Selected Readings in Applied Climatology [1 ed.] 9781443886567, 9781443875622

Citation preview

Selected Readings in Applied Climatology

Selected Readings in Applied Climatology Edited by

Robert V. Rohli and

T. Andrew Joyner

Selected Readings in Applied Climatology Edited by Robert V. Rohli and T. Andrew Joyner This book first published 2015 Cambridge Scholars Publishing Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Copyright © 2015 by Robert V. Rohli, T. Andrew Joyner, and contributors All rights for this book reserved. No part of this book 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 permission of the copyright owner. ISBN (10): 1-4438-7562-7 ISBN (13): 978-1-4438-7562-2

TABLE OF CONTENTS

Acknowledgments ....................................................................................... x Section I: Applied Climatology: An Overview Chapter One ................................................................................................. 2 What Is Applied Climatology? Robert V. Rohli Section II: Applied Climatology and Atmospheric Circulation Variability Chapter Two .............................................................................................. 12 Overview of Applied Climatology and Atmospheric Circulation Variability T. Andrew Joyner Chapter Three ............................................................................................ 20 Low Temperature Events in Central Michigan: The Seasonal Role of Migratory Hudson Bay Cold Pools Gregory Bierly and Randall Repic (1994) Chapter Four .............................................................................................. 32 Impact of ENSO Events on U.S. Climate Anomalies Kent M. McGregor (1996) Chapter Five .............................................................................................. 40 Impact of ENSO Events on U.S. Climate Anomalies: Update since 1996 Kent M. McGregor Chapter Six ................................................................................................ 46 Northern Hemisphere Flow Anomalies and U.S. Temperature and Precipitation Variability Anthony J. Vega, C.H. Sui and K.M. Lau (1997)

vi

Table of Contents

Chapter Seven............................................................................................ 54 A Climatological Analysis of Lower Atmospheric Ozone Transport across Phoenix, Arizona Mark L. Hildebrandt (2000) Chapter Eight ............................................................................................. 66 The Synoptic Climatology of Lower Atmospheric Ozone Exceedances in St. Louis, Missouri Mark L. Hildebrandt (2001) Chapter Nine.............................................................................................. 74 Seasonal Trends in Antarctic Temperature Reanalysis Jennifer M. Collins, David R. Roache, Edgar W. Kopp IV, Douglas Lunsford (2012) Section III: Applied Climatology and the Biosphere Chapter Ten ............................................................................................... 90 Overview of Applied Climatology and the Biosphere T. Andrew Joyner Chapter Eleven .......................................................................................... 96 Analysis of the Relationship between NDVI and Rainfall in Subhumid Agricultural Areas and the Influence of Environmental Factors Yongyi Tao, Zhenhuan Chi, John Harrington Jr., Brad Rundquist and M. Duane Nellis (1997) Chapter Twelve ....................................................................................... 109 The Effects of Climatic Factors on Vegetation Dynamics across Homogeneous Land Cover Bradley C. Rundquist (1998) Chapter Thirteen ...................................................................................... 120 Climate as a Control in Landscape Evolution Chris Welchhans (1998) Chapter Fourteen ..................................................................................... 130 Impact of Global Warming on Ecosystems: A Case Study of Biomass at Konza Prairie Biological Station Jincheng Gao and John A. Harrington Jr. (2002)

Selected Readings in Applied Climatology

vii

Section IV: Applied Climatology and Water/Energy Resources Chapter Fifteen ........................................................................................ 144 Overview of Applied Climatology and Water/Energy Resources Robert V. Rohli Chapter Sixteen ....................................................................................... 156 Climate Variations and the Water Supplies for Lake Superior Waltraud A.R. Brinkmann (1979) Chapter Seventeen ................................................................................... 169 Relationship of Rio Grande Headwaters Precipitation and Discharge to the Southern Oscillation Index Richard A. Earl and John A. Harrington Jr. (1994) Chapter Eighteen ..................................................................................... 181 Assessing the Water Quality Impacts of Global Climate Change in Southwestern Ohio, U.S.A. Amy J. Liu, Susanna T.Y. Tong and James A. Goodrich (2002) Chapter Nineteen ..................................................................................... 194 Water Supply and Climate Change: Rethinking Growth Regimes in Texas Kevin Romig and Laura Stroup (2009) Chapter Twenty ....................................................................................... 204 Applicability of “Optimal Climate Normals” for Use in the Energy Industry in the Southern United States: A Preliminary Analysis Robert V. Rohli and John M. Grymes III (2000) Section V: Applied Climatology and Agriculture Chapter Twenty-One ............................................................................... 218 Overview of Applied Climatology and Agriculture Robert V. Rohli Chapter Twenty-Two............................................................................... 222 The Perception of Climatic Variability and Its Role in Agricultural Decision-Making in West Central Alberta Catherine Hooey (1998, updated in 2015)

viii

Table of Contents

Chapter Twenty-Three............................................................................. 233 Assessing the Impact of Heat and Humidity on Livestock: Development of an Hourly THI Climatology John A. Harrington Jr. and Erik Bowles (2002) Chapter Twenty-Four .............................................................................. 241 The Water Budget, Climate Variability, and Climate Impacts Assessment in Northeast Kansas Iris E. Wilson, John A. Harrington Jr., Kendra McLauchlan, Elias Martinson and Stacy L. Hutchinson (2009) Section VI: Applied Climatology and Human Health, Comfort, and Behavior Chapter Twenty-Five ............................................................................... 254 Overview of Climate and Human Health, Comfort, and Behavior T. Andrew Joyner Chapter Twenty-Six................................................................................. 257 Assault and Deviations from Mean Temperatures: An Inter-Year Comparison in Dallas Stephen J. Stadler and Keith D. Harries (1985) Chapter Twenty-Seven ............................................................................ 270 Applied Macroanalysis of Golf Course Environments Stephen J. Stadler and Michael A. Simone (1988) Chapter Twenty-Eight ............................................................................. 283 The Assessment of Clo as an Alternative Weather Stress Index Yuk Y. Yan and John E. Oliver (1994) Chapter Twenty-Nine .............................................................................. 292 Impact of ENSO Phase on Australian Shark Attack Frequency Richard W. Dixon (2003) Chapter Thirty ......................................................................................... 298 Classifying Heat Stress Events in the Central United States Erik H. Bowles (2006)

Selected Readings in Applied Climatology

ix

Section VII: Communicating Climate to the Public Chapter Thirty-One ................................................................................. 310 Overview of Communicating Climate to the Public Robert V. Rohli Chapter Thirty-Two ................................................................................. 312 The Perils of Projection Earl Cook (1981) Chapter Thirty-Three ............................................................................... 317 Stakeholder Adaptation to Climatic Change in Kansas: What Have We Learned? John A. Harrington Jr., Lisa K. Tabor and Iris E. Wilson (2011) Chapter Thirty-Four................................................................................. 330 Developing Strategies to Convey Climate Science to Kansas Stakeholders: Evolution and Approach John A. Harrington Jr. (2012) Chapter Thirty-Five ................................................................................. 344 The Prospect for Applied Climatology John E. Oliver (1996)

ACKNOWLEDGMENTS

The authors express sincere gratitude to Joseph B. Harris, Suzanne Rohli, and Eric Rohli for their generous assistance in reproducing the figures and tables as closely as possible to their original forms. We also appreciate the proofreading assistance of Kristen Rohli and Neely MartinWhitaker, and the wisdom and support of Sam Baker and colleagues at Cambridge Scholars Publishing. Finally, we thank Jay Lee of Applied Geography Conferences, Inc. and Kent State University for the opportunity to participate in this project and for generous support during the process.

SECTION I: APPLIED CLIMATOLOGY: AN OVERVIEW

CHAPTER ONE WHAT IS APPLIED CLIMATOLOGY? ROBERT V. ROHLI

1. Introduction Climate consists of the long-term patterns of weather, across space and time. These patterns include not only the average weather conditions from place to place and from one period in history to another but also the extremes, variability in those long-term weather conditions, and the frequency with which those extremes occur at a place, or from place to place. Climatology is the scientific study of climate. Climatology differs from meteorology in that meteorology involves the study of weather – the instantaneous condition of the atmosphere at a specific time and place. Climatology involves the generalized characterization of the weather conditions over long time periods, in and across space too. If someone says that you’re acting grumpy today, then that’s a different type of observation than someone saying that today you’re in the grumpiest mood that you’ve displayed in the last six months, or that you’re a grumpy person. In the same way, weather involves the observation at a given instant, while climatology involves the general characterization of weather, and an assessment of how unusual or extreme the weather is, in the bigger picture. Most climatologists study the causes of climate. Why does San Francisco get little snow but yet feel so chilly for much of the year? What causes Kent, Ohio, to get snowfall totals that are as high as 250 centimeters in some years and as low as 60 centimeters in other years? What factors contribute to the fact that western Russia may experience several winters in a row of brutal sub-zero temperatures but then several winters in a row with much milder temperatures? Often the roots of these kinds of questions involve some type of influence by broad-scale

What Is Applied Climatology?

3

circulation patterns in the atmosphere and/or ocean which undergo variation for reasons that are partially but not completely understood. Applied climatology approaches the study of climate from the opposite perspective. Rather than emphasizing the causes of climate, applied climatology is concerned with the effects of climate on other economic, ecological, social, or recreational facets of society. These include, but are not limited to, agriculture, forestry, ecosystems and biomes, fisheries, architecture, energy supply and demand, human health, transportation, and economic and political activities and events. Even arts and literature are influenced by climate. For example, some art historians have noted that paintings from certain historical periods are more likely or less likely than paintings from other periods to show gloomy skies. To what extent are these differences attributable to differences in climate across the two periods versus differences in the preferred artistic motifs? The frequency, variability, and extremes of weather events tend to exert more effects than the means, as society is designed to function under “normal” conditions. How does the frequency of severe weather events impact tourism in Florida? How often can we expect 6 centimeters of rain to fall within a 2-hour period in Baton Rouge, Louisiana, where a culvert cannot handle more than this rate of input of water? What is the probability that a citrus grove planted in Orange County, California, will experience no killing frosts for the next 15 consecutive years, so that the planter can recover his investment and yield a profit? Note that these are “how often” questions rather than “why” questions.

