, in Yeung, H.H. and Collier, M., eds. Selected Essays from the On Walking Conference, 28–29 June, University of Sunderland, Sunderland: Art Editions North, 322–332, available: http://issuu.com/stereographic/docs/walkonconference. von Haaren, C., Warren-Kretzschmarb, B., Milosc, C. and Werthmann, C. (2014) ‘Opportunities for design approaches in landscape planning’, Landscape and Urban Planning, 130, 159–170. von Seggern, H. (2000) ‘“Alles Kunst” – Soziale Differenzierung, Polarisierung und öffentlicher Raum – Ein Plädoyer für komplexe Experimente’, in Harth, A., Scheller, G. and Tessin, W., eds. Stadt und Soziale Ungleichheit, Opladen: Leske+Budrich, 310–321. von Seggern, H. (2008) ‘Entwerfen als integrierender Entwurfsprozess’, in Von Seggern, H., Werner, J. and GrosseBächle, L., eds. Creating Knowledge: Innovationsstrategien im Entwerfen urbaner Landschaften – innovation strategies for designing urban landscapes, Berlin: Jovis, 212–237. Wechsler, D. (2014) ‘Crowdsourcing as a method of transdisciplinary research: tapping the full potential of participants’, Futures, 60, 14–22. Whyte, W. (1980) The Social Life of Small Urban Spaces, New York: Project for Public Spaces. Wunderlich, F.M. (2008) ‘Walking and rhythmicity: Sensing urban space’, Journal of Urban Design, 13(1), 125– 139. Wylie J. (2005) ‘A single day’s walking: Narrating self and landscape on the South West Coast Path’, Transactions of the Institute of British Geographers, 30(2), 234–247.

193 …

Chapter 12: Design guidelines Martin Prominski

INTRODUCTION When landscape architects commence a new design project they usually start by seeking relevant and up to date knowledge in order to meet the brief, because relying solely on their experience is rarely sufficient, especially when projects are complex. For example, if the project involves designing a new urban riverfront, designers might search for technical information such as the robustness of certain materials to withstand hydrological processes, for the best and most appropriate plants for the local ecological conditions, or for good examples of spatial design to achieve contact by people with the water. Designers will probably also be looking for similar completed projects as precedents and inspiration. However, due to the limited resources available to a typical office in terms of time to carry out a comprehensive search for applicable data, the results of search processes necessarily remain somewhat arbitrary – sometimes the results are good and appropriate knowledge can be assembled, sometimes the results are inadequate and affect the quality of the final design. Since the aim of design is to produce high quality projects (functionally, spatially, aesthetically and ecologically etc.) this search process should be made more efficient and effective. One way of achieving this is by the use of design guidelines. One of the challenges facing researchers is how to ensure that the result of their work – often hidden away in academic papers in journals to which most practitioners have little access, or presented in ways that make the interpretation of the results difficult – make a difference on the ground. Evidence-based design is now a key watchword in many fields such as architecture and engineering and it should also be the case in landscape architecture. One key means of achieving a stronger link between research and design is by translating the research findings into design guidelines. These might be suitable for general application, for example in relation to recommendations for improving access to parks by older people (see Ward Thompson – Chapter 14), while others may apply in a more restricted way or to a limited set of conditions. In this chapter the concept of design guidelines is introduced on the assumption that these would provide the knowledge basis needed for systematic design-related search processes and, by doing so, would make the design process itself more efficient and produce better results. In addition, the specific category of design research as a method is explored and critically evaluated. According to the Oxford English Dictionary (1996), a guideline is ‘a principle or criterion guiding or directing action’. Transferred to design this definition suggests that a design guideline gives guidance for design action, meaning that it suggests a specific direction by excluding many other

194 …

Design guidelines „

possible, and by implication, less suitable ones. A guideline also offers transferable knowledge because a principle is an abstraction (from a set of data or experiences) which works beyond a specific case to a more generalisable set of situations. The fact that design, by definition, deals with concrete and specific situations leads to the main question of this chapter: ‘How to develop abstracted, transferable guidelines that inform (and improve) landscape design?’ To answer this question, this chapter starts by developing a theoretical framework that establishes the position of guidelines in the design process. Then three research projects on design guidelines are presented and the differences between the projects and the strengths and weaknesses of the approaches used are addressed. Then the position of this type of research is discussed within the wider field of design research. This chapter ends with some concluding remarks about aspects of research for developing evidence-based design guidelines.

THEORETICAL FRAMEWORK The hypothesis presented above is that design guidelines provide knowledge which help to improve the design process. To verify this hypothesis, it is necessary, first, to address the question about the position design guidelines could or should occupy in the context of the design process. One key argument can be found in Donald Schön’s theory of the design process as a reflective practice. In the 1980s Donald Schön, Professor of Urban Studies and Education at the Massachusetts Institute of Technology (MIT), recognised that design is an irreplaceable method for finding solutions to complex problems in the real world and he examined design by looking at the work of many designers (Schön 1983). He describes two intertwining elements of the design process: first, the establishment of an idea or concept derived from the subjective ‘appreciative context’ of the designer; second, the objective evaluation of the established concept in terms of the extent to which it fulfils the requirements of ‘normative design domains’ such as function, local conditions, spatial organisation, technology or cost (Schön 1983, p.96). A complex network of logical ‘if … then’ chains emerges from these two intertwined elements of the design process and the decisions made at various nodes in this network determine the course of the design process. To illustrate this Schön uses the example of a design for a school and its yard (Schön 1983, pp.79–93). If the idea is proposed to build L-shaped classrooms into the hillside, then retaining walls must be built. If retaining walls are built then it is only possible to enter the rooms from one side. If entrance is only feasible from one point then the costs will increase and so on. The course of the design process changes according to decisions made at various nodes in this network. In the end, the designer’s concept and the objective evaluation criteria ought to be as coherent as possible. The – wonderful – challenge of this design process lies in the fact that the number of possible ideas or concepts from the subjective ‘appreciative context’ of the designer is as unlimited as the number of normative design domains, and reaching a coherent result is a demanding task. To deal with this complex challenge, designers often search for precedents or best cases, because such examples have shown a successful conclusion to reflective practice. However, precedents or best cases are, in principle, specific to their location and objectives and it is difficult to apply them to a new design task which has its own unique circumstances. Here, the potential of design guidelines comes into play because, according to the definition mentioned above, they provide principles or strategies rather than one specific solution. Abstracted design principles express a strategic corridor of possible directions which guides designers in their search for solutions that fit a specific situation.

195 …

Martin Prominski

„

Thus, design guidelines are neither totally specific nor completely universal and represent structured knowledge bundles at an intermediate level. These characteristics indicate two advantages of design guidelines within the design process. First, they advance and speed up reflective practice through preselected, recommended suggestions and avoiding culs-de-sac in the design process: they are comparable to ‘enzymes’ which act as catalysers to accelerate processes. Second, they allow designers to apply the principle to any specific design case where individual creativity is mandatory. This openness is important, because the alternative of closed, prescribed solutions would render design unnecessary and lead away from site-specific, unique results. In summary, taking research results and using further research to develop design guidelines creates applicable knowledge for design practitioners. The guidelines themselves are not designs; they only serve as an ‘enzyme’ which designers may use in the design process for the purpose of gaining efficiency and higher quality. They act as an intermediate step between the mass of research evidence and its application in a complex situation. Having addressed the position of design guidelines within the context of the design process, the research question about the development of abstracted, transferable guidelines remains to be addressed. From the meaning of the word ‘abstraction’ it is clear that first there must be something concrete which then undergoes the process of abstraction. To obtain successfully abstracted design guidelines, I have experimented with two concrete starting points: best-practice examples and test designs. Research projects on design guidelines that start from two points first have to define the research questions and the theoretical context of the research topic (obtained from, for example, a literature review of research evidence and theoretical models). This foundation leads to criteria for selecting best-practice examples or for doing test designs. Once the criteria are set, the abstraction process can start. This approach has surprising similarities to Donald Schön´s reflective practice: abstracted design principles, strategies or measures are proposed as ideas and continuously reflected against the theoretical framework and best-practice examples or the test designs. In the end, the aim is to obtain an ordered structure of coherent design guidelines which comprehensively cover the range of possible design questions regarding a specific topic. In the following section I describe three examples of such research projects.

RIVER. SPACE. DESIGN. PROCESS-ORIENTATED DESIGN OF URBAN RIVER SPACES The research project, ‘Process-Orientated Design of Urban River Spaces’ (project leaders: Martin Prominski and Antje Stokman; funded by the German Research Foundation (DFG) from 2008 to 2011), originated in the context of the European Water Framework Directive (2000) and the European Flood Directive (2006) which demand an integrative management plan for water systems as well as flood risk management plans from all European Union (EU) member states. Our assumption was that concentrating exclusively on flood protection and ecology – a pragmatic approach that many communities were likely to follow – meant that the open space quality of urban river spaces would be increasingly neglected. There are already many examples where dikes or other protection devices have been made higher, blocking or obscuring visual and physical contact between the urban fabric and the river and we wanted to avoid this if possible. From this perspective, we were interested in a multifunctional approach and asked the following research question: ‘What process-orientated design strategies are suited to meet the demands of flood protection, ecology and attractive open spaces together?’ We decided to use best-practice

196 …

Design guidelines „

projects as the basis for addressing this question. The most crucial issue is a careful definition of selection criteria for best-practice projects – the more comprehensive the choice of projects, the more convincing the resulting design guidelines would be. To qualify as a best-practice project in our research, an urban river project had to address three principal aims: flood protection, ecological enhancement and amenity. After this main filter was applied, we took care that the set of projects covered a broad spectrum of scales – from small-scale interventions such as steps down to the water through to large projects such as the revitalisation of streams which had been culverted into pipes for several kilometres or the creation of several hectares of retention space to buffer cities against flood surges. As many different design approaches as possible should be presented, which meant that not only particularly spectacular projects should be included. If one project illustrated, for instance, an aspect that appeared in no other example, then this finding was reason enough to include it in the selection. In the end 45 extremely diverse projects were selected, which were all visited and most of them were also discussed on site with planners, ecologists and representatives of the responsible public authorities. After this phase, the most difficult part of the research began: how to organise and abstract the different aspects of the designs so that design guidelines as transferable knowledge could be developed from them? Since water processes played such a crucial role, we started by clarifying what processes actually occur in river spaces. We proposed a fundamental division into two processes: morpho-dynamic processes which actually change the river space by sedimentation and erosion, and non-morpho-dynamic processes which only involve a change in water level without changing the river space itself. In each urban river space, these two processes were more-or-less controlled by setting limits for floodwater and channel development. The two process types along with these two limits gave us a specific point of view from which to analyse the selected 45 urban river spaces. This analysis resulted in a typology of what we called ‘Process Spaces’ consisting of five different urban river space types, or strategies to deal with water processes (Figure 12.1). In Process Space A, ‘Embankment Walls and Promenades’, the banks are very steep and there is hardly any flood area available. For this reason fluctuations in watercourse conditions are mainly vertical and morpho-dynamic processes are consequently excluded. In Process Space B, ‘Dikes and Flood Walls’, large vertical elements limit the flood area at some distance from the normal watercourse. Both horizontal and vertical fluctuations can occur in the watercourse, since the borders of this Process Space only permit very small-scale morpho-dynamic processes. Process Space C, ‘Flood Areas’, comprise spaces near the watercourse that are regularly submerged under its horizontal expansion and present processes within which spatial design has to work. In these three Process Spaces A–C no alterations to the water space itself is intended; water flow fluctuations alone bring about their constantly changing appearance. In Process Spaces D and E, by contrast, morpho-dynamic processes dominate, such as the shifting of sediment or changes to the course of the river; the fluvial dynamics can be read not only in the changing water level but also in changes to the river itself. In Process Space D, ‘Riverbeds and Currents’, when the river is not sealed in place reversible aggradation and erosion processes can happen along the riverbed, with consequences for the form of the riverbed as well as the banks. Process Space E, ‘Dynamic River Landscapes’, is shaped by processes that are to be found in natural watercourses. By including the flood areas in the erosion and aggradation processes, the river can shift its entire course. This abstraction from 45 urban river spaces to five process spaces systematised the broad range of specific solutions, and was the most crucial phase in the research project. It took many months,

197 …

Martin Prominski

„

Figure 12.1 ‘River. Space. Design.’ Design guidelines are organised in five groups of design strategies; each group collects strategies suitable for the respective process space. The five process spaces have been developed in the theoretical part (‘Fundamentals’). Within each strategy, there are between two and nine design tools which are drawn in an abstracted way to allow for a creative transfer to the specific design case of the user. (Source: Prominski et al. 2012, pp.44–45)

198 …

Design guidelines „

Figure 12.2 ‘River. Space. Design.’ Three design tools from the design strategy ‘A1-Linear spatial expansion’ as one example from ‘Process Space A, Embankment Walls and Promenades’. For the strategy, combinations with other strategies are offered, and for the tools, reference projects from the best-practice catalogue in Part B of the book are given. (Source: Prominski et al. 2012, pp.52–53. Note: in the book, additional explanations are given as text, which are left out here.)

with several failed attempts, to arrive at this typology – it was like design, where you envisage different solutions until you finally arrive at a coherent proposal that fulfils the programme in an elegant way. For each of the five Process Spaces, we proposed different design strategies, again by analysing and abstracting from the 45 selected design projects. Each design strategy has its specific design tools and design measures illustrated in a conceptual way, giving future designers freedom to transfer these approaches to their specific cases (Figure 12.2). In summary, the selection and analysis of existing projects aimed to answer an original research question and included a systematic reflection on the basic design strategies used. The result is an abstraction of process spaces with specific design guidelines that can be transferred to future design cases of urban river spaces (Prominski et al. 2012). In the end the extracted abstracted design guidelines should facilitate multifunctional solutions for the many new urban river spaces which will have to be designed in the near future in Europe and beyond.

DESIGN GUIDELINES FOR SHRINKING CITIES The starting point of the applied research project, ‘Design Strategies for Using Common Local Vegetation to Revitalise Unused Space in Shrinking Cities’ (project leaders: Norbert Kühn and Martin Prominski; funded by Deutsche Bundesstiftung Umwelt (DBU) from 2006 to 2008), was the abundant vacant space within settled areas, particularly in the east of Germany where two million residential units were marked for demolition. Usually, shrinking cities have neither a use nor money for maintaining such areas. Furthermore, the cleared spaces must, in many cases, remain zoned for building development in order to avoid existence-threatening asset value loss for the owners. Establishing permanent vegetation such as tree-planting or converting them to parks is not possible. These circumstances led

199 …

Martin Prominski

„

to this research question: ‘How can unused space in shrinking cities be improved in terms of open space use and ecological value in the face of difficult economic and legal conditions?’ The objective of the research was to develop a toolbox with transferable design guidelines that would help many shrinking cities in East Germany and beyond which face the same challenges. The funding institution DBU requires cooperation with practice partners and the East German cities of Wolfen, Dessau and Chemnitz were willing to provide ‘test sites’. These sites belonged either to public housing associations or to the city and the municipal administrations had a strong interest in developing innovative concepts for their vacant areas. The research started with an extensive literature review of existing approaches dealing with vacant areas in shrinking cities. To organise these literature reviews thematically, we used the working metaphor of ‘high-bay storage’. We identified five key issues relevant to our research question: area availability; cost reduction; vegetation types; nature protection strategies; establishment and maintenance. Each of them served as a ‘high-bay rack’, and within these racks thematic shelves were organised; for example, within the rack of ‘cost reduction’ there are the four shelves: ‘innovative planting methods’, ‘cultivation for returns’, ‘maintenance-extensive models’ and ‘procurement of external funding’. Each shelf contained concrete design tools: for example, within the shelf ‘cultivation for returns’ there are the five design tools, ‘urban agriculture’, ‘urban forestry’, ‘renewable energies’, ‘water treatment’ and ‘gardening’, which all have a short description and recommendations for their application. In addition, for each design tool we offered links to best-practice examples from our literature review and useful design tools within other parts of the high-bay racks (Figure 12.3). The research work was characterised by a continuous back and forth flow between the development of site-independent, transferable guidelines and their testing in the site-specific design scenarios. With a first version of this ‘design knowledge storage’, we carried out one-day design workshops with each of our three partners, where we developed three alternative designs for each site. With this ‘designing phase’ we were able to check the applicability of our guideline system and to develop it further by improving existing or adding new tools. Because the resulting high-bay racking was characterised by many links within each design tool, referring to other thematically close tools or to best-practice projects, we decided to present the research report as a PDF file with hyperlinks, making it accessible to all owners of vacant lots who might have similar problems and who might have commissioned landscape architects with specific design projects (Kühn et al. 2008; Prominski et al. 2008).

DESIGN GUIDELINES TO HELP COMBINE NATURE PROTECTION AND OPEN SPACE USE In Germany, nature compensation measures resulting from environmental impact assessments may lead to open spaces devoted solely to nature protection and excluding human uses. This leads to conflicts, especially in the urban realm, due to the limited availability of space, and sometimes even that (the aim of) nature protection encounters societal opposition. Thus, the research project, ‘Design Guide for Nature Compensation Measures in Urban Areas’, asked the following research question: ‘What design guidelines could facilitate the combination of nature protection and open space use in multifunctional urban spaces?’ The project was funded by DBU from 2012 to 2013. We had two project partners: the cities of Bremen (in German, Freie Hansestadt Bremen, Senator für Umwelt, Bau und Verkehr) and Hamburg (in German, Freie und Hansestadt Hamburg, Behörde für Stadtentwicklung und Umwelt). Each partner

200 …

Design guidelines „

Figure 12.3 The design guideline system of high-bay racking in the project on shrinking cities: top, there are the five thematic high-bay racks. Bottom left illustrates the content of the high-bay rack ‘Cost reduction’. It has four areas, for example ‘Cultivation for Returns’; in each area there are shelves (e.g. ‘Urban Agriculture’) which contain the single design tools. Bottom right displays the content of the design component ‘Use of urban meadows’, which contains a description, notes for application and links to best-practice examples in the annex or useful design tools within other parts of the high-bay racks. (Source: Prominski et al. 2008)

201 …

Martin Prominski

„

had specific areas and spaces where they wanted to apply the preliminary results of the project. In Bremen this was the area called ‘Waller Fleet’, a huge allotment garden located to the north of a working-class city district. This area is separated from the city by railway tracks and a motorway, and due to such barriers and other disadvantages an increasing number of lots are falling vacant. The city was following a strategy to pool its compensation measures in the ‘Waller Fleet’ and wanted to create an area where leisure and nature protection form a productive symbiosis. The task in Hamburg was completely different: it related to difficulties establishing habitat connectivity. According to Sections 20 and 21 of the German Nature Protection Law, networks of flora and fauna habitats are required to be developed all over the country. Because of high urban density, Hamburg cannot rely solely on nature protection areas to achieve such a network; it also must try to integrate existing green open spaces. Thus, the task here included the identification of areas where compensation measures could enhance habitat networks without decreasing the amenity values of existing open spaces. The project started, in a similar way to the urban river research, by identifying and analysing bestpractice examples. To be evaluated as best practice, projects in the urban realm had to demonstrate a multifunctional approach by fulfilling clearly defined nature conservation goals and allowing open space use at least in part of the project. The design strategies and tools we derived from these examples were provisionally systematised and abstracted into design guidelines to serve as a basis for the designs in Bremen and Hamburg. In designing these two projects, we began by initially testing a broad spectrum of design strategies. Our approach was to improve and expand on provisional design guidelines as well as to invent completely new ones. Since the two design tasks in Bremen and Hamburg were broad and complex, they could play a significant role in developing transferable design guidelines. After this ‘research through designing’ phase we took the key step of systematising and abstracting the multifunctional design strategies and tools derived from both existing and proposed designs. Similar to the two previously described research projects, the most difficult step was to develop a coherent order which is able to integrate the design guidelines developed from the diverse designpractice examples. This order should be easy to understand and to apply by future users who want to integrate nature protection and open space use. First, we tried to relate all design strategies and tools to two main categories – nature protection and open space use – and to develop hybrid clusters from them. This did not work, so in the second attempt we developed categories from nature protection theory such as continuity, dynamics, resources or network but again this did not work. The reason for failure was always the fact that the design strategies and tools found in the examples could not be logically integrated into the proposed order. Either the strategies and tools did not fit or a category was not clear enough and many different strategies would fit into one category, which causes confusion. After running through many unsuccessful propositions we came up with nine design strategies (in German Entwurfsfelder) as top categories, and each included three to four design tools (in German Entwurfswege) as subcategories (Figure 12.4; Prominski et al. 2014).

DISCUSSION Comparison of the three projects The three research projects presented above all started with a research question, which led to a selection of best-practice examples according to criteria and the reflection of the theoretical context

202 …

Design guidelines „

Figure 12.4 Diagram of the research method or process to develop design guidelines. The research question and the reflection of the theoretical context are the foundation for the later phases, which consist of research into design (Analysing best-practice examples) or research through design (Test designs). A combination of both is also possible, and the relation of research into and through design depends on the research subject. (Source: Prominski et al. 2014)

203 …

Martin Prominski

„

of the research topic. In research terms these examples formed a data set, a selective sample the analysis of which against criteria from literature and so on was used to develop or abstract sets of design guidelines. The design guidelines with design strategies and tools were thus abstracted from the best-practice examples and put into a coherent system in a way which we think is applicable to the application process used by practitioners. In the two examples ‘Shrinking Cities’ and ‘Nature Protection and Open Space Use’ the development of design guidelines was supported by ‘test designs’ which served as a valuable feedback loop to improve the system. However, the successful result of the example ‘River. Space. Design.’ indicates that test designs are not mandatory. In terms of making the research findings public, we decided for all three projects to present not only the abstracted guidelines, but also the best-practice examples used to inform them. The reason for this was that in our experience it is useful to see how a single design guideline, for example ‘Combining habitat networks and open space system’, is integrated with other aspects in a concrete design project. Thus, users of our research output would be able to switch between the guidelines and best-practice projects. From the project ‘Shrinking Cities’ we learned that the selection of a metaphor for the system of design guidelines has to be done carefully – our choice of ‘high-bay racking’ led to misunderstandings because some people first understood it literally and thought we would propose a physical high-bay rack in their shrinking city!

Strengths and weaknesses of the approach Reflecting on our research into design guidelines, and starting from the three research projects presented above, there are a number of strengths and weaknesses from the perspective of the user, that is practicing landscape architects (or other designers). The strengths of design guidelines are, first and foremost, the improvement and acceleration of future design processes, although this claim requires further research to support it. By selecting best-practice examples, reflecting on the existing theoretical context and, finally, offering an applicable system of design guidelines, the research gives a solid foundation for reflective practice in the design of specific cases. It also saves practitioners much time, because scanning the literature and the internet for sound theory and best-practice examples needs time which practitioners might put to better use. Without guidelines, it is more an issue of chance if practitioners find the right theoretical sources and best-practice projects. A second strength is the abstract character of the guidelines. This was the most demanding part of the research, but it was also extremely important because only then might the necessary openness for the application to specific design cases be guaranteed. A weakness of the method is the fact that it is fixed in time. By necessity, there has to be a selection only of current best-practice projects as well as the reflection on the current status of knowledge and theory. However, the design world moves on, new research emerges and creativity continues, while theory also continues to develop. Thus, guidelines need regular updates in terms of projects and theory, a task for which funding could be difficult to acquire because, as a ‘second edition’, it would not be sufficiently new and innovative. However, this weakness does not justify discarding the development of design guidelines as a valuable tool. The online publication of guidelines such as those for the ‘Shrinking Cities’ project makes it easier to add novel practices and to update guidelines. After all, ideally, design guidelines inspire practitioners to designs which are more innovative than the best-practice projects used in the research and thus the guidelines’ success may contribute to their becoming outdated. Future research might use systems such as crowdsourcing to

204 …

Design guidelines „

help in keeping design guidelines abreast of current developments. More research is also needed to understand better whether and how design guidelines contribute to the improvement, acceleration and creativity of designing in practice, and the concordance between a designer’s concepts and the objective evaluation criteria.

Design guidelines and the larger research landscape Where does research for purposes of developing design guidelines sit within the wider research landscape? Does it fulfil the standard research criteria which funding institutions set for all sciences, or is it a special kind of research which needs ‘softer’ criteria? For example, the DFG, the most prestigious German research funding institution, places the following demands on all proposals, including those coming from the sciences, the humanities, engineering and others (DFG 2014, p.4). First, to be original – every proposal needs to pose a new research question. Then, this question has to be relevant inasmuch that responding to it will result in an advancement of knowledge. This knowledge has to be of scientific significance, thus the proposal has to reflect the scientific context in a critical way and set itself in relation to it. Finally, a broader impact (i.e. on society at large) has to be demonstrated by showing how the knowledge can be communicated and generally transferred. When we reflect on the three projects, all of them posed a research question which was original. The scientific context was reflected in a critical way, which contributed to scientific significance, for example the concept of nature in relation to nature protection was developed further in ‘Nature Protection and Open Space Use’. In the project, ‘Process-Orientated Design of Urban River Spaces’, an observation of river types and processes led to a new categorisation of urban river spaces. The design guidelines themselves express an advancement of knowledge, because their systematic abstraction had never been undertaken before for the respective subjects. Finally, the results have the potential for a broader impact because design guidelines are abstracted in a way that is transferable to a wide field of future design tasks within the respective topic and they are published in a format which is easily accessible by the target audience. This assessment indicates that research used to develop design guidelines is able to fulfil the same tough criteria set for other scientific research. If broken down to the basic aspects, design research does not seem to differ from other science.

Design guidelines and the design research landscape In the context of design research, the division of research about, for and through design (Frayling 1993; Jonas 2012) is still the most influential and productive categorisation. According to Jonas (2012, p.23), research about design reflects design work from a distance without changing it, for example design history, theory or criticism. Research for design also operates from a distance and researchers are suppliers of knowledge. In terms of landscape architecture, these two categories of research are often and usually performed by members of the history and theory departments and also by the construction, planning-related sociology, and by ecology or vegetation departments. The third category, research through design, is the type of research in which the act of designing is the essential component of the research. This is the most specific and interesting type of design research, but also the most difficult. The writings of design research theorists like Frayling (1993), Borgdorff (2007) and Jonas et al. (2013) include a few realised examples with rather vague descriptions. An important contribution has been made by Lenzholzer et al. (2013), who redefine the category as

205 …

Martin Prominski

„

‘research through designing’. By using the active verb, they stress the design process itself as the decisive quality of this type of research. They also give some good completed examples of research through designing which clarify the link between landscape architecture’s core action and research (see also Lenzholzer et al. – Chapter 4). Where is the position of design guidelines research in these three design research categories? It is definitely research about design, because the three research projects described above all reflected the theoretical context of the general research subject as well as several best-practice examples. Two of the research projects, ‘Shrinking Cities’ and ‘Nature Protection and Open Space Use’, also worked with elements of research through designing. In ‘Nature Protection and Open Space Use’, test designs for sites of the two practice partners in Hamburg and Bremen influenced the resulting design guidelines in the same way as did intensive designing work for the three cities of Wolfen, Dessau and Chemnitz in the ‘Shrinking Cities’ project. Design guidelines are a supplier of knowledge and thus also to be included in research for design. This might sound a bit stretched, but in the end it is difficult to falsify. In fact, all design research is, in the end, research for design. Thus a more precise wording of the description of the research-for-design category within the design research discourse could be considered. This reflection on the three design research categories indicates that it is impossible to relate research for developing design guidelines exclusively to one of the three categories. The categories interact, and it can be concluded that for the case of design guidelines, it is hard to imagine research through designing being done without elements of research about design. This does not mean that the three-category model of design research should be abandoned, but it does suggest that a strict separation of the categories does not make sense. Instead, a mix of the categories seems productive.

CONCLUSION I would like to conclude this chapter on research for developing design guidelines by proposing five components: s

s

s

Research question. As shown in the three research projects presented above, a strong research question is fundamental, because it serves as the focal point for the whole research process. The research question guides not only the choice of best-practice examples or design sites, but especially it guides the development of the structure for the design guidelines. If a decision is to be made on the coherence of a proposed order of guidelines or tools, the research question is an important reference point. ‘The truth is on the pitch’ (in German Die Wahrheit ist auf dem Platz), a famous saying by Sepp Herberger who coached Germany’s world cup winning football team in 1954. When we talk about evidence-based design guidelines, the truth is also on the pitch, that is in the projects. Design guidelines cannot be developed out of thin air, but have to relate to actual existing design work. They could be developed from best-practice projects and/or from test designs as demonstrated in this chapter. Theoretical context. Although the truth is on the pitch, a neglect of theory would be misleading. An understanding of the contemporary theoretical context, for example for ecology, floodwater management or nature protection, allows a better selection of innovative projects as well as a future-orientated framing of the categories for the design guidelines.

206 …

Design guidelines „

s

s

Criteria for choosing best-practice examples or test-design sites. The choice of best-practice examples or design sites is crucial for the future applicability of the design guidelines. Research will be most successful if the guidelines are helpful for all conceivable design cases which relate to the research question. Thus, a certain diversity in terms of scale, geographic situation, approaches and so on has to be covered by the examples in order to develop widely applicable design guidelines. Another criterion may be their innovative character. It is important to mention that it is impossible to provide a list of criteria which is valid for all design guideline research projects, because each research question demands a different set and range of criteria. Abstraction with an open character. The way design guidelines are developed and communicated graphically or verbally should make clear that they are not a recipe which guarantees success independent of where you cook. A design is always specific and guidelines can only support it. In practice, they have to be adapted to the specific situation of the site. Thus, design guidelines never automatically produce good designs and they cannot serve non-designers as a recipe book – it needs skilled landscape architects to apply design guidelines. To make this clear, the presentation of design guidelines needs a degree of abstraction which, while giving guidance, still leaves it open to adapt them to the specifics of a design situation. The pictograms in ‘River. Space. Design.’ are a good example of this openness by abstraction.

In summary, a core for research for the development of design guidelines has become visible from the description and discussion of the three research projects described above and enables meeting the same application funding criteria as for other sciences. Some variations within this core are possible – when few existing projects are available, more research through designing by means of test designs will be necessary. The resulting design guidelines serve practitioners – they are like variously applicable ‘design enzymes’ which expand the knowledge base of individual design disciplines and their trans-disciplinary intersections, thus becoming a worthwhile goal of future paths in design research.

REFERENCES Borgdorff, H. (2007) ‘The debate on research in the arts’, Sensuous knowledge: Focus on Artistic Research and Development, 2, Bergen: Bergen National Academy of the Arts. DFG (2014) Guidelines for the Written Review, Bonn: DFG form 10.02, available: http://www.dfg.de/ formulare/10_20/10_20_en.pdf (accessed: 06.11.2015). Frayling, C. (1993) ‘Research in art and design’, Royal College of Art Research Paper, 1(1), 1–5. Jonas, W. (2012) ‘Exploring the swampy ground’, in Grand, S. and Jonas, W., eds. Mapping Design Research, Basel: Birkhäuser, 11–41. Jonas, W., Chow, R. and Grand, S. (2013) ‘Alternative design doctorates as drivers for new forms of research’, in Engels-Schwarzpaul. A.-Chr. and Peters, M.A., eds. Of Other Thoughts: Non-Traditional Ways to the Doctorate: A Guidebook for Candidates and Supervisors, Rotterdam: Sense Publishers, 183–202. Kühn, N., Prominski, M., von Birgelen, A. and Langner, S. (2008) Hochregallager: Entwurfsbausteine, Berlin and Hannover, available: http://www.freiraum.uni-hannover.de/fileadmin/freiraum/Forschung/080819_ HRL.pdf (accessed: 06.07.2016). Lenzholzer, S., Duchart, I. and Koh, J. (2013) ‘“Research through designing” in landscape architecture’, Landscape and Urban Planning, 113, 120–127. Oxford English Dictionary (1996) 2nd ed., Oxford: Oxford University Press.

207 …

Martin Prominski

„

Prominski, M., Langner, S., Kühn, N., von Birgelen, A. (2008) Aufwertung ungenutzter Freiflächen im Stadtumbau. Handbuch zur Verwendung des “Hochregallagers“ als entwurfsorientiertem Wissensspeicher. Hannover and Berlin, available: http://www.freiraum.uni-hannover.de/fileadmin/ freiraum/Forschung/080924b_Handbuch.pdf (accessed: 06.07.2016). Prominski, M., Stokman, A., Zeller, S., Stimberg, D. and Voermanek, H. (2012) River. Space. Design.: Planning Strategies, Methods and Projects for Urban Rivers, Basel: Birkhäuser. Prominski, M., Maaß, M. and Funke, L. (2014) Urbane Natur gestalten: Entwurfsperspektiven zur Verbindung von Naturschutz und Freiraumnutzung, Basel: Birkhäuser. Schön, D.A. (1983) The Reflective Practitioner: How Professionals Think in Action, New York: Basic Books.

208 …

PART IV ADDRESSING SOME OF THE GRAND CHALLENGES

This page intentionally left blank

Chapter 13: Cultural landscape meanings and values Ken Taylor

Any landscape is a condition of the spirit. (Henri Frédéric Amiel)

INTRODUCTION This chapter explores the contemporary challenge of recognising, protecting and managing intangible values of cultural heritage with specific reference to cultural landscapes. This is not to deny that cultural landscapes have physical, tangible shape. More importantly for this chapter they demonstrate associative intangible values related to the meaning of a landscape for the landscape makers, owners and visitors/users, that is, for the community in its widest sense and their engagement with an active use of the heritage. Such an approach to cultural heritage generally, and cultural landscapes in particular, marks a divergence from the elite connoisseurship approach to cultural heritage – the Authorized Heritage Discourse (AHD) (Smith 2006) – focusing on famous archaeological remains and buildings that held singular sway until the mid-1980s. Appreciation of associative values – like a breath of fresh air – presents an opportunity for landscape architecture research to make a distinctive contribution to the field of study and practice of understanding intangible aspects of landscape and to the process of how human meanings and values accumulate. The emergence of the idea of values and meanings is now inextricably linked to the concept of cultural heritage and prompts the question: ‘What is meant by the conjunction of the two words “cultural heritage”?’ The word ‘cultural’ derives from ‘culture’ in the way that Horne (1986) nicely phrased as: ‘the repertoire of collective habits of thinking and acting that give particular meanings to existence’. In this sense it can be legitimately claimed that all heritage is intangible (Smith 2011). Byrne succinctly encapsulates this: Those of us who have pushed for recognition of ‘the intangible’ in heritage work are also those who tend to stress the ‘cultural’ in cultural heritage. We try to resist the tendency of heritage discourse to reduce culture to things, we try to counter its privileging of physical fabric over social life. (Byrne 2009, p.229)

The chapter enlarges on the points raised by Byrne to present an overview of cultural landscape meanings and values and associated research approaches that address intangible aspects and community values. The chapter then examines the way in which research knowledge informs

211 …

Ken Taylor

„

methods used in practice to unravel landscape meaning particularly from the point of view that landscape is not simply a product of human endeavour, ‘not as object to be seen or a text to be read, but as a process by which identities are formed’ (Mitchell 1994, p.1). Landscape is not simply what is seen as an assembly of physical components and natural elements, but rather, as Cosgrove proposes, it is a way of seeing that has its own history, but a history that can be understood only as part of a wider history of economy and society; that has its own assumptions and consequences, but assumptions and consequences whose origins and implications extend well beyond the use and perception of land; that has its own techniques of expression, but techniques which it shares with other areas of cultural practice. (Cosgrove 1984, p.1)

Pivotal to critical research inquiry into these lines of thinking prompts the question: ‘Whose values are we addressing and whose heritage is it?’ (Taylor 2014, p.1939). It is this research question that informs the inquiry set out in this chapter. A subsidiary question is: ‘What does this mean for landscape architects who may find themselves undertaking research in the field of historic cultural landscapes?’

