131 95 27MB
English Pages 300 [321] Year 2013
Air Pollution XXI
WIT Press publishes leading books in Science and Technology. Visit our website for the current list of titles. www.witpress.com
WITeLibrary
Home of the Transactions of the Wessex Institute. Papers presented at Air Pollution XXI are archived in the WIT eLibrary in volume 174 of WIT Transactions on Ecology and the Environment (ISSN 1743-3541). The WIT eLibrary provides the international scientific community with immediate and permanent access to individual papers presented at WIT conferences. http://library.witpress.com
Twenty-first International Conference on Modelling, Monitoring and Management of Air Pollution
AIR POLLUTION XXI Conference Chairmen J.W.S. Longhurst
University of the West of England, UK
C.A. Brebbia
Wessex Institute of Technology, UK
International Scientific Advisory Committee C. Borrego U. Franck M. Kastek M. Khare M. Lopes S. Nagendra F. Patania S. Pongpiachan A. Seroji C. Trozzi Y. Yu P. Zannetti
Organised by
Wessex Institute of Technology, UK University of the West of England, UK University of Siena, Italy
Sponsored by
WIT Transactions on Ecology and the Environment International Journal of Sustainable Development and Planning
WIT Transactions Transactions Editor Carlos Brebbia Wessex Institute of Technology Ashurst Lodge, Ashurst Southampton SO40 7AA, UK
Editorial Board B Abersek University of Maribor, Slovenia Y N Abousleiman University of Oklahoma,
G Belingardi Politecnico di Torino, Italy R Belmans Katholieke Universiteit Leuven,
P L Aguilar University of Extremadura, Spain K S Al Jabri Sultan Qaboos University, Oman E Alarcon Universidad Politecnica de Madrid,
C D Bertram The University of New South
USA
Spain
A Aldama IMTA, Mexico C Alessandri Universita di Ferrara, Italy D Almorza Gomar University of Cadiz, Spain B Alzahabi Kettering University, USA J A C Ambrosio IDMEC, Portugal A M Amer Cairo University, Egypt S A Anagnostopoulos University of Patras, Greece
M Andretta Montecatini, Italy E Angelino A.R.P.A. Lombardia, Italy H Antes Technische Universitat Braunschweig, Germany
M A Atherton South Bank University, UK A G Atkins University of Reading, UK D Aubry Ecole Centrale de Paris, France J Augutis Vytautas Magnus University, Lithuania
H Azegami Toyohashi University of Technology, Japan
A F M Azevedo University of Porto, Portugal J Baish Bucknell University, USA J M Baldasano Universitat Politecnica de Catalunya, Spain J G Bartzis Institute of Nuclear Technology, Greece S Basbas Aristotle University of Thessaloniki, Greece A Bejan Duke University, USA M P Bekakos Democritus University of Thrace, Greece
Belgium
Wales, Australia
D E Beskos University of Patras, Greece S K Bhattacharyya Indian Institute of Technology, India
E Blums Latvian Academy of Sciences, Latvia J Boarder Cartref Consulting Systems, UK B Bobee Institut National de la Recherche Scientifique, Canada
H Boileau ESIGEC, France J J Bommer Imperial College London, UK M Bonnet Ecole Polytechnique, France C A Borrego University of Aveiro, Portugal A R Bretones University of Granada, Spain J A Bryant University of Exeter, UK F-G Buchholz Universitat Gesanthochschule Paderborn, Germany
M B Bush The University of Western Australia, Australia
F Butera Politecnico di Milano, Italy W Cantwell Liverpool University, UK D J Cartwright Bucknell University, USA P G Carydis National Technical University of Athens, Greece
J J Casares Long Universidad de Santiago de Compostela, Spain
M A Celia Princeton University, USA A Chakrabarti Indian Institute of Science, India
J-T Chen National Taiwan Ocean University, Taiwan
A H-D Cheng University of Mississippi, USA J Chilton University of Lincoln, UK
C-L Chiu University of Pittsburgh, USA H Choi Kangnung National University, Korea A Cieslak Technical University of Lodz, Poland S Clement Transport System Centre, Australia M W Collins Brunel University, UK J J Connor Massachusetts Institute of Technology, USA
M C Constantinou State University of New York at Buffalo, USA
D E Cormack University of Toronto, Canada M Costantino Royal Bank of Scotland, UK D F Cutler Royal Botanic Gardens, UK W Czyczula Krakow University of Technology, Poland
M da Conceicao Cunha University of Coimbra, Portugal
L Dávid Károly Róbert College, Hungary A Davies University of Hertfordshire, UK M Davis Temple University, USA A B de Almeida Instituto Superior Tecnico, Portugal
E R de Arantes e Oliveira Instituto Superior Tecnico, Portugal
L De Biase University of Milan, Italy R de Borst Delft University of Technology, Netherlands
K Dorow Pacific Northwest National Laboratory, USA
W Dover University College London, UK C Dowlen South Bank University, UK J P du Plessis University of Stellenbosch, South Africa
R Duffell University of Hertfordshire, UK N A Dumont PUC-Rio, Brazil A Ebel University of Cologne, Germany E E Edoutos Democritus University of Thrace, Greece
G K Egan Monash University, Australia K M Elawadly Alexandria University, Egypt K-H Elmer Universitat Hannover, Germany D Elms University of Canterbury, New Zealand M E M El-Sayed Kettering University, USA D M Elsom Oxford Brookes University, UK F Erdogan Lehigh University, USA D J Evans Nottingham Trent University, UK J W Everett Rowan University, USA M Faghri University of Rhode Island, USA R A Falconer Cardiff University, UK M N Fardis University of Patras, Greece P Fedelinski Silesian Technical University, Poland
H J S Fernando Arizona State University, USA
G De Mey University of Ghent, Belgium
S Finger Carnegie Mellon University, USA
A De Montis Universita di Cagliari, Italy
E M M Fonseca Instituto Politécnico de
A De Naeyer Universiteit Ghent, Belgium W P De Wilde Vrije Universiteit Brussel, Belgium
D De Wrachien State University of Milan, Italy
L Debnath University of Texas-Pan American, USA
G Degrande Katholieke Universiteit Leuven, Belgium
E del Giudice University of Milan, Italy S del Giudice University of Udine, Italy G Deplano Universita di Cagliari, Italy I Doltsinis University of Stuttgart, Germany M Domaszewski Universite de Technologie de Belfort-Montbeliard, France
J Dominguez University of Seville, Spain
Bragança, Portugal
J I Frankel University of Tennessee, USA D M Fraser University of Cape Town, South Africa M J Fritzler University of Calgary, Canada T Futagami Hiroshima Institute of Technology, Japan
U Gabbert Otto-von-Guericke Universitat Magdeburg, Germany