2. The Rise, Fall, and Rise of Applied Climatology The formal, organized study of weather and climate has only existed for about 150 years – the work of several brilliant individuals such as Benjamin Franklin notwithstanding. The formation of the United States Weather Bureau in 1870 represented the first organized attempt at the federal level to measure, collect, store, and disseminate weather and climate information, to predict weather, and to sponsor weather and climate research, in any country in history. Although the U.S. Weather Bureau was moved to different cabinet departments a few times in its history, its mission remained the same: To protect life and property. Applied climatology has played a major part in protecting life and property, because of its emphasis on links between climate and human

4

Chapter One

health, livestock, and agriculture. However, in the early days of weather and climate research, in both the United States and abroad, much more emphasis was placed on understanding the causes of weather, rather than the effects of climate. Climate was relegated to secondary importance to such an extent that even by the dawn of the 20th century, few atmospheric scientists even considered long-range weather patterns or understood that climates change over time. The few that did recognize climate change at the turn of the century, such as Robert Ward, were marginalized in the discipline. Instead, climatologists were seen as little more than the recordkeepers, with the most important work being done by meteorologists. By contrast, developments in telecommunications, such as the telegraph in 1857 and the telephone in 1876, attracted the cutting-edge scientific minds of the day toward improved understanding of the simultaneous weather conditions over large distances, and the tracking of those systems as they moved across space. One example of the results of such developments in meteorology included Isaac Cline’s prediction of the landfall of the great Galveston hurricane of 1900, despite his earlier disbelief that such a hurricane could ever happen in Galveston. Another example was the set of breakthroughs in the 1910s and 1920s, mostly by the Bergen School of Meteorology in Norway, in understanding how midlatitude storms and their associated cold fronts and warm fronts form, grow, and die. These fronts, which separated large masses of air of different properties, reminded Wilhelm Bjerknes and his colleagues at the Bergen School of the trenches which separated large masses of troops, so they borrowed the term “front” from their recent horrific experience in World War I. Still though, the balance of knowledge remained far tilted in favor of understanding the causes of weather rather than the effects of weather and climate. The need for understanding the causes of weather became even more urgent during World War II. For the first time, the world had seen combat that had occurred on land, sea, and sky, at the global scale, simultaneously. Never before had there been such a need to understand weather thousands of miles away from home. Whichever side could develop the better understanding of the principles governing atmospheric and oceanic circulation patterns, invent better instruments to detect and communicate those atmospheric and oceanic conditions, and apply those principles to forecasts of weather, would have a huge advantage. These demands further reinforced the importance of the causes of weather, rather than the effects of weather and (especially) climate. The polar front jet stream was

What Is Applied Climatology?

5

discovered as airplanes flying toward the Pacific theater from the United States observed strong headwinds and ran short of fuel, but planes returning home from the Pacific experienced consistent tailwinds and fuel savings. Meteorology knowledge and a skillful weather forecast played a major and perhaps decisive role in the D-Day invasion. The lack of such knowledge perhaps saved the U.S.S.R., as Hitler opted to invade on the erroneous assumption that three consecutive harsh Russian winters would be very unlikely. The immediate post-war period saw continued emphasis on the causes of weather. The war effort had created tools such as radar and sonar that, although designed for military uses, also had civilian applications in understanding the causes of atmospheric and oceanic circulations. As the Cold War and Space Race ensued, the emphasis on pure rather than applied science persisted. One positive impact of these efforts was continued advancement on the nature of the Earth-ocean-atmosphere system and principles that are still used today in geoscience. Advances in magnetometry, for example, revealed from clues in the iron-rich mineral magnetite on the ocean floor that the Earth’s polarity shifted many times in the past. These shifts were then dated chronologically and used to estimate not only the age of the Earth but also the rate at which its tectonic plates were moving. It gave support for a previously-rejected hypothesis that the continents “drift” slowly over time. Post-war technology in meteorology provided new insights into the causes of weather. In 1948 the first regular, systematic, synchronous launching of weather balloons began. The Americans and the Soviets even collaborated to sponsor the “International Geophysical Year” of 1957, to not only showcase but also share their recently gained knowledge. Subsequently, the first meteorological satellite, TIROS, was launched in 1960. But through all of these innovations in understanding the causes of weather and celebrations of their newfound knowledge, scientists had mostly overlooked the impacts of weather, and especially climate. By the 1960s, the paradigm had begun to shift. The publication of Silent Spring by Rachel Carson in 1962 had shifted the public mindset into realizing that impacts of our development and technology could not be ignored. The book increased public awareness of the effects of the pesticide DDT on birds, with an inferred influence on society. Although the book was immensely successful at generating awareness of the environmental impacts from DDT and spurred a ban on its agricultural use in many countries around the world, the impacts of the ban itself were not

6

Chapter One

understood, even by Carson. For instance, millions died from malaria because of a reduction in the use of DDT. Other pesticides were substituted for DDT, which themselves had harmful or potentially harmful impacts. It was only decades later that the cascade of impacts of our activities began to be understood more fully (Mandavilli, 2006). Once again, a more acute awareness of, and foresight regarding, the potential impacts of pure science could have averted the consequences of science developed with the best of intentions. The modern environmental movement continued throughout the 1960s, culminating in the passage of the Clean Air Act of 1970 in the United States. For the first time in human history, a federal government placed limits on acceptable thresholds of concentrations of pollutants in the air and water and enforced those limits. The thresholds were set to correspond to the concentrations of each pollutant that began seriously impacting life and property. So applied science was beginning to take hold. Rather than asking, “What is the cause of the pollutant?” scientists were asking, “What is the effect of the pollutant?” And applied science was taking hold elsewhere as well. While the lunar landing in 1969 proved that we can land on the Moon, we don’t continue to settle the Moon today because we don’t reason that the benefit exceeds the cost. In other words, we don’t believe that the effects of settling the Moon make it a worthwhile endeavor. At about the same time that applied science was gaining traction among scientific thought, the study of climatology was beginning to gain increased respect as a viable and important field within the atmospheric sciences. Several factors explain this paradigm shift. First, the exponential increase in computational power and increasingly sophisticated numerical models had begun to allow for the simulations of not only weather systems for daily forecasts but also climatic patterns that might have occurred in the Earth’s distant past or future. Second, the archived data from weather balloons, satellites, buoys, and other sources had become extensive enough to reveal long-term patterns of observed climate. The arrival of geostationary weather-observing satellites in the 1970s provides one example of the wealth of continuous environmental observations which were available over much of the Western world, and later, the planet. Third, the development of multivariate statistical techniques, mostly borrowed from other fields such as psychology and made possible by computer-based analyses, helped to identify underlying patterns in these

What Is Applied Climatology?

7

datasets that may not have been otherwise apparent. And finally, concerns in the 1970s about the possibility of global cooling and, later, of global warming, attracted increasing scientific attention and raised the public consciousness of the importance of understanding climates and the causes and effects of climatic changes. Consequently, climatologists were elevated as equals within the field of atmospheric sciences and meteorologists were no longer considered the favored or superior scientists. The labor pool was able to satisfy the demand for analyzing the impacts of climate on other facets of society, in addition to the causes of weather. World War II had created an overabundance of trained weather forecasters and analysts; many had failed to find use for those skills as civilian meteorologists after the war. Veterans and others were eager to apply their skills in private consulting companies which opened in response to the need for monitoring long-term weather patterns and their impacts, largely as a result of the regulations imposed by the Clean Air Act and its amendments and the opportunities afforded by additional weather observations and increased computing power. According to Dixon (2013), this modern, enlightened perspective on the importance of applied science, the attention to impacts of phenomena, and the rise of climatology may have found its roots in Gove Hambridge’s edited volume, Climate and Man: The 1941 Yearbook of Agriculture. But the modern movement in climatology only began to develop in earnest since the 1970s. One feature of applied scientific inquiry is consideration that systems are interconnected. “Systems science” emerged over this time as an interdisciplinary field of study that examines problems from a variety of disciplinary perspectives at a range of scales simultaneously. In the Earth and environmental sciences, including climatology, systems science has taken the form of “Earth systems science,” which emphasizes the interconnections between processes occurring as parts of larger cycles, above Earth’s surface (the atmosphere), in its water bodies (the hydrosphere), within the highest 100 or so kilometers of its land surface (the lithosphere), and within the living creatures above, on, and below the surface (the biosphere). Earth systems science and climatology also emphasizes the interactions between the four “spheres” and between the spheres and human systems. Those interconnections and interactions often include feedbacks – responses to a perturbation that may either amplify (positive feedback) or dampen (negative feedback) the initial perturbation. For example, long-term global warming could warm a small part of the

8

Chapter One

polar ice-covered area, causing ice and snow to melt (initial perturbation). But the melted snow and ice may expose dark-colored surfaces, which would absorb further radiation, increasing the temperature further, and enhancing the probability of more ice and snow melting, which would further expose dark-colored surfaces, etc. This example is a positive feedback, as the initial perturbation is enhanced by the response to that perturbation. As another example, one impact of increasing surface temperatures (the initial perturbation) on a warm summer day is the rising of warm air from the surface to form a cloud which may precipitate. But the fallen precipitation will cool the surface, dampening the initial perturbation (negative feedback). Another feature of applied Earth and environmental science, including climatology, is the increased recognition that many of the impacts are not predicted, and perhaps even not predictable. The nature of feedbacks, and even feedbacks on feedbacks, is inherently complicated. Furthermore, when natural systems interface with human systems, additional uncertainty is involved, because human behavior is inherently illogical, inconsistent across individuals, and even inconsistent in the same individual over time. Because of the realization that impacts may play out in unforeseen and sometimes seemingly unforeseeable ways, the perspectives of applied climatologists have become more important than ever. The Technology and Information Revolution has provided unprecedented availability of environmental monitoring equipment and the collection of data from such equipment. For example, the (U.S.) National Weather Service’s Automated Surface Observation System (ASOS) collects an array of atmospheric data on a near-real-time basis at hundreds of socalled “first-order” weather stations around the country. Radar-derived estimates of precipitation are now collected and used by hydrologic modelers to provide timely estimates of flood potential. The Oklahoma Mesonet, a collaborative endeavor by universities across the state, collects and archives a suite of atmospheric data from a dense network of 138 surface sites. Applied climatologists have played an important role in these and similar efforts. Paradoxically, such work has taken applied climatologists back to their roots as “record keepers,” but this time the loop has been closed, with the applied climatologists themselves formulating research hypotheses and analyzing and interpreting the data that they have helped to collect, and therefore contributing directly to the advancement of science.