CONCEPTUAL FRAMEWORK What is cultural landscape? Fundamental to an examination of values, meanings and cultural heritage processes is the word ‘landscape’ itself and then its conjunction with the word ‘cultural’ to give us ‘cultural landscapes’. Why is this? It is because: Inextricably linked to a cultural concept of landscape is the understanding that one of our deepest needs is for a sense of identity and belonging and that a common denominator in this is human attachment to landscape and how we find identity in landscape and place. Cultural landscape study has also been coincidental with a widening interest in the public history movement and everyday landscapes. It underpins the notion that landscapes reflecting everyday ways of life, the ideologies that compel people to create places, and the sequence or rhythm of life over time tell the story of people, events and places through time, offering a sense of continuity: a sense of the stream of time. They also offer the context for broader concepts and understandings of cultural heritage than monuments and sites … The cultural landscape concept is therefore intended to increase awareness that heritage places are not isolated islands and that there is interdependence between people, their social structures and ecosystems, and landscape conservation. (Taylor 2015b, p.1)

If, as the above quote suggests, there is an immutable link between cultural landscapes and modern thinking on cultural heritage, it is useful to look at a definition of cultural landscape. Here I refer readers to ‘Cultural Landscape: Dreadful Phrase, Great Concept’ in which Fowler (2001) includes a number of definitions. The definition I quote below, and why it is quoted, is because it is succinct. Like Fowler I find it theoretically and professionally workable: the last sentence expressing the very essence of what we mean by ‘cultural landscapes’ with ‘a brevity beguiling its profundity’:

212 …

Cultural landscape meanings and values „

Cultural landscapes reflect the interactions between people and their natural environment over space and time. Nature, in this context, is the counterpart to human society; both are dynamic forces, shaping the landscapes … A cultural landscape is a complex phenomenon with a tangible and intangible identity. The intangible component arises from ideas and interactions which have an impact on the perceptions and shaping of a landscape, such as sacred beliefs closely linked to the landscape and the way it has been perceived over time. Cultural landscapes mirror the cultures which created them. (Plachter and Rössler 1995, p.15)

The definition evolved from discussions at a United Nations Educational, Scientific and Cultural Organization (UNESCO) international experts’ meeting held at Templin in Germany in 1993 following UNESCO’s 1992 initiative of recognising three categories of cultural landscapes for World Heritage listing purposes. The three categories below evolved out of increasing interest in the cultural landscape concept during the 1980s and early 1990s so that, as it gathered momentum, it permeated cultural heritage management and planning theory and practice: s s

s

Clearly defined landscapes designed and intentionally created by man; Organically evolved landscapes in two categories: ƕ A relict or fossil landscape in which an evolutionary process has come to an end but where its distinguishing features are still visible; ƕ Continuing landscape which retains an active social role in contemporary society associated with a traditional way of life and in which the evolutionary process is still in progress and where it exhibits significant material evidence of its evolution over time; Associative cultural landscapes: the inclusion of such landscapes is justifiable by virtue of the powerful religious, artistic, or cultural associations of the natural element rather than the material cultural evidence (UNESCO 2008, Annex 3).

Critical to the 1990s movement were the 1960s and 1970s scholarly writings of cultural geographers like David Lowenthal, Peirce Lewis, Donald Meining, J.B. Jackson with his inimitable essays on the everyday American scene, Dennis Cosgrove in Britain or Dennis Jeans in Australia. They built on the late nineteenth century German tradition of Otto Schlüter’s Kulturlandschaft with landscape morphology seen as a cultural outcome, and Franz Boas who championed the idea that different cultures adjusted to similar environments and taught the historicist mode of conceptualising environment. Boas argued that it was important to understand cultural traits of societies – their behaviours, beliefs, and symbols – and the necessity of examining them in their local context. He also understood that as people migrate from one place to another, and as the cultural context changes over time, the elements of a culture, and their meanings, will change, which led him to emphasise the importance of local histories for an analysis of cultures (Livingstone 1992; see also http://en.wikipedia. org/wiki/Franz_Boas). His teachings and ideas in social anthropology and geography remain central to present-day interest in the cultural landscape idea where landscape is a clue to culture. At this point in the chapter it is germane to acknowledge that a fundamental question or dilemma facing any critical discussion on cultural landscape is whether the term ‘cultural’ is in reality redundant. Why use it if discourse on ‘landscape’ is inextricably linked to aspects of culture, nature, diversity and human identity leading to the idea that all landscape is culturally defined? Is cultural landscape a problematic term? The Chinese scholar and landscape architect, Feng Han (Han 2006) believes

213 …

Ken Taylor

„

so. She argues for example that in China the term is problematic. She posits that people are part of the landscape experience and that landscape in the context of nature has its specific meanings which, she argues, contrast with Western notions, including inter alia that it is humanistic rather than religious; it is aesthetic rather than scientific; travelling in nature aims to be enjoyable, instead of solitude oriented; artistic rebuilt nature is more beautiful than the original (Taylor 2009). As Greffe (2010, p.1) clearly articulates ‘we may wonder if there are really any landscapes that are not cultural’. Similarly regarding the term ‘natural landscape’, one may pose the question of whether such a thing exists. Most landscapes, rural and urban, include natural elements that physically shape the land that humans occupy. This is land which we settle, modify, and imbue with meanings and values so that through time we can recognise layers of human occupation and change. It is the process of human change that creates the thing we call landscape and how landscapes vary not just through time but according to different cultural values and ideologies. The more one experiences other people’s landscapes the more one appreciates the remarkable cultural diversity that is represented in the world’s landscapes and how biodiversity is a fundamental aspect of many cultural landscapes. Such is the case with the International Union for Conservation of Nature’s (IUCN’s) Category V Protected Landscapes as outlined for example by Dudley and Stolton (2012) in Protected Landscapes and Wild Biodiversity.

Brief history of the term ‘landscape’ Landscape is a ubiquitous word in English and other European languages with origins in AngloGerman language dating back to c.500 AD in Europe. The words – landskipe or landscaef – and the notions implied were taken to Britain by Anglo-Saxon settlers (Jackson 1984). The meaning was a clearing in the forest with animals, huts, fields, fences. It was essentially a peasant landscape carved out of the original forest or weald, that is out of the wilderness with interconnections to patterns of occupation and associated customs and ways of doing things. Landscape from its beginnings, therefore, has meant a man-made artefact with associated cultural process values. Here is a holistic view of landscape as a way of seeing – its morphology resulting from the interplay between cultural values, customs and land-use practices – critically explored by Wylie (2007) in his book Landscape; Olwig (2007) nicely summarises what we mean by landscape as ‘an active scene of practice’. Jackson (1984) further indicates there is an equivalent meaning in Latin-based languages derived from the Latin pagus, meaning a defined rural district. He notes that this gives the French words pays and paysage, but that there are other French words for landscape including campagne deriving from champagne meaning a countryside of fields; the English equivalent once being ‘champion’. But what is ‘landscape’ and what are its connections with human memory? On the first question it is instructive to go back to two mid-twentieth century pioneering teachers of landscape study, J.B. Jackson and W.G. Hoskins. Jackson (1984, p.8) in his reflections on what landscape is quotes what he calls ‘the old fashioned but surprisingly persistent definition of landscape: “A portion of the earth’s surface that can be comprehended at a glance”’. He saw landscape as “A rich and beautiful book [that] is always open before us. We have but to learn to read it” (Jackson 1951, p.5). Hoskins (1955, p.14) asserted the significance of landscape with the proposal that ‘The ... landscape itself, to those who know how to read it aright is the richest historical record we possess.’ What Hoskins and Jackson were contending forms the modern foundation for landscape study. This is where landscape is not looked on as simply a pretty picture or as a static text: rather it is the expression of landscape as cultural process. The connections, therefore, between landscape and

214 …

Cultural landscape meanings and values „

identity and hence memory, thought and comprehension are fundamental to understanding of landscape and human sense of place. People see and make landscapes as a result of our shared system of beliefs and ideologies (Biger 2006). In this way landscape is a cultural construct, a mirror of our memories and myths encoded with meanings which can be read and interpreted. Over thirty years ago Meinig (1979, pp.1–3) proposed that ‘Landscape is an attractive, important, and ambiguous term [Meinig’s emphasis] [that] encompasses an ensemble of ordinary features which constitute an extraordinarily rich exhibit of the course and character of any society’ and that ‘Landscape is defined by our vision and interpreted by our minds’. In other words, to understand ourselves we need to look searchingly at ‘landscapes as a clue to culture’ (Lewis 1979, p.15), and our ordinary everyday landscapes at that, not just the national icons. It is what geographers refer to as reading the landscape. Notably there have been, and remain, tensions in various schools of thought on how we see landscape and study it. Wylie (2007, pp.1–2) posits that the tension is ‘between proximity and distance, body and mind, sensuous immersion and detached observation. Is landscape the world we are living in, or a scene we are looking at, from afar ... a set of visual strategies and devices for distancing and observing?’ Here is the tension between our lived-in world concept and landscape as an artistic and historical genre, and both have relevance to the human experience of landscape. The European Landscape Convention is instructive in thinking on landscape as a construct. It presents a simple but decisive definition: ‘Landscape means an area, as perceived by people, whose character is the result of the action and interaction of natural and/or human factors’, that is a cultural landscape. In the modern idiom landscape is viewed as humanistic where culture/ nature are not divided. This culture–nature link is also a fundamental principle in the World Heritage cultural landscape categories. The old Germanic/English landscaef connotation has in effect been revitalised. What is also significant about the Convention is that it ‘recognises the potential value of all landscapes to communities’ (Taylor et al. 2015, p.4) including the ordinary, everyday or vernacular landscape. In the convention it can be seen that: In particular, the ordinary landscapes where most people live are seen as having potential value to someone, even though the quality may be low in terms of many of the commonly identified indicators, such as scenic beauty, biodiversity rating, range of use and accessibility. The emphasis here is very much on the value to ‘someone’ (communities, cultures and individuals). (Roe and Taylor 2014, p.8)

Values ‘Conservation of cultural heritage in all its forms and historical periods is rooted in the values attributed to the heritage’ (ICOMOS 1994, p.46). Related to the concept of values is that of meaning of heritage places, including landscapes. It is important to note in relation to meaning that it will change over time as cultural traditions and values change. The elucidation of meanings and values is, therefore, fundamental to the cultural heritage process, and indeed, a values-based approach to research and practice is international in scope. There are internationally available typologies of heritage values. ‘By use of such a typology – a framework that breaks down significance into constituent kinds of heritage value – the views of experts, citizens, communities, governments, and other stakeholders can be voiced and

215 …

Ken Taylor

„

compared more effectively’ (Mason 2002, p.9). What is critical to the process is that ‘the respect due to all cultures requires that heritage properties must be considered and judged within the cultural contexts to which they belong’ (ICOMOS 1994, p.47). ICOMOS (International Council on Monuments and Sites) is a major international agency in the field of cultural heritage conservation. Established in 1965 ICOMOS is a non-governmental organisation with headquarters in Paris dedicated to the conservation of the world’s historic monuments and sites. It provides a forum for professional dialogue and a vehicle for the collection, evaluation and dissemination of information on conservation principles, techniques and policies. It also advises UNESCO on World Heritage cultural matters. There are 108 national committees of ICOMOS established in countries around the world that are members of UNESCO. A sample of values is shown in Table 13.1 (see also Mason 2002). The values-based approach to heritage conservation including addressing intangible heritage is, for example, central to Australian theory and practice through the Burra Charter (Australia ICOMOS 2013). Practice Note 1 in the Charter refers to aesthetic, historic, scientific, social and spiritual values and how these inform the notion of significance expressed in a statement of significance: Cultural significance is the sum of the qualities or values that a place has, including the five values—aesthetic, historic, scientific, social and spiritual—that are listed in Article 1.2 of the Burra Charter. Through the processes of investigating the place and assessing each of these values, we can clearly describe why a place is important. This is the first step towards ensuring that our decisions and actions do not diminish its significance. (Diefenthaller 2014, p.8)

Table 13.1 Examples of heritage value typologies

216 …

Cultural landscape meanings and values „

In practice values are normally assessed and analysed so that a statement of significance can be prepared for the heritage resource under study and its management. For World Heritage properties significance is set out in the statement of Outstanding Universal Value (OUV). It is therefore critical that the process of research and also that of documentation used in practice, aiming to unravel values and significance of landscapes, follows a framework: s s

s

That is apparent and understandable by other researchers, clients, community and other stakeholders; Includes methods that are replicable so that their application may be tested elsewhere and modified where appropriate; archival and on-site research will be vital components (see next section on ‘Reading the landscape’) in the context of informing rigorous practice; Allows evaluation of proposed decisions or recommendations and be appropriate to the country and culture in which you are working.

Aesthetics At this point I propose to venture into the realm of aesthetics and aesthetic value, not least because aesthetics is a term regularly used in landscape architecture. Hence what is meant by aesthetics, is it high art or context? Aesthetics has assumed a high profile since the eighteenth century in the way humans see the world. Places and objects are referred to as aesthetically pleasing or valuable to convey a particular idea and ideal. But it may not always be clear what the idea or ideal exactly are. In heritage practice, as in everyday use, it can be a misused term, too often applied superficially to a predetermined idea of what is assumed to be an ideal. In this vein the Burra Charter value that I find difficult is that of aesthetic value. The Charter refers to criteria to do with sensory perception: form, scale, colour, texture and material of the fabric which appear woolly to me and too preoccupied with how something looks. Equally the reference to beauty and aesthetic ideals is challenging: what does it mean? The process becomes confused with the Western history of aesthetics and particularly the eighteenth century notion of aesthetic being equated with beauty and good taste. For me aesthetic concerns are equally those dealing with experience and this can and does cover the ordinary everyday places that we may not usually refer to as beautiful. But why not? They are the places imbued for many through experience with a sense of belonging and sense of place where knowledge of ways of doing things is critical. I contend that the China Principles document (China ICOMOS 2002) with its description of artistic value is more comprehensive and embracing than the Burra Charter, particularly in the way it addresses intangible cultural context aspects of artistic values. Nevertheless aesthetic value can be significant where it is expressed in architectural or landscape design terms as an achievement of a recognised high order of cultural excellence. Examples would include Versailles Gardens in France, the Taj Mahal complex in India or the imperial gardens in China (Souzhou) and Japan. I think it is necessary to think of aesthetics as a style of perception as suggested by Bullock and Stallybrass in a definition of the term: The philosophical study of art and also of nature to the extent that we take the same attitude to it as we do to art. The notion of an aesthetic attitude is thus of central importance. It is commonly held to be a

217 …

Ken Taylor

„

style of PERCEPTION [original emphasis] concerned neither with factual information to be gained from the things perceived, nor with their practical uses, but rather with the immediate qualities of the contemplative experience itself. (Bullock and Stallybrass 1978, p.10)

Eagleton (1995, p.13), in his critique of the modern discourse on aesthetic thought, proposes that ‘Aesthetics is born as a discourse of the body’ and that ‘the term refers not in the first place to art, but to the whole region of human perception and sensation in contrast to the more rarefied domain of conceptual thought’. Eagleton suggests further that aesthetics is to do with affections and aversions, the whole of our sensate life, which post-Cartesian philosophy has managed to overlook. In this sense aesthetics concerns sensory perception as in the Burra Charter, but in all of the ways in which we refer to the world around us through experience and association of ideas. Eagleton (1995, p.16) sees this as ‘simply the name given to that hybrid form of cognition which can clarify the raw stuff of perception and historical practice disclosing the inner structure of the concrete’. The aesthetic encompasses the idea of the beautiful ‘where objects stand out in a sort of perfection dimly akin to reason’ (Eagleton 1995, p.17). Unfortunately ‘aesthetic’ became a substitute for ‘beautiful’ in early nineteenth century English, denoting good taste (and hence bad taste as the opposite, or even tastelessness, where we do not expect a place to exhibit aesthetic value). For example, one commonly expressed aesthetic value of landscape has become entangled with purely visual images related to ideas of picturesque scenes/scenery corresponding with Eagleton’s comment on objects standing out in a sort of perfection. This is not an argument against the appeal of the picturesque, far from it. What has happened is that it is an extension of the eighteenth century concept of beauty and the picturesque, but often ignores the fact that beauty and utility were inseparable in the concept and that landscape and place are a cultural construct, a way of seeing, with attendant affections and aversions. Whereas, there has been a tendency, by some practitioners in the design professions (including landscape architecture), to abrogate the meaning of aesthetic value to be limited to formal abstract qualities. The relevance for landscape architecture and research possibilities regarding aesthetics is summarised by Berleant (2010a) in an online essay where he observes: Expanding the scope of aesthetics raises challenging questions about the experience of appreciation. Traditional accounts of aesthetic appreciation are inadequate to identify and illuminate the perceptual satisfactions that these new applications evoke. But not only does an enlarged range of aesthetic appreciation recognize beauties beyond the arts. It also must account for the range of aesthetic perception into the oneiric, the bizarre, and the terrible, while the social and political significance of aesthetic values has led to the recognition of a wide range of such values, not all of them positive. (Berleant 2010a, Introduction, para. 2; see also Berleant 2010b)

Matters of aesthetic response to, and social value of, places or objects involve critical engagement, that is participatory transaction or interest between people and place/object. In an urban setting engagement concerns what Hayden (1995) refers to as: ‘the power of place, the power ... to nurture citizens’ public memory, to encompass shared time in the form of shared territory’. Aesthetic and social values are abundant in social and cultural meaning and sense of identity. Of note in Hayden’s discourse is the reference to people who shape places, particularly understanding the role of people

218 …

Cultural landscape meanings and values „

who experience, live, work, recreate in urban places. This notion is commensurate with the idea of layers through time inherent in cultural landscapes and reflecting values of people who inhabit them. Hence such places are repositories of social history and community values. Coincidental to scholarly interest in urban place making is the current emergence of challenges to the long held orthodoxy of the focus of urban conservation that has historically been on architectural fabric and planning ensembles with an emphasis all too often on famous buildings or monuments. This is the point of departure for the new Historic Urban Landscape (HUL) concept which views urban areas as cultural landscapes (Taylor 2015a) where conservation of historic urban landscapes needs to be seen within the paradigm of understanding cities as communities of people changing through time, not merely architectural fabric to be selectively conserved (Bandarin and van Oers 2012). Here there are connections with the concept of landscape urbanism. Bandarin (2015, p.2) points to the way ‘urban heritage can no longer be conceived of as a separate reality, a walled precinct protected from external forces of change by plans and regulations’. He critically reviews how the ‘domains of Ecological Urbanism and Landscape Urbanism open up a new dimension for urban conservation’. Thompson (2012) in a pithy interpretation of landscape urbanism probes, amongst other topics, landscape urbanism’s attack on the urban–rural binary – what he calls the opposition between city and landscape – and whether landscape urbanism offers any contribution to the challenges of worldwide urbanisation. Both authors notably make a link with the seminal work of Ian McHarg in his 1969 book Design with Nature which injected the idea of landscape as critical to the urban management process.

READING THE LANDSCAPE: IDENTIFICATION, DOCUMENTATION, ASSESSMENT, ANALYSIS AND EVALUATION The process and policy model for undertaking a cultural heritage landscape study is set out in Figure 13.1. Very often this process is linked to that of preparing a Conservation Management Plan (CMP), whether it be a designed landscape such as a park or garden (public or private), an everyday living rural landscape or a historic urban landscape. In order to prepare conservation management and sustainable development plans, a systematic approach to reading and understanding the landscape is recommended in order to address the steps in Figure 13.1 of involving key interest groups, documenting history, and significance assessment. Methods used must give adequate consideration to intangible values of landscape as well as physical fabric so that the primary focus for future planning is grounded firmly in aspects of cultural sustainability and cultural diversity as well as environmental sustainability and biodiversity. In undertaking a study of a cultural landscape I find it helpful to address the following specific research questions: s s s s s

What has occurred, When did it occur, Where, Who has been involved in shaping the landscape over time, Why did they do what they did to shape the landscape and continue to do so?

The first three items address tangible data, whilst the latter two address aspects of intangible values involving deciphering the ideologies and cultural traditions that have been critical to the process of landscape making. Figure 13.1 and the five questions propose a set of principles to guide

219 …

Ken Taylor

„

Figure 13.1 Planning model for heritage conservation management policy

conservation action – policy – within which specific archival and on-site research related to each landscape being studied is critical and individual to each study (Documentation step, Figure 13.1). It is the outcome of specific research informing us why the landscape takes the shape it does which distinguishes cultural landscape studies from a visual assessment that tells us what the landscape

220 …

Cultural landscape meanings and values „

looks like, but not why. Such research will of necessity be multidisciplinary, involving, dependent on the parameters of each study, various specialists such as a landscape historian, archaeologist, architectural historian, anthropologist, landscape architect. In effect what is taking place is improvement of methods used in practice through research, rather than applying a set recipe. Uzzell (2009, p.333) validates the multidisciplinary approach: ‘one of the reasons why we undertake [it] is to communicate and engage with others in order to develop and employ methodologies in an informed way to understand the heritage’. The process of landscape architects being involved in a rigorous landscape research process offers potential to develop and add to the discipline’s body of knowledge that can inform and enrich the profession. It does, of course, require time, patience and dedication to the task of finding out about a landscape: why does it, the landscape, take a particular form rather than merely describing the form without understanding its context? During the documentation stage (Figure 13.1), through archival and on-site research outcomes, framed within the parameters of the five questions, it becomes possible to decipher the story of events, people and places through time represented in the current landscape to convey its meanings, bearing in mind that: ‘The meaning of heritage will vary over time and for different groups of people. It serves social, cultural and political functions. But the heritage during this process does not remain static and unchanged … We use the heritage in the creation of our own individual, group and national identities’ (Uzzell 2009, pp.326–327). Addressing the questions will also provide interpretative information that will not simply or primarily consist of instruction, but provocation, that is it will provoke people – locals, planners, visitors – to think and make connections with the landscape. Provocation, not instruction, as the chief aim of interpretation is one of Tilden’s six principles of interpretation (Tilden 1977, p.9): 1 2 3 4 5 6

Any interpretation that does not somehow relate what is being displayed or described to something within the personality or experience of the visitor will be sterile; Information, as such, is not interpretation. Interpretation is revelation based upon information. But they are entirely different things. However all interpretation includes information; Interpretation is an art, which combines many arts, whether the materials presented are scientific, historical or architectural. Any art is in some degree teachable; The chief aim of interpretation is not instruction, but provocation; Interpretation should aim to present a whole rather than a part, and must address itself to the whole man rather than any phase; Interpretation addressed to children (say, up to the age of twelve) should not be a dilution of the presentation to adults, but should follow a fundamentally different approach. To be at its best it will require a separate program.

I recommend these be addressed by landscape architects when considering, for example, how to present their designs to the public or clients. Indeed one may pose the question, ‘Can modern designed landscapes innately convey meaning and have significance from their inception in the way in which we understand examples from history?’ (Taylor 1997, p.18). I suggest that designers ought to think seriously about how they ‘may see themselves as enticing users to share the designer’s artifice and discover meaning’ (Taylor 1997, p.3). A case that comes to mind is High Line in New York where visitors are made aware of the context, meaning and significance of the design: an exercise that heightens their experience.

221 …

Ken Taylor

„

Addressing the above five essential questions will involve the following steps: Identification and documentation of cultural landscape resources, Assessment of landscape characteristics, and Analysis/evaluation. These steps align with the process set out in Figure 13.1.

Step 1: identification and documentation of cultural landscape resources 1 2

Identify the type of cultural landscape and its setting; Document landscape history.

This step covers the description of site/landscape history based on research of archival primary and secondary sources, historical information and on-site examination and assessment of the landscape character. Primary sources include, inter alia, journals, diaries, private and public records, family archives, photographs, newspapers, historic plans and maps. Secondary sources include, inter alia, books, reports and academic material such as theses. Information gleaned then helps orientate site observation work to start to suggest the how and the why of the landscape and will, to an expert, experienced observer in the field, prompt questions and focus attention on seeing what is there rather than merely looking at the landscape. The all-important early research work identifies and helps develop an understanding of the historic context of the landscape (see also Step 3b evaluating cultural significance).

Step 2: assessment of landscape characteristics Once the background identification and documentation stage has been completed it is possible to address the task of assessment of the particular characteristics of a landscape. A comprehensive system for organising presentation of information on landscape characteristics, values and significance is advisable given that cultural landscapes are a montage, or series of layers through time. These layers are created by the forces of human intervention over time and the natural forces (processes) that have shaped the physical landscape in a continuum where human forces – including cultural values and ideologies – modify, change and adapt the physical landscape for human occupation, thereby creating the cultural landscape consisting of: s s s s s

Land uses and activities Patterns resulting from periods of landscape making Cultural traditions Natural elements Individual components.

Processes involved in landscape making

Within the cultural landscape under investigation there will be particular tangible and intangible landscape characteristics that create landscape character. Table 13.2 is a typical example list of characteristics for a rural landscape based on the work of Page et al. (1998) for the US National Park Service, which has lasting application. There is a range of scholarly work available grounded in research into relevant aspects of landscape characterisation assessment covering landscapes from local to national scales and based on specific research models (see for example Kikuchi et al. 2014; Warnock and Griffiths 2015; Kapfer et al. 2015; Heritage Council of Victoria 2015). The one with which I am familiar, Kikuchi et al. 2014, is based on a research project conducted with local people in the Hungduan district of the Philippine Cordilleras World Heritage Rice Terraces. Following the

222 …

Cultural landscape meanings and values „

Table 13.2 Example landscape characteristics for rural landscapes (source: Page et al. 1998)

Table 13.3 Example landscape characteristics for urban landscapes

research being conducted, the project team involved published a report outlining research methods and outcomes (Okahashi et al. 2012) giving a detailed and instructive insight into how a local group of indigenous people value their landscape, which accords with the research question for this chapter of whose values need to be addressed. For an urban cultural landscape the criteria in Table 13.3 will be applicable. An important and useful tool for handling large areas is the subdivision of a study area into landscape character units (LCUs). The scale of the units will vary according to landscape use and activity. LCUs should not be regarded as land units based on physical criteria or visual catchment zones, although these factors will be taken into account (see also e.g. Swanwick 2002; Fairclough et al. 2016, in prep.). Rather they are defined by landscape patterns informed by historical settlement data essential to an understanding of landscape making.

Step 3: analysis / evaluation This is the stage of fully understanding a cultural landscape through its inherent cultural values and establishing its significance as a guide to conservation and future sustainable use. Historical documentation, field survey, and assessment of the forces that have created the landscape form

223 …

Ken Taylor

„

the basis for understanding its cultural values. The following factors applied to each LCU need to be addressed: 1 2

Analyse inherent values; Evaluate cultural significance.

Analysis of values such as aesthetic, historic, scientific, social and spiritual or others are set out in Table 13.1. Evaluation of cultural significance involves synthesising information gleaned from research and analysis of values to develop a clear statement of how and why the landscape is valued. This leads to development of a statement of significance for each LCU and an overall statement for the complete study area. The statement should set out (Heritage Council Victoria 2015, p.27): s s s

The cultural heritage values represented in the landscape Who holds these values, and; How important they are to the groups involved.

Critical to this stage of the process will be an articulation of the historic context of a landscape through academic work based on scholarly research which has been undertaken in the documentation stage (Figure 13.1). Historic context underpins inquiry into how the landscape under study is significant: it is the framework which informs the meaning of the landscape in relation to its setting, cultural traditions and values through time. Co-ordinating research information and site observation to understand the landscape is a particularly satisfying task. It involves researching information for its own sake as a means of advancing the body of knowledge on the concept and meaning of landscape as well as aiding in improving methods used in practice. Related to the latter point is that future management options will be based on the statement. Many countries and associated heritage agencies use the device of a statement of significance to summarise heritage significance whereby: ‘A Statement of Significance (SOS) is a declaration of value that briefly explains what a historic place is and why it is important. The SOS identifies key aspects of the place that must be protected in order for the historic place to continue to be important’ (Canadian Register of Historic Places in Kalman 2014, p.212). A statement of significance is based on analysis of each aspect of significance in respect to a set of values (Table 13.1) or criteria recognised by a particular agency having jurisdiction to list heritage places. Many countries worldwide have a set of national criteria as do many local governments; similarly for World Heritage purposes there are set criteria (Table 13.1). Hence it is possible to recognise significance at local, national and international levels and some heritage places may meet more than one set of criteria, so that, for example, in Australia a heritage place must be entered on the National Heritage List before it may be nominated for World Heritage listing.

HISTORIC LANDSCAPE ASSESSMENT: WINGECARRIBEE SHIRE, NEW SOUTH WALES, AUSTRALIA This is an example where the approach and methods presented above were used. It is a process applied to a practical case in which research inquiry involved archival and on-site investigations. The research inquiry resulted in valuable outcomes relating to understanding landscape meanings

224 …

Cultural landscape meanings and values „

– important in their own right – and their application to guide the conduct of the study and recommendations for future action.

Context Wingecarribee Shire, 2702 square kilometres (sq. km) in extent, is part of the Tablelands region in New South Wales (NSW) located mid-way between Sydney and Canberra (Figure 13.2). The Shire has a large number of items and places of heritage value which are significant locally, regionally and nationally, including its rural landscapes and historic country towns. As a result the area is a main tourist destination. The present population is around 44,000. Development pressures have

Figure 13.2 Location of Wingecarribee Shire

225 …

Ken Taylor

„

been experienced due to spill-over growth from the Sydney metropolitan area. Concern at the likely changes over time and loss of rural character prompted mounting community representations on the need for planning controls in the latter half of the 1980s. In response to mounting concern at growth pressures and anxious to ensure that the historic character of the Shire is not eroded, the Shire sought and gained funding for a heritage study in co-operation with the State Heritage Branch. This is a study done to inform practice and real-world decisions. Methods used are recommended as applicable in the context of landscape architecture research. The co-operative nature of the study recognised the local and state significance of the heritage resources of Wingecarribee. The main aims of the study included those of identifying and analysing the environmental heritage of the Shire and providing practical recommendations, statutory and non-statutory, for the conservation and management of its heritage resources. A major component of the overall study was a historic cultural landscape assessment. Historically the area played a significant role in early European exploration and then rural settlement of the colony of NSW. The history of European exploration and subsequent settlement and land-use developments from the early 1800s onwards have left a distinctive landscape pattern with a series of layers representing various periods through history telling the story of human occupation of the land and attitudes to it. The central part of the Shire, the focus of the study, is a rural landscape supporting extensive grazing with some crop growing in the eastern portion. It is about 40 km x 20 km in extent occupying an undulating hill and valley topography with an altitude range of about 650–850 metres above sea level bisected by the Wingecarribee River. The undulating topography is punctuated by a number of landmark hills and historically significant towns and villages. The area is also famous for its historic gardens. This central rural area setting is surrounded by eucalypt-clad forested hills underlain by sandstone. The encircling forested landscape consists of extensive national park land, state forest and Sydney metropolitan water catchment land. Study method The sequence of steps followed in the study method is outlined in Figure 13.3. The assessment approach used the characteristics set out in Table 13.2 above, but because it is a broad scale study it amalgamated these characteristics into four categories: 1 2 3 4

Overall landscape patterns; Building clusters, structures, special features; Circulation routes; Historical associations.

It is stressed that the study method is not a visual assessment of what is seen in the landscape. Rather it is a reading and interpretation of the landscape with emphasis on historic values and social values inherent in the landscape informed by historical research. At no stage were there any comments made in the assessment of any vague visual scenic values, aesthetic scenic values or preference tests leading to vacuous scenic value ratings. Following an initial review of historical data and site visits it was possible to distinguish patterns in the landscape – what the landscape looks like – related to settlement themes to provide historic context for the study area. Settlement themes were:

226 …

Cultural landscape meanings and values „

Figure 13.3 Wingecarribee Study Model

s s s s s

Early exploration and settlement, 1800–1840s; Consolidation of major pastoral holdings, 1840s–1860; Post-1860 rural extensions; Community development post-1860 (development of small country towns); Tourism and recreation, 1868 to present.

Landscape patterns are a cogent way of understanding historic trends and significant cultural landscape elements. Much of the structural framework of today’s landscape patterns was established in the nineteenth century, including roads, tracks, boundaries, clearing of trees, sites of townships and villages, pastoral holdings and buildings. Significant for the landscape patterns of the study area are the important imprints from the early to mid-1800s. This pattern overlies the earlier Aboriginal cultural landscape pattern resulting in a rich mosaic of layers. From the patterns a series of settlement themes based on settlement phases through time post1800 can be distinguished. These provide a framework for understanding the present landscape, 227 …

Ken Taylor

„

its making, its meaning and its cultural values. The themes inevitably overlap and are not a simple chronology of events, rather they encompass the phases of the history of the process of landscape making and highlight significant landscape reminders from the past. The themes were applied in the assessment and analysis stages to each cultural landscape unit as part of reading and interpreting the cultural landscape.

Documentation, assessment, analysis Primary (e.g. journals, diaries, historic photographs/paintings/drawings, newspapers, maps) and secondary (e.g. reports, books, guides, biographies, theses, journal papers) research sources were central to understanding settlement patterns and documenting the history of landscape making for the study team consisting of landscape historian (author), landscape architect and an archaeologist supported by a historian who sought out archival maps and plans. Particularly helpful were Portion Plans from the early to mid-nineteenth century. Following colonial government instructions in 1825 a general survey of the colony of NSW and its division into counties and parishes was undertaken. These formed a basis for ongoing mapping in what are called Portion Plans depicting portions of the landscape. Portion Plans show original grantees or lessees of land in parishes and can include place names, notes, boundaries, roads/tracks and later improvements to create grazing areas. Often surveyors made notes of the landscape character such as ‘well grassed open forest’ referring to the picturesque open park-like landscape created by millennia of Aboriginal burning which appealed to the new settlers because of its grazing potential. Later mapping might refer to grassy paddock (field) where some tree clearing had taken place to create more open grazing after about 1860 when fencing was introduced. Other historical information came from diaries, local history books, and records such as those of the governor of NSW, Lachlan Macquarie, travelling through the area in 1820: We met a numerous herd of about 400 head of cattle belonging to Mr Throsby feeding in a fine rich meadow … The grounds adjoining Mr Throsby’s hut are extremely pretty, gentle hills and dales with an extensive rich valley in his front, the whole having a very park-like appearance, being very thinly wooded. (Macquarie 1836, 18 October)

In 1836 colonial artist Conrad Martens painted the view described by Macquarie leaving a remarkable pictorial image of this nineteenth century cultural landscape (Figure 13.4). The view shows an open grassy area surrounded by open forest merging into thicker forest on the surrounding hills. It is possible therefore to track the changes in the view to the present day (Figure 13.5) and map the various layers in the landscape. Piecing together primary and secondary source archival information with on-site observation the study team was able to locate specific areas of historic significance. For example, in the early nineteenth century the area was renowned for its wheat paddocks with wheat sent to Sydney to support the new colonial centre of NSW. Additionally it is recorded that ploughing competitions took place. The team located these from descriptions of them being on a hill slope located at one of the early nineteenth century land grants. Knowing that ploughing was done by bullock teams with single blade ploughs and that this method created ridge and furrow plough marks, these were located after a day in the field during a late winter afternoon where the low sun angle highlighted the marks running uphill (Figure 13.6).

228 …

Cultural landscape meanings and values „

Figure 13.4 Throsby Park, a painting by Conrad Martens

Figure 13.5 Present day view of Throsby Park

229 …

Ken Taylor

„

Figure 13.6 Paddock showing nineteenth century ridge and furrow plough marks

A series of ten cultural landscape units was delineated from field survey assessment of landscape patterns and historical settlement data overlaying natural landscape elements. Subdividing the landscape into units is informed by the fact that the cultural landscape is a montage or composite picture created by the inter-relationships between various cultural and natural elements in the landscape. Cultural elements are representations of various historic periods of landscape making which result in layers and patterns of development. They also mark change over time. The units, therefore, are indicative of various periods of landscape making and act as windows onto the past and are a montage of the forces which define cultural landscape form: s s s

Natural features including landform, vegetation and water forms and how these have been used in the creation of particular cultural patterns; Land-use patterns created through time; Particular cultural elements and components.

The units delineated resulted from field assessment including land-use patterns and vegetation and particular components such as building types, paddock boundaries overlain by historic settlement data and the way in which natural elements have been treated so that each unit is distinctive. It is stressed that the units are neither land units based on physical criteria nor visual catchment units based on visual boundaries, rather they are defined by cultural landscape patterns informed by historical/archival settlement data essential to the understanding of landscape making. Establishing cultural significance from the identification/documentation, assessment and analysis/evaluation process is the basis for the formulation of conservation policy and management

230 …

Cultural landscape meanings and values „

recommendations. It was decided in this study not to rank significance, but to attempt to delineate those culturally significant landscape units where the interpretative and associative values and landscape integrity are critical to the history and heritage of Wingecarribee. These are the units where subdivision by such uses as hobby farms or rural residential would lead to an irretrievable loss to the historic authenticity of the landscape and a loss to the Shire’s sense of place and cultural significance. Therefore a decision was taken to establish and to recommend key historic units. As a result of the assessment and analysis inputs, four key historic cultural landscape units were identified. These are cultural landscapes particularly able to inform interpretation of the history of Wingecarribee, promote a sense of place, and relate this to the people involved in the landscape making. The key units display a high level of intactness and authenticity where past historic landscape patterns are still visible within later layers of settlement. One of the units included the ridge and furrow plough marks (Figure 13.6). In addition four key historic villages/towns were identified. The key units include examples of landscape types from the early major holdings and from the post1860 dairying era overlain with twentieth century layers. In response to community comment, a supplementary landscape study was undertaken. As a result, an additional key historic cultural landscape unit was added to the recommended list, making a total of five units occupying about 50 per cent of the study area. The key historic units contain significant historic elements from past landscape making which remain comprehendible in existing landscape patterns. The historic elements promote a particular sense of continuity in the landscape and cogent links with the past, serve to enhance present-day landscape values by a knowledge and understanding of the actions of past inhabitants, and provoke a sense of human participation in landscape making and sense of place. Wingecarribee Shire Council created a Draft Local Environment Plan through the Environmental Planning and Assessment Act. This outlined a Landscape Protection Zone covering the key historic units. The action recognised the cultural significance of these landscapes to the present generation and future generations and restricted land use to rural activities in keeping with the existing character and scale.