G Gambolati Universita di Padova, Italy C J Gantes National Technical University of Athens, Greece
L Gaul Universitat Stuttgart, Germany A Genco University of Palermo, Italy N Georgantzis Universitat Jaume I, Spain P Giudici Universita di Pavia, Italy L M C Godinho University of Coimbra, Portugal
F Gomez Universidad Politecnica de Valencia, Spain
R Gomez Martin University of Granada, Spain
D Goulias University of Maryland, USA K G Goulias Pennsylvania State University, USA
F Grandori Politecnico di Milano, Italy W E Grant Texas A & M University, USA
S Grilli University of Rhode Island, USA R H J Grimshaw Loughborough University, UK
D Gross Technische Hochschule Darmstadt, Germany
R Grundmann Technische Universitat Dresden, Germany
A Gualtierotti IDHEAP, Switzerland O T Gudmestad University of Stavanger, Norway
Y Jaluria Rutgers University, USA C M Jefferson University of the West of England, UK
M K Jha Morgan State University, USA P R Johnston Griffith University, Australia D R H Jones University of Cambridge, UK N Jones University of Liverpool, UK N Jovanovic CSIR, South Africa D Kaliampakos National Technical University of Athens, Greece
N Kamiya Nagoya University, Japan D L Karabalis University of Patras, Greece A Karageorghis University of Cyprus M Karlsson Linkoping University, Sweden T Katayama Doshisha University, Japan K L Katsifarakis Aristotle University of Thessaloniki, Greece
J T Katsikadelis National Technical University of Athens, Greece
R C Gupta National University of Singapore,
E Kausel Massachusetts Institute of
J M Hale University of Newcastle, UK K Hameyer Katholieke Universiteit Leuven,
H Kawashima The University of Tokyo, Japan B A Kazimee Washington State University,
C Hanke Danish Technical University,
S Kim University of Wisconsin-Madison, USA D Kirkland Nicholas Grimshaw & Partners
Singapore
Belgium
Denmark
K Hayami University of Toyko, Japan Y Hayashi Nagoya University, Japan L Haydock Newage International Limited, UK A H Hendrickx Free University of Brussels, Belgium
C Herman John Hopkins University, USA I Hideaki Nagoya University, Japan D A Hills University of Oxford, UK W F Huebner Southwest Research Institute, USA
J A C Humphrey Bucknell University, USA M Y Hussaini Florida State University, USA W Hutchinson Edith Cowan University, Australia
T H Hyde University of Nottingham, UK M Iguchi Science University of Tokyo, Japan D B Ingham University of Leeds, UK L Int Panis VITO Expertisecentrum IMS, Belgium
N Ishikawa National Defence Academy, Japan J Jaafar UiTm, Malaysia W Jager Technical University of Dresden, Germany
Technology, USA
USA
Ltd, UK
E Kita Nagoya University, Japan A S Kobayashi University of Washington, USA
T Kobayashi University of Tokyo, Japan D Koga Saga University, Japan S Kotake University of Tokyo, Japan A N Kounadis National Technical University of Athens, Greece
W B Kratzig Ruhr Universitat Bochum, Germany
T Krauthammer Penn State University, USA C-H Lai University of Greenwich, UK M Langseth Norwegian University of Science and Technology, Norway
B S Larsen Technical University of Denmark, Denmark
F Lattarulo Politecnico di Bari, Italy A Lebedev Moscow State University, Russia L J Leon University of Montreal, Canada D Lesnic University of Leeds, UK D Lewis Mississippi State University, USA S lghobashi University of California Irvine, USA
K-C Lin University of New Brunswick, Canada A A Liolios Democritus University of Thrace, Greece
S Lomov Katholieke Universiteit Leuven, Belgium
J W S Longhurst University of the West of
England, UK G Loo The University of Auckland, New Zealand J Lourenco Universidade do Minho, Portugal J E Luco University of California at San Diego, USA H Lui State Seismological Bureau Harbin, China C J Lumsden University of Toronto, Canada L Lundqvist Division of Transport and Location Analysis, Sweden T Lyons Murdoch University, Australia Y-W Mai University of Sydney, Australia M Majowiecki University of Bologna, Italy D Malerba Università degli Studi di Bari, Italy G Manara University of Pisa, Italy S Mambretti Politecnico di Milano, Italy B N Mandal Indian Statistical Institute, India Ü Mander University of Tartu, Estonia H A Mang Technische Universitat Wien, Austria G D Manolis Aristotle University of Thessaloniki, Greece W J Mansur COPPE/UFRJ, Brazil N Marchettini University of Siena, Italy J D M Marsh Griffith University, Australia J F Martin-Duque Universidad Complutense, Spain T Matsui Nagoya University, Japan G Mattrisch DaimlerChrysler AG, Germany F M Mazzolani University of Naples “Federico II”, Italy K McManis University of New Orleans, USA A C Mendes Universidade de Beira Interior, Portugal R A Meric Research Institute for Basic Sciences, Turkey J Mikielewicz Polish Academy of Sciences, Poland N Milic-Frayling Microsoft Research Ltd, UK R A W Mines University of Liverpool, UK C A Mitchell University of Sydney, Australia K Miura Kajima Corporation, Japan A Miyamoto Yamaguchi University, Japan
T Miyoshi Kobe University, Japan G Molinari University of Genoa, Italy T B Moodie University of Alberta, Canada D B Murray Trinity College Dublin, Ireland G Nakhaeizadeh DaimlerChrysler AG, Germany
M B Neace Mercer University, USA D Necsulescu University of Ottawa, Canada F Neumann University of Vienna, Austria S-I Nishida Saga University, Japan H Nisitani Kyushu Sangyo University, Japan B Notaros University of Massachusetts, USA P O’Donoghue University College Dublin, Ireland
R O O’Neill Oak Ridge National Laboratory, USA
M Ohkusu Kyushu University, Japan G Oliveto Universitá di Catania, Italy R Olsen Camp Dresser & McKee Inc., USA E Oñate Universitat Politecnica de Catalunya, Spain
K Onishi Ibaraki University, Japan P H Oosthuizen Queens University, Canada E L Ortiz Imperial College London, UK E Outa Waseda University, Japan A S Papageorgiou Rensselaer Polytechnic Institute, USA
J Park Seoul National University, Korea G Passerini Universita delle Marche, Italy F Patania University of Catania, Italy B C Patten University of Georgia, USA G Pelosi University of Florence, Italy G G Penelis Aristotle University of Thessaloniki, Greece
W Perrie Bedford Institute of Oceanography, Canada
R Pietrabissa Politecnico di Milano, Italy H Pina Instituto Superior Tecnico, Portugal M F Platzer Naval Postgraduate School, USA D Poljak University of Split, Croatia V Popov Wessex Institute of Technology, UK H Power University of Nottingham, UK D Prandle Proudman Oceanographic Laboratory, UK