What Is Applied Climatology?

9

3. Role of the Applied Geography Conference in Modern Applied Climatology The Applied Geography Conference was founded in 1977, in response to the need for geographers to participate in the “applied science revolution.” Geography’s traditional emphases on spatial and holistic features of the natural and human environment, the interconnectedness of the “spheres,” and the intricate and complicated nature of human-nature relationships made the Applied Geography Conferences a welcoming home for applied climatologists. At every Applied Geography Conference since its founding, applied climatologists have participated and presented new ideas and perspectives. In many cases, applied climatologists have fertilized geography by sharing ideas borrowed from or expanded upon meteorological theory. They have also brought inherently geographical ideas, perhaps gained from Applied Geography Conferences, to the meteorology community. They have participated actively in what climatologist Stanley Changnon (2005) referred to as “the foundation upon which the world’s weather-sensitive activities and infrastructure have been developed.” The purpose of this book is to trace the development of applied climatology seen through the lens of the Applied Geography Conferences. While it is recognized that applied climatology developed well beyond the radar of the Applied Geography Conferences, the use of the Applied Geography Conferences as the common denominator for characterizing applied climatology’s development is appropriate, for two reasons: 1) the conference’s history overlaps elegantly with the “golden age” (Changnon, 2005) of applied climatology; and 2) it is unlikely that major breakthroughs have escaped discussion at this conference. The rest of this book is organized topically, with an essay introducing each of the book’s sections that traces the role of applied climatology and connects the ideas within a specific topic area. Manuscripts from the Proceedings of the Applied Geography Conferences are included as examples of the work of applied geographers throughout this “golden age.” Graphics have been updated to conform to contemporary standards, obvious typographical errors have been corrected, and other grammatical and organizational constructions were standardized across the manuscripts, but otherwise the works have been unedited from their original format; the editors preserved the denotations and connotations of original authors.

10

Chapter One

Facts and opinions presented by the original authors are not necessarily endorsed by the editors. Section II features several papers that represent the prevailing thought at the time regarding how broad-scale circulation drives climate and climate variability. Though some of these papers are recent, it was the recognition of the importance of broad-scale circulation patterns that provided the early link from the causes of weather to the effects of climate. The impact of climate on the biosphere, water and energy resources, agriculture, and human systems is the topic of Sections III through VI, respectively. Finally, Section VII contains essays that present applied climatologists’ viewpoints of how climatic variability and change are communicated to and perceived by the public which in turn drives the course of their research.

References Changnon, S.A. 2005. Applied climatology – The golden age has begun. Bulletin of the American Meteorological Society 86, 915–919. Dixon, R.W. 2013. Gove Hambidge (ed.), Climate and Man: The 1941 Yearbook of Agriculture. Washington, DC: United States Government Printing Office. Progress in Physical Geography 37, 562–566. Mandavilli, A. 2006. Health agency backs use of DDT against malaria. Nature 443(7109), 250–251.

SECTION II: APPLIED CLIMATOLOGY AND ATMOSPHERIC CIRCULATION VARIABILITY

CHAPTER TWO OVERVIEW OF APPLIED CLIMATOLOGY AND ATMOSPHERIC CIRCULATION VARIABILITY T. ANDREW JOYNER

1. General Circulation of the Atmosphere The globally-interconnected atmospheric circulation system is one of the most important features of weather and climate. Not only does it move air around the planet, but it also moves whatever matter is in the air, such as water vapor, ice crystals, liquid water droplets, soot particles, pollutants, and salt crystals that become suspended every time a wave breaks on a shoreline. When this so-called “general circulation” causes air to move over surfaces that have more moisture than the surfaces over which they sat, the air can evaporate water from the new surface. This water can later be condensed into clouds and form precipitation. The general circulation can also push air over surfaces where water is less abundant. In such cases, water already condensed elsewhere can fall on the drier surface. Similarly, the general circulation can bring warmer, colder, more polluted, or less polluted air into an area. It also steers the weather systems that bring the day-to-day changes in weather, which collectively comprise climate. Although locally-generated circulations can also be important, the general circulation can reduce or eliminate local-scale effects in the same way that a river’s current can overwhelm smaller circulations within a local part of the river. The amazingly powerful, omnipresent, and complicated natural wonder known as the general circulation of the atmosphere ultimately results from two simple, fundamental laws of nature. The first law of thermodynamics says that energy can never magically appear or disappear. Instead it can only be transformed from one form to another. Virtually all of the energy that drives the general circulation of the atmosphere comes from the Sun and is transformed into energy of motion, or kinetic energy, which moves

Overview of Applied Climatology and Atmospheric Circulation Variability 13

the atmosphere. Thus, the first law of thermodynamics implies that the more energy Earth gets from the Sun, the more energetic the system of circulation will be. On bodies that do not have matter in the form of an atmosphere, such as the Moon, the energy received from the Sun must be transformed in other ways. One form of the second law of thermodynamics says that the energy in a system must be redistributed from places where it is more abundant to places where it is less abundant in the system. In the Earth-oceanatmosphere system, the tropical parts of the system receive the most direct impact of energy from the Sun because of their position relative to the Sun’s incoming rays. So the general circulation moves this surplus of energy toward the polar parts of the system, where less energy is received, in fulfillment of the second law of thermodynamics. It is upon these two premises that the general circulation of the atmosphere is based, but because of the Earth’s size, rotation, and distribution of land and water features, the redistribution of energy from the equatorial areas to the polar areas is more complicated. Warmed air near the surface of the equatorial areas rises, because warmer air is less dense than the surrounding air. As it rises, it cools, and its water vapor can condense. So the equatorial areas are characterized by a more-or-less continuous belt of low surface atmospheric pressure (called the equatorial trough, or intertropical convergence zone) caused by the release of pressure at the surface by the rising air, extensive and persistent cloud cover, and abundant rainfall. The air aloft then proceeds laterally poleward, both in the Northern and Southern Hemispheres. But by the time the air is about one-third of the way toward the poles in each hemisphere, it sinks back down toward the surface. This sinking air creates a semi-permanent belt of enclosed high-pressure systems (anticyclones), cloud suppression, and generally dry conditions about one-third of the way toward the poles (around 30°N and 30°S latitude) in many places around the Earth. At the same time, the frigid air over the polar areas sinks, because cold air is so dense. This sinking creates a permanent surface anticyclone near the polar surface. The second law of thermodynamics implies that air must move from areas of higher pressure to areas of lower pressure. So the air with high pressure around 30° of latitude moves toward lower pressure on its equatorward side and its polar side, while polar high-pressure air must also move toward lower pressure (i.e., away from the pole). So, air moving

14

Chapter Two

poleward from 30° of latitude meets air moving equatorward from the pole (in each hemisphere) near 60°N and 60°S latitude. This converging air is forced to rise, in each hemisphere, which creates a series of enclosed areas of low pressure (cyclones) near 60°N and 60°S latitude and the attendant cooling, condensation of water vapor, cloud cover, and abundant precipitation. So, many, but not all, areas near 60°N and 60°S latitude around the world are characterized by cloudy, damp conditions. The rotation of the Earth causes the wind patterns created by the highto-low-pressure circulations to be apparently deflected from their true directions. In the belt between the equator and 30°N, the near-surface wind is bent so that it usually comes from the northeast. Between the equator and 30°S, the near-surface wind is bent so that it usually comes from the southeast. In the belt between 30°N and 60°N and between 30°S and 60°S, the surface winds near the surface usually come from the west. Most of us who live in these mid-latitudes have noticed that storm systems and cold fronts are steered by these west-to-east currents of air. And in the belt between 60°N and the north pole, and between 60°S and the south pole, the surface winds near the surface usually come from the east. Aloft, the deflection due to the rotation of the Earth generally causes west-to-east flow across most of the Earth, in both hemispheres. As with other features of weather and climate, the general circulation of the atmosphere undergoes variability on various time scales. For example, during some winters the upper-level general west-to-east flow aloft takes many small dips equatorward (troughs) and poleward (ridges) on its general west-to-east flow. During other winters, there are few but intense troughs and/or ridges. During still other winters there are few troughs and ridges and those that exist are weak. The variability in this feature and other features of the general circulation may occur on a wide range of time scales as well. Atmospheric scientists usually consider the hemispheric-scale flow pattern of the middle-to-upper troposphere as a single, continuous, organized system that circumnavigates the pole, largely from west-to-east in each hemisphere, bounded on the equatorward side by the polar front jet stream, which exists near the leading edge of the polar air masses. The sharp contrast of air mass properties at this leading edge leads to sharp differences in pressure, which triggers the particularly rapid wind flow characteristic of the polar front jet stream. The entire continuous system, known as the circumpolar vortex, typically expands in winter, as the cold

Overview of Applied Climatology and Atmospheric Circulation Variability 15

pool poleward of the polar front jet stream expands, and contracts in summer, as the warmer air masses advance poleward.