DISCUSSION AND CONCLUSION In practice not all landscape architects are frequently involved in dealing with heritage landscape conservation management aspects and values, or want to be involved. Where they are involved in such a task and briefed to address aspects of landscape assessment – as for example in the Wingecarribee study or review of an existing piece of urban open space and its history – it is critical that the study method used reaches beyond being merely a visual appraisal. Thorough research through interdisciplinary teamwork will be instrumental in guiding the choice of the method to address the research question posed at the beginning of this chapter and the pivotal idea of defining cultural and historic contexts. I would also point out that addressing these principles is not the end of the project as there is always the need to continue to consult with stakeholders in order to evaluate whether the process has worked and/or if it needs fine-tuning. A corollary to this line of thought raises the question of future research and what direction(s) this might take for landscape architects. One direction that occurs to me as a fruitful one connects with the idea of the relationship between people and place expressed through the special characteristics of the concept of landscape where ‘one of our deepest needs is for a sense of identity and belonging … and how we find identity in landscape and place’ (Taylor 2015b).

231 …

Ken Taylor

„

Related to this line of thought is the ever growing global urbanisation of societies. I therefore suggest that here is a distinctive opportunity for landscape architects to focus on designed urban spaces and parks to connect with communities of people through the intellectual concept of landscape and meaning, a field of research that has focused on the landscape and meaning narrative since the endeavours of Franz Boas and subsequent generations of geographers and anthropologists. Its application to urban studies could be particularly fruitful and engaging. Linked to this seemingly rich area of landscape architecture research for me is that of designed urban spaces under two categories: historic ones and new designs. Inspiration could be taken from the 2015 UNESCO World Heritage Centre listing for Roberto Burle Marx’s work, ‘the Walter Burle Marx Site’ (UNESCO 2015), General Conference Recommendation on Historic Urban Landscape [HUL] (UNESCO 2012) and Nishimura’s work on urban conservation planning (Nishimura 2004). Critical inquiry could engage such questions as: what is their intellectual foundation, what makes them work, whose values do they represent?

SUGGESTED FURTHER READING Gillman, G. (2010) The Idea of Cultural Heritage, 2nd ed., New York: Cambridge University Press. Howard, P., Thompson, I. and Waterton, E., eds. (2013) Routledge Companion to Landscape Studies, Abingdon and New York: Routledge. Smith, L. (2006) The Uses of Heritage, Abingdon and New York: Routledge. UNESCO (2013) New Life for Historic Cities: The Historic Urban Landscape Approach Explained, Paris: UNESCO, available: http://whc.unesco.org/en/activities/727/. Verschuuren, B., Subramaniam, S. and Hiemstra, W., eds. (2014) Community Well-Being in Biocultural Landscapes: Are We Living Well?, Rugby: Practical Action Publishing Ltd. Waterton, E. and Watson, S., eds. (2010) Culture, Heritage and Representation: Perspectives on Visuality and the Past, Farnham: Ashgate Publishing Limited.

REFERENCES Australia ICOMOS (2013) The Burra Charter: The Australia ICOMOS Charter for Places of Cultural Significance, available: http://australia.icomos.org/wp-content/uploads/The-Burra-Charter-2013Adopted-31.10.2013.pdf. Bandarin, F. (2015) ‘Urban conservation and the end of planning’, in Bandarin, F. and van Oers, R., eds. Reconnecting the City: The Historic Urban Landscape Approach and the Future of Urban Heritage, Chichester: John Wiley. Bandarin, F. and van Oers, R. (2012) The Historic Urban Landscape: Managing Heritage in an Urban Century, Chichester: John Wiley. Berleant, A. (2010a) Aesthetic Sensibility, available: http://www.autograff.com/berleant/pages/Aesthetic%20 Sensibility.draft%208.htm. Berleant, A. (2010b) Sensibility and Sense: The Aesthetic Transformation of the Human World, Exeter: Imprint Academic. Biger, G. (2006) ‘Introduction: On ideology and landscape’, in Baker, A.R.H. and Biger, G., eds. Ideology and Landscape in Historical Perspective: Essays on the Meanings of Some Places in the Past, Cambridge: Cambridge University Press, 1–15. Bullock, A. and Stallybrass, O., eds. (1978) The Fontana Dictionary of Modern Thought, London: Fontana/Collins. Byrne, D. (2009) ‘A critique of unfeeling heritage’, in Smith, L. and Akagawa, N., eds. Intangible Heritage, London and New York: Routledge, 229–252.

232 …

Cultural landscape meanings and values „

China ICOMOS (2002) Principles for the Conservation of Heritage Sites in China, translation edited by Agnew, N. and Denis, M., Los Angeles, CA: Getty Institute, available: http://www.getty.edu/conservation/ publications_resources/pdf_publications/pdf/china_prin_heritage_sites.pdf. Cosgrove, D.E. (1984) Social Formation and Symbolic Landscape, London and Sydney: Croom Helm. Diefenthaller, I. (2014) Carter’s Farm, Shudy Camps: Change in Tenure of the Later Granary, Building ‘F’, Heritage Impact Assessment, available: http://plan.scambs.gov.uk/swiftlg/MediaTemp/1133538-513248.pdf. Dudley, N. and Stolton, S., eds. (2012) Protected Landscapes and Wild Biodiversity, Volume 3 in the Values of Protected Landscapes and Seascapes Series, Gland: IUCN. Eagleton, T. (1995) The Ideology of the Aesthetic, 2nd ed., Oxford: Blackwell. Fairclough, G., Sarlöv-Herlin, I. and Swanwick, C., eds. (2016) Routledge Handbook of Landscape Character Assessment: Current Approaches to Characterisation and Assessment, Abingdon: Routledge. Fowler, P. (2001) Cultural Landscape: Dreadful Phrase, Great Concept, London: UK ICOMOS, available: http:// archaeologydataservice.ac.uk/archiveDS/archiveDownload?t=arch-302-1/dissemination/pdf/fwp88/ fwp88.pdf. Greffe, X. (2010) ‘Urban cultural landscapes: An economic approach’, Working Paper 1/2010, Turin: Department of Economics, University of Turin. Han, F. (2006) The Chinese View of Nature: Tourism in China’s Scenic and Historic Interest Areas, (PhD), School of Design, Queensland University of Technology, Brisbane. Hayden, D. (1995) The Power of Place: Urban Landscapes as Public History, Cambridge, MA: The MIT Press. Heritage Council of Victoria (2015) Landscapes of Cultural Heritage Significance: Assessment Guidelines, Heritage Council of Victoria, available: http://heritagecouncil.vic.gov.au/research-projects/landscapesof-cultural-heritage-significance-assessment-guidelines/. Horne, D. (1986) The Public Culture: The Triumph of Industrialism, London: Pluto Press. Hoskins, W.G. (1955) The Making of the English Landscape, London: Hodder and Stoughton. ICOMOS (1994) ‘The Nara Document on Authenticity’, in Lemaire, R. and Stove, H., eds. Proceedings of the Nara Conference on Authenticity in Relation to the World Heritage Convention, Nara, Japan, 1–6 November, 46–48, available: http://www.icomos.org/charters/nara-e.pdf. Jackson, J.B. (1951) Landscape, 1 (Spring 1951), 5. Jackson, J.B. (1984) Discovering the Vernacular Landscape, New Haven, CT and London: Yale University Press. Kalman, H. (2014) Heritage Planning: Principles and Process, Abingdon: Routledge. Kapfer, M., Zieselb, S. and Kantelhardta, J. (2015) ‘Modelling individual farm behaviour and landscape appearance’, Landscape Research, 40(5), 530–554. Kikuchi, Y., Sasaki, Y., Yoshino, H., Okahashi, J., Yoshida, M. and Inaba, N. (2014) ‘Local visions of the landscape: Participatory photographic survey of the World Heritage Site, the Rice Terraces of the Philippine Cordillera’, Landscape Research, 39(4), 387–401. Lewis, P.F. (1979) ‘Axioms for reading the landscape: Some guides to the American scene’, The Interpretation of Ordinary Landscapes: Geographical Essays, 11–32. Livingstone, D.N. (1992) The Geographical Tradition: Episodes in the History of a Contested Enterprise, Oxford: Wiley-Blackwell. Macquarie, L. (1836) Lachlan Macquarie, Governor of New South Wales, Journals of His Tours in New South Wales and Van Diemen’s Land 1810–1822. Reprint (1956) Sydney: Trustees of the Public Library of New South Wales. Mason, R. (2002) ‘Assessing values in conservation planning: Methodological issues and choices’, in de la Torre, M., ed. Assessing the Values of Cultural Heritage Research Report, Los Angeles, CA: The Getty Conservation Institute, 5–30. Meinig, D.W. (1979) ‘Introduction’, in Meinig, D.W., ed. The Interpretation of Ordinary Landscapes: Geographic Essays, New York: Oxford University Press, 1–9. Mitchell, W.J.T., ed. (1994) Landscape and Power, Chicago, IL: The University of Chicago Press. Nishimura, Y. (2004) Urban Conservation Planning, Tokyo: University of Tokyo Press. Okahashi, J., Yoshida, M., Iniba, N., Kikuchi, Y., Yoshino, H. and Sasaki, Y. (2012) What is ‘Heritage’ for the Hungduan People? Significance of a World Heritage Landscape for Local Lives, Global Negotiation

233 …

Ken Taylor

„

Programme (GNP) Project Report, Tokyo: University of Tsukuba, 1–100, available: http://s3.amazonaws. com/zanran_storage/www.tulips.tsukuba.ac.jp/ContentPages/2546481789.pdf. Olwig, K.R. (2007) ‘The practice of landscape “conventions” and the just landscape: The case of the European Landscape Convention’, Landscape Research, 32(5), 579–594. Page, R.R., Gilbert, C.A. and Dolan, S.A. (1998) A Guide to Cultural Landscape Reports: Contents, Process, and Techniques, Washington, DC: U.S. Dept. of the Interior National Park Service. Plachter, H. and Rössler, M. (1995) ‘Cultural landscapes: Reconnecting culture and nature’, in von Droste B., Plachter, H. and Rössler, M., eds. Cultural Landscapes of Universal Value: Components of a Global Strategy, New York and Stuttgart: Gustav Fischer Verlag, 15–19. Roe, M. and Taylor, K., eds. (2014) New Cultural Landscapes, London and New York: Routledge. Smith, L.. (2006) Uses of Heritage, Abingdon: Routledge. Smith, L. (2011) All Heritage is Intangible: Critical Heritage Studies and Museums, Amsterdam School of the Arts: Reinwardt Academy, available: https://www.ahk.nl/fileadmin/download/reinwardt/lectoraat/ All_heritage_is_intangible.pdf. Swanwick, C. (2002) Landscape Character Assessment Guidance – Topic Paper 1: Recent Practice and the Evolution of Landscape Character Assessment, Cheltenham: Countryside Agency, and Edinburgh: Scottish Natural Heritage. Taylor, K. (1997) ‘Design with meaning’, Landscape Review, 3(2), 3–21. Taylor, K. (2009) ‘Cultural landscapes and Asia: Reconciling international and Southeast Asian regional values’, Landscape Research, 34(1), 7–31. Taylor, K. (2014) ‘Cultural heritage management: International practice and regional applications’, in Smith, C., ed. Encyclopedia of Global Archaeology, Vol 3, New York: Springer, 1939–1951. Taylor, K. (2015a) ‘Cities as cultural landscapes’, in Bandarin, F. and van Oers, R., eds. Reconnecting the City: The Historic Urban Landscape Approach and the Future of Urban Heritage, Chichester: Wiley Blackwell, 197–202. Taylor, K. (2015b) ‘Ken Taylor on the human side of heritage conservation’, Routledge, available: http://www. bitly.com/158PwVA. Taylor, K., St. Clair, A. and Mitchell, N. (2015) ‘Introduction: Cultural landscapes. Twenty-first century conservation opportunities and challenges’, in Taylor, K., St. Clair, A. and Mitchell, N., eds. Conserving Cultural Landscapes: Challenges and New Directions, New York and Abingdon: Routledge, 1–19. Thompson, I.H. (2012) ‘Ten tenets and six questions for landscape urbanism’, Landscape Research, 37(1), 7–26. Tilden, F. (1977) Interpreting Our Heritage: Principles and Practices for Visitor Services in Parks, Museums, and Historic Places, 3rd ed., Chapel Hill, NC: University of North Carolina Press. UNESCO (2008) Operational Guidelines for the Implementation of the World Heritage Convention, World Heritage Centre, Paris, 8 January, available: http://whc.unesco.org/archive/opguide08-en.pdf#annex3. UNESCO (2012) ‘Recommendation on the Historic Urban Landscape’, in Records of the General Conference: Volume 1 Resolutions, Paris, 25 October – 11 November 2011, Paris: UNESCO, 50–55. UNESCO (2015) Sítio Roberto Burle Marx: Description, available: http://whc.unesco.org/en/tentativelists/6001/. Uzzell, D. (2009) ‘Where is the discipline in heritage studies? A view from environmental psychology’, in Sørensen, M.L.S. and Carman, J., eds. Heritage Studies: Methods and Approaches, Abingdon and New York: Routledge, 326–333. Warnock, S. and Griffiths, G. (2015) ‘Landscape characterisation: The living landscapes approach in the UK’, Landscape Research, 40(3), 261–278. Wylie, J. (2007) Landscape: Key ideas in Geography, Abingdon and New York: Routledge.

234 …

Chapter 14: Landscape and health Catharine Ward Thompson

INTRODUCTION Environmental design is of interest to planners and policy-makers because of its potential contribution to addressing the current health crises in the western world. There are alarmingly rapid rises in levels of obesity, Type 2 diabetes, cardiovascular disease, cancer and mental illness, and they have practical consequences not only for individual wellbeing but also for the cost of healthcare and the productivity of the workforce. Many such illnesses are not the result of exposure to pollutants or organic disease vectors but are partly the consequence of availability and choice in what food people eat, how and where they spend their leisure time and, in addition, the increasingly sedentary nature of most jobs and work contexts in the developed world. We now know that sedentary behaviour is an independent risk factor for health, above and beyond the effect of poor diet and low levels of physical activity (Sugiyama et al. 2009). All of this indicates that individual preference and decision making, as well as the nature of the socio-ecological context in which they occur (Evans and Stoddart 1990; Sallis and Owen 2002; Shortt et al. 2014), have a large part to play in improving public health. Under the European Union’s Horizon 2020 six-year research programme, launched in 2014, there has been a focus on six grand challenges faced globally and in need of urgent action (European Commission 2014). One of these is ‘health, demographic change and wellbeing’. While there are many aspects of individual and community life, and its societal context, which impact on this grand challenge, environmental design has recently been highlighted as an area of opportunity for research. This is partly because it has the potential to address problems ‘upstream’, in a preventative way, rather than simply focusing on dealing with the ‘downstream’ consequences of ill health in an ageing society (Morris et al. 2006). Aspects of other grand challenges are also of relevance to landscape architecture: inclusive societies; green transport (including active travel); and climate action to address the impacts of climate change. In all of this, a major interest lies in understanding how we can design ‘salutogenic’ landscapes (salutogenic as defined by Antonovsky 1996) – landscapes that support health in mind and body, that support active lifestyles and mental wellbeing, while maintaining equitable and inclusive access to these benefits in creating attractive and healthy places for people’s day-to-day lives. Of course, the idea is not new that environmental design and good quality landscapes such as public parks are essential for good health (Ward Thompson 2011). Environmental interventions to enhance public health were central to nineteenth-century improvements in urban areas but

235 …

Catharine Ward Thompson

„

had become marginalised in the pharmaceutically focused and high-technology world of postwar twentieth-century medicine (Morris and Robertson 2003). The renewed interest in physical environment is now focused on identifying and understanding environments that support healthy behaviours and responses, recognising that such environments may have more permanent and population-wide effects than other forms of public health interventions targeted at individuals (Saelens et al. 2003; Owen et al. 2004). Landscape architecture alone cannot address global health problems and prevent illness in today’s society, but it is possible that a salutogenic environment, one that supports health, is a necessary condition for other, perhaps more individually targeted, interventions to succeed. A supportive environment may also be necessary as a first step towards any behaviour change in the direction of healthier lifestyles and it is certainly likely to make any such change more sustainable (Sugiyama et al. 2013). What, then, are the research methods that might be appropriate for landscape architecture in this context? There will be aspects of interest to landscape architects across the spectrum of scales, from national and regional landscape planning to the details of urban, park and garden design and to specification at the finest grain. The scale at which my research has focused, by and large, is one which offers perhaps the greatest potential to see links between landscape and health – the neighbourhood or local community scale. It is work at the community scale, therefore, that research methods described here are designed to address, seeking to understand how different landscape characteristics, and people’s varying access to them, may be associated with different kinds of health behaviours and/or outcomes. In the context of research into health and environment, it is important to be explicit about the robustness of the method. If we are to have influence beyond our own discipline and, vitally, to be taken seriously by public health researchers and policy-makers, we need to use methods that are demonstrably up to the task, that provide a satisfactory level of evidence. This doesn’t mean we always have to produce the equivalent of what is considered the ‘gold standard’ of evidence in clinical medicine – randomised, controlled trials. Indeed, the level of proof possible to demonstrate a causal relationship between people’s health and their local environment may only be weak at best, given the complex and dynamic relationship between health, environment and individual (Morris et al. 2006). Equally, a carefully considered change to recreational land use policy or park design principles, for example, is unlikely to be life-threatening, or directly responsible for major harms to human health. Nonetheless, we need to pay attention to the kinds of theoretical bases that have credibility with public health experts, and to the measures indicative of health, wellbeing and quality of life that are recognised by health professionals, if we are to make claims about links between landscape and health that will be taken seriously in policy-making and development of legislation, guidance and practice. In this spirit, the examples of methods that are described here include a brief description of the theories and potential causal pathways that underlie the research approach and indicate the particular health measures used. They illustrate the strength of mixed methods approaches that draw on a range of techniques to respond to a research question. The findings from this kind of research are often targeted at national and local authority planning policy, as well as public health policy-making, with the aim of informing planning and design decisions at the community or neighbourhood scale and providing the evidence needed to make the case for landscape interventions. This kind of research may not tell experienced landscape architects how to design differently (that is not the aim), but may offer valuable support to such professionals in arguing the case for their contribution within the wider policy context.

236 …

Landscape and health „

What follows is a series of studies that exemplify the use of different methods to research links between landscape design and health and wellbeing. They take different contexts and different subgroups in the population as their focus, recognising that different people, and at different stages in their life course, will have varying needs and will experience the landscape in different ways. These studies illustrate the research methods’ strengths, weaknesses and challenges and identify the detailed issues that need to be taken into account in considering the use of each.

CROSS-SECTIONAL RESEARCH ON ENVIRONMENTS THAT SUPPORT OUTDOOR ACCESS IN AN AGEING SOCIETY The research methods illustrated first have been part of a multidisciplinary and collaborative project undertaken over many years (2003–2011), through the consortium entitled I’DGO: Inclusive Design for Getting Outdoors. More details of the theoretical background, methods and results can be accessed via the I’DGO website (http://www.idgo.ac.uk). Presented here are examples of research contributions from my own research centre, OPENspace.

Theoretical foundations The theoretical basis for the research initially drew on projective approaches to understanding relationships between individuals and their environment. It assumes that there is a transactional relationship between people and place: how we perceive and experience the environment is influenced by what we want to do in it, and how well it serves our needs of the moment, in the context of memories and past experience (Little 2000). Lawton and Nahemow (1973) developed an ecological model of ageing that introduced the concept of environmental press – the differential effect of the environment on behaviour that relates to the capabilities and characteristics of the individual. Building on this and transactional theory, models of environmental fit (Lawton 1980; Kahana et al. 2003) have been developed to describe how the environment can become a limiting factor on people’s mobility as their functional capabilities change in old age (Iwarsson 2005). We follow Oliver’s social model of disability in our work, which considers the individual as an active being, capable of coping successfully with daily living and major life changes if personal and environmental contexts are appropriate and enabling (Oliver and Sapey 2006). Related to this, the concept of environmental support draws on the work of Kelly (1955) and Little (1983; 2010) to focus on environmental quality in relation to the activities older people want to undertake; in other words, how well the environment makes it easy and enjoyable to do things, or frustrates and hinders action (Sugiyama and Ward Thompson 2007a). As a concept, it links environmental attributes with people’s perceptions of them in relation to their own, idiosyncratic desired and necessary activities.

Research questions These theoretical perspectives informed a series of research questions and methods to address them that took as their underlying premise the following notion: that older people’s physical, mental and social health, indeed their quality of life, will be better if they live in a neighbourhood whose outdoor environment makes their activities easy and pleasant to undertake, compared with those whose local environment makes it hard to undertake their activities. We were interested therefore

237 …

Catharine Ward Thompson

„

in outcomes that measured aspects of activity, wellbeing and quality of life and related these to outdoor environmental attributes. The measures we used are discussed in more detail below but inform the research questions we developed, as follows: a. b.

Do older adults’ perceptions of the qualities of their physical neighbourhood environment predict their quality of life? What qualities of a local park or open space are associated with older people’s varying levels of outdoor activity and quality of life?

Research ethics It is important to note that most research, and particularly any research involving human participants, requires attention to ethical standards and procedures; universities and major funders will normally require ethics approval before a study is allowed to proceed. For our research, focused on older people who might have physical, cognitive or sensory impairments and lower levels of stamina than younger adults, it was important that every stage of our research was carefully reviewed, as some of our participants might be considered vulnerable. Any interviews or engagement with participants required each person to be told about the purpose for collecting the data and reassured about the confidentiality with which data would be held and its secure storage (in line with UK and European personal data legislation). They were also asked to indicate that they understood what would be involved (i.e. they had the capacity to give informed consent) and to sign a form agreeing to participate, while acknowledging their right to decline to continue participating at any stage without penalty or the need to give a reason. In addition, an information sheet was given to each participant to keep, giving researchers’ contact details and allowing the opportunity to ask questions about the study at any subsequent stage. Such procedures are standard good practice now.

Research design Our research was a mixed methods approach and involved several phases, starting with focus groups and then a wider survey based on a questionnaire.

Focus groups Initially, in order to explore what the important elements and dimensions of the outdoor environment were for older people, we ran eight focus group discussions with older people in a range of urban, suburban and rural contexts in England and Scotland. These groups, which included both frail and comparatively fit older people, were recruited from older people’s interest groups such as Edinburgh’s ‘City for All Ages’ group. We followed accepted good practice for focus groups (Krueger and Casey 2000; Barbour 2007; Brookfield et al. 2013). The topics included what aspects of local outdoor environments people liked or disliked, what they found attractive or made things difficult when going out, what experiences were associated with going to different outdoor environments, what they would change about the environment if they could, and what made a difference to their quality of life. Such focus groups need to be carefully managed to allow each participant a voice, to keep to time and to cover the necessary themes without constraining the discussion too much. The

238 …

Landscape and health „

strengths are that themes may be developed and expanded upon that elucidate the reasons behind certain attitudes or behaviours, and the free-response format ensures that what matters to people is articulated rather than being constrained by researchers’ preconceptions. Weaknesses include the possibility of one or two people dominating conversations (hence the need for skilled facilitation) and the discursive nature of the outcomes, which may take some time to analyse as a result. With participants’ permission, we audio recorded the focus groups and analysed them by repeatedly reviewing the discourse to draw out themes and sub-themes in common. Eventually, we arrived at a list of aspects of quality of life mentioned by participants (such as ‘opportunities to get fresh air’, ‘interacting with others in the neighbourhood’) and the attributes of the environment (physical, social or psychological) that were important, whether positive or negative, for quality of life and access outdoors (such as ‘good pavements’, ‘somewhere to sit’, ‘feeling unsafe’). We were confident that we had elicited most key aspects of interest or concern when we found our list of themes reached saturation point (i.e. no new ones were emerging) and confirmed this by comparing our findings with those of collaborators also working on I’DGO and with previous literature as a way of triangulation. In moving to the next phase of research, that is to develop fixed-response questionnaires, this gave us confidence that we would not omit anything important to the study.

Developing the questionnaire As with most questionnaires, our I’DGO study involved a potentially vast array of issues and themes that we wanted to explore. One of the challenges is to limit the length and complexity of any such survey so that potential participants are not discouraged or over-burdened. At the same time, it is important to include anything likely to be critical in subsequent analysis. If possible, it is always helpful to be focused on a very limited number of key outcomes to be served by the questionnaire data, and this assists in refining the questionnaire content. It is also helpful to postulate some of the likely pathways between the aspects of interest under study (in our case aspects of the physical, outdoor environment) and the outcomes (in our case, quality of life), and to consider what might influence these pathways (potential mediators or confounders, for example). A diagram or logic model to illustrate what is being considered helps to clarify at this stage, such as that shown in Figure 14.1, the first diagrammatic representation developed for our early I’DGO work.

Figure 14.1 Quality of Life (QOL) predicted by a) Environmental Support (ES), b) Outdoor Activity (OA) and Environmental Support (ES), and c) Outdoor Activity (OA), Environmental Support (ES) and Personal Factors (PF) (source: first published in Sugiyama and Ward Thompson 2007a)

239 …

Catharine Ward Thompson

„

As Figure 14.1a shows, we were initially focused on quality of life as the key outcome, with outdoor environmental support leading directly to a better quality of life (Sugiyama and Ward Thompson 2007a). However, we postulated that such support would make most difference when it benefits outdoor activity (Figure 14.1b), offering an additional but indirect link between environmental support and quality of life, where outdoor activity (e.g. the frequency, nature or length of outdoor activities) is the mediator. Further, considerable evidence shows that personal characteristics, from personal relationships and income level to mobility, health status or individual attitudes and experience, influence quality of life. Figure 14.1c therefore shows the model revised to take into account personal factors such as these, which might influence what activities people undertake, and even how supportive the environment is considered to be, as well as directly influencing quality of life. Our questionnaire was initially developed based on this model. We thus needed to include sections or modules within the questionnaire that captured the following: a. b. c. d.

A range of potentially relevant personal factors; People’s perceptions of their local neighbourhood environment; The nature and typical frequency of activities that took people outdoors, as well as the amount of time spent outdoors; People’s quality of life.

Most research which breaks new ground involves devising aspects of the research tools that are new and different but it is, nonetheless, always advisable to use existing, well-tried tools wherever possible, for a number of reasons. Firstly, existing questionnaire modules that have successfully been used by other, respected researchers will have been tested in different circumstances and so many potential problems that arise with new questions (such as participants not understanding the question, or finding it too difficult to answer) should have been addressed. Secondly, if data are collected that match those from wider surveys, for example national surveys or census data, it is possible to compare the study sample with this wider population, giving a useful context for interpretation of results. Thirdly, and especially important in researching links between landscape and health, it is essential that outcome measures such as those relating to health and wellbeing are recognised as valid by relevant professionals and policy-makers, as was mentioned earlier. Where scales or questionnaire modules developed by others are used, it is essential to acknowledge this, and also to use the same wording and fixed response categories as the original, or any refinement of the original. Authors are usually helpful in confirming the latest version of their module, or public agency websites such as the UK’s Medical Research Council (MRC n.d.) will offer updated recommendations. General advice on questionnaire design is covered by many publications in the social sciences (e.g. Allan and Skinner 1991; Gideon 2012); the text here focuses on aspects that are particularly relevant to environment and health research. In the case of I’DGO, our questionnaire was refined and amended in subsequent stages of the multi-phased research but the components identified above remained the essential core. Details of the questionnaire elements as developed and refined at various stages in the I’DGO project, along with their analyses, can be reviewed in Sugiyama and Ward Thompson (2006; 2007b; 2007c; 2008), Sugiyama et al. (2009), Ward Thompson et al. (2014) and Curl et al. (2015; 2016). Aforementioned components are:

240 …

Landscape and health „

a.

b.

c.

d.

e.

Questions on personal characteristics followed standard national census or international survey formats in almost all cases and covered such items as age, ethnicity, living arrangement, level of educational attainment and so on. Other items that seemed particularly relevant to outdoor activities were included or subsequently added, such as access to a private car, dog ownership and so on. In addition, we asked about people’s ability to undertake everyday activities, such as being able to walk a certain distance, climb stairs, see to read and so on, based on an instrumental activities of daily living scale (Jette et al. 1986), since this seemed likely to influence people’s ability to undertake outdoor activities and their response to the environment. We developed our own set of neighbourhood open space (NOS) attributes for the questionnaire as we were not aware of a set that had been developed specifically with older people in mind, and with a UK/northern European focus. We drew on our focus group analyses as well as prior studies (e.g. Saelens et al. 2003; Humpel et al. 2004) to identify as succinct a list as possible that included all the key elements relating to perception of the local outdoor environment. Initially there were 26 items but some were found to be unreliable and others were subsequently added to give us a later version of 38 items grouped under ‘outdoor places in and around your home’, ‘your local street or streets’, ‘your local open space’ and ‘your local neighbourhood’. These were presented as statements (such as: ‘There is a pleasant place to sit outside the home where I live’; ‘The local open space is safe to walk in after dark’; ‘steep hills and steps in my neighbourhood make it difficult to get around’) to which participants agreed or disagreed using a five-point Likert scale (from strongly disagree to strongly agree). Visits outdoors were measured by asking participants how often they went outdoors for different activities, whether functional or recreational, in a typical summer and a typical winter month, categorised as walking to get to places; walking for recreation; gardening; or other outdoor activities. We also asked how much time was typically spent when undertaking each of these activities. Health and quality of life were initially measured using two simple but validated methods. As a measure of general health, we asked participants to state on how many days in the previous month they were unhealthy, defined as being too unwell to look after themselves or leave the house. To measure quality of life, we used a five-item Satisfaction With Life Scale (SWLS) developed by Diener et al. (1985). We subsequently reconsidered the measures most appropriate for health and quality of life, the latter a term which has been much debated in recent years and which is considered to have many domains. In later I’DGO work, in addition to the single-item ‘unhealthy days’ question, we used EUROQOL (EQ-5D) (EUROQOL Group 1990), a self-rated health scale, assessing five aspects of health: mobility; self-care; usual activities; pain/discomfort; and anxiety/depression. We also used the Environmental Quality (EQ) Visual Analogue Scale (VAS), which allows participants to rate their health state on a 100-point scale from worst imaginable (0) to best imaginable (100). For quality of life, we used CASP-19 (Hyde et al. 2003), a measure of quality of life in older age based on a model of needs satisfaction and comprising four domains: comparison, autonomy, pleasure and self-realisation. An additional, innovative section was added to the front of the questionnaire to allow participants to identify their personal needs and desires in relation to going outdoors, which might vary considerably between individuals. This was based on the work of Little (1983; 2010) on personal projects, and allowed us to ask what the key personal tasks or projects were for participants that took them outside the home, how important each was to them,

241 …

Catharine Ward Thompson

„

and how easy or difficult the environment made it for them to do. It allowed us to assess environmental support for each individual on a basis that accommodated differences and made no assumptions about ‘normal’ activities.

Administering the questionnaire The common challenges in administering a questionnaire relate to obtaining a sample appropriate to the research questions being asked, both in terms of sample number and demographic profile. For many questions relating to the environment in which people live, it is advisable to survey a sample that represents as closely as possible the demographics of the population or sub-population(s) of interest living there, and this is no easy task. Our initial I’DGO surveys were undertaken by post, mailing out to a sample of over-65-yearolds’ addresses provided by a market survey company, and targeted at a range of different urban, suburban and rural locations across Britain, in an attempt to get a variety of responses from those in different physical and socio-economic environmental contexts. A key element in postal surveys is making the introductory information attractive, inviting interest, and the survey itself clear and easy to complete. For our I’DGO participants this meant ensuring the text was of adequate font size for people with some visual impairment, while offering to provide larger text versions if people requested them. Despite all such efforts, our response rate was below 20 per cent, which is not uncommon for postal surveys, but underlines the challenges of getting a good sample this way. We also found we had poor representation from minority ethnic groups and so added to the sample via two community-based sessions targeted at different ethnic groups, where participants were invited via a regular social event and help with translation was available. Learning from this, our second phase of the I’DGO project (described later) involved personal interviews in a more limited set of locations, where we estimate we achieved approximately 50 per cent participation – a response rate that gives greater confidence in its representativeness.

Analyses Analyses of the I’DGO data in the early stages of the project followed standard procedures for crosssectional data, that is data derived from a single survey at one time point, where analysis is looking for differences between people that might reflect differences in their environment. We used the Statistical Package for the Social Sciences (SPSS); details of tests and their appropriate use are not described here but are well covered in many publications such as Field (2013). Initial analyses confirmed an acceptable level for the internal consistency of the new NOS scale we had developed, and calculated a measure of supportiveness of the neighbourhood environment (SNE) based on each individual’s personal projects identified, their importance and how easy the environment made them to undertake. We looked at bivariate correlations among variables and between the items measuring personal factors and perceptions of environment (NOS or SNE) and outcome measures relating to outdoor activities and quality of life. Subsequent analyses reduced the number of environmental variables being considered by using principal component analysis on the NOS scale, and then using these components as variables to explore correlations with outcome variables. Where relationships were found, logistic regression analysis was then used to identify any environmental predictors for the quality of life outcomes, and for activity types and levels that might

242 …

Landscape and health „

then predict quality of life outcomes, taking into account personal factors that might confound the findings. Such analyses are important as there is much evidence to show that personal factors such as age, functional capability and socio-economic status are independent predictors of quality of life, so the regressions indicate what environmental factors might be associated with differences in quality of life above and beyond these personal attributes. Analyses in the second phase of the I’DGO project were more complex as this was based on a longitudinal design (described below) and involved newer approaches to regression (also described below) that are considered more robust for analyses with small sample sizes and large numbers of variables.

Results The detailed results of the various analyses undertaken for the I’DGO project are not presented here as they are well described in the individual publications cited in the previous pages. A brief overview from the first phase shows findings that support our initial model, shown in Figure 14.1, that is those who live in a supportive environment tended to walk more, and high-level walkers were more likely to be in good health, but there was also an association between supportiveness of the neighbourhood environment and health independent of activity. Additional findings include that the pleasantness and safety of open spaces, and their proximity, were associated with participants’ life satisfaction, whereas the quality of paths to open spaces was associated directly with walking behaviour. The pleasantness of a local open space and lack of nuisance in it were associated with walking for recreation, while good paths to reach the open space and good facilities within it were conducive to more walking for transport. Findings from the second phase of I’DGO are discussed under the following section, on longitudinal surveys.

LONGITUDINAL STUDIES TO EXPLORE ENVIRONMENTAL DESIGN INTERVENTIONS: RESIDENTIAL STREETS While cross-sectional surveys such as described above have been used in many contexts of social science and environmental research, and can provide useful indications of important associations between aspects of people’s characteristics and behaviour in relation to aspects of their environment, they have an unavoidable shortcoming. It is never possible to be completely confident about the direction of any association between one variable and another. To put it simply: does an environment with better quality landscape attributes, such as good footpaths and attractive green space, encourage people living there to feel better about their lives, and perhaps be more active outdoors, or is it the case that people who are already active outdoor enthusiasts with a good quality of life tend to choose to live in places where there is good quality landscape? While we can control for personal differences in education or income or age, for example, in our analyses, we cannot be sure that, other things being equal, we are seeing the influence of environment on people’s activities and quality of life, or the influence of people with certain dispositions on where they choose to live or visit. One way of addressing this is to see whether the same kinds of relationships are found repeatedly in different specific contexts, including those where people may have less opportunity to exercise choice over where they live. Recent epidemiological studies, often led by health geographers, have undertaken these kinds of analyses, where the sample sizes are sufficiently large (with respondent numbers in the

243 …

Catharine Ward Thompson

„

thousands, tens of thousands or more) to allow very sophisticated analyses (e.g. Mitchell and Popham 2007; 2008; Maas et al. 2009; Stigsdotter et al. 2010; White et al. 2013). Such studies have repeatedly shown some kind of relationship between access to green space and people’s longevity, health or wellbeing, and that this relationship is usually stronger where people live in contexts of poverty or multiple deprivations. Such findings add confidence (but not certainty) to interpretations suggesting that better access to green space may be a cause of better health or quality of life, particularly as those living in poverty or deprivation are less likely to have control over their local outdoor environment or what neighbourhood they live in. Such large-scale analyses allow the potential influence of access to green or natural environments (including coasts, for example) to be estimated with confidence but are less good at analysing the influence of specific landscape details on people’s activities and health. It is here that longitudinal studies, where an intervention to change the environment is assessed in terms of any influence on the local population, can provide the strongest evidence of causality. What follows is a description of the theoretical basis and research design for a longitudinal study of this sort undertaken by OPENspace as part of the second phase of the I’DGO project: I’DGO TOO.