M Predeleanu University Paris VI, France I S Putra Institute of Technology Bandung, Indonesia
Y A Pykh Russian Academy of Sciences, Russia
F Rachidi EMC Group, Switzerland M Rahman Dalhousie University, Canada K R Rajagopal Texas A & M University, USA T Rang Tallinn Technical University, Estonia J Rao Case Western Reserve University, USA J Ravnik University of Maribor, Slovenia A M Reinhorn State University of New York at Buffalo, USA G Reniers Universiteit Antwerpen, Belgium A D Rey McGill University, Canada D N Riahi University of Illinois at UrbanaChampaign, USA B Ribas Spanish National Centre for Environmental Health, Spain K Richter Graz University of Technology, Austria S Rinaldi Politecnico di Milano, Italy F Robuste Universitat Politecnica de Catalunya, Spain J Roddick Flinders University, Australia A C Rodrigues Universidade Nova de Lisboa, Portugal F Rodrigues Poly Institute of Porto, Portugal C W Roeder University of Washington, USA J M Roesset Texas A & M University, USA W Roetzel Universitaet der Bundeswehr Hamburg, Germany V Roje University of Split, Croatia R Rosset Laboratoire d’Aerologie, France J L Rubio Centro de Investigaciones sobre Desertificacion, Spain T J Rudolphi Iowa State University, USA S Russenchuck Magnet Group, Switzerland H Ryssel Fraunhofer Institut Integrierte Schaltungen, Germany S G Saad American University in Cairo, Egypt M Saiidi University of Nevada-Reno, USA R San Jose Technical University of Madrid, Spain F J Sanchez-Sesma Instituto Mexicano del Petroleo, Mexico B Sarler Nova Gorica Polytechnic, Slovenia S A Savidis Technische Universitat Berlin, Germany A Savini Universita de Pavia, Italy G Schmid Ruhr-Universitat Bochum, Germany R Schmidt RWTH Aachen, Germany B Scholtes Universitaet of Kassel, Germany W Schreiber University of Alabama, USA
A P S Selvadurai McGill University, Canada J J Sendra University of Seville, Spain J J Sharp Memorial University of Newfoundland, Canada
Q Shen Massachusetts Institute of Technology, USA
X Shixiong Fudan University, China G C Sih Lehigh University, USA L C Simoes University of Coimbra, Portugal A C Singhal Arizona State University, USA P Skerget University of Maribor, Slovenia J Sladek Slovak Academy of Sciences, Slovakia
V Sladek Slovak Academy of Sciences, Slovakia
A C M Sousa University of New Brunswick, Canada
H Sozer Illinois Institute of Technology, USA D B Spalding CHAM, UK P D Spanos Rice University, USA T Speck Albert-Ludwigs-Universitaet Freiburg, Germany
C C Spyrakos National Technical University of Athens, Greece
I V Stangeeva St Petersburg University, Russia J Stasiek Technical University of Gdansk, Poland
G E Swaters University of Alberta, Canada S Syngellakis Wessex Institute of Technology, UK
J Szmyd University of Mining and Metallurgy, Poland
S T Tadano Hokkaido University, Japan H Takemiya Okayama University, Japan I Takewaki Kyoto University, Japan C-L Tan Carleton University, Canada E Taniguchi Kyoto University, Japan S Tanimura Aichi University of Technology, Japan
J L Tassoulas University of Texas at Austin, USA
M A P Taylor University of South Australia, Australia
A Terranova Politecnico di Milano, Italy A G Tijhuis Technische Universiteit Eindhoven, Netherlands
T Tirabassi Institute FISBAT-CNR, Italy S Tkachenko Otto-von-Guericke-University, Germany
N Tosaka Nihon University, Japan
T Tran-Cong University of Southern
Queensland, Australia R Tremblay Ecole Polytechnique, Canada I Tsukrov University of New Hampshire, USA R Turra CINECA Interuniversity Computing Centre, Italy S G Tushinski Moscow State University, Russia J-L Uso Universitat Jaume I, Spain E Van den Bulck Katholieke Universiteit Leuven, Belgium D Van den Poel Ghent University, Belgium R van der Heijden Radboud University, Netherlands R van Duin Delft University of Technology, Netherlands P Vas University of Aberdeen, UK R Verhoeven Ghent University, Belgium A Viguri Universitat Jaume I, Spain Y Villacampa Esteve Universidad de Alicante, Spain F F V Vincent University of Bath, UK S Walker Imperial College, UK G Walters University of Exeter, UK B Weiss University of Vienna, Austria
H Westphal University of Magdeburg, Germany
J R Whiteman Brunel University, UK T W Wu University of Kentucky, USA Z-Y Yan Peking University, China S Yanniotis Agricultural University of Athens, Greece
A Yeh University of Hong Kong, China B W Yeigh SUNY Institute of Technology, USA
J Yoon Old Dominion University, USA K Yoshizato Hiroshima University, Japan T X Yu Hong Kong University of Science & Technology, Hong Kong
M Zador Technical University of Budapest, Hungary
K Zakrzewski Politechnika Lodzka, Poland M Zamir University of Western Ontario, Canada
G Zappalà CNR-IAMC, Italy R Zarnic University of Ljubljana, Slovenia G Zharkova Institute of Theoretical and Applied Mechanics, Russia
N Zhong Maebashi Institute of Technology, Japan
H G Zimmermann Siemens AG, Germany
Air Pollution XXI
Editors J.W.S. Longhurst
University of the West of England, UK
C.A. Brebbia
Wessex Institute of Technology, UK
Editors J.W.S. Longhurst University of the West of England, UK C.A. Brebbia Wessex Institute of Technology, UK
Published by WIT Press Ashurst Lodge, Ashurst, Southampton, SO40 7AA, UK Tel: 44 (0) 238 029 3223; Fax: 44 (0) 238 029 2853 E-Mail: [email protected] http://www.witpress.com For USA, Canada and Mexico WIT Press 25 Bridge Street, Billerica, MA 01821, USA Tel: 978 667 5841; Fax: 978 667 7582 E-Mail: [email protected] http://www.witpress.com British Library Cataloguing-in-Publication Data
A Catalogue record for this book is available from the British Library
ISBN: 978-1-84564-718-6 eISBN: 978-1-84564-719-3 ISSN: (print) 1746-448X ISSN: (on-line) 1743-3541 The texts of the papers in this volume were set individually by the authors or under their supervision.Only minor corrections to the text may have been carried out by the publisher. No responsibility is assumed by the Publisher, the Editors and Authors for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. The Publisher does not necessarily endorse the ideas held, or views expressed by the Editors or Authors of the material contained in its publications. © WIT Press 2013 Printed in Great Britain by Lightning Source, UK. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of the Publisher.