2. Links between Atmospheric Circulation and Oceanic Circulation The general circulation of the atmosphere is also interconnected to a general circulation of the ocean. Changes in the strength and/or position of the atmospheric circulation or some sub-components of it are inherently tied to variations in the strength and/or position of the surface ocean circulation adjacent to the atmosphere. These changes in surface ocean circulation then cascade downward to affect deeper circulation patterns in the ocean. Therefore, it is no surprise that a firm foundation in understanding atmospheric circulation is necessary for understanding the impacts to the ocean associated with applied climatology. Often, the link between the circulation and the applied impact is through synoptic meteorological or synoptic climatological research. An important and well-known link between atmospheric and oceanic properties involves the periodic oscillation of sea surface temperatures in the tropical equatorial Pacific Ocean. At the end of each year, waters tend to warm in the tropical Pacific, and even extend to the normally-coolocean-current-influenced eastern tropical Pacific near Peru. This natural, seasonal warming is anomalously intense for a period of several months approximately every three to seven years, in what is known as an El Niño event. El Niño events have tremendous environmental and ecological repercussions, largely because the oceanic upwelling of cold water near the Peruvian coast weakens and fails to move nutrients back upward where they can contribute to rich food webs. In other years, the normal, seasonal oceanic warming in the tropical equatorial Pacific leaves the eastern Pacific unaffected, producing anomalously cold sea surface temperatures and even more intense upwelling than normal near the Peruvian coast for a period of several months in a La Niña event. El Niño and La Niña events are also linked to anomalies in the atmospheric general circulation, particularly in the tropics, where the alteration between circulation patterns associated with El Niño and La Niña is known as the Southern Oscillation. However, the atmospheric circulation anomalies linked to the Southern Oscillation ripple beyond the tropics.

16

Chapter Two

The combined, interdependent oscillation of oceanic and atmospheric circulation anomalies associated with these features is known as El Niño – Southern Oscillation (ENSO). Because ENSO-related variability impacts weather patterns and weather-dependent endeavors, it has traditionally been an area of focus by geographers and particularly by applied climatologists. These impacts are on the atmosphere’s so-called synoptic circulation.

3. Synoptic Meteorology and Synoptic Climatology The analysis of weather features such as cyclones, anticyclones, fronts, and jet streams, necessarily takes on broad scales, because these features are relatively large. The word “synoptic” comes from “syn” which means “same,” as in a synonym, and “optic,” which means “to see.” So “synoptic” refers to “seeing at the same time,” or a snapshot of the broadscale – the “synoptic scale” – condition of the atmosphere – its fronts, pressure patterns, jet streams, and similar features – at an instant in time. A typical weather map as shown on your favorite news program would be an example of a synoptic weather map. Synoptic meteorologists study these types of weather features, how they form and interact with other features, and how they produce impacts to life and property. At first glance, “synoptic climatology” would seem to refer to the long-term pattern and variability associated with features like the jet streams, the semi-permanent cyclones and anticyclones, migrating warm fronts and cold fronts, and their impacts to life and property. In the beginning, that was the domain of synoptic climatology. But over time, the synoptic meteorologists have increasingly left the “impacts” sides of the research to the synoptic climatologists and have concentrated instead more on the interplay between atmospheric dynamics and the synoptic conditions of the atmosphere. At the same time, synoptic climatologists seized the opportunity to increasingly emphasize the impacts and in the process tightened their links to geography with its traditional emphasis on human-environment interactions. As a result, in recent decades, synoptic climatology has come to refer to the relationship between the broad-scale atmospheric circulation features and the surface environment. Here, “surface environment” is defined broadly to include not only environmental quality per se but also other features of the surface environment, such as the biosphere, water and energy availability, agriculture – the subjects of other sections of this book, and more.

Overview of Applied Climatology and Atmospheric Circulation Variability 17

Early announcements of this “new synoptic climatology” had been made since the 1970s. Werner Terjung’s (1976) call for geographicalclimatological research of the highest form – physical-human-processresponse systems, that recognizes the cascade of linkages between and within systems, is an early example. Later, climatologist Brent Yarnal called for climate research as an integrated, interactive system (Yarnal et al., 1987), and Andrew Carleton (1999) saw the onset of synoptic applications to weather prediction. The latter suggested that all geographical climatology has at least an implicitly applied component. The fact that all three of these papers were published by flagship journals in American geography testifies to the importance of these viewpoints within the community of geographers. Two subsequent papers published in Progress in Physical Geography tend to stand out in terms of their emphasis on the synoptic approach. Sturman (2000) saw the importance of air pollution studies as an obvious topic emanating from the synoptic approach per se and through its linkage to mesoscale circulations. Sheridan and Lee (2012) emphasized the importance of a synoptic climatological approach in understanding spatial and temporal variability in the semi-permanent pressure patterns that drive circulation – the study of atmospheric teleconnections. However, the seminal work on the “new synoptic climatology” was Yarnal’s (1993) Synoptic Climatology in Environmental Analysis: A Primer. This volume clarified the role of synoptic climatology as an applied subdiscipline and identified three major types of synoptic climatological analysis (synoptic typing, map pattern classification, and regionalization). Perhaps more importantly, Yarnal (1993) also provided worked examples of quantitative methods that should be part of any synoptic climatologist’s toolkit. These include a comprehensive review of the capabilities and limitations of manual classification methods, correlation-based analyses, eigenvector-based techniques, compositing, indexing, and specification. Yarnal (1993) emphasized the contrast in method between what he called “circulation-to-environment” vs. “environment-to-circulation.” In the former, the categorization of climate is done first, with all temporal entities in the analysis (i.e., days, months, etc.) included in the classification, and subsequently, an analysis of the percentage of those days that meet the environmental criterion of the research study is determined. By contrast, the “environment-to-circulation” approach

18

Chapter Two

(Yarnal, 1993) involves the pre-selection of only those temporal entities (i.e., days, months, etc.) that meet the environmental threshold of the study in the categorization of atmospheric circulation. For example, Rohli et al. (2004) use elements of both approaches in categorizing atmospheric circulation associated with excessive tropospheric ozone in Louisiana. Carrying out the blueprint for this “new synoptic climatology” has played to the advantage of its practitioners in the last decade. “Integrated environmental assessments,” particularly those that forecast impacts of changes in climate involving large research groups, have increasingly been funded and published. New research centers such as the Department of the Interior’s Climate Science Centers and the National Oceanic and Atmospheric Administration’s Regional Integrated Science Assessments (RISA) increasingly favor the broad, holistic perspective of the new synoptic climatology.

4. Synoptic Climatology and the Applied Geography Conferences Applied climatologists at the Applied Geography Conferences have contributed in important ways to our understanding of the general circulation of the atmosphere through synoptic climatological approaches. The seven papers comprising the remaining chapters in Section II, organized chronologically, provide early to recent representative examples of the new synoptic climatology. Gregory Bierly and Randall Repic (1994) used an environment-to-circulation approach to trace the evolution of the cold pools that are part of the digging troughs in the circumpolar vortex and their role in generating extreme low temperatures that impact corn in Michigan. Kent McGregor’s (1996) paper on ENSO and its impacts follows, and it is followed by McGregor’s update since 1996. Its rich literature review makes it an excellent example of the early modern knowledge of ENSO and its impacts on water availability, and it also nicely contrasts what was known in 1996 with what is known today. A paper by Anthony Vega, C.H. Sui, and K.M. Lau (1997) then is included to tie together the circulation variability associated with ENSO using a synoptic climatological approach. Two environment-to-circulation-based papers by Mark Hildebrandt (2000, 2001) focusing on mesoscale circulation and their impacts on temperature and tropospheric ozone follow, in the spirit of Terjung’s (1976) emphasis on the “cascade of linkages” and the recognition that pollutant transport is obviously dependent on atmospheric circulation and its variability. The foci of this research are two

Overview of Applied Climatology and Atmospheric Circulation Variability 19

metropolitan areas that are well-studied for their localized warm signal known as the urban heat island: Phoenix (e.g., Brazel et al., 2007) and St. Louis (Arnfield, 2003). The section concludes with a contemporary example of the use of the National Centers for Environmental Prediction (NCEP)/National Center for Atmospheric Research (NCAR) Reanalysis (NNR) dataset, by Jennifer Collins, David Roache, and Edgar Kopp (2012). The use of the model-based NNR dataset provides an excellent example of an important source of contemporary data used in studying atmospheric circulation variability.

References Arnfield, A.J. 2003. Two decades of urban climate research: A review of turbulence, exchanges of energy and water, and the urban heat island. International Journal of Climatology 23, 1–26. Brazel, A., Gober, P., Lee, S-.J., Grossman-Clarke, S., Zehnder, J., Hedquist, B., and Comparri, E. 2007. Determinants of changes in the regional urban heat island in metropolitan Phoenix (Arizona, USA) between 1990 and 2004. Climate Research 33, 171–182. Carleton, A.M. 1999. Methodology in climatology. Annals of the Association of American Geographers 89, 713–735. Rohli, R.V., Russo, M.M., Vega, A.J., and Cole, J.B. 2004. Tropospheric ozone in Louisiana and synoptic circulation. Journal of Applied Meteorology 43, 1438–1451. Sheridan, S. and Lee, C.C. 2012. Synoptic climatology and the analysis of atmospheric teleconnections. Progress in Physical Geography 36, 548–557. Sturman, A.P. 2000. Applied climatology. Progress in Physical Geography 24, 129–139. Terjung, W.H. 1976. Climatology for geographers. Annals of the Association of American Geographers 66, 199–220. Yarnal, B. 1993. Synoptic Climatology in Environmental Analysis. London, UK: Belhaven Press. Yarnal, B., Crane, R.G., Carleton, A.M., and Kalkstein, L.S. 1987. A new challenge for climate studies in geography. Professional Geographer 39, 465–473.