Research question Phase 2 of our I’DGO project sought to determine the effect of changes to the design of the residential street on which people lived, in order to make them more pedestrian-friendly and reduce the dominance of motorised traffic. We drew on the same theoretical framework as described earlier, and on emphasis within public health literature on the need to consider social ecological models of behaviour as a way to understand the role of the physical environment in relation to activity (Scottish Government 2008; Bull et al. 2010). Such models recognise that individual characteristics and preferences are active within the context of socio-economic, political, cultural and environmental factors that operate at different scales, from household and community to wider geographic levels (Barton and Grant 2006). Our main research question was: ‘Do environmental changes to the local residential street, to enhance the pedestrian experience, make a difference to older people’s quality of life?’ We wanted therefore to investigate whether interventions to enhance residential streets for local people would result in better quality of life for older people living on those streets, whether they would go out more often or spend more time outside in the local environment, and whether they would have better social networks as a result. Although updated for phase two, we essentially used the same questionnaire modules as for phase one, described above. We added a few additional questions to enquire about how lonely our participants felt (considered a sign of social isolation, especially for older people) and whether, post-intervention, they felt more active or better connected with their neighbours.

The street design intervention One of the challenges of longitudinal research of this sort, aimed at assessing the effects of an intervention to the physical environment, is to identify sites where appropriate changes will, with certainty, take place within a specified time period such that residents can be surveyed before and after the environmental changes, and all within the time frame of a single research project (the strengths, weaknesses and challenges of this method are discussed further in the concluding section). The design intervention chosen was a programme that followed on from a three-year

244 …

Landscape and health „

Department for Transport (DfT) funded pilot of ‘Home Zone’ style street improvements (based on Dutch woonerven principles, Clayden et al. 2006). These were aimed at reducing the dominance of motorised vehicular traffic on residential streets. The pilot Home Zone interventions were well funded and claims have been made for their success in enhancing local residents’ quality of life (Department for Transport 2005). For the phase two I’DGO project, DfT funding was no longer available and, although the principles of Home Zone type improvements continued to be promoted by government, funding support for such projects was more limited. We identified a programme run by a sustainable transport charity, Sustrans, as fitting the time frame of our study (2007–2011) where pilot ‘DIY Streets’ (Do-It-Yourself) interventions were planned in collaboration with a number of local communities, many of which were in very deprived urban communities in England, Wales and Scotland (Sustrans n.d.). We initially identified nine locations where a particular street was part of such an intervention (one in Scotland that was not part of the Sustrans scheme), and based our longitudinal study on these; in the event, only seven sites had an intervention completed in time for the second wave of survey, post-intervention, in 2010 or 2011. The type of changes that were eventually made within different DIY sites were not radical or comprehensive redesigns of the street environment, unlike the earlier Home Zone work supported by DfT, and were constrained by a more limited budget and more cautious community engagement process. As researchers, we were not involved in either the community participation work or in the design of the interventions; on the one hand, this meant we were able to be truly independent in our evaluation of the results, on the other, it meant we were unable to suggest designs that might have offered greater support for older people’s outdoor activity.

Research design The ideal research design for a longitudinal intervention study is to follow a cohort of participants from pre- to post-intervention, interviewing the same people before and afterwards to see what changes. Such a study requires control or comparison sites, matched as closely as possible in terms of environmental and participant characteristics, where no intervention takes place, so that wider societal influences and changes can be taken into account. Our I’DGO study design aimed to follow this principle, with each DIY Street intervention site matched as closely as possible with a comparison street in the same urban area, with similar built environment and socio-economic characteristics, but where there was no intervention. We undertook a baseline survey, before any changes to the environment, and a second wave, between three and six months after completion of the intervention. Such a study is often termed a natural experiment, or a quasi-experiment, with pre–post test design. All data were collected between May and September, the summer months in Britain, to minimise effects of variation in season. Ward Thompson et al. (2014) describes the study design and initial analyses in more detail.

Recruitment of participants We aimed to contact all people aged 65 or older on the relevant intervention and comparison streets, using a variety of recruitment techniques, to interview both before and after the intervention. Because there were only a limited number of sites, we administered our updated questionnaire by interview, face to face, but we still faced challenges in recruiting. We used prior leafleting of the streets, community meetings, newsletters and other contacts via community facilitators, to invite participants,

245 …

Catharine Ward Thompson

„

whom we interviewed either in their homes or at a local community centre. The benefits of this are the greater likelihood of a good response rate, and interviewers are able to ensure the questionnaire is properly completed, but it is a much more time-consuming process than a postal questionnaire and needs to be undertaken by a group of trained field-workers. We estimate we obtained a 50 per cent or so response rate for people aged 65 or older living in the targeted streets. For a number of reasons, principally the reduced number of sites where interventions had been completed by 2011, we had more participants in the baseline (pre-intervention) survey (n=96) than after (n=61). Furthermore, although we were aiming to maintain a cohort of the same individuals pre- and post-survey, in practice we only achieved a total cohort sample of 36, of which 20 were in the intervention sites and 16 in comparison sites. This cohort was nested within the two waves of cross-sectional survey. The low retention of participants reflects the difficulties of recruiting and retaining cohorts over several years of study, especially in an older population and one of high deprivation in some sites – often associated with higher turnover of housing. Additional measures used In addition to the questionnaire survey, the second phase of I’DGO involved a number of other measures. Participants were asked to wear accelerometers for seven consecutive days, and to keep activity diaries which recorded the timing, location and purpose of any activity that took them out of the house on each day that they wore the accelerometer. As is typical in such surveys, not all those who completed the questionnaire also agreed to wear an accelerometer and keep a diary, and adherence to the protocol for these was variable, with some people failing to comply after the first few days, so that we had a maximum (but varying) sample of 47 pre-intervention and 22 postsurvey for these measures. Figure 14.2 shows the participant numbers in each part of our study.

Figure 14.2 Number of participants in each part of the study (source: first published in Ward Thompson et al. 2014)

246 …

Landscape and health „

In addition, the research team undertook street audits before and after the interventions, to ensure we recorded the physical environment consistently and noted any changes to these environments. This used a version of the Scottish Walkability Assessment Tool (SWAT) developed by OPENspace for another project assessing the walkability of people’s local streets (Millington et al. 2009; Robertson et al. 2012). It was amended to include the kinds of items likely to be introduced in ‘DIY Streets’ interventions. Behaviour observations of use of the streets before and after interventions were also undertaken, recording activities for set morning and afternoon periods of 30 minutes, on different days of the week. This was designed to capture typical activities being undertaken, including walking and cycling, and whether or not these involved social interactions with companions or others encountered, as well as the estimated age of those observed. While time did not allow for a fully comprehensive set of observations, the methods used were broadly in line with those described in GoliĞnik and Ward Thompson (2010) and recorded onto a geographic information system (GIS) base map. Questionnaire analysis The initial analysis using SPSS, for both the wider, cross-sectional data and for the cohort, pre- and post-intervention, focused on what changes, if any were associated with the intervention sites but not the comparison sites. Initially, simple, paired t-tests were used to explore difference over time. The difference in the degree of change over time between intervention and comparison groups was also examined for each variable. Because of the complexity of the data, the items in the NOS scale (38 in this survey) were initially reduced using factor analysis (maximum likelihood method), producing nine factors to assist in exploratory analysis. Although the sample size was smaller than usually recommended for such factor analysis, the Kaiser Meyer Olkin measure of sampling adequacy was satisfactory and the factor structure gave us the opportunity to undertake initial exploration of the data more readily. Subsequently, a more innovative analysis, better suited to a high number of variables and a low sample size, was carried out on the entire set of variables for the cohort via Correlated Component Regression, using the CORExpress statistical package (Magidson 2013; Curl et al. 2015). Questionnaire results Our cohort study found that participants in the intervention group perceived their street as more walkable, and that they were more active, post-intervention, yet their self-report levels of activity did not increase over time. We did not find positive changes in other health, wellbeing or social connectedness measures, some of which declined over time in both intervention and comparison sites. The key result was that participants in the intervention sites perceived it was easier to walk on the street near home, post-intervention, and this was not found in the comparison sites. Many of the findings from the two waves of cross-sectional survey over time reinforced those of our earlier phase of I’DGO research. They underline the importance of footways that are attractive and easy to use to access local open space as factors in remaining active into old age and influencing overall quality of life. Other data analyses and results The analysis and results from the other methods of data collection continue to be explored. The use of accelerometers produced a considerable volume of data which requires expert advice to analyse and interpret correctly, and the small sample size means that we have had to be cautious in any conclusions

247 …

Catharine Ward Thompson

„

that might be drawn, recognising the results may be useful as illustrations of activity patterns for particular individuals of a certain age and living in certain urban conditions, rather than generalisable beyond our sample. The diary data are potentially valuable in helping to interpret results but, again, produced a large volume of data which requires careful processing and classification to be amenable to quantitative analysis. It is worth noting that the improved availability of Global Positioning System (GPS) recorders now, in terms of cost and wearability, compared to what was available at the research design stage of our I’DGO phase 2 study in 2006/7, means that this might be considered an alternative or addition to the activity diaries in any future such study. However, the challenges of interpreting the data and preparing them for meaningful analysis would remain. Our behaviour observations across all street users at different times of day (n=3859) indicated little change in patterns pre- and post-intervention, perhaps to be expected given the limited nature of the interventions. Some aspects of this are being explored further but, in terms of physical activity, the majority of people observed were passing through the street fairly rapidly, with most walking but 10 per cent cycling and 3 per cent running. Only 1 per cent were observed sitting, reflecting the lack of seating provision even post-intervention, and activity values were similar for the morning and afternoon observations. However, the frequency of social contact was higher during the afternoon (37 per cent of those observed), compared to only 19 per cent in the morning sessions. The street audits indicated changes that had occurred as expected but there appeared to be little association between the particular design details and new or altered behaviour by street users. In general, it is worth noting that the benefits of mixed method approaches, where different types of data may be used to check the robustness of findings, is often accompanied by greater complexity in analysis and thus it may take considerable time and/or effort to fully explore the findings from the data. The challenges in collecting and analysing longitudinal data based on questionnaires alone are considerable, despite their value in pointing to causal relationships between environment and health. Non-self-report measures such as accelerometers and behaviour observations may offer opportunities to better characterise or explain any behaviour change that has taken place but each additional method requires its own expertise to undertake and interpret. In our I’DGO phase 2 study, it has taken considerable time after the data were collected to analyse them adequately and the results suggest only very limited effects from the interventions. Nonetheless, our research design was strong and well placed to identify potential causal relationships between environmental interventions and changes in activity, sociability or quality of life, and remains sound despite the somewhat limited results reported to date. A more recent study is also using a longitudinal approach to investigate the wellbeing benefits of woodlands improvements near deprived urban communities. The logic model developed for this (see Figure 14.3) illustrates the approach to the study design, which has met the requirements of a health research funding agency, described in Silveirinha de Oliveira et al. (2013). Such longitudinal research offers opportunities for better evidencing links between landscape and health. A different kind of innovative approach being used by OPENspace, and described below, can also help planners and designers prioritise different kinds of environmental interventions.

CONJOINT ANALYSIS: AN APPROACH TO LEARNING ABOUT THE COMPARATIVE IMPORTANCE OF DIFFERENT ENVIRONMENTAL ELEMENTS The final research method illustrated here, conjoint analysis, is a ‘stated preference’ technique that is useful for learning about the relative importance of different attributes under study, for a particular

248 …

Landscape and health „

Figure 14.3 Logic model of a longitudinal approach to study design (source: first published in Silveirinha de Oliveira et al. 2013)

population. This is a relevant issue where environmental design is concerned: clients often need to know which of several alternative intervention strategies might yield the best results. The method arises out of market research, and has only recently been applied to studies on the comparative importance of different elements in the environment (e.g. Aspinall 2007; Bullock 2008).

249 …

Catharine Ward Thompson

„

Conjoint analysis has been used across a number of OPENspace projects. It has value in healthrelated research in that it offers the opportunity to understand where changes might have greatest effect in relation to a desired health or wellbeing outcome. It is a form of discrete choice methodology recommended by the UK National Institute for Health and Care Excellence (NICE) as a means of determining the best allocation and use of resources as well as to inform policy (Ryan 2004). Conjoint analysis involves choice tasks that are seen as realistic and comparable to the way people make choices in real world situations. Respondents to a conjoint task are presented with profile characteristics of objects or situations which consist of several attributes, and varying values for each attribute. Their task is simply to choose which is preferred within a given set. From a number of comparison tasks, conjoint analysis generates ‘utilities’, which are values placed on the different attributes of the situation being examined. Since choice is made by comparing different values in multiple attributes, the utilities generated in conjoint analysis are based on relative considerations of all attributes. The utilities thus tend to reflect the way people weigh up pros and cons in real life, and offer a more reliable way of judging comparative importance, or predicting people’s response to different opportunities, than conventional questionnaires that ask about each attribute separately. As with other methods described in this book, space does not allow more than an overview, but it is helpful to briefly illustrate how the initial conjoint questionnaire was developed and to demonstrate the kinds of results and scenario-modelling that subsequent analysis offers. The example is based on a separate survey that formed part of the I’DGO project, and is based on choicebased conjoint analysis, a form of the method that allows preparation of a set of questionnaires that can then be administered by post (as in the case of our study). Details are described in Alves et al. (2008) and Aspinall et al. (2010).

Developing the conjoint questionnaire Since any conjoint survey presents a series of options for choice, involving a predetermined set of attributes, in order to determine the relative importance of each, it is vital to ensure that the attributes cover all the aspects under study that participants might consider important. For this reason, it is usual to base a conjoint survey on prior research that is likely to have elicited the key aspects of relevance from the same population under study. Qualitative work such as with focus groups is valuable as well as more conventional questionnaires, and our I’DGO study drew on focus groups and accompanied walks with older participants, as well as the first phase questionnaire described earlier, in order to devise the conjoint questionnaire. The challenge then is to devise a survey using as limited a set of issues (called ‘attributes’) as possible (to ease the burden on participants), but still covering all the attributes, and different options or ‘levels’ for each, that might be most important to someone in making a choice between alternative scenarios. In our study, recognising that good access to open space was a predictor of older people’s activity levels and quality of life, we wanted to know which local park or open space would be preferred as a place to visit for our older participants. A total of 15 attributes relevant to neighbourhood parks and open space were chosen, including eight items relating to the route to a local park, such as distance to the park, pavement quality, presence of trees or seats along footpaths, and levels of traffic. Seven attributes related to the environment within a local park, such as trees and plants, facilities provided, seating available, level of park maintenance and

250 …

Landscape and health „

aspects of nuisance, with levels defined according to, for example, the density of trees and plants or the availability of facilities such as car parks, cafés and toilets. Each attribute had between two and four levels. The CBC questionnaire was constructed using Sawtooth Software (2008). CBC stands for Choice-Based Conjoint (CBC) studies. Rather than showing choices that compared variations across all 15 attributes at a time, which would create extremely difficult decision-making tasks, we used a partial-profile design, where the number of attributes shown in each comparison task was limited to four, to make the task easier. The software allowed for this by producing different versions of the questionnaire and calculating the number of different versions required in order to achieve effective coverage over all the options. Thus, we ended up with 15 different versions of the questionnaire, each of which offered 11 paired option tasks, but each respondent would receive only one such version of the questionnaire. An example of two of the option tasks offered is shown in Figure 14.4.

Figure 14.4 An example of two of the option tasks in the questionnaire

251 …

Catharine Ward Thompson

„

The results and scenario modelling options The most important attributes overall, for our 237 participants, were nuisances (such as signs of vandalism or dog fouling), facilities (such as toilets or a café) and trees and plants (density of vegetation), followed by levels of traffic en route, things to watch (such as views, wildlife) and levels of maintenance. There were two individual characteristics that made a significant difference to how participants prioritised attributes: whether people had mobility impairments that made it difficult to get around and whether people lived alone or with someone else. For example, those who lived with someone else placed relatively greater importance on the provision of facilities and a car park, while those who lived alone placed relatively greater importance on distance to the local park. The major benefit of the conjoint analysis for planners and decision makers is that the analysis produces a dynamic tool that can be used to model different scenarios, both for the sample as a whole and for sub-samples, as indicated above. This allows exploration of trade-offs, such as ‘what if we make this rather than that change?’ For example, if medium traffic levels en route to the park are reduced to light levels, does that increase preference more than adding a toilet to a park without one at present? In our study, the answer to this was ‘no’. Alternatively, adding ‘some trees and plants’ to a park with no existing trees was more than twice as valued as adding a café to a park with existing toilet provision. Although the results of conjoint analysis are always constrained by the options offered, and the population sampled, such scenario modelling would allow a planner to have confidence in how best to invest with limited resources. For example, the results could inform how to intervene at a certain point in time to maximise park preference or use by the local population, or sub-populations, providing the model was based on an appropriate survey and sample. This example is just one illustration of the conjoint analysis approach which has been used by OPENspace researchers and PhD students at the University of Edinburgh in a range of contexts. Other examples include research using different forms of conjoint analysis (such as adaptive conjoint) to explore people’s preferences for moving to different neighbourhoods in the peri-urban landscape in a range of European contexts (Nilsson et al. 2013). PhD studies have used conjoint analysis to investigate how local communities judge the scale effects of different wind farm proposals and how residents of cities in southern Europe consider which urban environments are most desirable to live in. One aspect of interest in developing conjoint analysis questionnaires is whether or not to use images as part of the presentation of alternative options for participants’ choice. Laing et al. (2005), for example, have studied residents’ streetscape preferences using three dimensional (3D) computergenerated visualisations and Priestley (2005) used photomontages of different green spaces to explore adult preference for different facilities or features. The strength of such approaches lies in the ease with which respondents can assess and respond to visually presented alternatives. However, they suffer from difficulties in producing images that accurately reflect the particular combinations of attributes assigned to each option, and in determining whether such attributes have been noticed by respondents in coming to a decision. In addition, a respondent’s decisions may be overly influenced by lighting, composition or people featured in the presented image that may have comparatively low priority in real-life decision making (Laing et al. 2009). The outdoor natural environment also presents many more challenges to realistic representation than indoor or built spaces, given the variability of season, weather, atmosphere and vegetation state (Laing et al. 2005). For this reason, OPENspace work has usually avoided such difficulties by using written descriptions

252 …

Landscape and health „

rather than images, which give clarity about what aspects of the environment are to be considered and allow participants to envisage environments that are familiar to them. However, recent PhD work at Edinburgh (ongoing at the time of writing) has also successfully used a combination of monochrome line drawings of landscape options (which allow for only the relevant features to be marked) and written descriptions in conjoint analysis. Such an approach makes use of landscape architecture skills in innovative ways for research (Stanton 2015).

DISCUSSION OF RESEARCH METHODS THAT MIGHT BE USED TO CONSIDER OPEN SPACE DESIGN FOR SALUTOGENIC LANDSCAPES Alternative research methods The methods described above are by no means an exhaustive illustration of what might be used to explore links between landscape design and people’s health and wellbeing. There are, in particular, more in-depth approaches such as interviews and accompanied walks or ‘go-along’ methods that can provide very detailed insights on how people respond to the landscape and what benefits they experience, especially if conducted over a period of time. The work of Patrik Grahn, Ulrika Stigsdotter and colleagues, based at the Alnarp Rehabilitation Garden, draws on such methods to provide a rich and detailed analysis of the qualities of garden space needed for different stages of recovery from stress-related illness (Grahn and Stigsdotter 2010; Grahn et al. 2010). Their research has resulted in a ‘quality evaluation tool’ to be used in the process of designing outdoor environments in healthcare settings (Bengtsson and Grahn 2014), which offers valuable insights for design of therapeutic gardens. While quantitative analysis such as conjoint, described earlier, and in-depth, qualitative research of the sort undertaken by the Alnarp team can both yield relevant guidance on detailed design for landscape architects, there are other relevant approaches to consider. These include behaviour observation, which can offer opportunities for both qualitative and quantitative analysis and reveal valuable findings for landscape planners and designers. Robin Moore, Nilda Cosco and colleagues have used these methods, based on theories of affordance, to undertake research on ways in which playgrounds encourage different kinds of activities in children (e.g. Moore 1974; Cosco 2007; Moore and Cosco 2007; 2010), while GoliĞnik and Ward Thompson (2010) illustrate the use of behaviour observation in the wider context of public parks and open space in the city. Space does not allow the opportunity to expand here on the full range of methods that have been effective in researching the relationship between landscape architecture and health, in its widest interpretation, but Ward Thompson et al. (2010) and Ward Thompson (2013) outline in more detail some of the approaches that have been used, particularly by designers, to investigate links between activity and design of the physical environment. The strengths, weaknesses and challenges in such methods are briefly discussed below, in concluding this chapter.

Objectivity, subjectivity and the use of self-report measures A fundamental challenge in researching people’s engagement with, and response to, their environment is understanding how much value to place on empirical measures of environment that might be considered independent of personal bias, such as mapping of vegetation based on

253 …

Catharine Ward Thompson

„

remote sensing or photography, and how much to place on people’s reports and perceptions of environments. In the latter, it may be harder to discern the difference between objective (i.e. reliable and unbiased) measures and those that are clearly subjective. An affordance theory viewpoint would claim that there is a false dichotomy in thinking about ‘objective’ versus ‘subjective’ measures, since every assessment is partly dependent on the perception of the assessor and the activity and purpose they have in mind. Nonetheless, for landscape architects and environmental designers, there is undoubtedly some value in using professional assessment in measuring quality and availability of landscape elements such as parks, streets or urban squares in research, as such professionals use these kinds of judgements in their work all the time. Audit tools such as the woodland audit tool developed by OPENspace (Silveirinha de Oliveira et al. in prep.) offer this kind of measure and can provide a level of landscape insight into research studies that is rigorous and reliable, adding confidence in the comparative objectivity of results. To do so, it is important to demonstrate that a trained auditor would give the same results if they were to audit the same, unchanging environment on two different occasions and, equally, that two different trained auditors would give the same results for that environment. For this reason, it is advisable to assess new audit tools to ensure that test–retest and inter-rater reliability are within acceptable limits. Millington et al. (2009) illustrate how this has been done for our Scottish Walkability Assessment Tool. It is important to recognise, however, that it is the local community’s own perceptions of their environment that will influence how they experience and respond to it, regardless of how expert auditors assess it. This is particularly true of aspects such as how safe an environment feels, but also whether the nearest green space is perceived as easy to access, or whether there are ‘enough’ benches en route. Someone’s perception of a local park being too far away, or not safe to use, for example, may deter them from ever visiting it, regardless of other measures such as the mapped distance or incidence of crime in the place. Such subjective perceptions are likely to vary among community members according to gender, age and a number of other factors. For this reason, if we want to understand what it is about any environment that is salutogenic, we usually need to understand users’ perceptions. These can be assessed by group methods, such as focus groups and community audits, or individual interviews and questionnaires, all of which have been described in the examples above. The most valuable research for planners and designers usually involves one of these methods as well as any non-self-report or objective assessment of environment, recognising the value of both but also accepting that the resulting assessments may differ. In a similar way, non-self-report measures of other aspects of people’s lives, and in particular their activity and health, may be valuable but give different answers to those that people report themselves. Behaviour observation is a long-established technique that offers valuable insights into how people use particular places in different ways; it remains one of the simplest but most valuable tools for landscape architects to understand response to environment. However, while it offers insights into what people do, it is unable to explain why they do it. Hence it is always useful to engage with people directly and ask them questions as well. In the arena of physical activity and health research, it is well known that certain self-report measures give different answers from other measures of (apparently) the same thing. For example, people generally tend to overestimate how active they are. Thus, while the International Physical Activity Questionnaire (IPAQ 2002) may have been developed to provide a reasonably accurate and reliable measure of people’s level of activity, including moderate and vigorous activity, in the

254 …

Landscape and health „

previous week or month, it is accepted that IPAQ results may well not tally with activity levels found if participants are asked to wear an accelerometer for a week. Both may be good at measuring comparative levels of activity between people but both have their weaknesses, for example the time it takes to ask the questions in IPAQ, and possibly inaccurate recall of respondents, or the challenge when different participants wear an accelerometer for different lengths of time throughout the day, or forget to wear it one day. As with assessments of the environment, therefore, it is often best to combine non-self-report and self-report measures of activity, health and wellbeing if possible. Equally, sometimes a simple but valid and reliable question (such as one on perceived general health) can be useful and prove a good predictor of more elaborate measures of health, especially where time and resources are constrained. For researchers working in Europe, EQ-5D is increasingly seen as important and useful because it is a very brief (five-item) self-report measure of overall health but also one that can be used as a basis for health economics and measures of ‘Quality Life Years’ (QALYs) – the unit by which health benefits are often assessed. As has been suggested above, one challenge of non-self-report measures, such as by accelerometry or GPS-derived geospatial data, is the volume of data generated. If participants are asked to wear a GPS monitor combined with an accelerometer, to provide details of where they’ve been, and how active, minute by minute, over days or even weeks, there will be an enormous volume of data generated that needs to be checked, cleaned and then analysed. This usually involves categorising the data into meaningful classes that can then be subjected to quantitative analysis. It thus requires input from researchers with expertise in the data gathering tools and the manipulation and interpretation of results. There are new means for gathering data remotely and digitally that are developed every year, for example using mobile phones or mobile neuroheadsets (OPENspace has been involved in the latter, see Aspinall et al. 2013), all of which are exciting and offer new opportunities but bring new challenges with them. Some of OPENspace’s most innovative work in recent years has involved testing salivary cortisol several times during the course of the day to measure diurnal cortisol patterns as indicators of stress. While this has proved a useful non-selfreport measure of stress that can be undertaken in more ecologically valid contexts (i.e. people going about their everyday lives) than the usual laboratory or field experiments involving biomarkers, it was only possible because of collaboration with an expert in measuring and interpreting cortisol as well as experts in health geography and psychology. Our findings have been widely reported and add to understandings of links between green space and health (see Ward Thompson et al. 2012 and Roe et al. 2013) but point to the necessity for interdisciplinary collaboration.

Questionnaires and area-level versus individual data Many other research methods publications offer details of the challenges, strengths and weaknesses of using questionnaires, and the different ways of administering them. For landscape architects, a current challenge in landscape and health research is to find meaningful but consistent measures of different aspects of the landscape that are of interest, and that can be used in different contexts and (ideally) different countries as well. There are increasing numbers of questions relating to perceptions and use of local green space, for example in UK national surveys undertaken by different public authorities, and contributions to testing and refining these can be of value, as can the development of more specific and detailed questions relating to particular aspects of the landscape that may relate to health and wellbeing.

255 …

Catharine Ward Thompson

„

Where new surveys are undertaken by questionnaire, the age-old challenges of sample size, response rate and representativeness remain but we are expected to meet the rigorous standards of public health research if our findings are to be taken seriously in that context. An alternative is to take the data being analysed from secondary sources. The growing enthusiasm of research councils and policy-makers for ‘big data’, which are increasingly available and amenable to complex analyses that would have been unthinkable pre-computers, means it is possible to access and use many different sources of information. Some is based on individual-level data, which can be most useful but involves complex mechanisms to access anonymised versions of it in compliance with data protection legislation. Other data may be compiled and made available at area level, such as the Index of Multiple Deprivation available (but calculated slightly differently) in Scotland, Wales, England and Northern Ireland. It takes national census data on, for example, housing, employment, social class and car ownership to create a single measure of how deprived an area is. The virtue of using such data is that it avoids primary data collection, and can be combined with GIS-derived data such as area or percentage of green space or woodland, or distance from the centre of each data zone to the nearest green space. However, such area-level data will not pick up on individual variations. By contrast, primary data collection allows for very idiosyncratic differences in how people might want to use the outdoor environment to be expressed and recorded (e.g. Curl et al. 2016). In our increasingly multicultural society, it may be important to allow for such differences in understanding what may or may not count as a salutogenic environment. A very different approach, using conjoint analysis as a form of discrete choice experiment, has also been shown in our work to be of value to planners of the environment, offering ways to explore prioritisation of resources and sensitive to different sub-groups within the population under study. As with all primary data collection, attention must be paid to adequate sample size, appropriate response rate and representativeness of sample in gathering and interpreting the data, where limitations on generalisability may pertain.

Longitudinal research – opportunities and challenges Longitudinal study designs have enormous potential for researching landscapes, health behaviours and quality of life. A well-founded longitudinal study that (crucially) involves an environmental intervention, and a control or comparison without an intervention, and a pre–post-intervention design, remains one of the best ways to investigate environmental influence and to reduce the suggestion of confirmation bias. It avoids the flaws found in many studies where local residents are surveyed only after some change has been implemented, and without comparison to any other site; such studies almost invariably show local people’s enthusiasm for the change in their environment, which may simply reflect enthusiasm for anyone paying attention to their community, investing in it and asking them for their opinions of it. By contrast, a well-designed project, such as the second phase I’DGO study illustrated above, has the potential at least to offer causal evidence for the benefit (or otherwise) of changes to the landscape. Such studies are termed ‘natural experiments’ in psychology to differentiate them from experiments in controlled conditions where participants are recruited and asked to spend time and undertake specified tasks in defined locations (whether laboratories or predetermined locations such as a park or an urban street). A natural experiment involves an intervention happening in the real world, and so is more likely to reflect people’s everyday lives and the usual settings for their everyday activities. It therefore has greater ecological validity than artificially designed experiments, which are considerably

256 …

Landscape and health „

easier to set up and control (and thus often used in environmental psychology) but not always possible or desirable ethically. However, as the examples show, it is extremely difficult to undertake a natural experiment based on a major intervention in the landscape (whether urban, peri-urban or rural) and to get the timing of funding, planning, intervention, and both pre- and post-survey right. In practice, such a study requires considerable forward planning and an excellent working relationship with those funding and implementing the environmental changes in order to secure funding for the research element of the study. Research support, which is usually funded from a different source than the intervention itself, needs to be applied for in confidence that it will not be awarded too late to undertake the baseline survey, but also that the intervention will happen as planned, in a timely fashion, and the post-intervention survey can be undertaken before the research funding runs out. In practice, all of these aspects are vulnerable to delays and changes of plan. In addition, maintaining a cohort of the same individuals to survey both before and after the intervention (the ideal study design) is also vulnerable to practical difficulties in retaining participation, especially if the research is undertaken in areas of poverty, deprivation and/or high turnover of residents. Such difficulties beset many a well-designed longitudinal research project and it is only through establishing good working relationships with all organisations and funders involved that any prospect of success is possible. This might suggest that it is better for a landscape architect to be involved both in designing the research study and in designing and supervising implementation of the environmental intervention. Such involvement might ease some of the problems but is rarely possible in practice as the skills and time commitment needed for one may be at odds with the other. It is also the case that acting as an independent researcher allows the study to be completely unbiased and less susceptible to criticism of conflict of interests. As stated earlier, the research examples described in this chapter point to ways of gathering evidence in which landscape architects as researchers have a great deal to offer, but which are likely to require working within a larger team with a considerable range of expertise, including qualitative and quantitative techniques and often very sophisticated statistical analysis. The findings are also likely to offer results of great value to landscape architects because they provide an evidential base on which to promote the need for our profession in planning and designing our living environments at all scales, from city level to local community and street. Our role as experts in designing spaces that support health and wellbeing may be recognised anew, drawing on such research, providing we undertake it in ways that answer twenty-first-century public health needs for robust and meaningful evidence.

SUGGESTED FURTHER READING Allan, G. and Skinner, C., eds. (1991) Handbook for Research Students in the Social Sciences, London: The Falmer Press. Aimed at helping research students, working full- or part-time in the social sciences. Aspinall, P.A. (2007) ‘On quality of life, analysis and evidence-based belief’, in Ward Thompson, C. and Travlou, P., eds. Open Space: People Space. Abingdon: Routledge, 181–194. This offers a useful introduction to the principles and virtues of conjoint analysis. Aspinall, P.A. (2010) ‘On environmental preference: Applying conjoint analysis to visiting parks and buying houses’, in Ward Thompson, C., Aspinall, P.A. and Bell, S., eds. Innovative Approaches to Researching Landscape and Health, Open Space: People Space 2, Abingdon: Routledge, 179–208. This provides an applied example of conjoint analysis offering insights that are of practical use in decision making about investing in environments.

257 …

Catharine Ward Thompson

„

Field, A. (2013) Discovering Statistics using IBM SPSS Statistics, 4th ed., London: Sage. An excellent and very readable guide to correct use and interpretation of statistical analyses based on SPSS. Gideon, L., ed. (2012) Handbook of Survey Methodology for the Social Sciences, New York: Springer. This provides a comprehensive overview of survey methodology in the social sciences. Sallis, J.F. and Owen, N. (2002) ‘Ecological models of health behavior’, in Glanz, K., Rimer, B.K. and Lewis, F.M., eds. Health Behavior and Health Education: Theory, Research, and Practice, 3rd ed., San Francisco, CA: Jossey-Bass, 462–484. An introduction to health models. Ward Thompson, C. and Travlou, P., eds. (2007) Open Space: People Space, Abingdon: Routledge. An introduction to ways of exploring landscape architecture, inclusive access to outdoor environments and quality of life. Ward Thompson, C., Aspinall, P.A. and Bell, S., eds. (2010) Innovative Approaches to Researching Landscape and Health: Open Space: People Space 2, Abingdon: Routledge. This introduces researchers in landscape architecture to new and relevant approaches to research methodology and practice.

REFERENCES Allan, G. and Skinner, C., eds. (1991) Handbook for Research Students in the Social Sciences, London: The Falmer Press. Alves, S., Aspinall, P.A., Ward Thompson, C., Sugiyama, T., Brice, R. and Vickers, A. (2008) ‘Preferences of older people for environmental attributes of local parks: The use of choice-based conjoint analysis’, Facilities, 26(11/12), 433–453. Antonovsky, A. (1996) ‘The salutogenic model as a theory to guide health promotion’, Health Promotion International, 11(1), 11–18. Aspinall, P.A. (2007) ‘On quality of life, analysis and evidence-based belief’, in Ward Thompson, C. and Travlou, P., eds. Open Space: People Space, Abingdon: Routledge, 181–194. Aspinall, P.A., Ward Thompson, C., Alves, S., Sugiyama, T., Brice, R. and Vickers, A. (2010) ‘Preference and relative importance for environmental attributes of neighbourhood open space in older people’, Environment and Planning B: Planning and Design, 37(6), 1022–1039. Aspinall, P.A., Pangiotis, M., Coyne, R. and Roe, J. (2013) ‘The urban brain: Analysing outdoor physical activity with mobile EEG’, British Journal of Sports Medicine, available: doi:10.1136/bjsports-2012-091877. Barbour, R. (2007) Doing Focus Groups, London: Sage. Barton, H. and Grant, M. (2006) ‘A health map for the local human habitat’, Journal of the Royal Society for the Promotion of Health, 126(6), 252–253. Bengtsson, A. and Grahn, P. (2014) ‘Outdoor environments in healthcare settings: A quality evaluation tool for use in designing healthcare gardens’, Urban Forestry & Urban Greening, 13(4), 878–891. Brookfield, K., Bloodworth, A. and Mohan, J. (2013) ‘Engaging residents’ groups in planning using focus groups’, Proceedings of the Institution of Civil Engineers – Engineering Sustainability, 166(2), 61–74. Bull, F., Giles-Corti, B. and Wood, L. (2010) ‘Active landscapes: The methodological challenges in developing the evidence on urban environments and physical activity’, in Ward Thompson, C., Aspinall, P.A. and Bell, S., eds. Innovative Approaches to Researching Landscape and Health: Open Space: People Space 2, Abingdon: Routledge, 97–119. Bullock, C.H. (2008) ‘Valuing urban green space: Hypothetical alternatives and the status quo’, Journal of Environmental Planning and Management, 51(1), 15–35. Clayden, A., McKoy, K. and Wild, A. (2006) ‘Improving residential liveability in the UK: Home zones and alternative approaches’, Journal of Urban Design, 11(1), 55–71. Cosco, N.G. (2007) ‘Developing evidence-based design: Environmental interventions for healthy development of young children in the outdoors’, in Ward Thompson, C. and Travlou, P., eds. Open Space, People Space, Abingdon: Routledge, 125–135. Curl, A., Ward Thompson, C. and Aspinall, P.A. (2015) ‘The effectiveness of “shared space” residential street interventions on self-reported activity levels and quality of life for older people’, Landscape and Urban Planning, 139, 117–125.