Preface
This volume contains the peer-reviewed papers accepted for the Twentyfirst International Conference on Modelling, Monitoring and Management of Air Pollution held in Siena, Italy in 2013 organised by the Wessex Institute of Technology and the University of the West of England. This successful international meeting builds upon the prestigious outcomes of the 20 preceding conferences beginning with Monterrey, Mexico in 1993 and most recently in Malta in 2011 and A Coruna, Spain in 2012. These meetings have attracted outstanding contributions from leading researchers from around the world. The papers selected for presentation and included in the Conference Proceedings have been permanently stored in the WIT eLibrary as Transactions of the Wessex Institute (see http://library.witpress.com). These collected papers provide an important record of the development of science and policy pertaining to air pollution. To have reached a total of twenty one successful conferences on Modelling, Monitoring and Management of Air Pollution is a significant achievement and the editors would like to thank the many colleagues in academia and practice around the world who have contributed to this series of meetings. Undoubtedly, the series has filled a significant need and the fact that the WIT series continues to grow and meet the demands of a discerning conference market is testament to the quality of the science and policy presented at the meetings. The twenty one conferences in this series have discussed and considered many important air pollution issues and the highly international nature of the attendees has ensured that of the conference findings and conclusions enjoy a wide and rapid dissemination amongst the air pollution science and policy communities. The conferences to date have concluded that despite the long history of attempts to manage the consequences of air pollution the issue remains one of the most challenging problems facing the international community. The series has demonstrated the wide spread nature of the air pollution phenomena and has explored in depth the impacts of air pollution on human health and the
environment. Conference presentations have explored the causes of air pollution from transport, manufactured goods and services and discussed the often unintended, but none the less real, impacts on the atmospheric environment at scales from the local to the global. A particular strength of the conference series has been the attention given to regulatory and, more recently, market solutions to air pollution management. Conference delegates have explored a range of regulatory successes in minimising such impacts but equally have recognised that the continuing development of the global economy bring new pressures upon the ability of the atmosphere to process pollutants and to safely dispose of them. The willingness of governmental authorities to move quickly to regulate air pollution is often balanced by concerns over the economic impact of such regulation. This frequently results in a lag between the scientific knowledge about the nature, scale and effect of air pollution and the implementation of appropriate, targeted and timely legislation. The conference series has consistently acknowledged that science remains the key to identifying the nature and scale of air pollution impacts and reaffirmed that science is essential in the formulation of policy for regulatory decisionmaking. The series also recognised, at a very early stage, that science alone will not improve a polluted atmosphere. The scientific knowledge derived from well designed studies needs to be allied with further technical and economic studies in order to ensure cost effective and efficient mitigation. In turn, the science, technology and economic outcomes are necessary but not sufficient. Increasingly, the conference series has recognised that the outcome of such research need to be contextualised within well formulated communication strategies that help policy makers and citizens to understand and appreciate the risks and rewards arising from air pollution management. Consequently, the series has enjoyed a wide range of high quality papers that develop the fundamental science of air pollution and an equally impressive range of presentations that places these new developments within the frame of mitigation and management of air pollution. The peer reviewed nature of the conference volumes enables policy makers to confidently use the new findings to formulate sustainable decisions and to build public acceptance and understanding of the nature and scale of the air pollution problem. This volume brings together contributions from scientists from around the world to present recent work on various aspects of the air pollution phenomena. Notable in each of the twenty preceding conferences in this series has been the opportunity to foster scientific exchange between participants. New collaborations amongst scientists and between scientists and policy makers or regulators have arisen through contacts made in this series and each meeting has provided a further opportunity for identifying new areas of air pollution science demanding collaborative investigation. Contributions in the twentyfirst volume in the Modelling, Monitoring and Management of Air Pollution
series continue to address a broad range of urgent scientific and technical developments in our understanding of the cause, consequence and management of air pollution. The Editors wish to thank the authors for their contributions and to acknowledge the assistance of the eminent members of the International Scientific Advisory Committee with the organisation of the conference and in particular for their support in peer reviewing the submitted papers. The papers in this book are archived in the eLibrary of the Wessex Institute (http://libary.witpress.com), where they are easily and permanently available to the international scientific community. J.W.S. Longhurst C.A. Brebbia Siena, 2013
This page intentionally left blank
Contents Section 1: Air pollution modelling Accident reconstruction and plume modeling of an unplanned ammonia release A. Daly, P. Zannetti & M. Jennings ..................................................................... 3 Association of weather and air pollution interactions on daily mortality in 12 Canadian cities J. K. Vanos, S. Cakmak & L. S. Kalkstein.......................................................... 15 Exhaust emissions from vehicles in real traffic conditions in the Poznan agglomeration J. Merkisz, J. Pielecha, P. Lijewski, A. Merkisz-Guranowska & M. Nowak...... 27 Evaluations of benzene impacts of a coke plant in a complex-topography urban area with the RAMS-CALMET-CALPUFF modelling system V. Valdenebro, E. Sáez de Cámara, G. Gangoiti, L. Alonso, J. A. García, J. L. Ilardia & N. González .......................................................... 39 Section 2: Monitoring and measuring Mapping anthropogenic and natural volatile organic compounds around Estarreja Chemical Industrial Complex T. Nunes, C. Poceiro, M. Evtyugina, M. Duarte, C. Borrego & M. Lopes ........ 55 Long-term trend of indoor volatile organic compounds (VOC) S. Matysik, P. Opitz & O. Herbarth ................................................................... 65 Monitoring airborne dust in an Italian basalt quarry: comparison between sampling methods G. Alfaro Degan, D. Lippiello & M. Pinzari ..................................................... 75
Identification of work-related exposure to carcinogenic substances in Germany S. Gabriel, M. Steinhausen & R. Van Gelder .................................................... 85 Ozone and nitrogen dioxide concentrations in a Holm oak urban park and an adjacent open area in Siena F. Fantozzi, F. Monaci, T. Blanusa & R. Bargagli .......................................... 103 Levels of particulate matter in Western UAE desert and factors affecting their distribution F. Al Jallad, E. Al Katheeri & M. Al Omar ..................................................... 111 Section 3: Aerosols and particles Transport of aerosols in the Mediterranean coastal zone J. Piazzola, A. Demoisson & G. Tedeschi........................................................ 125 Trace element concentrations of size-fractionated particulate matter in the atmosphere of Istanbul, Turkey Ü. Alver Şahin, B. Onat & G. Polat ................................................................. 137 PM2.5 characterisation by scanning electron microscopy (SEM) and its correlation with diverse particle emission sources Y. I. Falcón, A. M. Maubert, A. Cruz & E. A. Zavala ...................................... 149 Particle formation in the planetary boundary layer over Tokyo and its suburban areas H. Minoura ...................................................................................................... 157 Section 4: Emission studies Investigating the influence of highway traffic flow condition on pollutant emissions using driving simulators M. R. De Blasiis, M. Di Prete, C. Guattari, V. Veraldi, G. Chiatti & F. Palmieri ................................................................................................... 171 Impact of motorcycles on urban tropospheric ozone L. F. A. Garcia, S. M. Corrêa, R. Penteado, L. C. Daemme, L. V. Gatti & D. S. Alvim ................................................................................. 183 Characteristics of carbonaceous aerosols and its relationships between emission sources H. Minoura, K. Takahashi, T. Morikawa, A. Mizohata & K. Sakamoto .......... 195
The correlation of distribution of PM number emitted under actual conditions of operation by PC and HDV vehicles P. Fuc, L. Rymaniak & A. Ziolkowski .............................................................. 207 Emission tests of the F100-PW-229 turbine jet engine during pre-flight verification of the F-16 aircraft J. Merkisz, J. Markowski & J. Pielecha ........................................................... 219 Section 5: Regional studies Air quality study for Montenegro Pljevlja area C. Trozzi, S. Villa, J. Knezević, C. Leonardi, A. Pejović & R. Vaccaro .......... 233 Variability in metal deposition among industrial, rural and urban areas in the Cantabria Region (Northern Spain) M. Puente, I. Fernández-Olmo & A. Irabien ................................................... 245 Section 6: Air pollution problems in Thailand (Special session organised by S. Pongpiachan) Temporal and spatial distribution of mutagenic index in PM10 collected at Bangkok, Thailand S. Pongpiachan, C. Choochuay & C. Kositanon ............................................. 257 Chemical characterization of gaseous species from the pyrolysis process using scrap tires S. Pongnailert, T. Supadit, P. Poungsuk & S. Pongpiachan............................ 269 Enhancing public participation in air pollution management from coal-fired power plant projects in Thailand C. Chompunth .................................................................................................. 279 Application of Strategic Environmental Assessment (SEA) to alleviate air pollution and the other impacts from power plant development C. Poboon ........................................................................................................ 291 Author index .................................................................................................. 299
This page intentionally left blank
Section 1 Air pollution modelling
This page intentionally left blank
Air Pollution XXI
3
Accident reconstruction and plume modeling of an unplanned ammonia release A. Daly1, P. Zannetti1 & M. Jennings2 1
The EnviroComp Institute, Fremont, California, USA San Jose State University, San Jose, California, USA
2
Abstract This paper describes the study we performed in relation to an accidental release of ammonia at a plant in Luling, Louisiana. Our work involved the understanding of the dynamics of the accident, the evaluation of the emission scenario, the computer modeling of the ammonia plume transported by the wind, and the visualization of our results for the purpose of litigation support and presentation to a non-technical audience. The emission scenario was complicated because of a high-velocity jet release and stormy weather conditions. We used emissions and weather data available at the site, in addition to other local weather data. We also used an EPA-approved model to simulate the transport and dispersion of the ammonia cloud and calculate ground-level concentrations of ammonia over the period of concern (a 2-hour interval). We concluded that maximum hourly concentrations of ammonia, beyond the industrial fenceline in the local community, were around 0.20 ppm, which is far below existing levels of concern. However, our odor modeling calculations showed that short-term concentrations were potentially larger, and may have reached 5 ppm for very short periods. Our estimated concentrations of ammonia were validated by the locations and times of local odor complaints as well as in-situ measurements. Finally, we visualized the plume by applying a simplified version of our MONTECARLO particle model and creating a computer animation of plume concentrations, in order to illustrate the dynamics of the events to a non-technical audience. Keywords: accident reconstruction, air quality modelling, ISC3, MONTECARLO, plume modeling.