CHAPTER THREE LOW TEMPERATURE EVENTS IN CENTRAL MICHIGAN: THE SEASONAL ROLE OF MIGRATORY HUDSON BAY COLD POOLS GREGORY BIERLY AND RANDALL REPIC (1994)

1. Introduction The susceptibility of crops to injuries from low temperature exposure poses risks in midlatitude agriculture during the early growing season (Levitt, 1941). The various agricultural species respond quite differently to reduced temperatures, and there are numerous factors including soil moisture, suddenness and duration of freeze, extremity of temperature drop, and rapidity of subsequent thawing which control the level of injury sustained. Prior exposure to cold temperature, as well as the level of maturity, also govern the plants’ ability to resist injury (Levitt, 1980; Steward and Bidwell, 1991). South-central Michigan is an area vulnerable to early spring freeze events which may prove harmful to newly planted crops. For many plant species, however, freezing temperatures are not required to induce chilling injury or retard growth. Corn (Zea mays L.), the major crop in Michigan in terms of acreage sown and economic value, is tolerant of only a small range of temperatures. Temperatures below 47°F can prevent growth and cause chilling stress, and may temporarily inhibit the potential yield of the plant (Keeling and Greaves, 1990). Wallace and Bressman (1923) found that growth stopped entirely below 40°F and that germination and rate of growth were both affected by temperatures below 50°F. Although plant injury during seedling emergence has been a primary concern of previous

Low Temperature Events in Central Michigan

21

low-temperature stress studies, other adverse temperature-related conditions are possible, such as photoinhibition, the inability to utilize electromagnetic energy following low temperature stress (Grogan, 1970; Martinson, 1971; Keeling and Greaves, 1990). May and June, crucial months for crop development and later yields, are prone to rapid, possibly damaging, temperature fluctuations in central Michigan (United States Department of Commerce, 1971–1985a). The temperature events examined in this study will reflect the relatively low tolerance of corn, the primary cash and silage crop in Michigan.

2. Cold Air Movement Dallavalle and Bosart (1975) associated the penetration of cold air into southern latitudes with the southward movement of strong polar anticyclones from northwestern Canada. The 500 mb synoptic configuration which permits this southward advection of cold air is composed of a central ridge flanked by troughs upstream and downstream. The strengthening of the midtropospheric trough over the eastern coast of the United States dynamically favors amplification of the midcontinental ridge and produces pronounced northerly flow into the central United States. Warm air advection in advance of the upstream trough produces midtropospheric height rises over western Canada, further intensifying the central ridge (Johnson, 1948; Dallavalle and Bosart, 1975). During the spring season, air masses transported southward by anticyclones into lower Michigan tend to originate in the vicinity of Hudson Bay (Klein, 1957; Harman, 1987). This anticyclone track becomes more meridional in March, shifting eastward from the northwesterly winter track. The “Hudson Bay” track becomes fully realized in May and June, when the position of anticyclone frequency extends directly from the northwestern coast of Hudson Bay southward into the northern Great Plains, Great Lakes, and Ohio Valley states (Harman, 1987). Harman (1968) examined ten years of negative temperature departures at Urbana, Illinois, for the months of January, May, June, and August. Analysis of apparent airmass trajectories during these cold events revealed that the largest percentage of daily negative temperature departures occurred with Hudson Bay airmasses and that this preference was strongest during May and June. It was also demonstrated that cold events

22

Chapter Three

of greatest severity were dominated by the Hudson Bay group, particularly in June. Calculations of vertical motion and sensible heat fields in the Hudson Bay region (Carlson, 1974) suggest, in agreement with the anticyclone climatologies and with Harman’s (1968) findings regarding the importance of Hudson Bay in June cold events, that anticyclones should form over the western half of Hudson Bay during May and the central Hudson Bay surface itself in June. Based upon sensible heat characteristics, Hudson Bay is the strongest regional cold air source during these spring months. In addition, the spring season position and strength of subsidence patterns in central and eastern Canada agree strongly with the climatological findings describing stationary anticyclones and tracks of migratory high pressure centers (Klein, 1957; Harman, 1987). The thermal influence of Hudson Bay upon regional air masses has been documented for several locations (Lamont, 1949; Bello and Rouse, 1985; Rouse and Bello, 1985; Rouse et al., 1989; Silis et al., 1989). Lamont (1949) observed that Hudson Bay acts as a typical maritime influence when not ice-covered, from late May until October. The melting process begins during May, and is usually complete by late June. Burbidge (1951) demonstrated that Hudson Bay plays a significant role in the modification of continental air passing through the region, and Harman (1968) speculated that the extent to which the Bay remains ice-covered is key to interpreting its role as a cold air source and heat sink during the spring transition season. Concentrations of ice have been shown to be strongly related to colder temperatures in neighboring regions affected by onshore winds (Bello and Rouse, 1985). Rouse et al. (1989) suggested that ice on Hudson Bay directly modifies the growing season climate at sites over 60 km inland. During periods dominated by onshore winds (most common during the growing season) temperatures averaged 20°F colder than during offshore winds. The observation that onshore winds are most likely during the fast and floating ice periods of early and mid-spring agrees with Carlson’s (1974) findings of regional subsidence and outflow during this period (Rouse et al., 1989).

3. Objectives It has been shown (Harman, 1987) that anticyclones moving into the Great Lakes region during May and June prefer a track extending from a locally cold area, the decaying Hudson Bay ice surface, southward, with

Low Temperature Events in Central Michigan

23

an increasingly more meridional azimuth through the spring transition season. Thus, Michigan is the potential recipient of relatively cold high latitude air during the early corn growing season. During January, anticyclone paths originate in more westerly locations. The objectives of this study are: a) to determine if the spatial patterns of spring and winter cold outbreaks in Michigan are different, as the seasonal anticyclone climatologies suggest, b) to specifically characterize spring cold outbreaks because of their significance in Michigan agriculture, and c) to examine the 500 mb ridge positions during winter and spring cold outbreaks to infer the general synoptic wave arrangements during each type.

4. Methods Minimum temperature data were acquired for the months of January, May, and June 1971–1985 for the Lansing, Michigan, station (United States Department of Commerce, 1971–1985a). Because of the steady seasonal increase in daily minimum temperature during the spring period, weekly mean minimum temperatures (hereafter WMM) were calculated from daily minimum temperature data and used instead of monthly values. The low temperature event is defined as a day when the daily minimum temperature is 10°F or more below the WMM temperature. This range spans temperatures less than or equal to 30°F in the first week of May to those less than or equal to 50°F in the last week of June. The entire range during the spring months includes temperatures capable of inhibiting growth, potential yield, or in some way causing injury to corn plants. Once a low temperature event was identified, the associated cold pool (closed isotherm) at 850 mb was traced back in time at 12-hour intervals (soundings at 0 and 12 Z) to its apparent source (United States Department of Commerce, 1971–1985b). Based upon initial observations and source regions suggested in the literature, cold pools were assigned to one of three air mass source regions (Harman, 1968). The source region east of 90°W longitude is called the East Hudson Bay source region. Cold pools originating between 90°W and 110°W are assigned the West Hudson Bay source region, and the region west of 110°W is referred to as the West Canada source region. Coordinates at 12-hour intervals were plotted to a base map of similar projection. Divisions of the eastern and western portions of Hudson Bay are designed to reveal possible seasonal shifts in

Chapter Three

24

source region preference as suggested by previous research (Harman, 1968; Carlson, 1974). Event frequencies by initial source region were determined for each month. In addition, a 2.5 x 2.5° latitude/longitude grid was superimposed upon eastern North America, and cold pool frequencies were calculated to provide a more detailed analysis of the track patterns. Percentages of events in each source region by severity and duration were also considered. Finally, ridge axis positions at 50°N latitude during cold pool outbreaks were assessed for the day prior to the low temperature event.

5. Results and Discussion 5.1. Source Regions and Frequencies During January, 850 mb cold pools accompanying surface low temperature events originate most frequently (41% of total events) in the West Canada source area (Table 3-1). Percentages of total cold pools decrease as the source region shifts eastward, with 35% originating in the West Hudson Bay region and 24% in the East Hudson Bay region. Table 3-1: Percentage of Cold Pools Originating in Each Source Region by Month, 1971–1985 Month January May June

Cold Pool Source Region E.H. Bay W.H. Bay 24% 35% 58% 31% 59% 23%

W. Canada 41% 11% 18%

May low temperature events typically occur with East Hudson Bay cold pools (58%), with an additional 23% arriving from the West Hudson Bay area and 11% from the West Canada area (Table 3-1). The East Hudson Bay source region is still dominant during June (Table 3-1), providing 60% of the air masses causing surface low temperature events. The primary track appears to become more variable at this time, with similar percentages originating in the West Hudson Bay and West Canada source regions. These source region results are consistent with seasonal surface anticyclone climatologies. Harman (1987) found that the primary

Low Temperature Events in Central Michigan

25

anticyclone track was more westerly during the Northern Hemisphere cold season when a broad ridge is the common synoptic wave feature over western North America. This ridge provides a natural region for the development of anticyclones, which propagate downstream, transporting cold air from higher latitudes into the Great Lakes region (Dallavalle and Bosart, 1975). In addition, with relatively cold continental surfaces during January, Hudson Bay is not the outstanding source region for cold air masses that it becomes in spring. Continuing height rises over central North America during late spring may help explain the increasing frequency with which cold pools move into the Great Lakes region along more meridional paths. May cold pools producing surface low temperature events are common from both Hudson Bay source regions, reflecting the early spring presence of the Hudson Bay ice surface. As the ice surface decays in June, the East Hudson Bay surface remains dominant, but the number of West Canada events rises slightly when compared with the previous month. This may indicate the annual variation in the amount of ice cover present in June and the arrival of cold air from other sources. Cold pool frequencies calculated from the 2.5 x 2.5° latitude/longitude grid (without imposing predefined classes) reinforce the source region percentages. Frequencies during January indicate several strong modal tracks, including a meridional path which appears to originate in the West Hudson Bay region and a zonal maximum along the northern United States. Neither of the spring months exhibit this strong zonal track originating in the West Canada area. The May frequencies are greatest along the western coast of Hudson Bay, with a secondary maximum extending southward from the northeastern coast. The June frequencies do not suggest organized patterns and are distributed throughout much of the grid. Again, this increased variability may be an indicator of the year-to-year fluctuations in the relative significance of the June Hudson Bay ice surface as a regional cold air source. The weak modes of activity are meridional, with tracks extending directly southward from the Hudson Bay surface.