258 …

Landscape and health „

Curl, A., Ward Thompson, C. and Aspinall, P.A. (2016) ‘Outdoor environmental supportiveness and older people’s quality of life: A personal projects approach’, Journal of Housing for the Elderly, 30(1), 1–17. Department for Transport (2005) Home Zones: Challenging the Future of our Streets, available: http://www.rudi. net/files/homezones.pdf (accessed 7 July 2016). Diener, E., Emmons, R.A., Larsen, R.J. and Griffin, S. (1985) ‘The satisfaction with life scale’, Journal of Personality Assessment, 49(1), 71–75. European Commission (2014) Horizon 2020 The EU Framework Programme for Research and Innovation: Societal Challenges, available: http://ec.europa.eu/programmes/horizon2020/en/h2020-section/societalchallenges (accessed 16 September 2014). EuroQol Group (1990) EQ-5D Health Questionnaire, available: http://www.euroqol.org (accessed 29 April 2011). Evans, R.G. and Stoddart, G.L. (1990) ‘Producing health, consuming health care’, Social Science and Medicine, 31(12), 1347–1363. Field, A. (2013) Discovering Statistics using IBM SPSS Statistics, 4th ed., London: Sage. Gideon, L., ed. (2012) Handbook of Survey Methodology for the Social Sciences, New York: Springer. GoliĞnik, B. and Ward Thompson, C. (2010) ‘Emerging relationships between design and use of urban park spaces’, Landscape and Urban Planning, 94(1), 38–53. Grahn, P. and Stigsdotter, U.K. (2010) ‘The relation between perceived sensory dimensions of urban green space and stress restoration’, Landscape and Urban Planning, 94(3), 264–275. Grahn, P., Ivarsson, C.T., Stigsdotter, U.K. and Bengtsson, I. (2010) ‘Using affordances as health promoting tool in a therapeutic garden’, in Ward Thompson, C., Aspinall, P.A. and Bell, S., eds. Innovative Approaches in Researching Landscape and Health, Open Space: People Space 2, Abingdon: Routledge, 120–159. Humpel, N., Owen, N., Iverson, D., Leslie, E. and Bauman, A. (2004) ‘Perceived environment attributes, residential location, and walking for particular purposes’, American Journal of Preventive Medicine, 26(2), 119– 125. Hyde, M., Wiggins, R.D., Higgs, P. and Blane, D.B. (2003) ‘A measure of quality of life in early old age: The theory, development and properties of a needs satisfaction model (CASP-19)’, Aging & Mental Health, 7(3), 186–194. IPAQ (2002) International Physical Activity Questionnaire: Short Last 7 Days Telephone Format, available: https:// sites.google.com/site/theipaq/questionnaire_links (accessed 18 January 2015). Iwarsson, S. (2005) ‘A long-term perspective on person-environment fit and ADL dependence among older Swedish adults’, The Gerontologist, 45(3), 327–336. Jette, A.M., Davies, A.R., Cleary, P.D., Calkins, D.R., Rubenstein, L.V., Fink, A., Kosecoff, J., Young, T.R., Brook, R.H. and Delbanco, T.L. (1986) ‘The functional status questionnaire: Reliability and validity when used in primary care’, Journal of General Internal Medicine, 1(3), 143–149. Kahana, E., Lovegreen, L., Kahana, B. and Kahana, M. (2003) ‘Person, environment and person-environment fit as influences on residential satisfaction of elders’, Environment and Behaviour, 35(3), 434–453. Kelly, G.A. (1955) The Psychology of Personal Constructs, New York: Norton. Krueger, R.A. and Casey, M.A. (2000) Focus Groups: A Practical Guide for Applied Research, 3rd ed., Thousand Oaks, CA: Sage. Laing, R., Davies, A.-M. and Scott, S. (2005) ‘Combined use of computer-based visualization and contingent rating’, in Bullock, C., ed. Greenspace: Final Report, Dublin: Environment Centre, 51–56. Laing, R., Davies, A.-M., Miller, D., Conniff, A., Scott, S. and Morrice, J. (2009) ‘The application of visual environmental economics in the study of public preference and urban greenspace’, Environment and Planning B: Planning and Design, 36(2), 355–375. Lawton, M.P. (1980) Environment and Aging, Monterey, CA: Brooks/Cole. Lawton, M.P. and Nahemow, L. (1973) ‘Ecology and the aging process’, in Eisdorfer, C. and Lawton, P.M., eds. The Psychology of Adult Development and Aging, Washington, DC: American Psychological Association, 619–674. Little, B.R. (1983) ‘Personal projects: A rationale and method for investigation’, Environment and Behaviour, 15(3), 273–309. Little, B.R. (2000) ‘Persons, contexts, and personal projects: Assumptive themes of a methodological transactionalism’, in Wapner, S., Demick, J., Yamamoto, T. and Minami, H., eds. Theoretical Perspectives

259 …

Catharine Ward Thompson

„

in Environment-Behavior Research: Underlying Assumptions, Research Problems, and Methodologies, New York: Plenum, 79–88. Little, B.R. (2010) ‘Opening space for project pursuit: Affordance, restoration and chills’, in Ward Thompson, C., Aspinall, P.A. and Bell, S., eds. Innovative Approaches in Researching Landscape and Health, Open Space: People Space 2, Abingdon: Routledge, 163–178. Maas, J., Verheij, R.A., de Vries, S., Spreeuwenberg, P., Groenewegen, P.P. and Schellevis, G.S. (2009) ‘Morbidity is related to a green living environment’, Journal of Epidemiology and Community Health, 63(12), 967–973. Magidson, J. (2013) ‘Correlated component regression: Re-thinking regression in the presence of near collinearity’, in Abdi, H., Chin, W.W., Vinzi, V.E., Russolillo, G. and Trinchera, L., eds. New Perspectives in Partial Least Squares and Related Methods, New York: Springer, 66–67. Millington, C., Ward Thompson, C., Rowe, D., Aspinall, P.A., Fitzsimons, C., Nelson, N. and Mutrie, N. (2009) ‘Development of the Scottish Walkability Assessment Tool (SWAT)’, Health and Place, 15(2), 474–481. Mitchell, R. and Popham, F. (2007) ‘Greenspace, urbanity and health: Relationships in England’, Journal of Epidemiology and Community Health, 61(8), 681–683. Mitchell, R. and Popham, F. (2008) ‘Effect of exposure to natural environment on health inequalities: An observational population study’, The Lancet, 372(9650), 1655–1660. Moore, R.C. (1974) ‘Patterns of activity in time and space: The ecology of a neighbourhood playground’, in Canter, D. and Lee, L., eds. Psychology and the Built Environment, London: Architectural Press, 118–131. Moore, R.C. and Cosco, N.G. (2007) ‘What makes a park inclusive and universally designed? A multi-method approach’, in Ward Thompson, C. and Travlou, P., eds. Open Space: People Space, Abingdon: Routledge, 85–110. Moore, R.C and Cosco, N.G. (2010) ‘Using behaviour mapping to investigate healthy outdoor environments for children and families: Conceptual framework, procedures and applications’, in Ward Thompson, C., Aspinall P.A. and Bell, S., eds. Innovative Approaches to Researching Landscape and Health: Open space: People Space 2, Abingdon: Routledge, 33–73. Morris, G.P. and Robertson, R. (2003) Environmental Health in Scotland and the Health Improvement Challenge, Royal Environmental Health Institute of Scotland, available: http://www.rehis.com/sites/default/files/ Morris-Robertson%20Report.pdf (accessed 29 October 2010). Morris, G.P., Beck, S.A., Hanlon, P. and Robertson, R. (2006) ‘Getting strategic about the environment and health’, Public Health, 120(10), 889–903. MRC (n.d.) Physical Activity Assessment – Questionnaire, Medical Research Council, available: http://dapa-toolkit. mrc.ac.uk/physical-activity-assessment/methods/questionnaire/examples-and-links.php (accessed 7 January 2015). Nilsson, K., Pauleit, S., Bell, S., Aalbers, C.B.E.M. and Sick Nielsen, T.A., eds. (2013) Peri-urban Futures: Scenarios and Models for Land Use Change in Europe, Dordrecht: Springer. Oliver, M. and Sapey, B. (2006) Social Work with Disabled People, 3rd ed., Hampshire: Palgrave Macmillan. Owen, N., Humpel, N., Leslie, E., Bauman, A. and Sallis, J.F. (2004) ‘Understanding environmental influences on walking: Review and research agenda’, American Journal of Preventive Medicine, 27(1), 67–76. Priestley, G. (2005) ‘Issues in relation to green space valuation, preferences and use in Barcelona’, in Bullock, C., ed. Greenspace: Final Report, Dublin: Environment Centre, 45–49. Robertson, L.B., Ward Thompson, C., Aspinall, P.A., Millington, C., McAdam, C. and Mutrie, N. (2012) ‘The influence of the local neighbourhood environment on walking levels during the Walking for Wellbeing in the West pedometer-based community intervention’, Journal of Environmental and Public Health, available: doi:10.1155/2012/974786. Roe, J.J., Ward Thompson, C., Aspinall, P.A., Brewer, M.J., Duff, E.I., Miller, D., Mitchell, R. and Clow, A. (2013) ‘Green space and stress: Evidence from cortisol measures in deprived urban communities’, International Journal of Environmental Research and Public Health, 10(9), 4086–4103. Ryan, M. (2004) ‘Discrete choice experiments in health care’, British Medical Journal, 328(7436), 360–361. Saelens, B.E., Sallis, J.F., Black, J.B. and Chen, D. (2003) ‘Neighborhood-based differences in physical activity: An environment scale evaluation’, American Journal of Public Health, 93(9), 1552–1558.

260 …

Landscape and health „

Sallis, J.F. and Owen, N. (2002) ‘Ecological models of health behavior’, in Glanz, K., Rimer, B.K. and Lewis, F.M., eds. Health Behavior and Health Education: Theory, Research, and Practice, 3rd ed., San Francisco, CA: Jossey-Bass, 462–484. Sawtooth Software (2008) The CBC Advanced Design Module (ADM) Technical Paper, Sawtooth Software: Technical Paper Series, available: http://www.sawtoothsoftware.com/support/technical-papers/ sawtooth-software-products/cbc-advanced-design-module-technical-paper-2008 (accessed 18 January 2015). Scottish Government (2008) Good Places, Better Health: A New Approach to Environment and Health in Scotland – Implementation Plan, Edinburgh: The Scottish Government. Shortt, N.K., Rind, E., Pearce, J. and Mitchell, R. (2014) ‘Integrating environmental justice and socioecological models of health to understand population-level physical activity’, Environment and Planning A, 46(6), 1479–1495. Silveirinha de Oliveira, E., Aspinall, P.A., Briggs, A., Cummins, S., Leyland, A.H., Mitchell, R., Roe, J. and Ward Thompson, C. (2013) ‘How effective is the Forestry Commission Scotland’s woodland improvement programme – “Woods In and Around Towns” (WIAT) – at improving psychological well-being in deprived urban communities? A quasi-experimental study’, British Medical Journal Open, 3(8), available: doi:10.1136/bmjopen-2013- 003648. Silveirinha de Oliveira, E., Bell, S., Aspinall, P.A., Roe, J. and Ward Thompson, C. (in prep.) ‘Development and testing an environmental audit tool for woodlands’. Stanton, C. (2015) ‘Perception of the Scale Effects of Wind Turbines in the Scottish Landscape’, accepted for the European Conference of the Landscape Research Group, Dresden, 16–18 September. Stigsdotter, U.K., Ekholm, O., Schipperijn, J., Toftager, M., Kamper-Jorgensen, F. and Randrup, T.B. (2010) ‘Health promoting outdoor environments: Associations between green space, and health, health-related quality of life and stress based on a Danish national representative survey’, Scandinavian Journal of Public Health, 38, 411–417. Sugiyama, T. and Ward Thompson, C. (2006) ‘Is Older People’s Perception of Neighbourhood Open Space Associated with Patterns of Outdoor Activity?’ accepted at the 1st International Symposium on Environment, Behaviour and Society, University of Sydney, 9–11 February, available: http://www.idgo.ac.uk/PDFs/ Sugiyama_WardThompson_NOS_and_outdoor_activity_June2006.pdf (accessed 7 January 2015). Sugiyama, T. and Ward Thompson, C. (2007a) ‘Outdoor environments, activity and the well-being of older people: Conceptualising environmental support’, Environment and Planning A, 39(8), 1943–1960. Sugiyama, T. and Ward Thompson, C. (2007b) ‘Older people’s health, outdoor activity and supportiveness of neighbourhood environments’, Landscape and Urban Planning, 83(2), 168–175. Sugiyama, T. and Ward Thompson, C. (2007c) ‘Measuring the quality of the outdoor environment relevant to older people’s lives’, in Ward Thompson, C. and Travlou, P., eds. Open Space: People Space, Abingdon: Routledge, 153–162. Sugiyama, T. and Ward Thompson, C. (2008) ‘Associations between characteristics of neighbourhood open space and older people’s walking’, Urban Forestry & Urban Greening, 7(1), 41–51. Sugiyama, T., Ward Thompson, C. and Alves, S. (2009) ‘Associations between neighborhood open space attributes and quality of life for older people in Britain’, Environment and Behavior, 41(1), 3–21. Sugiyama, T., Giles-Corti, B., Summers, J., Du Toit, L., Leslie, E. and Owen, N. (2013) ‘Initiating and maintaining recreational walking: A longitudinal study on the influence of neighborhood green space’, Preventive Medicine, 57(3), 178–182. Sustrans (n.d.) Community Street Design, available: http://www.sustrans.org.uk/diystreets (accessed 10 January 2015). Ward Thompson, C. (2011) ‘Linking landscape and health: The recurring theme’, Landscape and Urban Planning, 99(3), 187–195. Ward Thompson, C. (2013) ‘Activity, exercise and the planning and design of outdoor spaces’, Journal of Environmental Psychology, 34, 79–96. Ward Thompson, C., Aspinall, P.A. and Bell, S., eds. (2010) Innovative Approaches to Researching Landscape and Health: Open Space: People Space 2, Abingdon: Routledge.

261 …

Catharine Ward Thompson

„

Ward Thompson, C., Roe, J., Aspinall, P.A., Mitchell, R., Clow, A. and Miller, D. (2012) ‘More green space is linked to less stress in deprived communities: Evidence from salivary cortisol patterns’, Landscape and Urban Planning, 105(3), 221–229. Ward Thompson, C., Curl, A., Aspinall, P.A., Alves, S. and Zuin, A. (2014) ‘Do changes to the local street environment alter behaviour and quality of life of older adults? The “DIY Streets” intervention’, British Journal of Sports Medicine, 48, 1059–1065. White, M.P., Alcock, I., Wheeler, B.W. and Depledge, M.H. (2013) ‘Would you be happier living in a greener urban area? A fixed-effects analysis of panel data’, Psychological Science, 24(6), 920–928.

262 …

Chapter 15: Thermally comfortable urban environments Robert D. Brown and Terry J. Gillespie

INTRODUCTION More than half the world population lives in cities, and that proportion is increasing. However, cities are becoming more climatically inhospitable environments because of two anthropogenic phenomena: global climate change (e.g. Seneviratne et al. 2012) and urban heat island intensification (e.g. Watkins et al. 2007; Chow et al. 2012; Mishra et al. 2015). The World Meteorological Organization (2015) has reported that (t)he first 12 years of the 21st century have seen record temperatures …, exceptional heat waves in Western Europe (2003) and Russia (2010) … Many other extremes were also experienced elsewhere in the world. The year 2013 has been marked by extreme heat in Australia, drought in Brazil and the United States, and record summer heat in parts of China.

Microclimates are everywhere. Indoor microclimates tend to be carefully controlled and are almost unnoticeable, but outdoors the microclimate can vary widely. As people move through the landscape they pass through a series of microclimates that might make them feel too warm or too cool, and might even be dangerously hot or cold. The 2003 European heat wave alone caused more than 70,000 deaths (Robine et al. 2008). Cities should be designed to create microclimates that ameliorate the effects of global climate change and urban heat island intensification and lead towards the creation of sustainable cities (e.g. McCarthy et al. 2010; Grimmond et al. 2010; Brown 2011; Brown et al. 2015). Research has indicated that appropriately designed green spaces can create park cool islands (PCI) through shading of surfaces and increased evapotranspiration (e.g. Spronken-Smith and Oke 1998; Declet-Barreto et al. 2013). Further research is needed to determine how to maximize the effect of PCIs and to evaluate whether they are effective in all climate zones and all urban environments (Brown et al. 2015). There is increasing evidence of the effects of urban microclimates on human health and well-being (e.g. Vanos et al. 2010; Wolch et al. 2011; Vanos 2015). Heat waves are becoming more frequent and more intense making people increasingly vulnerable to heat stress (Harlan et al. 2006). Reductions in the amount of radiation that people receive in urban areas can lower negative health effects that are often associated with urban heat islands (e.g. Thorsson et al. 2014). Air pollution can be reduced in urban areas when the urban heat island is mitigated (Harlan and Ruddell 2011) through natural rather than artificial cooling (e.g. Shashua-Bar et al. 2011; Chow et al. 2011). People need access to solar

263 …

Robert D. Brown and Terry J. Gillespie

„

radiation in order to satisfy their requirement for Vitamin D (e.g. Holick 2004) while not exposing themselves to the dangers of erythema (sunburn) and possible skin cancer. This is particularly important in landscapes where people are required to spend time in places such as school playgrounds (e.g. Vanos 2015). Research needs to identify how to optimize the amount of solar radiation that people experience in different environments. And perhaps most importantly, outdoor spaces that are designed to be more thermally comfortable are used more, thus encouraging people to include outdoor physical activities in their lifestyle – an important aspect of promoting a healthy lifestyle and reducing obesity rates. Wildlife is increasingly at risk of extirpation or extinction due to habitat loss and a key component of a species’ habitat is the microclimate. The Karner blue butterfly, for example, has been extirpated in Ontario, Canada, likely due to loss of habitat and fragmentation (Chan and Packer 2006). It requires very specific microclimatic conditions to survive and reproduce. Research has identified how landscapes can be designed to provide the necessary microclimate (Brown et al. 2011). Global climate change will result in more species at risk of population decline due to their inability to adapt to changes in their local habitat, including their surrounding microclimates. Research is required to understand the microclimate needs of both flora and fauna in urban natural areas so that future designs can provide appropriate habitat. Some diseases are spreading as the climate warms. For example, Lyme disease is spreading into new areas due to global climate change (Ostfeld and Brunner 2015) putting human health and well-being at risk. Research has indicated that microclimate modification can reduce the possibility of humans contracting the disease (e.g. Ward and Brown 2004). Lyme disease relies on black-legged ticks to inject the spirochete into humans, and the ticks need very specific microclimatic conditions to survive. If the microclimate is made inhospitable through microclimatic design, we might be able to remove the tick and reduce the possibility of humans contacting Lyme disease. This preliminary work requires follow-up studies to confirm its efficacy, and other infectious diseases are emerging or spreading (McMichael 2013) and require research into how urban design can reduce the impact of disease on human populations. It takes an increased amount of energy to cool buildings in urban areas during heat waves compared to normal summer weather. The landscape has been shown to affect the amount of cooling required in buildings (e.g. McPherson et al. 1989; Millward et al. 2014). Appropriately designed urban environments lower urban heat islands and stabilize the thermal environment of buildings, thus lessening energy demands and the draw on building heating and cooling systems. This in turn decreases the risk of municipal power outages where people might become more susceptible to heat stress (e.g. Salamanca et al. 2014). However, more research is needed to identify how to maximize building cooling in urban areas through landscape modification, particularly during heat waves. This is a small sample of the many roles that microclimate plays in the landscape. Virtually all research in the landscape should consider microclimate. For example, studies of people’s visual preference might be affected by the microclimate and whether or not study participants are thermally comfortable. In addition the diversity, abundance, and survival of plant species is also largely dependent on microclimate. A great deal of important research remains to be done and this chapter will describe both appropriate instruments for measurement and approaches to estimating microclimate components. The discussion and examples will focus primarily on the thermal comfort of people in the landscape, but many of the instruments and methods can be used to study other microclimate-related questions in other contexts. In addition to microclimate research that informs design directly, there is a need

264 …

Thermally comfortable urban environments „

for research that increases the knowledge base of landscape architecture as an academic discipline. The methods, instruments, and techniques described in this chapter can be used for both basic and applied microclimate research. Johansson et al. (2014) reported that the number of urban microclimate and human thermal comfort studies has been increasing in number but there is no standard approach either in terms of instrumentation or analysis. This chapter provides a standard approach to instrumentation and measurement that is based on micrometeorological theory and practice, and calls for further study into identifying appropriate methods for measuring and assessing human thermal comfort.

THEORETICAL FOUNDATION Whenever people are outdoors they are experiencing their environment as an exchange of energy between themselves and their local environment. If they are receiving as much energy as they are giving off they will feel thermally comfortable. But if the balance tips towards receiving too much energy a person will feel too warm, and if this tips too far they can experience hyperthermia. Similarly, receiving too little energy can lead to a person feeling too cool and possibly experiencing hypothermia. These are extremes but even a small imbalance can lead to thermal discomfort. The goal of the design of outdoor environments should be to provide thermally neutral environments for people in all seasons, and in all possible climate futures. The conceptual framework is fairly simple. Designs must be appropriate for the prevailing climate of a region. The macroclimate cannot be controlled through landscape modification. Nothing can be done to change the location of the sun in the sky or substantially change the temperature of an air mass. This macroclimate tends to be quite stable over time but in the past 100 years or so has slowly been changing, a phenomenon known as global climate change (GCC). Cities within these fairly stable climates often modify the conditions somewhat and have their own urban heat island (UHI) at the mesoclimate scale, where built-up areas tend to be warmer and drier than the surrounding countryside (e.g. Howard 1833; Oke 1982). It is at the macroclimate scale where regional design affects policy for making land use decisions. These decisions have quantitative (land development) and qualitative implications (land cover such as type of buildings, forest, open space, etc.). It’s at the microclimate scale that landscape architects can most affect conditions through urban design. It’s not clear what the world climate will be in the future, but urban areas must be designed so that they will provide thermally comfortable environments and minimize the impact of UHIs in all possible future climates (Brown 2011). There are two main research questions that need to be investigated. The first is ‘How do built environments affect the macroclimate to create microclimates?’ That is, how are temperature, humidity, wind, and radiation modified by buildings, plants, and surfaces in the landscape? The second question is ‘How do humans perceive their thermal comfort under different microclimatic conditions?’ As these two questions get answered in more and more detail, evidence-based landscape architecture (Brown and Corry 2011) will allow designers to create urban environments that modify the macroclimate to create microclimates that are both thermally comfortable under normal weather conditions, and safe during heat waves or extreme cold. Microclimate theory is founded on the concept of energy budgets. Energy flows into and through landscapes in a variety of forms. The main inputs of energy to a landscape are in the form of solar radiation (from the sun) and terrestrial radiation (from all surfaces on earth and the atmosphere).

265 …

Robert D. Brown and Terry J. Gillespie

„

Solar radiation can be reflected or absorbed by surfaces depending on the albedo of the surface. A light coloured surface has a high albedo and will reflect more solar radiation than a dark coloured surface which will absorb most of the solar radiation that it receives. Terrestrial radiation is almost entirely absorbed by most surfaces. Since energy can be neither created nor destroyed, the radiation that is absorbed by a surface has to go somewhere. The energy is consumed by four main streams: evaporation, convection, conduction, and emitted terrestrial radiation. For example, if a surface is wet some water will evaporate and carry energy away, cooling the surface. If a surface is dry this pathway is not available and all the energy will go into the other three streams. Some of the energy will be conducted into the material based on its thermal admittance. When wind passes over a heated surface the air will carry heat away through a process called convection. In addition, everything on earth emits terrestrial radiation based on its temperature. The hotter a surface, the more terrestrial radiation it emits. Energy budgets can also be applied to humans and equations can be written to estimate the flows of energy to and from a human body. There are three main inputs of energy: solar radiation, terrestrial radiation, and metabolic heat generated inside the body. The metabolic energy is dependent on the activity level of a person (e.g. sitting, walking, running). A person at rest generates a relatively small amount of internal heat, while a very active person can generate a large amount of internal energy. There are three main ways that energy is lost from a person’s body: evaporation, convection, and emitted terrestrial radiation. The overall energy budget of a person can be written as an equation as follows: Budget = [SR(abs) + TR(abs) + M] – [E + C + TR(emit)], in watts

[1]

Where: SR(abs) = solar radiation absorbed by the person TR(abs) = terrestrial radiation absorbed by the person M = metabolic energy generated within a person E = evaporative heat loss from a person C = convective heat loss from a person TR(emit) = terrestrial radiation emitted by a person The equation yields the energy balance, in watts. Writing an equation like this provides the opportunity to consider each stream of energy in turn and evaluate which stream might be causing over- or under-heating of a person. Equation [1] is the basis for several outdoor thermal comfort models such as COMFA (Brown and Gillespie, 1986), Physiological Equivalent Temperature (Höppe 1999) and Universal Thermal Comfort Index (UTCI, 2009). These models include equations for calculating convective and evaporative heat loss. Estimates of metabolic energy generated within a person at various activity levels and insulation value of clothing can be found in published tables (e.g. Brown and Gillespie 1995; ISO 2007). The theory of energy flows in the landscape is well-established. Measurement and testing over many years has supported the concepts and can be used as the basis for evidence-based landscape architecture. What is less well-known is how elements in the landscape affect the microclimate. The understanding of people’s perception of their thermal comfort is also less well-established and there are still many questions to be answered. There is evidence that a person’s perception of their thermal comfort can vary over time, particularly in different seasons. For example, people tend to acclimatize

266 …

Thermally comfortable urban environments „

over time so a warm day after a cold winter might encourage people to go outside in a T-shirt and shorts. But those same people might put on a sweater and long pants if they experienced that same warm day at the end of a hot summer. Johansson et al. (2014) reported that more than 100 energy budget models have been developed and many different questionnaires have been used in thermal comfort studies. There is no generally accepted approach to either modeling or field measurements so study results cannot be compared. Relatively little research has been done to test energy budget models against people’s perceptions of their outdoor thermal comfort. This is an area that should be investigated further to strengthen the evidence base of landscape architecture.

METHODS FOR MEASUREMENT AND ESTIMATION OF MICROCLIMATE ELEMENTS: HOW DO ELEMENTS IN THE LANDSCAPE AFFECT COMPONENTS OF MICROCLIMATE? The components of microclimate are: solar radiation, terrestrial radiation, air temperature, air humidity, and wind. Each component has specialized instrumentation and techniques to ensure accurate and precise measurements. While measurement might seem easy and straightforward, without extreme care and attention measurements are likely to be neither accurate nor precise and can lead to erroneous conclusions. The following sections outline appropriate methods for measuring and estimating microclimatic elements. Some of the more technical instruments might require the collaboration of a micrometeorologist, but by following the advice provided a landscape architecture researcher can collect accurate and precise measurements.

Radiation Solar radiation Measurement Solar radiation is measured with instruments called pyranometers (‘fire-meters’). The most accurate models are ‘thermal pyranometers’ which sense the strength of the solar radiation by measuring the elevation in temperature of a black disk exposed to the sun (e.g. Eppley; Kipp and Zonen; Huskvarna). A glass dome covers the sensor to protect it from the influence of wind and rain, and to block incoming terrestrial radiation. These most accurate instruments might be required when measuring solar radiation in a situation with a complex arrangement of buildings and vegetation. The output voltage from these sensors is usually recorded with a data logger (e.g. Campbell Scientific). A more portable option is the LP02-LI19 pyranometer (Hukseflux), which is a hand-held device. The ‘photocell pyranometer’ (e.g. LI-COR model LI-200) is a less expensive and easy to use option (see Figure 15.1). The photocell does not sense across the whole solar spectrum, but this shortcoming is mediated by using an appropriate calibration factor. However, this calibration factor is only valid for measurements under open sky and is not valid in complex settings where the solar spectrum has been significantly modified, such as by reflection from buildings or under a leaf canopy. Both hand-held meters and data loggers are used to read the signals from these sensors. Solar radiation data recorded at weather stations and published by national weather services may be taken from both photocell and thermal pyranometers.

267 …

Robert D. Brown and Terry J. Gillespie

„

Figure 15.1 A small LI-COR pyranometer can be used to measure incoming solar radiation in open areas (photo: T. Gillespie)

Estimating Estimates begin with the ‘solar constant’ which is the average power on a surface held perpendicular to the solar beam at the top of the atmosphere – about 1370 W/m2. This power is reduced during transmission through the atmosphere by the factor Am, where A is the transmissivity and m is a measure of the path length through the atmosphere. Thus, the power in the solar beam (Sb) on a horizontal surface at ground level is: Sb = 1370 Am · sin E, in W/m2

[2]

where E is the elevation angle of the sun above the horizontal. For sunny situations, which are often most relevant to designing for thermal comfort, A values average about 0.8 and range from 0.9 in a very clear atmosphere to 0.6 in a smoggy air mass (Oke 1988; Roumpakias et al. 2015). For elevation angles greater than ten degrees, the path length is computed from: m = P/(101.3 · sin E)

[3]

where P is the atmospheric pressure, in kilopascals. The elevation angle (in degrees) can be determined from diagrams (e.g. Brown and Gillespie 1995, Chap. 6) or from the following formula: sin E = sin L · sin D + cos L · cos D · cos 15 · (12 – t)

[4]

where L is the latitude of your location (in degrees), D is the latitude where the sun is directly overhead at noon on that day (the solar declination), and t is the time in hours and decimal hours

268 …

Thermally comfortable urban environments „

using a 24-hour clock. D is determined from selecting the appropriate value off a diagram that is readily available online or in Brown and Gillespie (1995, p.103). Some of the solar radiation that is scattered out of the direct beam by air molecules or particles as the beam travels through the atmosphere on a clear day arrives at the ground as ‘diffuse radiation’ (Sd) from the blue sky. This radiation can be estimated from (Campbell and Norman 1998): Sd = 0.3 · [1370 · (1 – Am ) · sin E], in W/m2

[5]

The total solar radiation arriving on a horizontal surface at the ground (SR) is simply the sum of the beam and diffuse components: SR = Sb + Sd, in W/m2

[6]

Under thick overcast cloud SR is calculated as 20% of equation [6]. Solar radiation measurements and estimates usually refer to a horizontal surface. For use in thermal comfort considerations these data must be modified because an upright human is like a vertical cylinder, not a horizontal plate. If the objective is to determine the influence of radiation on comfort in an existing landscape, the direct measurement of all radiation components using a radiation thermometer (see the ‘Total radiation’ section, below) is strongly recommended. However, if measured or estimated solar radiation on a horizontal surface is to be used in a comfort study, there is a procedure to convert these data to the radiation that would be received by a vertical cylinder (Kenny et al. 2008). The solar beam radiation on a vertical surface (Sbv) is: Sbv = Sb · cotan E, in W/m2

[7]

For thermal comfort estimations the data are usually required in watts, and therefore Sbv is multiplied by the cross-sectional area of the cylinder (length · width) because the beam radiation comes only from one specific direction. The solar diffuse radiation for a vertical cylinder (Sdv) under an open sky is: Sdv = 0.5 · Sd, in W/m2,

[8]

since any point on the cylinder views half the sky. For a vertical cylinder we must also add the solar radiation that reflects diffusely in all directions from the ground (Sdg). The ground occupies the other half of the cylinder’s view and has reflectivity Rg, therefore: Sdg = 0.5 Rg · SR, in W/m2

[9]

If buildings or trees occupy some of the sky hemisphere, then the view factor for the sky should be appropriately reduced below 0.5, and the ground view factor should be increased so the sum of the two view factors always equals 1. For both the above components of diffuse solar radiation, conversion to watts requires multiplication by the surface area of the cylinder (π · diameter · length, ignoring the small contribution from the cylinder top), since diffuse radiation comes from all directions. In summary, the total solar radiation received by a vertical cylinder is: SRv = Sbv · (cross-sectional area) + (Sdv + Sdg) · (surface area), in watts

[10]

269 …

Robert D. Brown and Terry J. Gillespie

„

For use in equation [1], this must be multiplied by the solar absorptivity of the person, which is the fraction of received solar energy that is not reflected and depends on the colour of the person’s clothing and skin.

Terrestrial radiation Measurement Terrestrial radiation is emitted by everything in the earth-atmosphere system, at wavelengths that are about ten times longer than the red light in the solar spectrum. Although we can’t see this radiation, we can sense it when we stand near a warm stove or a cold window, and it is a very important component in the energy balance of a person. Terrestrial radiation is measured by instruments called pyrgeometers (‘fire-earth-meters’). As with thermal pyranometers, the temperature of a black disk exposed to the radiation is monitored but the dome over this sensor is designed to transmit the longer terrestrial wavelengths and block the shorter-wavelength solar energy (e.g. EKO MS202 Pyrgeometer). These instruments are expensive and mainly used in research projects. Their voltage output signal is monitored by a data logger. If high quality measurements of both incoming and outgoing terrestrial and solar radiation are desired, instruments housing both upward and downward looking pyrgeometers and pynanometers (see Figure 15.2) are available (e.g. Kipp & Zonen model CNR1).

Estimating There are few sites where a pyrgeometer is being used continuously to monitor terrestrial radiation. Therefore is it usually necessary to estimate terrestrial radiation when exploring designs for urban

Figure 15.2 A Kipp & Zonen CNR1 includes both upward and downward looking pyranometers and pyrgeometers to simultaneously measure solar and terrestrial radiation (photo: T. Gillespie)

270 …

Thermally comfortable urban environments „

thermal comfort. The terrestrial radiation from a clear sky (TRs) can be estimated by (Monteith and Unsworth 1990): TRs = 213 + 5.5 · Ta, in W/m2

[11]

where Ta is the air temperature in 0C. For overcast thick cloud, the value computed from equation [11] should be multiplied by 1.25. A similar approach can be used to estimate the terrestrial radiation from objects on the ground (TRg) that are shaded from direct solar radiation: TRg = 320 + 5.2 · Ta, in W/m2

[12]

The temperature of objects receiving direct solar radiation should be measured (Tg) and the emitted radiation determined from (Oke 1988): TRg = E · 5.67 · 10-8 · (Tg + 273)4, in W/m2

[13]

where E is the emissivity, which is a factor that accounts for differing abilities of various surfaces to emit radiation. This factor is close to 0.95 for most surfaces. Terrestrial radiation from the sky or ground arrives from all directions in the appropriate hemisphere. Therefore, when representing a person as a vertical cylinder, the inputs of terrestrial radiation can be estimated in the same fashion as was described above for diffuse solar radiation. With an open sky, both ground and sky view factors are 0.5, so the total terrestrial radiation input on the vertical cylinder (TRv) is: TRv = (0.5 TRs + 0.5 TRg) · (surface area of cylinder), in watts

[14]

If ground objects occupy some of the sky hemisphere, the ground and sky view factors should be adjusted accordingly.

Total radiation As outlined above, the estimation of radiation received by a person in the outdoors from measurements made by horizontal sensors is complex. The use of a ‘cylindrical radiation thermometer’ (CRT) is a simple, accurate, and precise option. This device is a vertical cylinder that mimics the shape of a standing person. The cylinder we have used is about 11 cm tall with a diameter of 1 cm (or other similar height to diameter ratios could be used), and is painted to match the albedo of a person. The colour should match the skin colour of the population being modeled. When it reaches an equilibrium temperature, the absorbed radiation is equal to the sum of the long wave radiation emitted by the cylinder plus its convective heat loss to the wind. To obtain the long wave and convective losses, and therefore know the absorbed radiation, a temperature sensor is embedded in the cylinder, and the accompanying air temperature and wind speed are measured. Details on the construction of a CRT, and the calculation of the absorbed radiation, are given in Kenny et al. (2008). This device can be mounted on an instrument tripod (see Figure 15.3) or used as a portable hand-held device. The cylindrical shape of a CRT has an important advantage over the spherical shape of a globe radiation thermometer for work outdoors when some of the solar power is arriving as a direct beam. A spherical sensor presents the same cross-sectional area perpendicular to the direct beam, regardless of the sun’s location in the sky. However, the direct beam will be captured by a standing

271 …

Robert D. Brown and Terry J. Gillespie

„

Figure 15.3 A radiation shield should always accompany temperature and humidity sensors to eliminate radiation error (photo: T. Gillespie)

person or the sides of a vertical cylinder according to their cross-sectional area multiplied by the cosine of the sun’s elevation angle. This means the sphere increasingly overestimates the amount of beam radiation received by a person as the elevation angle of the sun increases. The cylinder correctly mimics the reduction of direct solar radiation on a person when the sun is in high in the sky, but the globe does not capture this effect.

Temperature and humidity Temperature and humidity sensors are often found together in instruments that can be purchased for field use. It is essential that these sensors be shielded from direct radiation (see Figure 15.3).

272 …

Thermally comfortable urban environments „

Radiation shielding is provided in some instruments, otherwise they must only be used in the shade. In addition, it is highly desirable to ventilate the sensors. Ventilation may be provided by a downstream fan (electrical or spring-driven), or by good exposure to the wind. Since temperature and humidity vary with distance away from the ground surface, it is best to make these measurements at a standard height so that observations at different locations or times are not confounded by differences due to exposure height of the sensors. For thermal comfort research a measurement level in the chest-to-head height range is recommended, similar to the standard 1.5 m sensor height used at weather stations.