WIT Transactions on Ecology and The Environment, Vol 174, © 2013 WIT Press www.witpress.com, ISSN 1743-3541 (on-line) doi:10.2495/AIR130011
4 Air Pollution XXI
1 Introduction On September 18, 1998, at about 10:40 AM local time, there was a line rupture in the ammonia production unit at Monsanto Company’s facility in Luling, Louisiana. This rupture was quickly discovered and isolated by the plant operators, and the result was an atmospheric release of approximately 2,227 pounds of gaseous ammonia over a period of about 2 hours. The release brought a class action litigation of local residents alleging exposure to air emissions. A law firm retained our services to investigate the accident, develop a model for the release, and visualize the possible concentration impacts at ground level for ammonia. These tasks first required an estimate of the release rate of ammonia from the fractured pipe. The second step involved the computation of the dispersion of ammonia in the immediate neighbourhood of the release using specialized computer models. The US Environmental Protection Agency (EPA) has developed and tested air pollution dispersion models (e.g., ISC3 and AERMOD) that are suitable for simulating the dispersion of emissions from various types of sources and the physical phenomena associated with transport and dispersion of chemicals in the atmosphere. At the time of our original analysis, ISC3 was the EPA’s preferred dispersion model for short-range releases over flat terrain (up to 50 kilometres), and this was our chosen model. Today ISC3 is listed as “alternative model” (http://www.epa.gov/ttn/scram/dispersion_alt.htm).
2
Incident review and emission calculations
Monsanto produces ammonia by reacting Hydrogen with Nitrogen in an Ammonia Converter. At the time of the incident, a Monsanto employee was working on a line valve connected to the Converter. Subsequently, the line ruptured and the contents of the Converter were released horizontally into the atmosphere. The release was a mixture of several gases, but ammonia was the chemical of concern. Within approximately 2 minutes, Monsanto operators isolated the ammonia line from the rest of the plant, and gas from the Converter continued to flow from the broken line for about 2 hours. During that time, Monsanto personnel sprayed water on the release to wash out some of the ammonia in the gas. Due to the small pipe diameter and the high pressure inside the Converter, the ammonia gas escaped as a choked-flow horizontal jet. As a result, the gas velocity leaving the pipe was constant at the speed of sound (about 2,170 feet per second or 660 m/s) for much of the leak period [1]. Gas temperature sharply decreases during choked-flow, and in this case the estimated temperature drop is about 185ºF in the pipe [1]. However, when the gas leaves the pipe and expands, it is diluted with surrounding air and quickly reaches ambient temperature. This initial expansion to ambient temperature created a cloud with an estimated volume of 125 m3 using empirical models [2, 3]. WIT Transactions on Ecology and The Environment, Vol 174, © 2013 WIT Press www.witpress.com, ISSN 1743-3541 (on-line)
Air Pollution XXI
5
Using data provided by Monsanto and formulas for choked-flow [1, 4], the emission rate for ammonia was calculated and is summarized in Figure 1. The sharp drop in the ammonia release rate occurred after the operators isolated the ammonia line. Figure 1 shows that it took about 2 hours (124 minutes) for the system to release all of the ammonia and leak down to atmospheric pressure (i.e. 0 psig). The total ammonia release to the atmosphere was estimated to be 2,227 lbs. This is conservative (i.e. an overestimate) because some ammonia was removed by the sprayed water, and also some ammonia remained in the Converter after the system reached atmospheric pressure.
2500
250
2000
200
1500
150
1000
100
500
50
0 0
20
40
60
80
100
120
NH3 Flowrate (lb/min)
System Pressure (psig)
Figure 1 - NH3 Release Rate and System Pressure
0 140
Time (Min) System Pressure
Figure 1:
NH3 Flowrate
Ammonia release rate and system pressure.
3 Local weather conditions for air dispersion calculations September 1998 was a very warm and wet month for Louisiana. This month was Louisiana’s 7th warmest September from 1895-1998 [5] and several tropical storms and hurricanes affected the state during this time. A tropical depression that formed in the Gulf of Mexico on September 17th strengthened to tropical storm Hermine on the morning of the 18th [6]. Hermine transported warm Gulf moisture to southern Louisiana, which developed into rain and thunderstorms. The three closest available meteorological stations to Monsanto were the following: 1. New Orleans International Airport (MSY); 2. Slidell Airport (6R0); 3. Monsanto’s onsite station (MIDAS). WIT Transactions on Ecology and The Environment, Vol 174, © 2013 WIT Press www.witpress.com, ISSN 1743-3541 (on-line)
6 Air Pollution XXI Figure 2 shows the locations of these stations. Data from MSY and MIDAS characterize meteorological conditions near the surface, while data from 6R0 gives a vertical profile of the regional atmosphere. Both types of data are needed to accurately model plume dispersion. Trinity Consultants [7] collected the data from MSY and 6R0 and processed it into a format (“ISC-ready” format) suitable for plume modeling.
Figure 2:
Local meteorological stations.
Figures 3 and 4 show the wind speeds and directions, respectively, for both MSY and MIDAS stations. MIDAS recorded data every five minutes while MSY recorded data once or a few times per hour, so MIDAS provides a more complete record of local conditions. The figures show significant variations in wind over time and also between the two weather stations, which was caused by thunderstorms passing through the area. To have the best available meteorological characterization for plume modeling, we used the summary of MSY and 6R0 data from Trinity Consultants, but substituted the wind speed, wind direction, and temperature with the onsite MIDAS data from Monsanto [4]. In our opinion, this was the most accurate representation of that morning’s weather for modeling plume dispersion in the Monsanto area. Table 1 summarizes the meteorological data that were used for plume modeling. The stabilities and mixing heights were derived from MSY and 6R0 data, and the other variables came from onsite MIDAS data.