5.2. Event Severity Low temperature event severity (defined as the difference between the actual outbreak temperature and the WMM), also appears to be related to

26

Chapter Three

the source region of the air mass and the seasonal shift in frequencies (Table 3-2). During January, low temperature events in the most severe category (WMM-20°F) are dominated by West Hudson Bay events (53%) despite the high total proportion of West Canada events. Cold pools from this region were also quite frequent (47%) in the severe category. Less extreme events, even during January, were most commonly caused by the approach of an East Hudson Bay cold pool (Table 3-2). May low temperature events in all severity classes are dominated by East Hudson Bay cold pools (Table 3-2). During events of the most severe negative temperature departures, 100% of the May events involved East Hudson Bay cold pools. During June, the severity of the event was again proportional to the percentage of East Hudson Bay cold pools, and air masses from some portion of the coastal area accounted for all events of the most severe nature.

5.3. Duration The duration of cold pool events (period of time the station temperature was below the threshold temperature) also appears strongly related to cold pool source area (Table 3-3). January low temperature events last an average of 3.7 days when originating in the seasonally preferred West Canada source region. With more easterly air masses, the duration of events becomes shorter. May low temperature events of longest duration are from the East Hudson Bay source region (2.4 days) and become shorter with more westerly cold pools (Table 3-3). By June, the East Hudson Bay region continues to produce the low temperature events of greatest duration (2.3 days), but with the seasonal variability of the ice surface, the West Hudson Bay and Western Canada regions produce events of similar duration (Table 3-3).

Low Temperature Events in Central Michigan

27

Table 3-2: Percentage of Events in Temperature Classes below the Weekly Mean Minimum. East Hudson Bay (E.H.), West Hudson Bay (W.H.), and Western Canada (W.C.)

Region

% of Events in Temperature Classes below the Weekly Mean Minimum (WMM) T-20 T-(15-19) T-(10-14)

January E.H. W.H. W.C.

0 53 47

50 17 33

46 16 38

100 0 0

66 17 17

50 42 8

100 0 0

60 40 0

56 25 19

May E.H. W.H. W.C. June E.H. W.H. W.C.

Table 3-3: Mean Low Temperature Event Duration (Days). Source Regions are the same as Table 3-2

Month January May June

Mean Low Temperature Event Duration (days) E.H. W.H. W. C. 1.6 3.0 3.7 2.4 1.7 1.0 2.3 1.8 1.7

All the preceding surface and 850 mb temperature data suggest, in agreement with Harman (1968), that air masses propagating from northwestern Canada comprise the majority of the January cold pool climatology. These West Canada cold pools produce a large percentage of the most severe and persistent winter low temperature events, although it appears that the lesser events commonly involve Hudson Bay air masses and that West Hudson Bay cold pools actually produce the largest percentage of events in the most severe temperature class. Hudson Bay air

28

Chapter Three

masses are most frequent during May and June cold events, and are responsible for the majority of spring events exhibiting extremely large negative temperature departures and durations.

5.4. Ridge Position The frequency distributions of ridge axis positions at 50°N latitude during low temperature events were determined for January and May (the two months exhibiting the strongest modal frequency behavior). January ridge axes reflect the climatological mean wave position, with a large number of locations clustered near 140°W longitude (Harman, 1991). This suggests that January cold events occur within the preferred cold season synoptic configuration and are more zonal than their spring counterparts because of the upper level trajectory downstream from the ridge. May ridge axes during low temperature events are modal between 110°W and 130°W. With waves positioned farther eastward during spring low temperature events, it is understandable that the May cold pool trajectories are more meridional. Because the mean May 500 mb pattern is more weakly expressed than that of January and does not indicate dramatic height rises in the western United States, the amplified ridging required for meridional transport of Hudson Bay cold pools into the Great Lakes area may be more likely to occur with migratory features (Harman, 1991).

6. Conclusions Patterns of cold pool migration during low temperature events were examined for two spring months, May and June, with the intention of better clarifying the importance and variations of air mass source regions during this time of agricultural development. A threshold temperature scheme was selected which would be adverse to the development of Zea mays L., Michigan’s primary crop, in the definition of the observed low temperature event. The criteria were then extended to a winter month, January, when cold outbreaks are often quite dramatic, to compare the relative importance of various airmass trajectories in determining the cold event climatology. The patterns of source region preference are not surprising when compared with North American anticyclone climatologies (Klein, 1957; Harman, 1987). The mode of greatest anticyclone frequency during January is a broad swath extending from Alaska southeastward toward the

Low Temperature Events in Central Michigan

29

Great Lakes. This frequency zone is very similar to the modal frequency of January 850 mb cold pools during the type of low temperature event defined above. The May pressure pattern is characterized by an eastward shift in the preferred region of anticyclone frequency. Similarly, the percentage of more meridional East and West Hudson Bay cold pools rises dramatically during this month. During June, the anticyclone frequency increases in the midcontinent in response to seasonal 500 mb height rises, and the modal area becomes quite broad, encompassing the northern Great Plains, Great Lakes, and Ohio Valley. Although the primary source region during June remains the East Hudson Bay area, slightly increased preferences for the West Canada area (although still minor) with respect to event severity and duration possibly reflect this broadening swath of anticyclone occurrence as well as the diminishing and variable nature of Hudson Bay ice cover. Ridge positions during low temperature events confirm the dependence of cold pool migration upon the 500 mb trajectory for January but are less clear for May. It is interesting that events of similar magnitudes are climatologically preferred during January (based upon the mean western ridge position) but may require a migratory ridge in the central U.S. to produce a similar cold event in spring. Finally, characteristics of cold events during spring in central Michigan are strongly determined by Hudson Bay, that provides a regional source of cold air which may be advected southward and also a thermally-induced surface high pressure area. The migratory waves of spring combine with this natural anticyclogenesis region to produce cold events dangerous to crop development. These events are also rather difficult to predict. Future studies could examine this topic more rigorously by developing a cold event climatology that assembles the complete interaction between synoptic forcing and low level thermal characteristics, and also assesses the degree to which cold events are caused by migrating disturbances in anomalous circulations. In addition, some measure of ice cover (perhaps satellite data) could be incorporated to describe the varying June source regions described here more fully.

30

Chapter Three

References Bello, R.L. and Rouse, W.R. 1985. Large-scale Indicators of the Impact of Hudson and James Bay on the Climate in the Hudson Bay Lowlands, Technical Assessment of Subarctic Ontario, Report No. 16, TASO Office, T.B. 20, McMaster University. Burbidge, F.E. 1951. The modification of continental polar air over Hudson Bay. Quarterly Journal of the Royal Meteorological Society 77, 361–374. Carlson, J.D. 1974. Springtime anticyclones over Hudson Bay and the Great Lakes: A study of the regional subsidence in relation to surface energy budget changes from April to June. Unpublished M.S. Thesis, Dept. of Meteorology, University of Wisconsin, Madison. Dallavalle, J.P. and Bosart, L.F. 1975. A synoptic investigation of anticyclones accompanying North American polar air outbreaks. Monthly Weather Review 103, 941–957. Grogan, C.O. 1970. Genetic variability in maize (Zea mays L.) for germination and seedling vigor at low temperatures. Proceedings of the 25th Annual Corn and Sorghum Research Conference. American Seed Trade Association. Chicago, Illinois, 90–97. Harman, J.R. 1968. The Hudson Bay air mass type: Some climatological implications. Water Resources Bulletin 4, 9–14. —. 1987. Mean North American frequencies, 1950–79. Monthly Weather Review 115, 2840–2848. —. 1991. Synoptic Climatology of the Westerlies: Process and Patterns. Association of American Geographers, Washington, DC. Johnson, C.B. 1948. Anticyclogenesis in eastern Canada during spring. Bulletin of the American Meteorological Society 29, 47–55. Keeling, P.L. and Greaves, J.A. 1990. Effects of temperature stresses on corn: Opportunities for breeding and biotechnology. Proceedings of the 45th Annual Corn and Sorghum Industry Research Conference. Chicago, IL. Klein, W.H. 1957. Principal tracks and mean frequencies of cyclones and anticyclones in the northern hemisphere. Weather Bureau Research Paper No. 40. U.S. Department of Commerce, NOAA, Washington, DC. Lamont, A.B. 1949. Ice conditions over Hudson Bay and related weather phenomena. Bulletin of the American Meteorological Society 30, 288– 289. Levitt, J. 1941. Frost Killing and Hardiness of Plants: A Critical Review. Burgess Publishing Company, Minneapolis, MN.

Low Temperature Events in Central Michigan

31

—. 1980. Responses of Plants to Environmental Stresses. Academic Press, NY. Martinson, C.A. 1971. Low-temperature imbibition injury in seeds and susceptibility to seedling diseases. Proceedings of the 26th Annual Corn and Sorghum Research Conference. American Seed Trade Association. Chicago, IL, 105–110. Rouse, W.R. and Bello, R.L. 1985. Impact of Hudson Bay on the energy balance in the Hudson Bay Lowlands and the potential for climate modification. Atmosphere-Ocean 23, 375–392. Rouse, W.R., Hardill, S., and Silis, A. 1989. Energy balance of the intertidal zone of western Hudson Bay: II. Ice-dominated periods and seasonal patterns. Atmosphere-Ocean 27, 346–366. Silis, A., Rouse, W.R., and Hardill, S. 1989. Energy balance of the intertidal zone of Western Hudson Bay: I. Ice-free period. AtmosphereOcean 27, 327–345. Steward, F.C. and Bidwell, R.G.S. (eds.). 1991. Plant Physiology. Academic Press Inc., San Diego, CA. United States Department of Commerce, 1971–1985: Climatological Data, Michigan. January, May, June. NOAA, NESDIS, NCDC. —. 1971–1985: North American Constant Pressure Charts, January, May, June. 500 mb and 850 mb level. NOAA, NMC. Wallace, H.A. and Bressman, E.N. 1923. Corn and Corn Growing. Wallace Publishing Company, Des Moines, IA.