Air temperature Measurement Mechanical sensors that display temperature on a dial usually have the pointer connected to a ‘bimetal’ coil. The coil is made from two metals with different thermal coefficients of expansion, so it twists with changing temperatures and moves the pointer. These sensors are usually inexpensive, but may not be accurate. The familiar liquid-in-glass thermometer is an obvious possible choice for measuring air temperature, but has fallen out of favour for field use because of its fragility. However, robust units are available that provide protection, radiation shielding and ventilation (using a spring-driven fan). Usually these instruments house two in-glass thermometers mounted side by side – one dry, and the other with a wet sock over the bulb for humidity measurement – and are called psychrometers (many options are available: search ‘Assmann Psychrometer’ on the internet) Electronic temperature sensors are now widely available in packages that are very suitable for field use (e.g. EXTECH RH 300; Davis Instruments; Omega). The sensor is usually a ‘thermistor’ whose electrical resistance varies with temperature, and a suitable circuit provides a digital readout. Typically, the sensor is not shielded or mechanically ventilated by the manufacturer, so care must be taken to provide shade and expose it to the wind. Louvered shields are available for this purpose (Figure 15.3). Specifications for these sensors often include a ‘time constant’. This is a measure of the time required for the sensor to reach about two-thirds of its response to a change in temperature. Since a sensor’s response slows as it approaches a new temperature, it is advisable to wait at least three time constants before taking a new reading when moving from one spot to another. All of the above temperature sensors require visual readout. If there is a requirement for measurements over extended time, or at several locations simultaneously, electronic sensors are available for use with data loggers (e.g. Campbell Scientific). These sensors are usually housed in a shield that provides radiation protection and allows ventilation by the wind.

Estimating Temperature data are often available from a nearby standard weather station operated by the national weather service. In some cases, the data may be available from an urban station, but stations are often located at sites like airports, to reduce or avoid urban influences. These data should be adjusted due to the UHI, based on the information provided in Table 15.1.

273 …

Robert D. Brown and Terry J. Gillespie

„

Table 15.1 Information for estimating temperature increase due to UHI

Surface temperature Measurement Sensors that detect the terrestrial radiation from an object and use these data to display the object’s surface temperature are often called ‘infra-red thermometers’ (IRTs) since this radiation lies in the infra-red spectral band. This is a very useful tool for quantifying the effect of variables such as colour, orientation to the sun, vegetation and construction materials on surface temperature. Surface temperature measurements can also be used in equation [14] for more accurate estimates of emitted radiation, rather than assuming that surfaces are at air temperature. The least expensive IRTs are designed to be used indoors for such things as detecting hot and cool spots on walls due to insulation differences. They may be successfully used outdoors but typically lose accuracy when surface temperatures drop below about 10 °C. More expensive versions are designed to give accurate results over a wider range of temperatures, if this is required (e.g. Omega). Operating range and accuracy can be found in the performance specifications from the manufacturer. These devices are simply pointed at the surface of interest and the temperature is read from a digital display, or models are available with a voltage output for use with a data logger. Users are cautioned that these devices will not give accurate results for shiny metal surfaces because these materials will reflect infra-red radiation from the sky or nearby surfaces into the IRT and spoil the reading. The ultimate tool for observing surface temperatures is the infra-red camera (e.g. FLIR). This device provides a digital picture of the objects in its view and displays their temperatures as different colours on the screen, along with a scale that relates colour to temperature (e.g. Omega, Apple). The images can be saved for later review and analysis.

274 …

Thermally comfortable urban environments „

Humidity Measurement The moisture content of the atmosphere is typically expressed as the relative humidity (RH) or dew point temperature (DPT). Relative humidity is the ratio of the current water vapour content of the air to the maximum possible content at the current temperature, expressed as a percentage. It therefore depends on both the water vapour content and the temperature. However, the temperature to which the air must be cooled in order for water vapour to begin condensing into liquid dew (the DPT) depends only on the air’s moisture content. The classic tool for measuring humidity is the psychrometer (see page 273). Drier air causes increased evaporative cooling of the wet bulb, so the temperature difference between the two thermometers can be used to determine the humidity. Tables to convert from wet and dry bulb temperatures to both RH and DPT are provided by the manufacturer of the instrument. Electronic sensors whose resistance or capacitance depends on RH are used in hand-held devices with a digital display, or provide an output suitable for use with a data logger (e.g. Campbell Scientific). These RH sensors are often paired with a temperature sensor at modest cost and are sufficiently accurate for landscape design work. DPT can be measured directly by devices that monitor the onset of dew on a cooled mirror, but these ‘dewpoint hygrometers’ are expensive and generally used only in meteorological research.

Estimating Many humidity sensors provide their output as RH. Humidity data from national weather station networks are usually expressed as DPT. You may find you have humidity data expressed in one of these formats but the application you wish to use requires the other format. Exact conversion from one of these measures to the other is quite mathematically complicated, but here are two simple conversion equations that are sufficiently accurate for many purposes (Lawrence 2005): DPT = T – ((100-RH) / 5), in ˚C

[15]

RH = 100 – 5 · (T – DPT), in %

[16]

With T in the range 0–30 °C, and RH in the range 40–100%, these simple conversions are accurate to within about 1 °C for DPT and 5% for RH. If more accurate conversions are needed, tables or calculators can be found on the internet (e.g. http://www.dpcalc.org/index.php). Measured urban/rural water vapour differences are often quite small, especially if there is some wind flow. There is a tendency for a slightly higher water vapour content in rural air during the daytime when evaporation from vegetation is strong, and lower rural vapour content at night when water loss from the vegetation shuts down (e.g. Oke 1988). If humidity data from a rural weather station must be used during the design stage of an urban project, it is reasonable to assume the dewpoint differences are small enough to be ignored. It is better to import rural dewpoint data rather than RH data, since DPT data depend only on vapour content while RH data may be confounded by their dependence on both vapour content and temperature.

275 …

Robert D. Brown and Terry J. Gillespie

„

Figure 15.4 An anemometer with three cups can be effectively used to measure wind speed in uniform flow (photo: T. Gillespie)

Wind Measurement Wind speed is measured by devices called ‘anemometers’ (see Figure 15.4). Wind direction data are provided by a ‘wind vane’ which pivots into the oncoming air flow while a sensor monitors the vane’s position relative to north. Anemometers intended to measure the average horizontal wind speed over several minutes, or longer time intervals, sense the rotation speed of a set of cups (e.g. Met-One) or a propeller (e.g. Gill; Campbell Scientific). A propeller-type anemometer must be mounted on a vane to continuously point it into the wind while the cup-type does not require pointing. Both these types of anemometers are available for mounting on a mast, with output

276 …

Thermally comfortable urban environments „

signals for use with data loggers. This arrangement has the advantage of providing unattended monitoring over an extended time period, but costs increase if several locations must be monitored. To survey the wind situation at a number of locations within a site, hand-held anemometers are available with digital readouts, often combined with temperature and humidity sensors (e.g. Kestrel Weather Meters; Omega; La Crosse Technology). Care must be taken to manually point a propellertype, hand-held unit into the wind. More complex instruments are available to monitor both the horizontal and vertical components of the wind, but this is generally not necessary for the purpose of landscape design since the horizontal wind is usually much larger than the vertical component.

Estimating Wind data from a standard weather station are taken at a height of 10 m above a smooth surface such as clipped grass. For use in a study that requires wind estimates near 1.5 m above the ground, the following formula describes the typical logarithmic decrease in wind speed from 10 m (W10) to 1.5 m (W1.5), provided the test site is also open to the wind and has vegetation or other roughness elements less than 1 m tall: W1.5 = W10 · {ln [(1.5-d)/Zs]} / 6.65, in m/s

[17]

Where: Zs = 0.13 · h represents the roughness length, in metres d = 0.67 · h represents the zero plane displacement, in metres h = vegetation height or height of roughness elements, in metres If the design site is open except for a windbreak, the 1.5 m wind from equation [17] can be used along with knowledge of the windbreak’s porosity to estimate the air flow downwind from the break (e.g. Brown and Gillespie 1995, Fig. 7.3). If an estimate of the wind under a proposed tree canopy is needed, find a similar existing canopy near an open area. Measure the wind first in the open, then under the canopy, and again in the open. Average the values from the two open site measurements and calculate the canopy-to-open wind speed ratio. This ratio, along with the open site equation [17], will allow the required canopy wind estimates for longer time periods to be made from nearby weather station data. If a design project involves an urban setting, it is not possible to simply estimate the wind speeds and directions near tall buildings and down urban canyons from rural standard weather station data. This is a very complex problem that is usually solved by building a scale model of the proposed project and measuring the flow patterns in a wind tunnel or a water flume.

Data collection Setting up instruments in the field The following guidelines are suggested when setting up instruments to make microclimatic measurements at an outdoor site. The standard height for temperature, humidity and radiation measurements at weather stations is 1.5 m, and this is also a satisfactory choice for gathering data to assess human comfort. Instrument manufacturers can supply a tripod or mast that supports their

277 …

Robert D. Brown and Terry J. Gillespie

„

Figure 15.5 A typical set-up of microclimate instruments includes a pyranometer, temperature, and humidity sensors inside a radiation shield, and an anemometer. In this case the anemometer is a propeller type and includes a wind-direction indicator. This station also includes two cylindrical radiation thermometers (CRT) on the left side. The anemometer is on the upwind side of the station and the pyranometer and CRTs are located so that they are not in any shadow. (photo: J. Vanos)

sensors with minimum interference to the local microclimate. Due to air transport by the wind, temperature, and humidity often vary little across a location. In this situation these sensors can be mounted on a fixed mast near the centre of the site (Figure 15.5), but temperature and humidity sensors must be placed in a louvered shield that blocks radiation and allows ventilation by the wind. Radiation and wind measurements usually require a mobile support because they can vary greatly at different spots across a site if elements that sometimes block sunshine or wind are present. It is important to ensure that none of the supports for instruments cast shadows on radiation sensors. Manufacturers usually specify a ‘time constant’ for a sensor, which is a measure of its speed of response to a changing variable. A good guideline is to sample the sensor output at least every two time constants in order to get true average values. It is often convenient for ease of data logging to mount all sensors on one mobile support. This can be manually transported to different spots in a location, or the site may be large enough that instrument transport by car or bicycle is desirable (e.g. Klemm et al. 2015). If measurements are made by stopping at various locations, it must be ensured that the transport vehicle is parked so it is downwind of the sensors. If measurements are made while the sensors are in motion, they must be mounted sufficiently ahead of the vehicle to sense the undisturbed microclimate. It is difficult to get proper wind data on a moving platform because the transport speed is added vectorially to the true wind. Vanos et al. (2012a) provide an example of a measurement package mounted on a bicycle for surveys along transects through parks. In this study the mobile station started and ended

278 …

Thermally comfortable urban environments „

in the same location to determine how much the prevailing conditions had changed during the test. The results provided a thermal transect of the temperature and humidity in the park as well as the immediately adjacent neighbourhoods.

METHODS FOR MEASUREMENT AND ESTIMATION OF HUMAN THERMAL COMFORT: HOW DO HUMANS PERCEIVE THEIR THERMAL COMFORT UNDER DIFFERENT MICROCLIMATIC CONDITIONS? Several different approaches have been used to identify the thermal comfort of a person in an outdoor environment. The most common and simplest is to ask people (ISO 1995). This has the advantage of being quick and easy, but as much as half of the variance between objective and subjective responses has been attributed to cultural effects (Knez and Thorsson 2006). Another approach is to observe where people choose to spend time in a landscape. The assumption would be that people will seek out thermally comfortable locations to spend their time. These subjective approaches focus on the question of ‘thermal comfort’. If an assumption is made that thermal neutrality leads to thermal comfort, then there are some objective approaches as well. Measurements can be taken of skin temperature or core temperature. There is evidence that a person’s core and/or skin temperature might provide an indicator of their thermal situation and this should be further investigated. The following sections describe various approaches to measuring thermal comfort and analyzing the results.

Measurement A simple questionnaire can be used to provide an estimate of people’s perceptions of their thermal comfort level (i.e. thermal sensation). A five-point scale is commonly used (ISO 1995) with the categories: I would prefer to be much warmer; I would prefer to be warmer; I would prefer no change; I would prefer to be cooler; I would prefer to be much cooler. A seven-point scale can also be used and would include the categories of I would prefer to be slightly warmer, and I would prefer to be slightly cooler in the above noted categories. Subjects can also be asked to rate their thermal sensation and the seven categories would be hot, warm, slightly warm, neutral, slightly cool, cool, and cold. Different studies use different scales so it can be difficult to compare results. There should be a standard approach (Johansson et al. 2014) and it remains an important research question to identify the most appropriate approach. Study participants are asked to remain in a predetermined location for a predetermined amount of time and at the end of that time are asked to select one of the possible responses listed above. The amount of time needed for a person to achieve thermal equilibrium varies by season (Johansson et al. 2014) and likely by other variables, leaving the required amount of time as an important research question. Simultaneously, microclimate measurements are taken at the participants’ site-specific location. When the data are analyzed the goal is to find a relationship between the perceived thermal sensation of the subjects and the energy budget calculations based on site-specific microclimatic conditions. A typical study would take subjects to a wide range of microclimatic conditions including a windy, shady spot through to a sunny spot with very low wind speed in order to expose participants to a range of thermal comfort levels.

279 …

Robert D. Brown and Terry J. Gillespie

„

A similar study that does not require study participants to consciously consider their perceived thermal sensation is an observational study. A landscape with a wide range of microclimatic conditions is selected. This might be a plaza that has an area of heavy shade that also receives strong winds, and an area that is in the full sun and has very little wind. A time is selected when the area is likely to have many visitors, say during lunch time, or on a weekend afternoon. The microclimates of the plaza are either measured using mobile stations, or estimated using procedures described earlier in this chapter. Then the locations of visitors are recorded on a map at some predetermined time interval. This depends somewhat on the size of the space and the number of visitors, and as in other aspects of thermal comfort research, the standardization of the time interval of measurement is an important research question that needs investigation. During observation it is important to note physical characteristics of individuals such as amount and colour of clothing and their activity level (e.g. whether they just completed a run or have been sitting quietly for several minutes).

Estimation More than 100 energy budget models for estimating the human thermal comfort level associated with various microclimatic conditions have been developed over the years (Johansson et al. 2014), most of which are for use in indoor situations. These models need to be validated through comparison of their output with the subjective responses of individuals (e.g. Kenny et al. 2009; Vanos et al. 2012b), and the output of the various models should be compared in order to determine the most accurate and precise model.

DISCUSSION Many things that people intuitively believe to be true about microclimate are proven wrong when carefully studied. For example, it’s common for someone to walk into the shade of a tree on a hot sunny day and proclaim that it’s about ten degrees cooler than in the sun. If air temperature was measured accurately and precisely it would reveal that, at a standard height of 1.5 m above the ground, the temperature is essentially the same in the shade and in the sun. A person feels much cooler in the shade because they are receiving less solar radiation. There are many examples of this kind of misunderstanding of microclimate that can lead to erroneous design decisions which can either have no positive effect or inadvertently make the microclimatic conditions worse. For example, wind can cool a person through convective heat transfer, but the amount of cooling depends on the temperature difference between a person’s skin and the air. If the air is much colder than a person’s skin, like it is in winter, the amount of convective cooling can be large. But during heat waves when the air temperature might be the same as a person’s skin temperature there is no convective cooling. And if the air is hotter than a person’s skin, wind can actually add heat to a person (Brown et al. 2015). If the humidity of the air is high there will be little opportunity for a person to lose heat through evaporation. So while it might seem logical to give a person a fan during a heat wave, the effect of having hot air pass over a person’s body during a heat wave with high humidity could actually exacerbate one’s vulnerability to heat stress and lead to hyperthermia. Most of the components of microclimate are invisible to the human eye and instruments are needed to measure them. It is essential that the correct instrument is used in the correct manner. For

280 …

Thermally comfortable urban environments „

example, human eyes only see about half of the radiation emitted by the sun. The other half is the completely invisible near infrared (NIR), which has the same amount of energy as the visible portion. A light meter only measures the visible portion, so must not be used to measure solar radiation. If air temperatures are measured without an adequate radiation shield and appropriate ventilation, the resultant values will not actually be measurements of air temperature, but some combination of air temperature and radiation absorbed by the instrument. The amount of radiation and wind experienced by a person in the landscape can be substantially modified over very short distances through landscape architectural design, while air temperature and air humidity cannot typically be modified very much over short distances (Brown and Gillespie 1995). Radiation and wind have different levels of effect on thermal comfort in different seasons. During cold winter conditions the convective cooling effect of the wind can be much larger than the input of radiation. The opposite is the case in summer, when convective cooling is minimized and radiant input is maximized (Brown and Gillespie 1995). This means that the priority for microclimate modification research should focus on radiation levels for summer situations, and wind in winter. Air temperature and humidity can be modified at the mesoscale but require large interventions. For example, a large well-shaded urban park can cool the air as it moves over well-watered, shaded surfaces. The relationship between how specific landscape forms promote cooling (e.g. large greenspaces and shade spaces) is inherently complex and requires further research to identify the most effective ways for landscape architectural design to maximize the cooling in urban areas. Recent advancements in the area of ‘research through designing’ (Lenzholzer et al. 2013) provide opportunities for an additional method of microclimate research involving experiments. Some experimental methods make invisible microclimate elements visible and can be valuable in communicating microclimate to the public. For example, a model of a landscape can be put in a wind tunnel and smoke can be used to illustrate wind flow. The effect of various possible interventions on wind flow can be illustrated. Other experiments could include full-scale temporary installations where the effect on the solar radiation, wind, and surface temperatures can be measured. It’s essential that the correct instruments be used in the proper manner and that the resulting data are analyzed using correct and appropriate methods. Landscape architectural researchers need to ensure that the data they collect are both accurate and precise. If they are unfamiliar with the use of technical information it can be valuable to collaborate with micrometeorological and/or microclimatological researchers. If this is done, the field of microclimate modification through urban design has the potential to substantially ameliorate the effects of GCC and UHI intensification on human health and well-being. Individual landscape architects can design urban environments that modify the climate (and future climates) to create safe, thermally comfortable places for people. This design must be supported with solid, scientific, defensible information that is learned through the methods outlined in this chapter.

CONCLUSIONS Microclimatic design research is inherently complex yet has important implications for human health and well-being. Considerably more research needs to be done to more deeply understand the complex interactions between landscape elements, meso- and microscale climatic factors, human thermal comfort, and heat stress across different climate zones. High quality instruments are available for this research, but must be used in a proper manner to assure accurate and precise

281 …

Robert D. Brown and Terry J. Gillespie

„

measurements. The field of outdoor human thermal comfort research is quite new and there is no standard method for measuring, estimating, and assessing information. Research should focus on two questions: understanding the complex relationship between urban landscape and urban microclimate, and understanding how microclimate affects human thermal comfort and vulnerability to heat stress in outdoor environments. The results of this research are needed to inform urban design so that human thermal comfort and microclimatic design can be considered, along with socio-economic, ecological and political considerations, when developing or modifying urban environments.

ACKNOWLEDGEMENTS Thanks to Dr. Jennifer Vanos and Dr. Natasha Kenny for their valuable comments and suggestions. This chapter was substantially improved by the comments and suggestions of the editors. SUGGESTED FURTHER READING Brown, R.D. (2010) Design with Microclimate: The Secret to Comfortable Outdoor Space, 2nd ed., Washington DC: Island Press. Brown, R.D. (2011) ‘Ameliorating the effects of climate change: Modifying microclimates through design’, Landscape and Urban Planning, 100(4), 372–374. Brown, R.D. and Gillespie T.J. (1995) Microclimatic Landscape Design: Creating Thermal Comfort and Energy Efficiency, New York: John Wiley. Campbell, G. and Norman, J. (1998) An Introduction to Environmental Biophysics, 2nd ed., New York: Springer. Erell, E., Pearlmutter, D. and Williamson, T. (2011) Urban Microclimate: Designing the Spaces between Buildings, London: Earthscan. Johansson, E., Thorsson, S., Emmanual R. and Kruger, E. (2014) ‘Instruments and methods in outdoor thermal comfort studies – the need for standardization’, Urban Climate, 10(2), 346–366. Klemm, W., Heusinkveld, B.G., Lenzholzer, S., Jacobs, M.H. and van Hove, B. (2015) ‘Psychological and physical impact of urban green spaces on outdoor thermal comfort during summertime in the Netherlands’, Building and Environment, 83(83), 120–128. Lenzholzer, S. (2015) Weather in the City: How Design Shapes the Urban Climate, Rotterdam: nai010 Publishers. Monteith J.L. and Unsworth, M.H. (2013) Principles of Environmental Physics: Plants, Animals and the Atmosphere, 4th ed., Oxford: Academic Press. Oke, T.R. (1988) Boundary Layer Climates, 2nd ed., London: Routledge.

REFERENCES Brown, R.D. (2011) ‘Ameliorating the effects of climate change: Modifying microclimates through design’, Landscape and Urban Planning, 100(4), 372–374. Brown, R.D. and Gillespie, T.J. (1986) ‘Estimating outdoor thermal comfort using a cylindrical radiation thermometer and an energy budget model’, International Journal of Biometeorology, 30(1), 43–52. Brown, R.D. and Gillespie, T.J. (1995) Microclimatic Landscape Design: Creating Thermal Comfort and Energy Efficiency, New York: John Wiley. Brown, R.D. and Corry, R.C. (2011) ‘Evidence-based landscape architecture: The maturing of a profession’, Landscape and Urban Planning, 100(4), 327–329. Brown, R.D., Kenny, N.A. and Corry, R.C. (2011) ‘Testing the microclimatic habitat design framework in abandoned sand and gravel extraction sites using the Karner blue butterfly’, Ecological Restoration, 29(1–2), 52–63. Brown, R.D., Vanos, J., Kenny, N. and Lenzholzer, S. (2015) ‘Designing urban parks that ameliorate the effects of climate change’, Landscape and Urban Planning, 138, 118–131. Campbell, G. and Norman, J. (1998) An Introduction to Environmental Biophysics, 2nd ed., New York: Springer.

282 …

Thermally comfortable urban environments „

Chan, P.K. and Packer, L. (2006) ‘Assessment of potential Karner blue butterfly (Lycaeides melissa samuelis) (Family: Lycanidae) reintroduction sites in Ontario, Canada’, Restoration Ecology, 14(4), 645–652. Chow, W.T.L., Pope, R.L., Martin, C.A. and Brazel, A.J. (2011) ‘Observing and modeling the nocturnal park cool island of an arid city: Horizontal and vertical impacts’, Theoretical and Applied Climatology, 103(1–2), 197–211. Chow, W.T.L., Brennan, D. and Brazel, A.J. (2012) ‘Urban heat island research in Phoenix, Arizona: Theoretical contributions and policy applications’, Bulletin of the American Meteorological Society, 93(4), 517–530. Declet-Barreto, J., Brazel, A.J., Martin, C.A., Chow, W.T.L. and Harlan, S.L. (2013) ‘Creating the park cool island in an inner-city neighbourhood: Heat mitigation strategies for Phoenix, AZ’, Urban Ecology, 16(3), 617– 635. Grimmond, C.S.B., Roth, M., Oke, T.R., Au, Y.C., Best, M., Betts, R. and Voogt, J. (2010) ‘Climate and more sustainable cities: Climate information for improved planning and management of cities (producers/ capabilities perspective)’, Procedia Environmental Sciences, 1, 247–274. Harlan, S.L. and Ruddell, D.M. (2011) ‘Climate change and health in cities: Impacts of heat and air pollution and potential co-benefits from mitigation and adaptation’, Current Opinion in Environmental Sustainability, 3(3), 126–134. Harlan, S.L., Brazel, A.J., Prashad, L., Stefanov, W.L. and Larsen, L. (2006) ‘Neighborhood microclimates and vulnerability to heat stress’, Social Science and Medicine, 63(11), 2847–2863. Holick, M.F. (2004) ‘Sunlight and vitamin D for bone health and prevention of autoimmune diseases, cancers, and cardiovascular disease’, American Journal of Clinical Nutrition, 80(6)(Suppl.), 1678S–1688S. Höppe, P. (1999) ‘The physiological equivalent temperature: A universal index for the biometeorological assessment of the thermal environment’, International Journal of Biometeorology, 43(2), 71–75. Howard, L. (1833) The Climate of London: Deduced from Meteorological Observations Made in the Metropolis and Various Places Around It, London: Joseph Rickerby. ISO (International Organization for Standardization) (1995) ISO 10551:1995, Ergonomics of the thermal environment – assessment of the influence of the thermal environment using subjective judgement scales, Geneva: International Organization for Standardization. ISO (International Organization for Standardization) (2007) ISO 9920, Ergonomics of the thermal environment – estimation of thermal insulation and water vapour resistance of a clothing ensemble, Geneva: International Organization for Standardization. Johansson, E., Thorsson, S., Emmanual, R. and Kruger, E. (2014) ‘Instruments and methods in outdoor thermal comfort studies: The need for standardization’, Urban Climate, 10(2), 346–366. Kenny, N.A., Warland, J.S., Brown, R.D. and Gillespie T.J. (2008) ‘Estimating radiation absorbed by a human’, International Journal of Biometeorology, 52(6), 491–503. Kenny, N.A., Warland, J.S., Brown, R.D. and Gillespie T.J. (2009) ‘Part A: Assessing the performance of the COMFA outdoor thermal comfort model on subjects performing physical activity’, International Journal of Biometeorology, 53(5), 415–428. Klemm, W., Heusinkveld, B.G., Lenzholzer, S., Jacobs, M.H. and van Hove, B. (2015) ‘Psychological and physical impact of urban green spaces on outdoor thermal comfort during summertime in tNetherlands’, Building and Environment, 83(83), 120–128. Knez, I. and Thorsson, S. (2006) ‘Influences of culture and environmental attitude on thermal, emotional and perceptual evaluations of a public square’, International Journal of Biometeorology, 50(5), 258–268. Lawrence, M.G. (2005) ‘The relationship between relative humidity and the dewpoint temperature in moist air: A simple conversion and applications’, Bulletin of the American Meteorological Society, 86(2), 225–233. Lenzholzer, S., Duchhart, I. and Koh, J. (2013) ‘“Research through designing” in landscape architecture’, Landscape and Urban Planning, 113, 120–127. McCarthy, M.P., Best, M.J. and Betts, R.A. (2010) ‘Climate change in cities due to global warming and urban effects’, Geophysical Research Letters, 37(9), DOI: 10.1029/2010GL042845. McMichael, A.J. (2013) ‘Globalization, climate change, and human health’, New England Journal of Medicine, 368(14), 1335–1343. McPherson, E.G., Simpson, J.R. and Livingston, M. (1989) ‘Effects of three landscape treatments on residential energy and water use in Tucson, Arizona’, Energy and Buildings, 13(2), 127–138.

283 …

Robert D. Brown and Terry J. Gillespie

„

Millward, A.A., Torchia, M., Laursen, A.E. and Rothman, L.D. (2014) ‘Vegetation placement for summer built surface temperature moderation in an urban microclimate’, Environmental Management, 53(6), 1043– 1057. Mishra, V., Ganguly, A.R., Nijssen, B. and Lettenmaier, D.P. (2015) ‘Changes in observed climate extremes in global urban areas’, Environmental Research Letters, 10(2), 024005. Monteith J.L. and Unsworth, M.H. (1990) Principles of Environmental Physics, 2nd ed., New York: Routledge. Oke, T.R. (1982) ‘The energetic basis of the urban heat island’, Quarterly Journal of the Royal Meteorological Society, 108(455), 1–24. Oke. T.R. (1988) Boundary Layer Climates, 2nd ed., London: Routledge. Ostfeld, R.S. and Brunner, J.L. (2015) ‘Climate change and Ixodes tick-borne diseases of humans’, Philosophical Transactions of the Royal Society B: Biological Sciences, 370(1665), DOI: 10.1098/rstb.2014.0051. Robine, J.-M., Cheung, S.L.K., Le Roy, S., Van Oyen, H., Griffiths, C., Michel, J.-P. and Herrmann, F.R. (2008) ‘Death toll exceeded 70,000 in Europe during the summer of 2003. Plus de 70 000 décès en Europe au cours de l’été 2003’, Comptes Rendus Biologies, 331(2), 171–178. Roumpakias, E., Zogou, O. and Stamatelos, A. (2015) ‘Correlation of actual efficiency of photovoltaic panels with air mass’, Renewable Energy, 74, 70–77. Salamanca, F., Georgescu, M., Mahalov, A., Moustaoui, M. and Wang, M. (2014) ‘Anthropogenic heating of the urban environment due to air conditioning’, Journal of Geophysical Research: Atmospheres, 119(10), 5949–5965. Seneviratne, S.I., et al. (2012) ‘Changes in climate extremes and their impacts on the natural physical environment’, in Field, C.B., Barros, et al., eds. ‘Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation’, A Special Report of Working Groups I and II of the Intergovernmental Panel on Climate Change (IPCC), Cambridge: Cambridge University Press, 109–230. Shashua-Bar, L., Pearlmutter, D. and Erell, E. (2011) ‘The influence of trees and grass on outdoor thermal comfort in a hot-arid environment’, International Journal of Climatology, 31(10), 1498–1506. Spronken-Smith, R.A. and Oke, T.R. (1998) ‘The thermal regime of urban parks in two cities with different summer climates’, International Journal of Remote Sensing, 19(11), 2085–2104. Thorsson, S., Rocklöv, J., Konarska, J., Lindberg, F., Holmer, B., Dousset, B. and Rayner, D. (2014) ‘Mean radiant temperature: A predictor of heat related mortality’, Urban Climate, 10, 332–345. UTCI (2009) ‘UTCI – Universal Thermal Climate Index’, available: http://www.utci.org (accessed July 7, 2016). Vanos, J.K. (2015) ‘Children’s health and vulnerability in outdoor microclimates: A comprehensive review’, Environment International, 76, 1–15. Vanos, J.K., Warland, J.S., Gillespie, T.J. and Kenny, N.A. (2010) ‘Review of the physiology of human thermal comfort while exercising in urban landscapes and implications for bioclimatic design’, International Journal of Biometeorology, 54(4), 319–334. Vanos, J.K., Warland, J.S., Gillespie, T.J., Slater, G.A., Brown, R.D. and Kenny, N.A. (2012a) ‘Human energy budget modeling in urban parks in Toronto and applications to emergency heat stress preparedness’, Journal of Applied Meteorology and Climatology, 51(9), 1639–1653. Vanos, J.K., Warland, J.S., Gillespie, T.J. and Kenny, N.A. (2012b) ‘Improved predictive ability of climate–human– behaviour interactions with modifications to the COMFA outdoor energy budget model’, International Journal of Biometeorology, 56(6), 1065–1074. Ward, S.E. and Brown, R.D. (2004) ‘A framework for incorporating the prevention of Lyme disease transmission into the landscape planning and design process’, Landscape and Urban Planning, 66(2), 91–106. Watkins, R., Palmer, J. and Kolokotroni, M. (2007) ‘Increased temperature and intensification of the urban heat island: Implications for human comfort and urban design’, Built Environment, 33(1), 85–96. Wolch, J., Jerrett, M., Reynolds, K., McConnell, R., Chang, R., Dahmann, N., Brady, K., Gilliland, F., Su, J.G. and Berhane, K. (2011) ‘Childhood obesity and proximity to urban parks and recreational resources: A longitudinal cohort study’, Health & Place, 17(1), 207–214. World Meteorological Organization (2015) ‘Recent trends in extreme events are consistent with the expected impacts of climate change’, available: http://www.wmo.int/media/content/recent-trends-extremeevents-are-consistent-expected-impacts-climate-change (accessed March 18, 2015).

284 …

Chapter 16: The urban water challenge Antje Backhaus, Ole Fryd and Torben Dam

INTRODUCTION This chapter explores how landscape architecture research contributes to generating knowledge that helps to solve urban water challenges. The global urban population is expected to increase by 60 per cent from 2014 to 2050 (UN DESA 2014). The global urban area is expected to double or triple by 2030, in comparison to the year 2000 (Seto et al. 2011). These are changes that define hydrological challenges such as increasing water requirement and water shortages, insufficient infrastructure for wastewater and urban drainage, and greater flood risk. Local water supplies might become scarce, or the water quality might be poor. In some parts of the world cities are in decline and the dwindling number of users makes the existing urban water system expensive and/or infeasible to operate. Urban water challenges are exacerbated by climate change. In general, the wet regions on this planet are expected to get wetter while the dry regions will get drier (IPCC 2014). Many of the wet and wetter regions comprise relatively mature and fully built up cities that need to be retrofitted to manage future precipitation rates. Many of the dry and drier regions are subject to massive population growth. The new cities in these regions need to be able to provide water for their residents and preferably also reduce the ecological impact of urbanisation. Taken together, the growth, decline and transformation of cities across the world and the current and future climate changes call for innovative research and new approaches to water management. For more than a century engineers have provided the expertise for implementing water and sanitation infrastructure in cities. The conventional methods for sizing, costing and constructing infrastructure such as pipes, channels and dikes are now challenged by the uncertain magnitudes of climate change and also by the pace of urban growth and decline. As stand-alone solutions, engineered ‘grey’ infrastructures are criticised for being mono-functional, cost-intensive, ‘fail-safe’ rather than ‘safe-to-fail’ (Ahern 2011), and unable to provide synergy with other aspects of urban development such as promoting health and biodiversity (Fryd et al. 2010; Ahern 2011; Liao 2012). Integrated approaches to urban water management are expected to help to reduce ecological and hydrological impacts of urbanisation and also to help to increase the quantity and quality of urban green space. Integrated urban water management calls for new ways of collaborating and an ability to overcome sectorial and disciplinary divides. What is needed, then, are research methods that can help to increase knowledge about integrated urban water management. As a discipline, landscape architecture provides important expertise on the role of water in the urban landscape. In particular, landscape architecture is

285 …

Antje Backhaus, Ole Fryd and Torben Dam

„

expected to contribute evidence-based design expertise that can complement contributions other disciplines are making, and all together collaborate in their striving towards integrated responses to the global urban water challenge. One such integrated response is what we call ‘landscape based stormwater management’ (LSM), an approach that utilises natural processes and mimics the natural water cycle by promoting infiltration and evapotranspiration in the city while reducing the amount of surface runoff. LSM is also referred to as water sensitive urban design (WSUD) in Australia, low impact development (LID) in North America and sustainable urban drainage systems (SUDS) in the United Kingdom (Fletcher et al. 2015). The term LSM is used here to emphasise the potential of the role that urban landscapes play in relation to water management. LSM also aims to provide synergy between the urban water challenge and other aspects of urban development such as promoting urban greening, biodiversity and human health. In this chapter we discuss research approaches that help to generate knowledge to inform landscape architecture contributions to LSM. The main questions that guide our research are (a) what are relevant research approaches in landscape architecture that can help to address global urban water challenges, and (b) which methods and methodological considerations should be taken into account? Following this introduction, the chapter first provides a conceptual framework upon which our research is based. Here we introduce the specific focus and theoretical backdrop to our work on LSM. Second, we present and discuss three studies pertinent to this topic. Here we describe our approaches to study design, data collection, sampling and data analysis. In the third section, we discuss different aspects of our research approach and methods, such as the role of the researcher, and also aspects of scale and study design. In the final section, and based on a comparative discussion of our three research examples, the chapter outlines directions and perspectives for future research.

CONCEPTUAL FRAMEWORK The research methods illustrated in this chapter have been part of three collaborative studies undertaken between the years 2008 and 2010. The theoretical basis for the research draws on (a) understanding urban water challenges as ‘wicked problems’ and (b) on processual approaches to understanding relationships between research and design. It is anticipated that when cities try to respond to the urban water challenge they are faced with very complex tasks. They cannot rely on any linear relationships to exist between, on the one hand, climate change and rapid urbanisation, and consolidated approaches and techniques to improving urban water infrastructure on the other hand. First, any urban water solution faces competition for space above and below ground by residents and investors regarding transportation, ecological conservation and many other interests. Second, climate change and urban transformation processes put the urban water system under pressure, both alone and in tandem. Together, competition for space, climate change and urban transformation call for an innovative response. Addressing urban water problems is an example of being faced with a very complex challenge, one that has, in planning theory, also been called a ‘wicked problem’ (Rittel and Webber 1973). When, as was the case in the past, planning and research have adopted linear procedures that involve straightforward forms of data collection and analysis and problem solving, planners and researchers have tended to comprehend problems as ‘tame’ (Conklin 2005). Such seemingly simple understandings and methods are not suitable for problems like those posed by the urban water challenge. Wicked problems are characterised by a high level of complexity and uncertainty, and also by considerable value divergences amongst stakeholders (Head 2008).