WIT Transactions on Ecology and The Environment, Vol 174, © 2013 WIT Press www.witpress.com, ISSN 1743-3541 (on-line)
Air Pollution XXI
Figure 3:
Figure 4:
Local wind speeds.
Local wind directions.
WIT Transactions on Ecology and The Environment, Vol 174, © 2013 WIT Press www.witpress.com, ISSN 1743-3541 (on-line)
7
8 Air Pollution XXI Table 1: Hour (Local Time) 1 (10:38 AM to 11:38 AM) 2 (11:38 AM to 12:38 PM)
Meteorological data used for plume modeling. Urban Rural Mixing Mixing Height (m) Height (m) (MSY(MSY-6R0) 6R0)
Wind Flow (towards) (degrees) (MIDAS)
Wind Speed (m/s) (MIDAS)
Temperature (K) (MIDAS)
Stability Class (MSY-6R0)
236
3.3
300
4
331.3
824.7
294
1.9
303
3
410.7
780.8
4 Air dispersion calculations Using the emission and meteorological data discussed above, we ran a dispersion model to simulate the atmospheric concentration of ammonia generated in the area by the accidental release. We selected the Industrial Source Complex (ISC3) model [8] and ran it using the graphical user interface ISC-AERMOD VIEW provided by Lakes Environmental Software of Ontario, Canada. This model has been tested, calibrated, and validated during numerous field experiments and has been used by scientists and organizations throughout the world. This model is very suitable for simulating point, area, or volume sources and computing the concentration generated in the downwind regions. Concentrations are simulated as a series of steady-state hourly averages. Figures 5 and 6 present the results of the 2-hour ISC3 simulation of the accidental release. Figure 5 shows the average concentration between 10:38 AM and 11:38 AM, and Figure 6 between 11:38 AM and 12:38 PM.
Figure 5:
Computed levels of ammonia concentration, 10:38 AM to 11:38 AM, September 18, 1998.
WIT Transactions on Ecology and The Environment, Vol 174, © 2013 WIT Press www.witpress.com, ISSN 1743-3541 (on-line)
Air Pollution XXI
Figure 6:
9
Computed levels of ammonia concentration, 11:38 AM to 12:38 PM, September 18, 1998.
The contour lines in the figures are defined as follows: each contour line encircles an area inside which the one-hour average concentration exceeds the corresponding contour value. The lowest contour value shown in the figures is 0.20 ppm of NH3 – a level which can be used, as discussed in the next section, as a threshold level for possible odor perception.
5 Discussion of results Simulated concentrations should be compared with appropriate Levels of Concern (LOC) to ascertain whether or not the accidental release could have caused adverse effects in the community. As a reference for the simulated concentrations, we used the Emergency Response Planning Guideline (ERPG), which has established well-accepted 1-hour average levels of concern for a large number of chemicals [9]. As shown in Figure 5, the lowest ERPG value – the ERPG-1 value of 25 ppm for NH3 – is only exceeded in a small region close to the source. In the community, outside the Monsanto’s fenceline, ammonia concentrations are 100 times (or more) lower than the lowest ERPG value. The concentrations presented in Figures 5 and 6 are 1-hour average concentrations and are suitable for comparison with 1-hour average ERPG levels. However, short-term concentrations (e.g. 1-minute average concentrations) can be very different from the values shown in Figures 5 and 6. In fact, air quality measurements and atmospheric turbulence theory show that short-term concentrations do not remain constant and generally fluctuate above and below the hourly average. Odor perceptions are caused by short-term concentrations (e.g., 1-minute average concentrations). Formulas are available in the literature to estimate the
WIT Transactions on Ecology and The Environment, Vol 174, © 2013 WIT Press www.witpress.com, ISSN 1743-3541 (on-line)
10 Air Pollution XXI maximum short-term concentration during the hour from the hourly average value (i.e., a “peak-to-mean ratio”). For example, the formula below [10] has been used by several scientists, and is described by the Commonwealth Scientific and Industrial Research Organization (CSIRO): 0.35 C T1 T1 C T2 T2
(1)
where, for example, if T2 = 60 minutes and T1 = 1 minute, the formula says that the peak 1-minute average concentration CT1 during a 60-minute period is equal to 4.2 times the 60-minute average concentration CT2. Moreover, Figure 1 shows that the initial emission rate in the first few minutes of the release is higher (2-to-6 times higher) than the hourly average emission rate. Therefore, people affected at the beginning of the incident should have experienced short-term concentrations that were 2-to-6 times higher than the average values depicted in Figures 5 and 6. In conclusion, based upon the above discussion, the hourly average concentrations should be multiplied by 4.2 and by a number between 2 and 6, in order to represent the maximum, worst-case 1-minute concentration to characterize possible odor perception. Therefore, the 1-hour average 0.20 ppm contour line in Figure 5 actually represents a potential value of 2-to-5 ppm of ammonia, for a maximum, worst-case 1-minute exposure – a value at which sensitive people may experience an odor sensation. Figure 7 shows the modeling results for the first hour (10:38 AM–11:38 AM) along with plaintiff and complaint locations, which include the Boutte Christian
Figure 7:
Modeling results for the first hour (10:38 AM–11:38 AM) along with plaintiff and complaint locations.
WIT Transactions on Ecology and The Environment, Vol 174, © 2013 WIT Press www.witpress.com, ISSN 1743-3541 (on-line)
Air Pollution XXI
11
Academy (BCA) and the Credit Union (CU). The concentrations computed by the ISC3 model are in agreement with the record of complaints made to Monsanto on the day of the incident and NH3 monitoring performed by Monsanto technicians. In fact, a complaint at the Boutte Christian Academy (BCA) mentions an ammonia odor from 11:10 AM to 11:20 AM. The BCA location is at the plume centerline 2.5 miles away from the release. The wind is blowing at about 7 miles per hour and therefore this complaint reasonably matches the modeling results in location and time. It should be noted that the wind direction is blowing directly from the source of the release toward the BCA location from 10:38 AM to 12:10 PM. Yet, the odor complaint ends at 11:20 AM. This confirms that only at the beginning of the incident (e.g., the first 10 minutes) does the ammonia emission have enough strength to cause a small odor perception in the community. The fact that ammonia concentrations are low at BCA is further confirmed by the Drager Tube sample collected at 11:15 AM showing a non-zero concentration of ammonia below 5 ppm. This sample was taken at the center of the plume and yet the concentration is practically negligible. A faint odor of ammonia was also reported at 11:15 AM by a person parked at the Credit Union (CU). Again, a non-zero concentration of ammonia below 5 ppm was measured at the site using a Drager Tube. As shown in Figure 7, these two locations (BCA and CU) are directly downwind the release and, therefore, are the locations in the community where the maximum concentration impact is expected. The lack of other contemporaneous odor complaints and the non-zero measurements taken at other locations confirm that the plume trajectory affected a very limited area, in agreement with the modeling results. During the second hour of the release, the wind direction varies. However, the high-resolution 5-minute wind data collected at the Monsanto MIDAS station indicate that the wind flow remained almost constant (blowing from the location of the release directly towards BCA and CU) until 12:10 PM. After this time the emission from the release is practically negligible, as shown in Figure 1. Therefore, both the air pollution modeling calculations and the evidence of contemporaneous odor complaints and ammonia measurements show that the only plume impact (a short, faint ammonia odor) in the community was experienced around BCA and CU between 11:10 AM and 11:20 AM, at the time and location where the initial, larger release of the first few minutes (e.g., 10:38 AM to 10:48 AM) is expected to have caused the highest impact. Along with our ISC3 modeling results, we produced a MONTECARLO [11] animation showing how the ammonia plume traveled from the line rupture to the community in real time. Figure 8 shows an animation frame of the simulated particles. This animation used the onsite MIDAS wind data and agrees closely with the timing and locations of odor complaints in the community. The full animation, with different colors representing concentrations, can be examined at http://envirocomp.com/caps/projects/monsanto/v1.html.