CHAPTER FOUR IMPACT OF ENSO EVENTS ON U.S. CLIMATE ANOMALIES KENT M. MCGREGOR (1996)

1. Introduction The El Niño phenomenon is perhaps the single most important climatic anomaly that occurs on the Earth. The reversal of weather patterns across the central Pacific Ocean causes disasters not only in the Pacific Basin but throughout the tropical regions of the world. However the relation to unusual climatic events beyond the tropics is unclear. While some studies have shown indicative relations to U.S. weather patterns, others have not (Glantz et al., 1991; Diaz and Markgraf, 1993). The purpose of this study is to investigate the statistical association between the Southern Oscillation Index (SOI) and the history of climatic variation in the U.S. for the past 100 years. Climatic anomalies, such as droughts (Namias, 1982) and floods (Bell and Janowiak, 1995) are due to a variety of interacting factors. If the El Niño phenomena is related to these patterns, even in a limited way, then such information may provide better prediction of these and other unusual climatic events.

2. Background and Related Research El Niño research has become a major growth industry in science because it is linked directly to climatic anomalies in the tropical regions of the Earth (Barnett et al., 1988; 1994). El Niño refers to the warming of the ocean waters off the coast of Peru replacing the typically cooler waters in this area. This causes a reversal of pressure cells, wind systems, and ocean currents across the central Pacific Ocean. The high pressure cell near Peru is replaced by low pressure; the low pressure cell in northern Australia is

Impact of ENSO Events on U.S. Climate Anomalies

33

replaced by high pressure. This flip-flop of pressure cells across the Pacific Ocean is called the Southern Oscillation. The two terms are usually combined and called ENSO. This reversal causes rain in the deserts of Peru and severe drought in Australia. The reversal has also been linked to a variety of unusual climatic events around the world (Canby, 1984; Glantz et al., 1991), and consequently, the El Niño phenomenon has been blamed for practically every climatic anomaly on the planet—from global warming and increased hurricanes (Gray and Sheaffer, 1991), to drought in Africa and blizzards in the U.S. (Barnett, 1981; Bradley et al., 1987; Bunkers and Miller, 1996). ENSO events are especially interesting for two reasons. First, the phenomenon is aperiodic. While El Niño occurs on average once every 4 to 5 years, it is extremely irregular. A whole decade might pass with no occurrence, or it might be only a year. Since so much climatic phenomena is related to the annual cycle of solar insolation, this unusual behavior has forced a dramatic re-thinking of the climate system and climatic models. Secondly, it is no exaggeration to say that ENSO has become the great organizer and explainer of unusual weather events and climatic variability in tropical regions. Diaz noted that while El Niño associations are robust in the tropics (Kiladis and Diaz, 1989; Chu, 1995), extratropical teleconnections are much more difficult to identify and decipher (Ropelewski and Halpert, 1986; Ropelewski and Halpert, 1989; Graham and Barnett, 1995). Thus, research into the relationship of ENSO events to weather outside the tropics has produced a much more confused picture. In the context of the U.S., Namias (1978) made the first successful prediction of U.S. weather based on an ENSO event. He and his colleagues predicted that the 1976–77 winter would be colder than normal on the East Coast, the Gulf Coast wetter, and the West Coast warmer and drier. These conditions were similar to those of 1957 which was also an El Niño year. Barnett (1981) found a good correlation between ENSO events and winter temperatures on the West Coast and the Gulf coast. In contrast, he found no significant association in the interior of the U.S. However, Graham and Barnett (1995), in a later study, claimed that the strongest association between ENSO events and extra-tropical weather anomalies is in winter temperatures in the interior of North America. In reviewing the relationship to coastal precipitation, Stewart (1995) could account for 17 percent of the variance in Texas precipitation with the SOI. He went on to claim that the correlation with winter rains alone

34

Chapter Four

should be greater. He also believed that the amount of summer rains should increase, especially along the Gulf Coast in anti-El Niño (La Niña) years. Stahle and Cleveland (1988) also found the El Niño signal in the Texas tree ring record beginning in 1680. However, Haston and Michaelsen (1994) could find no clear relationship to precipitation in California based on the 600 year record from the rings of the big-cone spruce tree perhaps because some El Niño years are dry (1976–77) and some are wet (1994–95) in California. Gray and Sheaffer (1991) found that the incidence of Atlantic hurricanes increased during La Niña years. Trenberth and Branstator (1992) and Namias (1991) found that summer drought in the central U.S. was positively correlated with La Niña years. Based on computer model simulations, Atlas et al. (1993) found a relationship between tropical Pacific sea surface temperatures (SSTs) and reduced precipitation in the interior of the U.S. If reduced soil moisture was included in the model, then this caused an even greater reduction in precipitation and a significant increase in temperature. Bell and Janowiak (1995) related the severe 1993 floods in the Midwest to a strong El Niño event at the same time.

3. Methodology and Data The magnitude of the El Niño event is measured by the Southern Oscillation Index (SOI). This is based on the difference in pressure anomalies between Tahiti and Darwin, Australia. If both locations have normal pressure for that season then the values are near 0.0. During an El Niño event the values are negative. During a La Niña event the values are positive. The SOI has been calculated monthly for the period from 1900 until the present. Prevailing climatic conditions in the U.S. are measured by the Palmer Drought Index in its various forms. The index was developed by Wayne Palmer (1965) of the U.S. Weather Bureau. The index measures the departure from normal conditions for a particular place and time. It has a value near 0.0 when conditions are normal. A value of –2.0 indicates a moderate drought while a value of –4.0 indicates an extreme drought. In contrast, a value of +2.0 indicates moderately wet conditions and a value of +4.0 indicates extremely wet conditions. The index is calculated by climatic division, with some 344 of them in the 48 contiguous states. Monthly values are available from 1895 until the present. The Palmer Index has become the principal tool for measuring prevailing climatic conditions or anomalies in the U.S. There are two principal versions of the

Impact of EN NSO Events on U.S. Climate A Anomalies

35

index. The P Palmer Drougght Severity In ndex (PDSI) w was the originaal version derived by Wayne Palm mer. The Palmer Hydroloogical Droug ght Index (PHDI) wass developed laater and incorp porates longerr persistence and more accurately ddepicts the relaation to waterr resources. T The latter index x (PHDI) was used in this study. Correlatiion and regrression were used to deteermine the degree d of association between the SOI and PH HDI. The PHD DI from each h climatic division in tthe country was w correlated with the SOI for the 90 yeear period of record. The PHDI was lagged by tw wo months afteer the associatted value. Maps were pproduced show wing the r-vallues for each cclimatic divission using ATLAS-GIS S software.

4. Results R and d Discussion n Figure 44-1 is a time-sseries graph of o the SOI froom 1900 to 1990. The index showss a very high frequency f variiation with freequent movem ment from one extremee (El Niño) too another (La Niña). Figuree 4-2 is a similar timeseries graphh of the Palmeer Index for th he central Kaansas climatic division. The climaticc anomalies arre much moree persistent buut also have raapid shifts from one exxtreme to the other. o

Figure 4-1: Soouthern Oscillaation Index, 190 00–1990

36

Chapter Four

ndex Kansas D Division 5, Centrral Figure 4-2: Paalmer Hydrologgical Drought In

The mapp of the correelation coefficients (Figuree 4-3) indicatted much lower statisstical associaations than anticipated. a T The strongest positive correlation w was .24 and the t strongest negative corrrelation was –.29. – The coastal Paciffic Northwestt and the Tenn nessee, Kentuccky, and Westt Virginia region show wed generally positive p assocciations. This iindicates that when the SOI is negattive (El Niño)), these region ns have drier cconditions. In n contrast, the desert S Southwest, Teexas, and Florrida all show negative asso ociations. During El N Niño events, thhese locations are wetter thaan normal. For all thhe attention ENSO E events have receivedd, they have almost a no statistical poower to predict climatic an nomalies overr most of the U.S. The best that cann be said is thhat the associaations are highher in the sou uthern tier of states whhich are subttropical locatiions, and, theerefore, the ones o most likely to be aaffected by troopical patterns. In spite oof the low stattistical associaation betweenn the SOI and PHDI for the past cenntury, there arre, neverthelesss, some tantaalizing relatio onships in the record. For examplee, the very seevere droughtt of the 1950 0s in the central U.S. broke suddenly with very y heavy rains in the spring of 1957. During Januuary to Februuary 1957, th he SOI channged from pro onounced positive valuues to pronouunced negativ ve values. Thiis change wass perhaps the most rappid in the 100 year record off the SOI. In ccontrast, the 1930s

Impact of ENSO Events on U.S. Climate Anomalies

37

Figure 4-3: Correlation between Southern Oscillation Index and Palmer Hydrological Drought Index Two Month Lag, 1900–1990

drought occurred during a period of comparatively static SOI – one without serious positive or negative fluctuations. Thus, a profitable avenue for future research would be to focus analyses on shorter time periods relating to specific events, such as the 1950s drought in the central U.S. or the 1982–83 El Niño event (the strongest on record). Given the seasonal nature of El Niño relationships, it would also be profitable to focus on the anomalies during each season in specific locations.

5. Conclusions Weather patterns in the mid-latitudes are complex, and while they may be linked in some way to events in the tropics, they are not controlled by them. While the strongest associations between the SOI and the PDI did not account for more than 10 percent of the variance, the relationship

38

Chapter Four

could still be an important factor in understanding drought and floods. There is an important distinction between relating a specific drought event to a specific ENSO event versus relating all drought events to all ENSO events. Abnormal dry spells (or wet ones) in the U.S. have a multiplicity of causal forces. To the extent that the Southern Oscillation has any predictive value on the climatic variability of a particular place, it is important to know where, when, how much, and in what way.