286 …

The urban water challenge „

Figure 16.1 Sketch by Sven Ingvar Anderson (1994) showing the spiral of the design process with the amount of facts involved (stripes), the degree of determination in the design sketching (dots) and the design result (flower)

Wicked problems have no stopping rule, and the process of solving a wicked problem is identical to understanding the problem’s nature (Rittel and Webber 1973; Conklin 2005). Literature on wicked problems emphasises social learning and transdisciplinary research as a means to work with wicked problems (Roberts 2000; Head 2008). Actors from different sectors and disciplines must be capable of interacting with each other and also of learning to resolve the problem together. In doing so they should pursue an interactive process. In such a process, preliminary solutions are developed and reviewed first, and then again and again, each time with the aim to define new tasks and problems, which are resolved in continuous learning loops (Kline and Rosenberg 1986; Uhl-Bien et al. 2007; Fryd et al. 2010). Such a process is typical for a landscape architect working in practice and using the design process for problem solving. Understanding such a design process can be aided by following the lines seen in a sketch drawn by Anderson (1994) (Figure 16.1). This sketch is drawn in the shape of a circular cone, a conical helical line that spirals upwards while narrowing towards its apex. The sketch is meant to represent the phases of the design process and to illustrate the amount of facts involved (stripes) and the degree of determination in the design sketching (dots) that gradually informs the synthesis that is materialised in the final design. These phases confront, contextualise and integrate different kinds of technical, spatial, experimental or social knowledge in a spiralling process of consideration, action and reconsideration (e.g. Schön 1991; Stappers 2007; von Seggern et al. 2008). A designer such as a skilled landscape architect can focus on and develop solutions to a problem even if not all of the desirable information is available. In addition, one may realise that there are

287 …

Antje Backhaus, Ole Fryd and Torben Dam

„

always various solutions that would work in a given context (Steenbergen 2008). Hence, designing and the process of designing are found to be suitable approaches to conducting research that involves finding solutions to wicked problems. Besides conducting research using design, and trying to understand the processes behind design, there is an emerging field of research which uses the design process itself to find answers to research questions, thus doing ‘research through designing’ (see Lenzholzer et al. – Chapter 4). Research through designing elevates the design process from being the central working tool of a practitioner to becoming a research approach, one that employs several methods of continuous analysis and reflection, and one that makes the process of problem solving comprehensible and, to some degree, repeatable for others. In all of the three studies presented in this chapter, researchers are assuming the role of a ‘reflective practitioner’ (Schön 1991). As reflection is the key element within design-based research, the process of observation is of great importance. In the studies presented, the term ‘observation’ is understood in a rather broad sense, as described by John Zeisel (2006, p.41): ‘It means looking at phenomena connected to a problem by whatever means necessary: looking with one’s eyes, asking questions, using mechanical measurement devices, and so on.’ Some of the studies presented focus on understanding the process of designing LSM systems. They involve the landscape architect as a subject, and one who reflects on their practice. Other studies focus on the outcome of the process, that is the resulting landscape architecture project. Here the researcher is taking on the role of a more distant observer, analysing the specific design challenge from the outside. Both processes are forms of case study research (see Swaffield – Chapter 7). To put our studies into a broader context, we adopt the notions of sustainability transitions and ‘niches’. In transition literature, niches are defined as ‘protected spaces that allow nurturing and experimentation with the co-evolution of technology, user practices, and regulatory structures’ (Schot and Geels 2008, p.538). These niches can shortcut a transition process that may otherwise take decades. The specific technological transition discussed in this chapter is the move from conventional underground and linear water infrastructures managed more or less exclusively by engineers to the more holistic and circular LSM approach that aims to utilise the urban landscape and to provide synergy between multiple agendas and disciplines. ‘Niche experiments’ can serve as a learning room or a so called ‘transition arena’ where niche actors (e.g. innovators) interact with the actors of the existing regime (e.g. government officers) and third party stakeholders (e.g. community groups) with a view to influencing the ‘cognitive frames’ that may facilitate or impede novel solutions (Schot and Geels 2008). Several climate change researchers recommend approaching climate change adaptation as a learning process and a social and political process (e.g. Smit et al. 2000; Fünfgeld and McEvoy 2011). Transition theory indicates that strategic experimentation of design alternatives in protected niches is an approach relevant to informing decisions and facilitating the transfer of knowledge into practice. See Mguni et al. (2014) for a more comprehensive analysis of transition management in relation to LSM. Based on this theoretical backdrop, we found some aspects especially important in defining a study’s research approach, and for reflecting on the conceptual framework. We anticipate, that to work with urban water, the researcher needs to address and reflect on the spatial scale (e.g. centimetres to kilometres, or raindrop to catchment), the temporal scale (e.g. minutes to months, or short-term to long-term observations), the research objective and the technical research design, therein the role of the researcher, the object of study and the scale of the study. Spatial scale, temporal scale, the role of the researcher, research objective, analytical focus and the case study

288 …

The urban water challenge „

Figure 16.2 Conceptual framework, outlining variations in research approaches relevant to research on the urban water challenge. In each individual research project, the researchers decide on the parameters to be studied along a continuum between two extremes

design can be studied along a continuum between two extremes, as depicted by the double arrowheads in Figure 16.2. The continua include small scale and large scale (subject to a clearer definition of these scales), short term and long term (subject to definition), process oriented or outcome oriented study, and whether the study has a focus on living human subjects or non-living objects. All studies can be mapped along these continua. The aim is to help researchers to be mindful about the methodological choices and delineations made in the technical research design, especially when dealing with the urban water challenge.

THREE STUDIES ON THE URBAN WATER CHALLENGE In this section we introduce, compare and discuss three landscape architecture studies that address water related research questions. Here we describe our research design and methods used, including

289 …

Antje Backhaus, Ole Fryd and Torben Dam

„

aspects of data collection and data analysis, as well as specific study results. All of the three studies relate to our own previous research on LSM in Northern Europe. LSM refers to the practise of managing the urban stormwater runoff in the urban landscape through retention, infiltration and evapotranspiration, thus supporting a natural water cycle with a locally improved microclimate, groundwater recharge, water availability for vegetation and so on. The LSM approaches to water management comprise and use a range of different measures such as green roofs, porous pavement, bio-swales and ponds (see e.g. Woods-Ballard et al. 2007). This is in contrast to the conventional approach to urban drainage, which focuses on collecting, conveying and discharging stormwater away from the city as quickly as possible and with little concern for downstream environmental impacts. LSM captures the interface between different technologies and disciplines and highlights the conflicts and uncertainties related to ‘grey’ and ‘green’ responses to urban water management (Fryd et al. 2010). It is therefore a relevant study area to help understand the role that landscape architecture research plays in relation to the urban water challenge at large. The selected studies are: (1) a review of twenty LSM projects implemented in Northern Europe (Backhaus and Fryd 2013), (2) a first-loop catchment strategy for the retrofitting of LSM solutions in an existing built up area of Copenhagen (Backhaus and Fryd 2012) and (3) a design experiment that includes professional landscape architects (Backhaus et al. 2012). A summary of the three studies is included in Table 16.1. The three studies are analysed using the conceptual framework outlined in Figure 16.2. Below, the three studies are presented one by one. The three studies are compared and discussed in subsequent sections of this chapter.

Study 1: review of twenty LSM design projects in Northern Europe This review study forms the start of a longer research process on LSM design and was conducted to gain an overview of the current situation in built LSM projects. The purpose of the study was to complement the existing LSM literature which in 2008 was restricted to designers’ own project descriptions (e.g. Dreiseitl et al. 2001), technical analyses (e.g. Sieker et al. 2006) or specific geographic regions (e.g. Echols and Pennypacker 2008). The study was thus conducted with the aim of finding answers to the basic research question of this review: ‘What is the current state of the art in urban landscape stormwater management?’ The research question mentioned above was further refined during the running of the study and it evolved through a process of abductive reasoning (Lipton 1991), starting from the first general interest in learning about the ‘state of the art’ of stormwater management, leading towards a more specific investigation of the aesthetic performance of LSM systems. Thus, the study applied a normative multiple case study approach combining ‘empirical observation with normative assessment’ (Thacher 2006, p.1632). During a first intensive phase of data collection, LSM cases from Europe and North America were visited on site or reviewed in the literature. The decision to focus on those two regions was taken due to practical reasons, such as data and site accessibility for the authors. The collected data was compiled in a Microsoft Access database. From the database twenty case studies from Denmark, Germany, the Netherlands, Norway and Sweden were selected for detailed comparison. The criteria for case selection were the maximum variation in age, size, character, design premise and national context (Flyvbjerg 2006). In addition, the selection process was influenced by the authors

290 …

The urban water challenge „

Table 16.1 Main characteristics of the three studies presented in this chapter. Each project reflects the role of landscape architecture research in the context of broader transdisciplinary research projects addressing the urban water challenge

291 …

Antje Backhaus, Ole Fryd and Torben Dam

„

pre-existing site knowledge, recommendations from peers, and the body of literature describing individual projects available from libraries, the internet and in the form of grey publications from project stakeholders, such as landscape architects or municipal employees. Data was collected for each of the twenty case studies though a range of methods. Personal observations were made at site visits by at least one of the authors. During the site visits personal notes and photographs were taken for documentation. All photographs were systematically organised in a photo database for each site. Available literature such as reports, info sheets, articles and book chapters, as well as maps and design drawings were collected and reviewed in order to get the best possible overview of the design and construction of each project. These documents were used by the authors to develop short project descriptions. Semi-structured interviews were conducted with designers and other project stakeholders to gain insight into the projects’ genesis, design intentions and problems that have occurred during construction and use. Specifically, personal communication was used to shed light on questions that had arisen from the literature review and the site visits. The data collected was predominantly used to support the authors’ project understanding. The twenty selected cases were compared in a retrospective analysis. The analysis was structured around a series of parameters that were found relevant in previously published reviews (e.g. Beneke 2003; Mitchell 2006; Martin et al. 2007; Echols and Pennypacker 2008) and in other complementary sources (e.g. Dreiseitl et al. 2001; Wöbse 2002; Strom et al. 2009). Through a process of abductive reasoning and discussions with peers, the authors selected five key parameters to analyse and compare the twenty projects (see Table 16.2). Through a retrospective analysis of the twenty cases using the selected five key parameters, the study identified specific characteristics of stormwater management projects that might optimise or minimise a project’s aesthetic. Inspired by the method of a ‘close reading of the aesthetic language’ (Hauxner Table 16.2 Five key parameters selected for project analysis and comparison

292 …

The urban water challenge „

2003), the examination of ’aesthetic performance’ was done by the authors, who have advanced degrees in landscape architecture and urban design. Assessments were based on the authors’ empirical observations and normative preference ratings of the visual appearance of the selected projects. In relation to terrain changes, it was found that visible stormwater management can enhance the experience of local terrain and larger terrain changes, and thus should be implemented with sensitive respect for the users’ landscape experience. Area restrictions can force designers to prepare for massive terrain changes; these can lead to an ‘unsettled’ design that is not feasibly adapted to the site. In water dynamics and dimensioning, it was found that LSM can accentuate otherwise often unrecognised precipitation dynamics in urban areas, for example by keeping and staging stormwater runoff as a physical design element that would otherwise disappear fast into the sewers. Permanent open water surfaces and running water can be difficult to realise, if not combined with existing surface water bodies or through properly sized ponds. The appropriate sizing of single elements seems to be a common design challenge. As a specific example, LSM projects must be able to accommodate large water volumes on rare occasions, but at the same time the sites must be inviting and stimulate other activities during dry periods. Steep slopes and a strong basin shape may impede the multifunctional use of the space. A key finding of stormwater accentuation was that visible stormwater management accentuation ‘at any cost’ often results in unsettled designs. Furthermore, stormwater management is often too weak to be a sole connecting idea for a good and aesthetically pleasing design, as precipitation is irregular and the water element is mostly absent, despite being a main design feature. Simple designs using few key stormwater management elements appear to work better than systems with multiple water features. The analysis of construction and maintenance issues revealed that LSM systems may require intensive maintenance; this must be carefully considered in the design process. In addition, mistakes may occur during construction due to lack of experience that can lead to system failure. This calls for the sharing of knowledge amongst professionals and across sites. The study of site history and context confirmed that, for LSM, generalised textbook designs cannot meaningfully be transferred to new sites in a generic way.

Study 2: catchment-wide LSM strategy in Copenhagen Little knowledge exists on how to adapt existing urban drainage systems by use of LSM at the scale of entire sewer-sheds. The purpose of this study was to compare and assess two different water management options for reducing combined sewer overflows from a 15km2 fully built up sewer catchment in western Copenhagen. One option was to enlarge the existing underground sewer system through conventional engineering measures. This option would be a ‘tame’ problem in social and technical terms, yet the solution might be non-adaptive to the dynamics of climate change and urban transformation. Another option was to adopt an LSM approach and implement it in the existing urban environment. This option would be linked with more social and technical uncertainty, but also shed light on the opportunities and barriers for LSM retrofits. The two specific research questions were: (1) ‘Which “insights” and “setbacks” emerge in the process of designing a large scale LSM retrofit strategy?’ and (2) ‘How can these insights inform future design processes for similar LSM retrofitting strategies at the sewer-shed scale?’ The main approach taken in this study was to develop a draft master plan for the specific catchment area in Copenhagen and to analyse the design process in a retrospective manner. The

293 …

Antje Backhaus, Ole Fryd and Torben Dam

„

development of the master plan was executed by a group of eight PhD students spanning landscape architecture, economics, engineering, environmental chemistry, geology and urban planning, in collaboration with a group of relevant professional peers from the City of Copenhagen and from the water utility company HOFOR, as well as private engineering consultancies, and also with the PhD students’ academic supervisors. All participants were partners in the research project ‘Black, Blue & Green – Integrated infrastructure planning as key to sustainable urban water systems’, which was funded by the Danish Council for Strategic Research and followed a triple helix model of university– government–industry relations (see Etzkowitz and Leydesdorff 2000). The master plan was developed between October 2008 and December 2009, and the process was structured around two workshops and five joint meetings between academics and partners from the public and private sector. In addition to officially scheduled workshops and meetings, numerous meetings, interactions and data exchanges occurred between the individual PhD students as the catchment plan was shaped. The master plan was presented to key institutional stakeholders and professional peers at a national symposium in December 2009. The developed master plan and the process of developing the master plan in Copenhagen form an embedded single case study. The stages of the master planning were monitored and documented, and the process of designing was subjected to an analysis, where ‘insights’ and ‘setbacks’ were revealed and discussed to determine their importance in a successful design. The data used for the analysis comprised all graphic documents shared among the full group of PhD researchers. This included maps, sketches, photos and modelling results that were either presented at the joint meetings and workshops or which were circulated among the full group of PhD students. The data collection was done by two of the PhD students, who were part of the bigger team and had a background in landscape architecture and urban planning. The first data collected were background data available from the internet, municipal databases and policies and reports. Such data were, for example, aerial photos, topographic maps, historical maps or maps on soil conditions and land use types. During the research and design process new data was generated, for example in the form of groundwater maps (generated by a peer researcher in the team), data from site visits and observations (photos, notes and sketches), as well as maps produced by the team through analysis and layering of information and design ideas. For more detailed information on data collection and data analysis, please see Backhaus and Fryd (2012) and Fryd et al. (2013). A retrospective analysis of the design process was conducted to understand and communicate the lessons learned from the case study. Illustrations produced and shared by the researchers during the development of the master plan were collected and compiled in a chronological process diagram (see Figure 16.3) reflecting their spatial design scale and the date of circulation among the team of PhD students. By comparing the final master plan with each design step outlined in Figure 16.3, key ‘insights’ and ‘setbacks’ in the design process were identified. The insights were identified as moves that led more or less directly to a better problem understanding, and thus formed part of the final solution. The setbacks were referred to as phases in the process, which impeded the development of the solution. By mapping the design process it was identified that most initial studies were done at the catchment level, whereas the more detailed analyses and designs at the lot or neighbourhood level occurred towards the end of the design process. The intermediate district scale emerged as an important transition level, where catchment level strategies could be specified to frame site level interventions, and where detailed plot level studies could provide feedback on potential limitations of the overall

294 …

The urban water challenge „

Figure 16.3 Diagram of the design process. The timeline outlines illustrations shared by the research team, the type of illustration, their spatial scale (along the y-axis) and the time when they were produced in the design process (depicted along the x-axis). Important steps in the process are identified as ‘insights’ and ‘setbacks’. (source: Adapted from the full diagram published in Backhaus and Fryd 2012)

catchment strategy. Whilst most initial analyses and decisions were based on plan drawings (i.e. twodimensional top view plan drawings), the more fine-grained detailing emphasised the importance of section drawings as a tool for understanding and working with the urban water system. The two most important findings emerging from this study were the need to develop context specific solutions, and the utilisation of the underlying hydrological conditions as a generator of sustainable urban form. From a transdisciplinary research perspective, the development of a master plan that reflects a multi-level catchment strategy provided a key breakthrough for the team as a whole. During the initial stages, multiple analyses were done to find a problem definition and research question that everybody could relate to; but a shared understanding of the problem and the potential solutions were lacking. The catchment strategy suggested a range of sub-strategies and specific subquestions to be addressed. Specifically the catchment strategy outlined specific areas where LSM should be linked to the consolidation of urban green areas, areas where groundwater recharge should be in focus and areas where LSM should be avoided (Backhaus and Fryd 2012; Fryd et al. 2013). With the multi-level approach, each discipline, institution and individual could relate to at least one of the tasks at hand. The catchment strategy allowed researchers to work in parallel on different aspects and on different scales. The landscape design process became a medium for facilitating ownership, collaboration and knowledge exchange amongst all researchers involved in the project. The research

295 …

Antje Backhaus, Ole Fryd and Torben Dam

„

process showed that ‘scientific research’ (done by geologists, environmental chemists and engineers) and ‘design research’ (done by landscape architects) must be combined and is interdependent in the successful development of an integrated urban water management solution.

Study 3: Vanløse School design experiment Despite a number of potential benefits of using the urban landscape for managing stormwater runoff the adoption of LSM technology was slow in Denmark around the year 2010. As an emerging technology, LSM was linked to a process of building awareness, capacity and trust among stakeholder groups, including landscape architects (Fryd et al. 2010). This study was set up to answer the research question: ‘Which design challenges do landscape architects meet, when working with landscape based stormwater management?’ A three-week ‘design experiment’ (de Jong and van der Voordt 2005; Steenbergen 2008) was conducted and twenty professional landscape architects were invited with the aim to gain insight into the LSM design process and thus potentially support future design processes through provision of information and further research. The landscape architects taking part in the design experiment were gathered into six teams to work on a specific site in Copenhagen, Denmark. All of the landscape architects were selected from design practices in the cities of Aarhus and Copenhagen for their interest in urban stormwater management, their willingness to dedicate the necessary time and their assumed ability to work fruitfully with other design offices. Most of the participants were acquainted with the concept of LSM, but only one team had experience with completed (constructed and built) projects. As presented in Figure 16.4, the design experiment started off with a two-day training workshop on stormwater management to ensure all the participants had at least a basic knowledge of the subject. The participants were informed about the frontier of technical knowledge, for inspiration they were shown international examples of LSM design, and these examples were then discussed regarding their specific implications in the Danish context. The teams were then introduced to the assignment of designing a stormwater management solution for Vanløse Primary School in Copenhagen. The teams worked on the assignment separately in their home offices. After one week, all teams met to present their initial ideas and sketches, and to receive feedback from the other teams and also from a panel of experts, which included internationally acclaimed LSM practitioners as well as experienced senior university staff. The design experiment ended with a final presentation and discussion involving a public audience, key stakeholders and the expert panel. The workshop set-up allowed the researchers to observe the designers whilst working, and to gain different but comparable design proposals for one particular site. Data collection methods included video recordings, note taking, semi-structured interviews (e.g. Bryman 2008), design observations (e.g. Hannula et al. 2005; Zeisel 2006) and comparison of competition entries (e.g. Hauxner 2003). Video recordings were done during the full duration of the four workshop meetings (day one, day two, midway meeting and final meeting). Personal notes from the researchers and the expert panel supplemented the video recordings. Each team was visited once at their office by researchers to observe the design process and to interview the design teams. The interview questions were designed to reveal the teams’ progress and idea development. Questions included: Where do you see the biggest problem in your design at the moment? What do you think is the most important skill you need in order to complete this design task successfully?

296 …

The urban water challenge „

Figure 16.4 Set-up of the design experiment with two workshop days in the beginning, a 2.5 week design phase, where the teams worked individually with one midway meeting, and concluded with the final presentations

What could be the biggest chance for improvement of the area that is addressed in your solution? Finally, the six draft designs were submitted as A1 posters and PDF files at the midway meeting and the six final design proposals were submitted after the final presentation. See Backhaus et al. (2012) for more details about the research methods. Using multiple data sources was intended to get the best possible picture of the landscape architects’ design processes and to allow for triangulation between the different results. The video recordings were transcribed and coded. The coding was done manually using keywords and specific topics in order to reveal patterns, similarities and recurring themes. Personal notes were coded using the same system. For the purposes of analysis and comparison, the six pairs of proposals were normalised by redrawing, and described briefly in a consistent manner using the same terms for identical design elements. The observations of the teams’ design processes, the interviews and the draft final design proposals were compared in relation to themes arising from analysis of the different observations, as well as themes identified prior to the workshop from the authors’ general knowledge of LSM design aspects (see also Study 1 above). Microsoft Excel tables were used to sort and compare the data by teams and topics. Eleven key themes were identified (see Table 16.3). The required responses to the eleven themes can be summarised as a need for further research and knowledge on many topics (specifically A to D), a need for dissemination of already existing knowledge amongst students and practitioners (specifically B to D and I), a need for improved collaboration between experts, designers and administrators (specifically H, I, J and K) and a need for more practical experiences which are relevant in order to improve practice in more or less all of the identified topics.

297 …

Antje Backhaus, Ole Fryd and Torben Dam

„

Table 16.3 Key themes and recommendations identified from the authors’ general knowledge of LSM design aspects, supported by and specified according to the studies’ findings

SYNTHESIS AND DISCUSSION Study 1 adopted a multiple case study approach. The strength of the study is the rigour of the analysis. By using literature-derived key parameters for comparison, the study approach can be used in other contexts. Study 1 also provided a framework for project assessment with a focus on aesthetics. The value judgments made in the preference rating assessment is, however, less easily replicable than the overall approach itself. The researchers are acting as observers, describing their subjective understanding of the projects, herein possibly weighting assessment criteria differently from how others would. Arguably, subjectivity is a necessity to cope with aesthetical assessment and to allow influence from the poetic and symbolic aspects of design (see Hauxner 2003). Future research could expand on the number of disciplines, or include assessments conducted by non-professional actors. Study 2 confronted the water challenge by working with and suggesting alterations to the urban landscape, and we extracted guidelines for design processes from our own design process. While doing so, we worked towards an understanding of the problem through designing and attempting to solve the problem of large-scale LSM retrofitting which aligns with the literature on wicked problem management. By providing a draft solution, hence going beyond the purely descriptive layering of existing data, we generated a framework for discussion across disciplines and stakeholder groups and facilitated more detailed assessments of the design solution, its sub-components and the underlying problem understanding. The approach taken in this study highlights the value that research through designing can have in a transdisciplinary research setting, and it also highlights the capacity of landscape design as a means to translate multiple data sets into symbiotic, yet tangible

298 …

The urban water challenge „

design solutions. The thorough cross-examination of the drafted strategy, combining problem formulation, analysis and design solution, is where more in-depth research findings emerge; here specifically about insights and setbacks in a transdisciplinary design process. Through the mapping exercise summarised in Figure 16.3, we provided a systematic framework for interrogation that can be applied to other planning and design processes. Study 3 utilised the strengths of the second study by adopting a research through designing approach. This study addressed the urban water management challenge by first developing tentative solutions that were then systematically assessed and discussed; lessons learned are valuable resources for resolving similar challenges in the future. The study is site-specific in order to support knowledge generation and knowledge sharing, and to facilitate comparison in a retrospective design evaluation. The third study addressed the potential weaknesses of the second study in terms of the dependence on individual actors and their social relationship. It is less vulnerable to the risk of one team failing to complete the task. By having a larger number of teams working on the same problem, context dependent and context independent parts of the design solutions and of the occurring problems during the design process become obvious. Study 3 was more thoroughly engaged with the process of designing, and developed a kind of experimental design laboratory in which landscape architecture practitioners take roles as active objects of the research analysis. The dependence of the researcher on participants’ willingness and commitment to collaborate, and to allocate time and resources to the project, must be taken into account when choosing to adopt the experimental design approach. In this study, everything was done through in-kind contributions, motivated by the business advantage the participating landscape architects gained over their peers in terms of LSM specific knowledge. Future research grant applications might need to consider and specify the actual costs of running design experiments, in order to ensure that adequate resources are made available. Comparing the three studies presented above it is noticeable how three different research questions lead to three different research approaches. The asterisk, square and circle markings in Figure 16.5 indicate where we place the three studies on a continuum within the different aspects of the research. All three studies adopt a case study approach. Since learning from previous landscape designs are key to future problem solving and to the choices that landscape architects make in the process of designing, knowledge obtained through case studies is very important to inform practice. Such knowledge also serves as precedent-setting cases for designers when they consider new design solutions and adopt new design tendencies. Each of the three studies is working with a different number of cases. While the first study, which aimed to get an overview of the state of the art and analyse the consequences of LSM in built projects, is designed with 20 cases, the two later studies worked with fewer cases but focused on a more in-depth analysis. This seems quite typical, as an overview is needed in the beginning of a process. Here just a few key assessment criteria are used to review a representative range of cases with a view to identifying general trends. Focusing on only one or a few cases allows research on the full complexity of a chosen case. This also depends of course on the design of the research project as described in the conceptual framework of this chapter. In the three presented studies, the selected cases predominantly reflect the ‘usual’ within the field in order to be able to extract, as far as possible, generally valid statements. All three studies seem to prove the suitability of using the notion of ‘wicked problems’ to work with the urban water challenge. Especially in Studies 2 and 3, the involvement of human actors plays a major role. This greatly contributes to problem complexity and, unlike a laboratory set-

299 …

Antje Backhaus, Ole Fryd and Torben Dam

„

Figure 16.5 Based on Figure 16.2, this figure illustrates how the three studies presented in this chapter differ in focus and research approach. Each study is represented by a different marking (asterisk, square, circle). The dashed arrows indicate the span of scales covered by Study 2 and Study 3

up where studies are typically reduced to a few parameters, research involving human actors in real-world cases serves as an important setting for landscape architecture research. Research is done, in such cases, by a transdisciplinary group that acknowledges the nature of wicked problems as a premise for the research design. Based on real-world cases, approaches including research through designing and design experiments enable such research groups to divide research questions into manageable parts. Looking for design solutions thus becomes as a series of testing certain assumptions and hypotheses. Design-led studies are also based on understandings where accepting complexity is the norm. We argue that real-world settings result in real-world findings and realworld problem formulation. Such research can reflect on a problem derived from practice, with the aim of giving recommendations back to practitioners who, in this case, are addressing the urban water challenge. We are thus working as ‘reflective practitioners’ who observe and analyse, and who sometimes engage with the issue investigated.

300 …

The urban water challenge „

Figure 16.6 The different roles of the researcher in the three studies presented in this chapter

Figure 16.6 shows how the researcher’s role has changed in the different studies, from that of a purely distant observer in Study 1, to a reflective designer in Study 2 and, finally, to a facilitating experimenter in Study 3. Taking different roles allows researchers to study issues of interest from different perspectives. One result of being able to look at one thing from different angles can be that researchers gain insights through getting a more holistic view of a problem; this is achieved by taking a comprehensive look at the three studies of this chapter together. Another aspect of interest is that, when compared with research approaches commonly assumed in, say, natural science, researchers of all three studies of this chapter needed to handle subjectivity. We did so with different degrees of subjectivity. In the case of acting as a distant observer, we took on the most ‘classic’ role as critical analysts. Without obvious personal engagement and thus no feelings and strong personal preferences involved, we aimed to keep the research transparent and replicable. Subjective assessments, if gained through repeatable study processes, might well be accepted as scholarly sound and publishable research. If, on the other hand, researchers decide to engage with the design process themselves, they may well get intense insight into the researched subject but, at the same time, might find themselves needing to argue for wanting to include elements of subjectivity into an otherwise objective study. Research that includes elements of subjectivity, generated through creative design input, might, even though the design process is kept fully transparent, render a study with a lack of transferability, in particular if it is a single case study, as seen in Study 2. Researchers must be aware that any research including elements of subjectivity is potentially more susceptible to external critique. We argue, however, that if researchers state their intentions

301 …

Antje Backhaus, Ole Fryd and Torben Dam

„

clearly, provide comprehensive process documentation, and thoroughly reflect upon their own role as researchers (and designers) and the potential impact of their role on research results, the findings have much value for practitioners and the quality of the findings very much resembles the practitioners’ experiences. We argue further that these may very well be findings that are of much value for and in their quality very close to practitioners’ experiences. Finally, in cases where researchers take on, in one and the same study, two different roles, as done in Study 2 described above, the researcher and designer duality must be accounted for through and by the study design. While researchers, who are designers themselves, are investigating the work of other designers and doing so by looking at several designs in parallel, their study findings are expected to have the highest possible level of objectivity. The approach and methods of Study 3 is still not commonly applied for research purposes, and the inclusion of subjective elements in it might need to be addressed in future studies, studies that are done in a similar manner, but in different contexts and on other topics, and then comparatively assessed in concerted reviews and meta-studies. The specific characteristics of the design process presented in Studies 2 and 3 benefit from the hermeneutic qualities that are inherent to planning and designing. Where interpretative elements are accepted it is not necessary to define planning and designing as research activities that are seen in opposition to, say, natural science. A design-based research paradigm would then rather be a matter of including cooperation by and a combination of a number of different fields in one study. In such studies, the design process is, at the same time, studied as an ‘object’ and, simultaneously, used as a research method to generate new knowledge. When considering the design process as a mode of translating theory (e.g. on urban water management) into practice (Gänshirt 2007), and when including retrospective design evaluation (e.g. Deming and Swaffield 2011), design-based research becomes a method that proves to be a valuable means to approaching and understanding complex problems. Furthermore, design experiments aid researchers who need to be working dynamically across scales. This is a factor that is very important within urban water management, and where the consideration of watersheds is often of importance in understanding site-scale problems, and vice versa. Working in feedback loops that include small, intermediate and large-scale aspects is a special competence that designers possess and that they feed into the process of problem finding and problem solving. In our comparative study of twenty different European LSM cases we were able to compare projects of a more or less similar scale. Here scale is defined by the projects and thus mainly administrative. Scale is critically reviewed, in Study 1, as part of amongst others ‘Water dynamics and dimensioning’. Dealing with LSM projects, the watershed scale is always very important. The LSM projects need to integrate with the whole watershed, for example in relation to runoff control or receiving water body peak flows. However, the watershed scale is often larger than the administrative site scale, yet out of the direct research focus when comparing the designs ‘as built’. Having identified this as an issue, the following studies, Studies 2 and 3, changed the focus towards a multi-level approach, considering the design issues on the specific site scale in parallel to, for example, hydraulic issues (hydraulic modelling, catchment management strategies, water engineering), which are specifically important on the larger catchment scale. In the examples above we, as researchers, also worked across temporal scales. In our study comparing cases of built forms of LSM projects, the case selection included a variety of projects that were in their early stages and others that had reached mature stages. These different age examples were selected in order to be able to analyse aspects of wear and tear and long-term maintenance. Working across temporal scales proved to be very valuable. Where, on the other hand, examples

302 …

The urban water challenge „

such as presented in the design experiment in Study 3, researchers are able to only cover a ‘snapshot in time’, they are not able to consider effects that long-term processes might have. The planning and design approaches used in the presented studies were well suited to foster collaboration across different disciplines and stakeholder types. Working jointly on solving a planning problem led, in the studies presented above, not only to agreeing on a joint aim and creating a common working language, but also helped to define fields where further investigation is found to be necessary.

CONCLUSIONS There still is an insufficient understanding of the kind of knowledge that landscape architects need to have in order to successfully address the urban water challenge. This chapter presents three studies that demonstrate how landscape architecture contributes methods that can support research on urban water issues. All three studies have helped to generate new knowledge in ways that are closely linked to landscape architecture design traditions. When looking at the three studies together, and in their chronological order, readers may appreciate how we, the authors of this chapter, have ourselves evolved as researchers through engaging with the process of doing research. In this process we have dared to step out of the role of the purely descriptive and analytical researcher. In our desire to confront pressing problems, we have adopted an active and explorative role and used the designing process as a key research method. Using this method we were able to address the urban water challenge in close collaboration with ‘real-world’ partners, thus building bridges across disciplinary and sectoral boundaries. Through using designing for purposes of conducting research we have been able to generate new knowledge. In conclusion, three specific issues can be highlighted. First, design research benefits from using designed examples as precedents for new studies (see Prominski – Chapter 12). Studying precedents can help to summarise existing knowledge and provide a shared knowledge platform within and across disciplines. This, in turn, can help to inform future decisions and thus speed up the transition towards a more water sensitive urban environment. Second, design research can help to overcome any standstill where rational thinking struggles to define a problem or develop a solution in the face of uncertainty and complexity. As such, new research paradigms, such as research through designing, can overcome obstacles where classic deductive research methods fail to cope with the complex nature of wicked problems. Third, landscape architectural research approaches are well suited, by including processes of change, to actively harnessing uncertainty.

SUGGESTED FURTHER READING This book chapter is based on the authors’ earlier research on LSM. For more information about the three studies presented in this chapter please see the full papers by Backhaus and Fryd (2012), Backhaus and Fryd (2013) and Backhaus et al. (2012). For Study 2, see also Fryd et al. (2013). Antje Backhaus expands on the role of landscape architecture in relation to the urban water challenge in an essay in the bilingual German/English publication ‘Grüne Infrastruktur. Green Infrastructure’ (BDLA 2015). A British SUDS Manual can be downloaded free of charge from: http://www.ciria.org/Resources/ Free_publications/SuDS_manual_C753.aspx

303 …

Antje Backhaus, Ole Fryd and Torben Dam

„

Recommended research exploring the interface between landscape architecture and water management includes the work by Kelly Shannon and Bruno de Meulder in Vietnam and Belgium, Anuradha Mathur and Dilip da Cunha in India and the United States, Bernardo Secchi and Paola Vigano in Italy, and Antje Stokman, Hille von Seggern and Martin Prominski in Germany. Christian Nolf, Fransje Hooimeijer and Liao Kuei-Hsien are also notable researchers in this field, among many others. Leading international designers linking landscape architecture and urban water management include Kongjian Yu and Herbert Dreiseitl. The most recent books describing their works are Designed Ecologies: The Landscape Architecture of Kongjian Yu (Saunders 2012) and Waterscapes Innovation (Dreiseitl and Grau 2014). Landscape architecture plays an important role in ongoing urban water management transition processes across the world. See for example Copenhagen (e.g. Skt Kjelds Kvarter, http://www. klimakvarter.dk), Melbourne (e.g. Little Stringybark Creek, https://www.watersensitivecities.org. au), New Orleans (e.g. Greater New Orleans Urban Water Plan, http://www.livingwithwater.com), Rotterdam (e.g. Benthemplein, http://www.rotterdamclimateinitiative.nl) and Singapore (e.g. Bishan Park, http://www.pub.gov.sg/abcwaters). In April 2015, China launched a national initiative on ‘Sponge Cities’ which initially provides funding for pilot projects in sixteen cities.

REFERENCES Ahern, J. (2011) ‘From fail-safe to safe-to-fail: Sustainability and resilience in the new urban world’, Landscape and Urban Planning, 100(4), 341–343. Anderson, S.I. (1994) ‘Landscape architecture at the Royal Danish Academy of Fine Arts, Copenhagen, Denmark’, Landscape and Urban Planning, 30(3), 169–177. Backhaus, A. and Fryd, O. (2012) ‘Analyzing the first loop design process for large-scale sustainable urban drainage system retrofits in Copenhagen, Denmark’, Environment and Planning B: Planning and Design, 39(5), 820–837. Backhaus, A. and Fryd, O. (2013) ‘The aesthetic performance of urban landscape-based stormwater management systems: A review of twenty projects in Northern Europe’, Journal of Landscape Architecture, 8(2), 52–63. Backhaus, A., Dam, T. and Jensen, M.B. (2012) ‘Stormwater management challenges as revealed through a design experiment with professional landscape architects’, Urban Water Journal, 9(1), 29–43. BDLA, eds. (2015) Grüne Infrastruktur – Green Infrastructure, Bund Deutscher Landschaftarchitekten, Berlin: Birkhäuser Verlag. Beneke, G. (2003) Regenwasser in Stadt und Landschaft: Vom Stück-werk zur Raumentwicklung – Plädoyer für eine Umorientierung, Räumlichen Planung – Schriftenreihe des Fachbereichs Landschaftsarchitektur und Umweltentwicklung der Universität Hannover, Volume 70, Hannover: University of Hannover. Bryman, A. (2008) Social Research Methods, 3rd ed., New York: Oxford University Press. Chemetoff, A. (2009) ‘The projects of Genoble and Allonnes or the economy of means’, Journal of Landscape Architecture, 4(2), 82–89. Conklin, J. (2005) Dialogue Mapping: Building Shared Understanding of Wicked Problems, Hoboken, NJ: John Wiley. de Jong, T.M. and van der Voordt, D.J.M., eds. (2005) Ways to Study and Research Urban, Architectural and Technical Design, Delft: Delft University Press. Deming, E.M. and Swaffield, S. (2011) Landscape Architectural Research: Inquiry, Strategy, Design, Hoboken, NJ: John Wiley. Dreiseitl, H. and Grau, D. (2014) Waterscapes Innovation, London/Hong Kong: DesignMedia Publishing. Dreiseitl, H., Grau, D. and Ludwig, K.H.C., eds. (2001) Waterscapes: Planning, Building and Designing with Water, Basel: Birkhäuser Verlag.