WIT Transactions on Ecology and The Environment, Vol 174, © 2013 WIT Press www.witpress.com, ISSN 1743-3541 (on-line)
12 Air Pollution XXI
Figure 8:
Modeling results using MONTECARLO particles for the plume (animation frame).
6 Summary and conclusions About 2,227 pounds of gaseous ammonia were released from the Monsanto Plant in Luling, Louisiana from 10:38 AM to 12:38 AM on September 18, 1998. This gaseous release was simulated by the ISC3 dispersion model after calculating an emission timeline and identifying local weather conditions. We found that the concentration levels of ammonia in the community had an hourly concentration of 0.20 ppm or less, with a short-term peak of possibly 5 ppm. These values are well below any published level of concern for human health. Based on our simulation results, local measurements, and odor complaints, we concluded that a relatively small number of people could be affected by the plume and these people may have experienced a brief, faint odor sensation during the incident.
References [1] Perry, R.H. and Green, D.W., Perry’s Chemical Engineers’ Handbook, 7th Edition, McGraw-Hill: 1997. [2] Ignatius, T. J. and Rathakrishnan, E., Similarity Scales for Free and Impinging Jet Flow Fields. High Speed Jet Flows, FED-Vol. 214, ASME: 1995. [3] Panda, J. and Seasholtz, R. G., Velocity and Temperature Measurement in Supersonic Free Jets Using Spectrally Resolved Rayleigh Scattering, NASA TM 2003-212391, 2003. [4] Monsanto Corporation Counsel, Personal Communications, 2003-2004.
WIT Transactions on Ecology and The Environment, Vol 174, © 2013 WIT Press www.witpress.com, ISSN 1743-3541 (on-line)
Air Pollution XXI
13
[5] NOAA National Climatic Data Center Temperature Map Online. http://www.ncdc.noaa.gov/img/climate/research/1998/sep/SepmapC_Pg.gif NCDC Storm Record Details Online. http://www.ncdc.noaa.gov /stormevents/ftp.jsp Trinity Consultants, Personal Communication, 2003–2004. http://www.trinityconsultants.com [6] ISC3 Model Information Online. http://www.epa.gov/ttn/scram /dispersion_alt.htm#isc3 [7] ERPG Values Online. http://www.stb07.com/technical-safety/emergencyresponse-planning-guidelines.html [8] CSIRO Air Quality Peak-to-Mean Calculator Online. http://www.cmar. csiro.au/airquality/peaktomean.html [9] Zannetti, P. and Sire, R. MONTECARLO – A New, Fully-Integrated PC Software for the 3D Simulation and Visualization of Air Pollution Dispersion Using Monte Carlo Lagrangian Particle (MCLP) Techniques. Air Pollution 99, Stanford, California. WIT Publications: Ashurst, UK, July 1999. http://www.envirocomp.com/zcv/P.47.pdf
WIT Transactions on Ecology and The Environment, Vol 174, © 2013 WIT Press www.witpress.com, ISSN 1743-3541 (on-line)
This page intentionally left blank
Air Pollution XXI
15
Association of weather and air pollution interactions on daily mortality in 12 Canadian cities J. K. Vanos1, S. Cakmak1 & L. S. Kalkstein2 1
Environmental Health Research Bureau, Population Studies Division, Health Canada, Ottawa, Canada 2 Department of Geography and Regional Studies, University of Miami, USA
Abstract The overall composition of the troposphere, including meteorological attributes and air pollution concentrations, affect human health outcomes. Our objectives were first to determine the likelihood of extreme air pollution events in select weather types within 12 Canadian cities, and second to examine the association between daily mortality and daily concentrations of air pollutants, assessing both single- and two-pollutant interactions. Each pollutant is examined to determine the likelihood of an extreme air pollution episode occurring in a given weather type and city. Next, we use a distributed lag nonlinear model (DLNM) to estimate city-specific relative risks of mortality (RR) due to the single and twoway-interactive effects of nitrogen dioxide (NO2), ozone (O3), sulphur dioxide (SO2), and particulate matter 1.0 indicates that an extreme pollution episode is more likely to occur in the given weather type. A ratio > 2.0 indicates statistical significance of this likelihood [3]. 2.4.2 Distributed lag nonlinear modelling For each city and weather type, the RR due to exposure to each pollutant is modelled. In addition, the effects of exposure to two pollutants are modelled to determine the modifying effect on mortality using a DLNM [15]. We apply lags of 0–6 days for each pollutant examined to estimate the RR due to exposure in a single day (lag 0), and multiple lag days. We adjust for various time confounders using natural splines, accounting for serial autocorrelation. To do so, we use a categorical variable for day of week, and apply a natural cubic spline of time with one knot at each of 28, 56, 112, and 336 days of observation for annual, biannual, seasonal, and monthly time effects; therefore, the degrees of freedom (df) is 4. The DLNM is adjusted for mean air temperature, rather than minimum or maximum, as preliminary analysis demonstrated enhanced model strength based on model prediction values (or AIC). This was also found to be true based on AIC testing by Curriero et al. [5] in 11 U.S. cities, with further studies [9, 11] also using mean air temperature based on extensive work, and thereby better accounting for what people experience throughout the full day, rather than at one time (e.g., minimum or maximum temperature). Separate model runs are completed for exposure to single air pollutants, as well as adjustments for the single pollutant due to simultaneous exposure to the remaining three pollutants. Therefore, in total, for each weather type assessed for a city, we complete 28 regressions (4 knots, 7 lags) for single-pollutant exposure analysis to reflect independent contributions to raw mortality. An additional 84 regressions per pollutant, weather type, and city, are assessed based on all possible two-way pollutant interactions. From this, we select the optimal single or lagged model with the number of knots that either minimize the AIC, or maximize the evidence that the residuals do not display any type of structure. Finally, a pooled estimate for all 12 cities is generated across each synoptic weather type using a random-effects model, with 95% CI.