304 …

The urban water challenge „

Echols, S. and Pennypacker, E. (2008) ‘From stormwater management to artful rainwater design’, Landscape Journal, 27(2), 268–290. Etzkowitz, H. and Leydesdorff, L. (2000) ‘The dynamics of innovation: From national systems and “mode 2” to a triple helix of university–industry–government relations’, Research Policy, 29(2), 109–123. Fletcher, T.D., Shuster, W., Hunt, W.F., Ashley, R., Butler, D., Arthur, S., Trowsdale, S., Barraud, S., SemadeniDavies, A., Bertrand-Krajewski, J.-L., Mikkelsen, P.S., Rivard, G., Uhl, M., Dagenais, D. and Viklander, M. (2015) ‘SUDS, LID, BMPs, WSUD and more: The evolution and application of terminology surrounding urban drainage’, Urban Water Journal, 12(7), 525–542. Flyvbjerg, B. (2006) ‘Five misunderstandings about case-study research’, Qualitative Inquiry, 12(2), 219–245. Fryd, O., Jensen, M.B., Ingvertsen, S.T., Jeppesen, J. and Magid, J. (2010) ‘Doing the first loop of planning for sustainable urban drainage system retrofits: A case study from Odense, Denmark’, Urban Water Journal, 7(6), 367–378. Fryd, O., Backhaus, A., Birch, H., Fratini, C.F., Ingvertsen, S.T., Jeppesen, J., Panduro, T.E., Roldin, M. and Jensen, M.B. (2013) ‘Water sensitive urban design retrofits in Copenhagen: 40% to the sewer, 60% to the city’, Water Science and Technology, 67(9), 1945–1952. Fünfgeld, H. and McEvoy, D. (2011) ‘Framing climate change adaptation in policy and practice’, Working Paper 1, Melbourne: Victorian Centre for Climate Change Adaptation Research. Gänshirt, C. (2007) Werkzeuge für Idee: Einführung ins Architektonische Entwerfen, Berlin: Birkhäuser Verlag. Geldof, G.D. and Kluck, J. (2008) ‘The Three Points Approach’, Proceedings from the 11th International Conference on Urban Drainage, Edinburgh, Scotland, UK, available: http://web.sbe.hw.ac.uk/ staffprofiles/bdgsa/11th_International_Conference_on_Urban_Drainage_CD/ICUD08/pdfs/36.pdf. Hannula, M., Suoranta, J. and Vadén, T. (2005) Artistic Research: Theories, Methods and Practices, Helsinki/ Gothenburg: Helsinki Academy of Fine Arts and University of Gothenburg/ArtMonitor. Hauxner, M. (2003) Open to the Sky: The Second Phase of the Modern Breakthrough 1950–1970 – Building and Landscape, Spaces and Works, City Landscape, Copenhagen: Arkitektens Forlag. Hayden, D. (1997) The Power of Place: Urban Landscapes as Public History, Cambridge, MA: The MIT Press. Head, B.W. (2008) ‘Wicked problems in public policy’, Public Policy, 3(2), 101–118. Hintze, M. (2009) Personal communication, Allod Real Estate and Asset Management Company, Berlin Karow Nord. Hoyer, J., Dickhaut, W., Kronawitter, L. and Weber, B. (2011) Water Sensitive Urban Design: Principles and Inspirations for Sustainable Stormwater Management in the City of the Future, 018530-SWITCH, Berlin: Jovis Verlag. IPCC (2014) Climate Change 2014 Synthesis Report Summary for Policymakers, Intergovernmental Panel on Climate Change, available: https://www.ipcc.ch/pdf/assessment-report/ar5/syr/AR5_SYR_FINAL_SPM. pdf (accessed 9 July 2015). Kline, S.J. and Rosenberg, N. (1986) ‘An overview of innovation’, in Landau, R. and Rosenberg, N., eds. The Positive Sum Strategy: Harnessing Technology for Economic Growth, Washington DC: National Academy Press, 275–305. Liao, K.H. (2012) ‘A theory on urban resilience to floods: A basis for alternative planning practices’, Ecology and Society, 17(4), 48. Lipton, P. (1991) Inference to the Best Explanation, London: Routledge. Martin, C., Ruperd, Y. and Legret, M. (2007) ‘Urban stormwater drainage management: The development of a multicriteria decision aid approach for best management practices’, European Journal of Operational Research, 181(1), 338–349. Mguni, P., Herslund, L. and Jensen, M.B. (2014) ‘Green infrastructure for flood-risk management in Dar es Salaam and Copenhagen: Exploring the potential for transitions towards sustainable urban water management’, Water Policy, 17(1), 126–142. Mitchell, V.G. (2006) ‘Applying integrated urban water management concepts: A review of Australian experience’, Environmental Management, 37(5), 589–605. Rittel, H.W.J. and Webber, M.M. (1973) ‘Dilemmas in a general theory of planning’, Policy Sciences, 4(2), 155–169. Roberts, N. (2000) ‘Wicked problems and network approaches to resolution’, International Public Management Review, 1(1), electronic journal, available: http://journals.sfu.ca/ipmr/index.php/ipmr/article/view/175

305 …

Antje Backhaus, Ole Fryd and Torben Dam

„

Saunders, W., ed. (2012) Designed Ecologies: The Landscape Architecture of Kongjian Yu, Basel: Birkhäuser Verlag. Schön, D.A. (1991) The Reflective Practitioner: How Professionals Think in Action, 2nd ed., Aldershot: Ashgate. Schot, J. and Geels, F.W. (2008) ‘Strategic niche management and sustainable innovation journeys: Theory, findings, research agenda, and policy’, Technology Analysis & Strategic Management, 20(5), 537–554. Seto, K.C., Fragkias, M., Güneralp, B. and Reilly, M.K. (2011) ‘A meta-analysis of Global Urban Land Expansion’, PloS One, 6(8), available: doi:10.1371/journal.pone.0023777. Sieker, F., Kaiser, M. and Sieker, H. (2006) Dezentrale Regenwasserbewirtschaftung im privaten, gewerblichen und kommunalen Bereich: Grundlagen und Ausführungsbeispiele, Stuttgart: Fraunhofer IRB Verlag. Smit, B., Burton, I., Klein, R.J.T. and Wandel, J. (2000) ‘An anatomy of adaptation to climate change and variability’, Climatic Change, 45(1), 223–251. Stappers, P. J. (2007) ‘Doing design as part of doing research’, in Michel, R., ed. Design Research Now: Essays and Selected Projects, Berlin: Birkhäuser Verlag, 81–98. Steenbergen, C. (2008) Composing Landscapes: Analysis, Typology and Experiments for Design, Basel: Birkhäuser Verlag. Strom, S., Nathan, K. and Woland, J. (2009) Site Engineering for Landscape Architects, Hoboken, NJ: John Wiley. Thacher, D. (2006) ‘The normative case study’, American Journal of Sociology, 111(6), 1631–1676. Uhl-Bien, M., Marion, R. and McKelvey, B. (2007) ‘Complexity leadership theory: Shifting leadership from the industrial age to the knowledge era’, Leadership Quarterly, 18(4), 298–318. UN DESA (2014) World Urbanization Prospects: The 2014 Revision, United Nations Department of Economic and Social Affairs, Population Division, New York: United Nations, available: http://esa.un.org/unpd/wup/ Publications/Files/WUP2014-Report.pdf (accessed 9 July 2015). Von Seggern, H., Werner, J. and Grosse-Bächle, L., eds. (2008) Creating Knowledge: Innovation Strategies for Designing Urban Landscapes, Berlin: Jovis Verlag. Woods-Ballard, B., Kellagher, R., Martin, P., Jefferies, C., Bray, R. and Shaffer, P. (2007) The SUDS Manual, London: CIRIA Classic House. Wöbse, H. H. (2002) Landschaftsästhetik, Stuttgart: Eugen Ulmer Verlag. Zeisel, J. (2006) Inquiry by Design: Environment/Behavior/Neuroscience in Architecture, Interiors, Landscape and Planning, New York/London: W.W. Norton & Company.

306 …

Index

Page numbers followed by f indicate figures. Page numbers followed by n indicate note numbers. Page numbers followed by t indicate tables. Aarhus Convention on Access to Information 161 ABMs. See agent-based models Access to Justice on Environmental Matters 161 Aesthetic Creation Theory 28 aesthetics 217–19 agent-based models (ABMs) 130 agriculture/agricultural: landscapes 38, 109, 161, 166; mechanised 131; scientists 58; urban 200, 201f. See also Zürich case study AHD. See Authorised Heritage Discourse air pollution 263 Alnarp Rehabilitation Garden 253 Alport Valley case study 169–75; description of 169; long-term forest management plan 170; results 172–3, 174f; study design and protocol 171–2, 171f, 172f; study participants 170–1; visual stimuli 170 American Society of Landscape Architects (ASLA) 66, 91 Amiel, Henri Frédéric 211 Anderson, Sven Ingvar 287f anemometer 276–7, 276f Antrop, Marc 80 APIs. See application programming interfaces application programming interfaces (APIs) 141 architect/architecture. See case studies; Delphi method; landscape architecture; research; systematic review arts: theory and 40–1, 46–9 ASLA. See American Society of Landscape Architects Australia 48; National Heritage List 224; Throsby Park 228, 229f; Wingecarribee Shire, New South Wales, Australia 224–31 Authorised Heritage Discourse (AHD) 211 Backhaus, Antje x, 285–306

Bell, Simon x, 1–8, 11–23 Benthem, Roel 187 Berger, John 47 Berleant, A. 218 ‘Big Data’ 44 Bijhouwer, Jan 187 blogs 154 Boerenverstand project 125–7; assignment and research question 125; methods and materials 125–6; outcome 127; results 126 Böhme, Gernot 50 Borgdorff, Henk 191 Bosma, Koos x, 120–35 Bowring, Jacky 48 Boyle’s Law 41 British Mountaineering Council 171 Brown, Lancelot ‘Capability’ 11 Brown, Robert D. x, 263–84 Bruns, Diedrich x, 1–8, 11–23, 85–102, 136–60 Bryson, Bill 180 Bullock, A. 217–18 Burra Charter 217 Byrne, D. 211 Campaign to Protect Rural England (CPRE) 171 Campbell, Heather 76 Carter, Paul 48 case studies: Alport Valley case study 169–75; of an emerging body of theory 115; approach to research 299; based on types 114f; choosing the cases 108–10, 110f; comparison of 110–14, 111f, 112f; critical 109–10, 117; definition of 107; embedded geographical cases 112, 113f; evidence versus theory with 108; of exploratory nature 114–15; extreme 109; generalisation of research

307 …

Index

and 108; identification of type of case under investigation 116; I’DGO TOO 243–8; in landscape architecture 105–19; landscape based stormwater management design projects 290–3, 291t, 292t; landscape based stormwater management strategy in Copenhagen 291t, 293–6, 295f; L’DGO: Inclusive Design for Getting Outdoors 237–43; matching to research question 114–15; multiple analyses of 110–12, 111f, 112f; nature of case study research 107–8; overview 105–6; paradigmatic 108; of particular landscape settings 114; percentage of papers reporting use of 94–5, 94t; pragmatic approach to 106; prototypical 109; purposive samples of 108, 117; selection of 108; social media and 158; to test or challenge theoretical claims about complex situations 115; theoretical 109, 110f; ‘Under the Sky’ 106; of urban water challenge 289–303; Vanløse School design experiment 291t, 296–7, 297f, 298t; Wingecarribee Shire, New South Wales, Australia 224–31; within-case comparisons 116; Zürich case study 164–9 CASP-19 241 Catholicism 129 CBC. See Choice-Based Conjoint studies CELA. See Council of Educators in Landscape Architecture Cerwonka, Allaine 48 China 213–14; China Principles document 217 Choice-Based Conjoint (CBC) studies 251, 251f Christensen, K.S. 73 Christiaansen, Krijn 125 cities. design guidelines for 199–200; microclimates in 263–4; theoretical foundation 265–7; thermally comfortable urvan environments 263–84. See also microclimates citizen science 138 client–designer–contractor 11–12 CMP. See Conservation Management Plan commissioner–designer 11–12 concept: design 18–20, 19f; of landscape biography 121–3; research and 54–5; social media and 138–42; of urban water 286–9; of virtual environments 162–3; of walking 179–81 conjoint analysis 248–53, 249f; questionnaire development 250–1, 251f; results and scenario modelling options 252––253 Conservation Management Plan (CMP) 219, 220f construction and maintenance 292t constructivist landscape theory 144

„

Corner, James 50 Cosgrove, D.E. 212, 213 Council of Educators in Landscape Architecture (CELA) 87 CPRE. See Campaign to Protect Rural England crowdsourcing 137–8 CRT. See cylindrical radiation thermometer CSM. See walking, continuous/stop-motion walking culture, in landscape meanings and values 211–34; aesthetics of 217–19; analysis and evaluation of 223–4; assessment of landscape characteristics 222–3, 223t; conceptual framework 212–19; description of 212–14; heritage and 215–17, 216t; history of ‘landscape’ 214–15; identification and documentation of cultural landscape resources 222; overview 211–12; planning model for heritage conservation management policy 220f; Wingecarribee Shire, New South Wales, Australia 224–31 cylindrical radiation thermometer (CRT) 271–2, 272f Dam, Torben x, 285–306 data analysis: in process research 30–1 data collection: in process research 29–30 DBSCAN. See density-based clustering algorithms DBU. See Deutsche Bundesstiftung Umwelt de Certeau, Michel 128 de Jonge, Nico 187 Delphi method 86–9; academic perspectives and 87; competencies of landscape architect 88; defining characteristics of 86; description of 86, 87f; development of 86; feedback on 88; general procedure of Delphi study 87f; methodology of landscape architecture Delphi study 87–8; questionnaire designed for 88; research domain results 89t; results of Delphi study 88–9, 99–100; subjects of study 87–8 Deming, Elen 49–50 Denmark: landscape based stormwater management strategy in Copenhagen 291t, 293–6, 295f; Vanløse School design experiment 291t, 296–7, 297f, 298t dendrology 1 density-based clustering algorithms (DBSCAN) 148–9 de Roo, Gert 43–4 Descartes, Rene 48 design: conceptual 18–20, 19f; current meaning of research through 59–61; definition and basic concept of 54–5; evolution of research through 56–9; project design 16– 22; projective 58–9; relevance to theory 49–51; research and 11–14, 28–9, 54–64; research into, for and through 55–6; study and 28–9; technical 20–2, 21f

308 …

Index „

design guidelines 194–208; challenges in 194; components of 206–7; definition of 194–5; design research landscape 205–6; to help combine nature protection and open space use 200–2; larger research landscape and 205; overview 194–5; process-orientated design of urban river spaces 196–9, 198f, 199f; process to develop 203f; research about 206; for shrinking cities 199–200; strengths and weaknesses of 204–5; theoretical framework 195–6 Deutsche Bundesstiftung Umwelt (DBU) 199, 200 DFG. See German Research Foundation Diefenthaller, I. 216 digital terrain model (DTM) 170 directive. See European Flood Directive; European Water Framework Directive discipline. See landscape architecture; social media; theory, role in landscape architecture ‘DIY Streets’ 245 DMs. See dynamic models Dorling, Danny 44 DTM. See digital terrain model Duchhart, Ingrid xi, 54–64 dynamic models (DMs) 130 Eckbo, Garrett 40 ECLAS. See European Council of Landscape Architecture Schools ecology 196 ELC. See European Landscape Convention emoticons 154 energy, theory of 266–7 environment. See cities; concept; health, landscape and; landscape architecture; virtual environments EU. See European Union European Council of Landscape Architecture Schools (ECLAS) 2, 77, 87, 91 European Flood Directive 196 European Landscape Convention (ELC) 78, 161, 180 European Union (EU) 106, 130, 196; Horizon 2020 research programme 235 European Water Framework Directive 196 EUROQOL 241 experiments. See case studies; Denmark; systematic review; urban water, challenge of; walking explore. See case studies Facebook 145, 154 fields. See landscape architecture; theory, role in landscape architecture Flickr 151, 151t Forestry Commission 171 forests. See Alport Valley case study Forman, Richard T.T. 42

Foucault, Michel 50 FP7 HERCULES project 130 Francis, M. 106 Frayling, Christopher 191 Freud, Sigmund 47 Friends of the Peak District 171 Frontiers of Architectural Research 5 Fryd, Ole xi, 285–306 GCC. See global climate change Gehl, Jan 182 general research question (GRQ) 19 geographer/geography. See case studies; geographic information system; maps geographic information system (GIS) 130 GeoNames 147 geo-social information 137 geo-social media 141 Gephi 154 German Nature Protection Law 202 German Research Foundation (DFG) 196 Germany: Ostwürttemberg using WebGIS technologies 142–4 Germ Theory of Disease 41 Gillespie, Terry J. xi, 263–84 GIS. See geographic information system global climate change (GCC) 263, 265 Grahn, Patrik 253 Greffe, X. 214 Gros, Frédéric 180 GRQ. See general research question Han, Feng 213–14 Hartig, Terry 42 health, landscape and 235–62; conjoint analysis of 248–53, 249f; I’DGO TOO 243–8; L’DGO: Inclusive Design for Getting Outdoors 237– 43; overview 235–7; research methods for design of salutogenic landscapes 253–7; research on environments that support outdoor access in ageing society 237–43; residential streets 243–8 Hehl-Lange, Sigrid xi, 161–78 heritage management 131; planning model for heritage conservation management policy 220f high-bay racking 200, 201f Historic Urban Landscape (HUL) concept 219, 232 Home Zone 245 Horizon 2020 research programme 18 horticulture 11, 66 Howard, Peter 76 HUL. See Historic Urban Landscape concept humanities: theory and 40–1, 46–9 Husserl, Edmund 48, 180

309 …

Index

ICOMOS. See International Council on Monuments and Sites I’DGO TOO 243–8; additional measures used 246–7, 246f; other data analyses and results 247–8; questionnaire analysis 247; questionnaire results 247; recruitment of participants 245–6; research design 245; research questions 244; street design intervention 244–5 IFLA. See International Federation of Landscape Architects IFPRA 27 Information Processing Theory 41–2 infra-red camera 274 infra-red thermometers (IRTs) 274 Ingold, Tim 182 integrity: in process research 32–3 interdisciplinary 5, 24, 25, 26, 58, 66, 68, 72, 80, 81, 100, 120, 122, 231, 255 International Council on Monuments and Sites (ICOMOS) 216 Internationale Bauaustellung (International Building Exhibition) 131 International Federation of Landscape Architects (IFLA) 87, 91 International Physical Activity Questionnaire (IPAQ) 254–5 International Union for Conservation of Nature (IUCN): Category V Protected Landscapes 214 intervention: for street design 244–5 IPAQ. See International Physical Activity Questionnaire IRTs. See infra-red thermometers IUCN. See International Union for Conservation of Nature Jackson, J.B. 49, 213 Jacobs, Jane 182 Jacobs, Peter 80–1 Japan 217 Jeans, Dennis 213 Jellicoe, Sir Geoffrey 47 JoLA. See Journal of Landscape Architecture Jonas, Wolfgang 191 Journal of Design Research 5 Journal of Environmental Engineering and Landscape Management 90 Journal of Landscape Architecture (JoLA) 5, 77, 90, 106 Journal of Landscape Ecology 90 Journal of Urban Design 5 justice 32, 43, 69, 78 Kaiser Meyer Olkin measure of sampling 247 Kamiya, Gary 65, 81 Kaplan, Rachel 41 Kaplan, Stephen 41

„

Kinder and High Peak Advisory Committee 171 knowledge 182; constructivist 43; dualism of 47–8 Kolen, Jan xi, 120–35 Kühn, Norbert 199 Kuhn, Thomas 50 landscape: definition of 180; descriptions of 38, 180; health and 235–62; history of 214–15; representational models 39; rural 222, 223t; temporality of 49 Landscape and Ecological Engineering 90 Landscape and Urban Planning 5, 66, 90, 100 landscape architecture: academic versus projectbased research 2, 11–23; assessment of research priorities and quality 85–102; case study method of 70, 105–19; challenge of publication 65–84; constructivist landscape theory 144; context of ‘scientification’ 54; culture, in landscape meanings and values 211–34; definition of 4; design guidelines 194–208; discipline versus disciplinary field 51n1; educational process for 6; founders of 11; history of 1; key themes in 70; methodology of systematic review 90–1; ‘mode 2’ 66; multi-disciplinary origins of 2; overview 1–2; philosophy and 42–3; ‘post-academic science’ 66; process approach to research in 17–18, 17f, 24–34; professionals in the field 67; publication of research 65–84; purpose of 15; relationship between research and design 54–64; research basis 68–70, 69f; researcher 3–4; role of theory in 37–53; as a scholarly profession 71; social media and 136–60; systematic review of papers in 90–9; technological development for 12; thermally comfortable urban environments 263–84; urban water challenge 285–306; virtual environments and 161–78; walking and 179–93 landscape based stormwater management (LSM): description of 286; design projects 290–3, 291t, 292t; strategy in Copenhagen 291t, 293–6, 295f landscape biography 120–35; combined use of methods and sources for 123; conceptual framework of 121–3; ‘dwelling perspective’ of 123–4; examples using approach of 125–8; heritage management and 131; ‘historicising’ approach to 123; history of 120; human-made landscape and 123; next developments of 129–32; overview 120; principles for using as a research approach 123–4; rationale for using in landscape research 121; relevant historical research and reflection of 124;

310 …

Index „

sources and methods of 124–5; of urban landscapes 128–9 Landscape Change Model 39 landscape character units (LCUs) 223–4 landscape ecology 42 Landscape Ecology 90 Landscape Education Foundation 106 Landscape History 90, 91 Landscape Institute 91 Landscape Journal 5, 90, 100 Landscape Online 90, 91 Landscape Planning & Urban Ecology 66 Landscape Research 5, 70, 90 Lange, Eckard xi, 161–78 LCUs. See landscape character units L’DGO: Inclusive Design for Getting Outdoors 237– 43; analyses of data 242–3; focus groups 238–9; questionnaire administration 242; questionnaire development 239–42, 239f; research design 238–42; research ethics 238; research questions 237–8; results of 243; theoretical foundations of 237 Le Corbusier 56–7 Lefebvre, Henri 50 legislation: German Nature Protection Law 202 le Nôtre, André 11 LE:NOTRE Institute 106, 116 LE:NOTRE programme 106 Lenzholzer, Sanda xi, 54–64 Lewis, Peirce 213 LID. See low impact development Lieberman, Arthur 42 ‘The Limes nowadays – A biography’ 127–8; assignment and research question 127; methods and materials 127; outcome 128; results 128 Linguistic Inquiry and Word Count (LIWC) 154 LinkedIn 145, 146t liquid-in-glass thermometer 273 liveability 29, 180 Living Reviews in Landscape Research 90 LIWC. See Linguistic Inquiry and Word Count Lowenthal, David 213 low impact development (LID) 286. See also landscape based stormwater management Lyme disease 264 Lyotard, Jean-François 45 Macey, David 45 Macfarlane, Robert 180 Macquarie, Lachlan 228 management. See Alport Valley case study; case studies; culture, in landscape meanings and values; Denmark; heritage management; landscape based stormwater management; landscape biography; models; urban water, challenge of

maps: of embedded geographical cases 113f; of identified hotspots and tourist route densities 150f; of public input to landscape assessment 143f; Wingecarribee Shire, New South Wales, Australia 225f Martens, Conrad 228 Marxism 45, 47 material/materialist. See Boerenverstand project; ‘The Limes nowadays – A biography’ McHarg, Ian 57 McHargian landscape theory 41 media 171–5, 171f, 172f, 174f Meijering, Jurian xi, 85–102 Meinig, D.W. 215 Meining, Donald 213 mental 38, 40, 129, 141, 144, 186, 237 Merchant, Caroline 47–8 Merleau-Ponty, Maurice 48 Meyer, Elizabeth 46 microclimates 263; data collection 277–9; disease in 264; estimating 280; instruments 278f; measurement 279–80; methods for measurement and estimation of elements of 267–79; misunderstandings about 280–1; radiation 267–72; temperature and humidity 272–5, 272f; wind 276–9, 276f. See also cities models: agent-based 130; continuous/stop-motion walking 188f; digital terrain 170; dynamic 130; energy 267; of landscape 39; logic model of longitudinal approach to study design 249f; planning model for heritage conservation management policy 220f; virtual 165f; Wingecarribee Shire, New South Wales, Australia study model 227f Montens, Cathelijne 125 multi-disciplinary. See landscape architecture Murphy, Michael D. 37–8 National Institute for Health and Care Excellence (NICE) 250 natural sciences: theory and 40–3 nature protection 200–2 Naveh, Nev 42 neighbourhood open space (NOS) 241 Netherlands Organisation for Scientific Research (NWO) 120 Netvizz 154 networking 136 New Zealand 48, 109, 110f, 111–12 NICE. See National Institute for Health and Care Excellence Nicholson, Geoff 179–80 NodeXL 154 norms/normative 3, 13, 41, 45, 57, 70, 76, 108, 117, 153, 161, 195, 290, 293, 300 NOS. See neighbourhood open space

311 …

Index

NWO. See Netherlands Organisation for Scientific Research observations 11, 16, 20, 28, 29, 37, 44, 46, 65–6, 73, 80, 95–6, 99, 111, 157, 190, 215, 222, 224, 228, 247–8, 253–4, 280, 288, 290, 292–3, 296 OCs. See online communities Olmsted, Frederick Law 11 online communities (OCs) 155 OPENspace 247, 255 opinion mining 154 Orions, Gordon 43 Outstanding Universal Value (OUV) 217 OUV. See Outstanding Universal Value Pajek 154 PDNPA. See Peak District National Park Authority Peak District National Park 169. See also Alport Valley case study Peak District National Park Authority (PDNPA) 171 perspectives. See Delphi method; landscape biography phenomenology theory 180–1 philosophy: in landscape architecture 42–3 ‘Placebook Scotland’ 80 Plachter, H. 212–13 plants. benefits for 169; collecting expeditions for 11; ecological processes in 112; importing 48; innovative methods of planting 200; newly planted 164; rare 144; species of 28–9; survival of 264. See also vegetation positivism 42–3 Post Occupancy Evaluation 55 practitioners 1, 3, 4, 11, 27, 37, 46, 65–7, 71, 77–8, 90, 106, 128, 183, 194, 196, 204, 207, 218, 288, 296–7, 299–300, 302 process research: approach to 24–34; data analysis 30–1; data collection and sampling 29–30; formulating a research question 27–8; integrity 32–3; overview 17–18, 17f, 24–5; reporting 31–2; research design and study design 28–9; research methodology and process approach to 25–7 Prominski, Martin xi–xii, 180, 194–208 psychoanalytic theory 47 psychologist/psychology 6, 38, 40–2, 44, 55–6, 66, 85, 239, 255–6 psychometer 275 publication: challenge of 65–84; clarity in 76; direct quotations in 76–7; emerging themes 79t; key areas of research 77–80; keywords in 70; literature review 75; methods reporting 74, 74t; overview 65; peer-reviewed 76; quality of research and research papers 71–7; research paper defined 74–5; research ‘voices’ of 76; standards for

„

assessing quality of research 71, 71t; systematic review of papers in landscape architecture 90–9; trends in published research 77; in the United Kingdom 65–8 Public Participation in Decision-Making 161 pyranometers 267, 268f; CNR1 270f; photocell 267, 268f pyrgeometers 270, 270f QALYS. See Quality Life Years QDA Miner 154 QOL. See quality of life Quality Life Years (QALYS) 255 quality of life (QOL) 239–40, 239f questions: formulating a research question 27–8; general research 19 radiation: shielding 273; solar 267–72; terrestrial 270–1, 272f; total 271–2 REF2014 72–3, 74 Relph, Edward 46 Renes, Hans xii, 120–35 reporting: in process research 31–2 research: about design guidelines 206; academic research versus project-based research 2; academic versus non-academic 13; aspects of doing research 15f; assessment of priorities and quality 85–102; basis 68–70, 69f; case study approach to 299; challenge of publication 65–84; conceptual research design 18–20, 19f; current meaning of research through designing 59–61; definition and basic concept of 54–5; Delphi method as research method 86–9; design and 11–14, 54–64; designing a research project 16–22; emerging themes 79t; evolution through designing 56–9; integrated approach to 15; into, for and through design 55–6; key areas of 77–80; knowledge gain and 17; literature review 75; matching case studies to research question 114–15; methodology 24; methods reporting 74, 74t; paper defined 74–5; peer-reviewed journals 5; process approach to 24–34; process overview 17–18, 17f; project research versus research project 13; proposal checklist 14t; quality of papers 71–7; rationale for using landscape biography in landscape research 121; reporting 31–2; social media challenges and 147–8; standards for assessing quality of 71, 71t; technical research design 20–2, 21f; trends in published research 77; types of landscape architecture studies 14–16; ‘voices of’ 76 researcher 3–4; roles of 301, 301f Research Excellence Framework 71, 72

312 …

Index „

research through designing (RTD) 56, 60t; constructivist 61; current meaning of 59–61; evolution of 56–9; future of 61–2; transformative 61; types of 59, 60t residential streets 243–8 Riley, Robert 42 Rio Declaration on Environment and Development 161 River. Space. Design 196–9 Roe, Maggie xii, 65–84 Rose, Gillian 47 Rössler, M. 212–13 Rousseau, Jean-Jacques 180 S-A-D. See Survey-Analysis-Design salutogenic landscapes, research methods for 253–7; as alternative research methods 253; longitudinal research and 256–7; objectivity, subjectivity and use of selfreport measures 253–5; overview 253; questionnaires and area-level versus individual data 255–6 sampling: percentage of papers reporting use of random or non-random sampling method 95, 95t; procedure of 90; process in systematic review 92f; in process research 29–30 Savannah Theory 43 Schlüter, Otto 213 scholarly/scholarship: in landscape architecture as a profession 71 Schön, Donald 195 Schultz, Henrik xii, 179–93 science: description of 40 ‘scientification’ 54 scientism: theory and 43 SCOPUS: search query applied in 91t Scottish Walkability Assessment Tool (SWAT) 247 SDI. See spatial data infrastructures sentiment analysis 154 SentiStrength 154 SentiWordNet 154 Simon, Herbert 55 Sinclair, Ian 180 site history and context 292t SMA. See social media analytics SNA. See social network analysis SNE. See supportiveness of the neighbourhood environment social constructionists 141 social media 136–60. access rights and 153–4; big data and crowdsourcing 137–8; case studies and 158; classifications of 137; conceptual framework to study role of 138–42; content of 137; definitions of 136–7; extended transdisciplinary framework with interaction typology 140f;

false information and 157; methods for studying impacts of 153–4; outlook 157– 8; overview 136; as ‘place’ and interface of communication 139–41; research methods examples and 142–53; role in research 139, 139f, 140f; social discourse and ‘construction of reality’ 141–2; text options and 145; transdisciplinary framework with 139f; validation of data 156–7; validity of 156; value of 157; variety of 156; velocity of 137, 155; volume of 155. See also technology social media analytics (SMA) 153 social network analysis (SNA) 153 social sciences: theory and 40–1, 43–6 sociobiology 43 solar radiation 267–70; access to 264; estimating 268–70; measurement of 267, 268f SOS. See Statement of Significance space 163; accessibility of 190–1; open space use 200–2; urban river 196–9, 198f, 199f spatial data infrastructures (SDI) 129–30 specific research questions (SRQs) 19 SPSS. See Statement of Significance SRQs. See specific research questions Stallybrass, O. 217–18 Statement of Significance (SOS) 224 Statistical Package for the Social Sciences (SPSS) 242 Steinitz, Carl 58 Stemmer, Boris xii, 136–60 Stigsdotter, Ulrika 253 Stokman, Antje 196 stormwater accentuation 292t, 293 Studies in the History of Gardens & Designed Landscapes 90 SUDS. See sustainable urban drainage systems supportiveness of the neighbourhood environment (SNE) 242 Survey-Analysis-Design (S-A-D) 39 sustainable urban drainage systems (SUDS) 286 Swaffield, Simon xii, 49–50, 70, 105–19 SWAT. See Scottish Walkability Assessment Tool Switzerland: Zürich case study 164–9 systematic review: of case studies 116–17; formulation of research question codes 92–3; imagery as data source 99; methodology of 90–1; overview of sampling process 92f; of papers in landscape architecture 90–9; papers not following basic scientific reporting structure 99; percentage of papers related to a particular research domain 97, 98t; percentage of papers reporting quantitative and qualitative analysis of data 96–7, 97t; percentage of papers reporting research objective, question or hypothesis 93, 94t; percentage of papers

313 …

Index

reporting research objective or question 93, 94t; percentage of papers reporting use of case study, experiment or other study design 94–5, 94t; percentage of papers reporting use of random or non-random sampling method 95t; percentage of papers reporting use of various data collection methods 95–6, 96t; recommendations 98–9; results of 93, 94t, 95t, 96t, 97t, 98t; sampling collection codes 93; search query applied in SCOPUS 91t Taylor, Ken xii, 211–34 technology: for landscape architecture 12; in landscape biography 129–30. See also social media temperature and humidity 272–5, 272f; air temperature 273, 274t; electronic sensors for 273; humidity 275; surface temperature 274 terrain, changes in 292t, 293 Theile, Simone xii, 136–60 the Netherlands 130; Amsterdam tourism behaviour 148–53, 149t, 150f, 151t; Boerenverstand project 125–7; ‘The Limes nowadays – A biography’ 127–8; Zuid-Holland 145–8, 146t theory, role in landscape architecture 37–53; description of 37–9; as a discipline and profession 39–40; evidence versus 108; overlapping fields in 69f; procedural 40; relationship to natural sciences, social sciences, and arts and humanities 40–1; relevance to design 49–51; scientism and 43 theory of energy 266–7 thermometers: infra-red 274; liquid-in-glass 273 Thompson, Catharine Ward xiii, 42, 235–62 Thompson, Ian H. xii, 37–53 3D. See three dimensional landscape visualization three dimensional (3D) landscape visualization 162 Tilden, F. 221 Tobi, Hilde xiii, 1–8, 11–23, 24–34, 85–102 Tractatus Logico-Philosophicus 43 trans-disciplinary 3, 58, 121, 158, 207 transformation. See research through designing Twitter 145, 146t, 154 UCINET 154 UHI. See urban heat island ‘Under the Sky’ 106 UNESCO. See United Nations Educational, Scientific and Cultural Organization United Kingdom: Alport Valley case study 169–75; emerging themes identified in 78–9, 79t; National Institute for Health and Care

„

Excellence 250; research publication 65–8 United Nations Educational, Scientific and Cultural Organization (UNESCO): World Heritage Area 112, 122, 131, 232 urban heat island (UHI) 274t urbanism 56–7 urban river spaces 196–9, 198f, 199f urban water, challenge of 285–306; case studies 289–303, 300f, 301f; conceptual framework 286–9, 289f; design process 287, 287f; landscape based stormwater management design projects 290–3, 291t, 292t; landscape based stormwater management strategy in Copenhagen 291t, 293–6, 295f; overview 285–6; Vanløse School design experiment 291t, 296–7, 297f, 298t Uzzell, D. 221 van den Brink, Adri xiii, 1–8, 11–23, 24–34, 54–64, 85–102 van Etteger, Rudi xiii, 179–93 van Lammeren, Ron xiii, 136–60 Vanløse School design experiment 291t, 296–7, 297f, 298t vegetation 11, 124, 164, 170, 199–200, 205, 230, 252–4, 267, 274–5, 277, 290. See also plants vernacular 38, 215 virtual environments 161–78; Alport Valley case study 169–75; comparison of dynamic movement and static representation 163–75; conceptal framework of 162–3; history in landscape architecture 162; models 165f; overview 161–2; research question and hypothesis 163–4; space and 163; three dimensional landscape visualization 161; Zürich case study 164–9 Volunteered Geographic Information 137 von Seggern, Hille 180, 181 walking 179–93; conceptual framework 179–81; continuous/stop-motion walking 186–9, 188f, 190; engaging with the landscape 182–9; interplay of walking modes 184–5, 184f; overview 179; reflection and 189–91; ‘strollology’ experiments 180; wandering-method 183–6, 184f water. See urban water, challenge of water dynamics 292t water sensitive urban design (WSUD) 286. See also landscape based stormwater management WebGIS platform 142 Weddle, Arnold 66

314 …

Index „

Whyte, William 182 wildlife 264 Williams, Raymond 47 Wilson, E.O. 43 wind 276–9, 276f; estimating 277; measurement of 276–7 Wingecarribee Shire, New South Wales, Australia 224–31; context 225–6; documentation, assessment, and analysis 228–31, 229f; location of 225f; paddock 230f, 231; Portion Plans 228; study method 226–8, 227f WordStat 154

World Heritage Area 112, 122, 131, 232 World Meteorological Organization 263 WSUD. See water sensitive urban design Wylie, J. 215 Zeisel, John 57 Zürich case study 164–9; dynamic walkthroughs 164, 166f; results 167–8, 167f; static images 164, 165f; study design and protocol 166–7; study participants 164–6; visual stimuli 164

315 …