3 Results Across Canada, the prevalence of the mild and benign DM weather type in the summer season is the most prevalent (Table 1), with DT being the least common WIT Transactions on Ecology and The Environment, Vol 174, © 2013 WIT Press www.witpress.com, ISSN 1743-3541 (on-line)
Air Pollution XXI
19
weather type. The DT weather type has the highest mean air temperature coinciding with the highest air pollution concentrations of NO2, O3, and SO2. Moist tropical plus presents the second highest air temperatures, with the highest concentrations of PM2.5 in all cities experiencing this weather type. MT+ overnight temperatures are generally the highest of all the summer air masses. The extreme ambient conditions in the MT (and its extreme subset, MT+) weather type, and the associated high overnight temperatures, are responsible for its association with consistently elevated relative mortality, which was highest in 10 of the 12 cities. In addition, MT+ has the highest overall mean standardized mortality for all cities (1.92). The DT and MT weather types have an average mortality rate of 1.75 deaths per 100,000 people. The most benign weather types studied here are found to be DM and MM, which display the lowest standardized mortality and air temperatures, and have the lowest and comparable concentrations of all pollutants. Table 1:
Descriptive summertime statistics reporting the averagea for 12 Canadian urban areas in each of the five studied weather types (DM, DT, MM, MT, MT+).
SD
Ta (oC)
Windspeed (km hr-1)
Dew Point (oC)
Cloud Cover (10ths)
1.68
0.49
24.38
8.1
12.3
5.0
1.75
0.6
30.89
6.0
14
3.8
19.1
1.67
0.46
20.6
7.5
16.5
8.0
11.5
1.75
0.47
27.12
7.4
18.4
6.5
2.3
1.92
0.41
29.51
7.0
20.1
7.0
Weather Type
Freq (%)
Relative Mortalityb
DM
29.8
DT
2.5
MM MT MT+ a
4:00pm averages for each variable. b Mortality rate per 100,000 people, calculated based on yearly population.
3.1 Extreme air pollution We have identified the synoptic conditions that when present, have a higher likelihood of being associated with extreme pollution episodes that have been shown to harm human health [3] (Table 2). Results for all cities combined demonstrate that on average, extreme air pollution episodes (defined as elevated levels of NO2, O3, SO2 and PM2.5) are most likely to occur in the DT weather type, with relative likelihood ratios of 2.1, 6.0, 2.2, and 4.2, respectively. Significant combined city results are also prevalent in hot, humid weather. For MT, extreme pollution episodes due to O3 or PM2.5 are almost three times more likely than normal. For MT+, the resulting in ratios are 3.8 for O3 and 2.8 for PM2.5. Overall, 75% and 67% of the cities studied experience a statistically significant likelihood of extreme episodes of O3 and PM2.5 pollution, respectively, under the DT weather type.
WIT Transactions on Ecology and The Environment, Vol 174, © 2013 WIT Press www.witpress.com, ISSN 1743-3541 (on-line)
20 Air Pollution XXI Table 2:
Summertime (JJA) synoptic weather types per city reported based on frequency (%), and likelihood ratio to result in extreme pollution episodes. Select weather types chosen based on high ratios and weather type presence.
City
Calgary
Category
b
Edmontonb
Halifax
Montreal
Ottawa
Quebec City
Regina St. John'sc
Toronto Vancouverb,c Windsor
Winnipeg a
Pollutant Ratioa
Freq (%)
NO2
O3
SO2
PM2.5
DT
3.2
1.93
2.17
2.21
5.56
MT
0.5
1.54
1.54
3.13
3.95
DT
1.1
1.43
8.57
1.47
5.48
MT
0.7
0.00
2.35
0.00
6.01 5.73
DT
0.6
1.82
1.66
1.74
MT+
1.2
1.29
0.74
1.54
2.55
DT
1.3
1.82
10.91
2.43
8.49
MT+
4.4
0.89
2.83
0.53
3.97
DT
3.7
2.37
5.47
2.32
1.13
MT+
2.4
0.68
1.67
0.33
1.79
DT
0.7
2.88
5.23
2.82
NA
MT
16.1
1.07
2.3
0.58
2.95
MT+
2.9
0.3
3.04
0
3.19 2.07
DT
4.0
3.04
4.84
1.11
MT+
1.4
1.47
4.27
0.00
2.0
MT
6.9
0.5
2.01
0.56
2.86
MT+
1.3
0.00
0.98
0.00
3.09
DT
6.2
1.63
7.52
2.76
5.43
MT+
3.5
0.66
1.32
0.66
3.93
MT
0.7
3.15
13.64
2.1
5.88
DT
4.7
2.4
6.18
3.12
3.73
MT+
7.3
0.22
0.65
0.11
1.82
DT
1.9
1.63
7.86
1.88
0.0
MT+
2.9
0.00
2.35
0.00
2.5
Ratio = (% of days within the top 5% level of pollution):(overall \% of occurrence of the weather type in JJA). A ratio > 2.0 (bolded) identifies those synoptic categories where the occurrence of an extreme pollution episode is statistically significantly more likely to occur, being greater than expected [3]. bMT+ weather type not present. c DT weather type not present.
3.2 DLNM modelling results The pooled single- and two-pollutant adjusted models (95% CI) for each weather type are presented in Figure 1. Single-pollutant model results (Table 3) suggest a substantial health burden due to air pollution for all combinations of pollutants and weather types tested. We found the strongest risk for pollution-related mortality to be, on average, 1–3 days after exposure during warm-hot weather WIT Transactions on Ecology and The Environment, Vol 174, © 2013 WIT Press www.witpress.com, ISSN 1743-3541 (on-line)
Air Pollution XXI
Table 3:
Weather Type DM
DT
MM
MT
MT+
21
Relative risk of mortality (RR) and 95% CI associated with single pollutant models, calculated at pooled population weighted means (PWM), with standard error and average time-lag of all cities. Pollutant
PWM
Lag (d)
SE
RR
95% CI
NO2
13.25
2
0.000
1.041†
(1.032-1.051)
O3
12.58
2
0.000
1.032†
(1.023-1.041)
PM2.5
12.99
1
0.001
1.050†
(1.035-1.065)
SO2
12.05
2
0.001
1.059†
(1.042-1.076)
NO2
12.01
2
0.001
1.067†
(1.035-1.100)
O3
12.14
2
0.001
1.064†
(1.033-1.096)
PM2.5
13.00
2
0.004
1.191†
(1.076-1.319)
SO2
12.42
3
0.008
1.272†
(1.057-1.531)
NO2
12.59
1
0.000
1.036†
(1.024-1.049)
O3
11.99
3
0.000
1.036†
(1.026-1.045)
PM2.5
12.80
3
0.001
1.050†
(1.034-1.065)
SO2
13.11
2
0.001
1.041†
(1.027-1.055)
NO2
14.22
2
0.001
1.038†
(1.019-1.057)
O3
12.89
3
0.001
1.038†
(1.024-1.053)
PM2.5
14.46
3
0.001
1.051†
(1.008-1.097)
SO2
13.33
3
0.001
1.077†
(1.041-1.114)
NO2
13.44
3
0.002
1.117†
(1.053-1.186)
O3
12.52
3
0.002
1.065†
(1.024-1.108)
PM2.5
13.04
3
0.004
1.117†
(1.018-1.225)
SO2
12.75
1
0.005
1.176†
(1.030-1.342)
†Indicates statistical significance of the estimate (p8, 8-6.5, 6.5-5.2, 5.2-3.5, 3.5-2.6, 2.6-1.7, 1.7-1, 1-0.43 and