References Atlas, R., Wolfson, N. and Terry, J. 1993. The effect of SST and soil moisture anomalies on GLA model simulation of the 1988 U.S. summer drought. Journal of Climate 6, 2034–2048. Barnett, T.P. 1981. Statistical prediction of North American air temperatures from Pacific predictors. Monthly Weather Review 109, 1021–1041. Barnett, T.P., Bengtsson, L., Arpe, K., Flügel, M., Graham, N., Latif, M., Ritchie, J., Roeckner, E., Schlese, U., Schulzweida, U., and Tyree, M. 1994. Forecasting global ENSO-related climate anomalies. Tellus 46A, 381–397. Barnett, T.P., Graham, N., Cane, M., Zebiak, S., Dolan, S., O’Brien, J., and Legler, D. 1988. On the prediction of the El Niño of 1986–87. Science 239(4862), 192–196. Bell, G.D. and Janowiak, J.E. 1995. Atmospheric circulation associated with the Midwest floods of 1993. Bulletin of the American Meteorological Society 76, 681–695. Bradley, R.S., Diaz, H.F., and Kiladis, G.N. 1987. ENSO signal in continental temperature and precipitation records. Nature 327(6122), 497–501. Bunkers, M.J. and Miller Jr., J.R. 1996. An examination of El Niño-La Niña related precipitation and temperature anomalies across the northern plains. Journal of Climate 9, 147–160. Canby, T.Y. 1984. El Niño’s ill wind. National Geographic 165, 144–183. Chu, P.S. 1995. Hawaii rainfall anomalies and El Niño. Journal of Climate 8, 1697–1703. Diaz, H.F. and Markgraf, V. (eds.). 1993. El Niño: Historical and Paleoclimatic Aspects of the Southern Oscillation. Cambridge, UK: Cambridge University Press. Glantz, M.H., Katz. R.W., and Nicholls, N. (eds.). 1991. Teleconnections Linking Worldwide Climate Anomalies. Cambridge, UK: Cambridge University Press.

Impact of ENSO Events on U.S. Climate Anomalies

39

Graham, N.E. and Barnett, T.P. 1995. ENSO and ENSO-related predictability. Part II: Northern hemisphere 700-mb height prediction based on hybrid coupled ENSO model. Journal of Climate 8, 544–549. Gray, W.M. and Sheaffer, J.D. 1991. El Niño and QBO influences on tropical cyclone activity. In M.H. Glantz, R.W. Katz and N. Nicholls (eds.), Teleconnections Linking Worldwide Climate Anomalies: Scientific Basis and Societal Impact. Cambridge: Cambridge University Press, 157–284. Haston, L. and Michaelsen, J. 1994. Long-term central coastal California precipitation variability and relationships to El Niño-Southern Oscillation. Journal of Climate 7, 1373–1387. Kiladis, G.N. and Diaz, H.F. 1989. Global climatic anomalies associated with extremes in the Southern Oscillation. Journal of Climate 2, 1069– 1090. Namias, J. 1978. Multiple causes of the North American abnormal winter of 1976–77. Monthly Weather Review 106, 279–295. —. 1991. Spring and summer 1988 drought over the contiguous United States - causes and prediction. Journal of Climate 4, 54–65. —. 1982. Anatomy of Great Plains protracted heatwaves (especially the 1980 summer drought). Monthly Weather Review 110, 824–838. Palmer, W.C. 1965. Meteorological Drought. Research Paper No. 45, Washington DC: U.S. Weather Bureau. Ropelewski, C.F. and Halpert, M.S. 1986. North American precipitation and temperature patterns associated with El Niño-Southern Oscillation. Monthly Weather Review 114, 1101–1106. Ropelewski, C.F. and Halpert, M.S. 1989. Precipitation patterns associated with the high index phase of SO. Journal of Climate 2, 268–284. Stahle, D.W. and Cleveland, M.K. 1988. Texas drought history reconstructed and analyzed from 1698 to 1980. Journal of Climate 1, 59–74. Stewart, R.H. 1995. Predictability of Texas rainfall patterns on time scales of six to twelve months: A review. In Norwine, J., Giardino, J.R., North, G.R., and Valdes, J.B. (eds.), The Changing Climate of Texas, College Station, TX: Geobooks, 39–48. Trenberth, K.E. and Branstator, G.W. 1992. Issues establishing causes of the 1988 drought over North America. Journal of Climate 5, 159–172.

CHAPTER FIVE IMPACT OF ENSO EVENTS ON U.S. CLIMATE ANOMALIES: UPDATE SINCE 1996 KENT M. MCGREGOR

1. Many Climate Indices Available In 1996 the Southern Oscillation Index (SOI) was perhaps the most common measure of conditions in the tropical Pacific and the status of El Niño/La Niña. The SOI is based on the pressure differences between Tahiti and Darwin, Australia. Since then, superior measures have become available based on sea surface temperatures (SSTs). These indices typically measure the temperature (anomaly) in degrees Celsius in some portion of the tropical Pacific Ocean. Figure 5-1 shows a map of these regions. Figure 5-2 is a map of the correlation coefficients between the Niño 4 index and precipitation from January to March for all 344 climate divisions from 1982 to 2000. El Niño/Southern Oscillation (ENSO) events tend to have their greatest impact during the winter months, usually increasing precipitation along the southern tier of states and reducing precipitation in the Northwest (Mo et al., 2009; McGregor, 2010). As is often the case, stronger correlations occurred when the precipitation lagged the index date by 2–3 months, so the Niño 4 index leads the precipitation data by a season (3 months). The highest positive correlations, above .5, occurred in the in the Southwest, the coastal Southeast and where Nebraska, South Dakota, and Iowa meet. The most negative correlations, below –.5, occurred in the Ohio River Valley and southern Montana, but were not as extensive in area as the locations of strongest positive correlations. While other climate oscillations known before 1996, much of the focus was on ENSO when perhaps the strongest event of the century occurred in

Impact oof ENSO Eventts on U.S. Clim mate Anomalies : Update since 1996

41

Regions in the Pacific Ocean fo or Four ENSO IIndices Figure 5-1: R

Figure 5-2: C Correlation betw ween Niño 4 an nd January to M March Precipitaation Index Leads Precipiitation by Threee Months

1997–98. S Since then, the t National Oceanographhic and Atm mospheric Administration (NOAA) and a other scieentists have foound quite a nu umber of

42

Chapter Five

additional oscillations in the Earth’s climate system. It could almost be said that oscillation hunting has become a growth industry in climate science. NOAA’s Physical Sciences Division (PSD) has made data available for many oscillation indices on their websites and also provided analytical tools that run in real time (NOAA, 2014a). Table 5-1 lists many of the indices but is by no means a complete list. Table 5-1: Indices of Climate Oscillations Teleconnections:

Atmosphere:

Pacific North American (PNA) West Pacific (WP) North Atlantic Oscillation (NAO) North Atlantic Oscillation (NAO) Jones Eastern Asia/Western Russia (EA/WR) North Pacific (NP) Northern Oscillation (NOI) Pacific Decadal Oscillation (PDO) ENSO:

Quasi-Biennial Oscillation (QBO) Global Angular Momentum Southern Oscillation Index (SOI) Antarctic Oscillation (AAO)

Multivariate Enso Index (MEI) Oceanic Niño Index (ONI) Niño 1+2

Under SST: Atlantic Tropical Southern Atlantic Index Tropical Northern Atlantic Index (TNA) Tropical Southern Atlantic Index (TSA) Atlantic Tripole Atlantic Multi-decadal Oscillation (AMO) Atlantic Meridional Mode (AMM) North Tropical Atlantic Index (NTA) Caribbean Index (CAR)

Niño 3 Niño 3.4 Niño 4 Trans-Niño Index (TNI) Western Hemisphere Warm Pool (WHWP) Pacific Warm Pool

Arctic Oscillation (AO) Madden-Julian Oscillation (MJO)

Tropical Pacific EOF:

For example, Figure 5-3 shows the correlation coefficients across the U.S. between the West Pacific (WP) Index and winter precipitation from 1948 to 2011. The index leads precipitation values by two months. The correlations are not as strong as those in Figure 5-2 indicating that ENSO

Impact oof ENSO Eventts on U.S. Clim mate Anomalies : Update since 1996

43

Figure 5-3: Correlation beetween West Pacific Index and January to March wo Months Precipitation, 1948–2011 Inddex Leads Preccipitation by Tw

has more im mpact on U.S. precipitation patterns tthan the Wesst Pacific Oscillation. The areas of highest possitive associattions (.3) weere in the coastal Souutheast and also a in south hwest Kansass and the Oklahoma O Panhandle. T The most neggative associaations (–.2) weere found in the t lower Ohio River Valley and a band from southeastern s W Washington across a the Idaho Panhaandle and intoo southwestern n Montana. Thhe spatial disttributions across the U U.S. look sim milar on both maps, indicaating that the Southern Oscillation pprobably influuences the Weest Pacific Osccillation.

2. Development of thee Reanalysiis Model Another important innovation sin nce 1996 is the develop pment of NOAA’s Reeanalysis Moddel (Kalnay ett al., 1996). T This atmospherric model and the dataa processed byy it are truly tributes to m modern technollogy. The original Reaanalysis was a comprehensive global m meteorologicall data set from 1948 uuntil the presennt. It is based on both obserrvational meassurements

44

Chapter Five

and model calculations (also called the assimilation system). The observational data range from the usual surface station observations, to upper air soundings from weather balloons, to measurements from satellites. The variables are classified as “A” variables, based solely on observations, “B” variables primarily based on observations but with a small component of model calculations, and “C” variables (such as precipitable water and surface energy fluxes) which are based solely on model calculations. Since all of the data are processed through the same model, the results are directly comparable to each other. For example, it is then possible to compare ENSO events that occurred decades apart. Many published studies have demonstrated that the outputs are reliable estimates of actual meteorological conditions, and even the ones with the greatest error are generally accurate within a 5 –10% range of error. For example, Figure 5-4 was constructed with data from the reanalysis model (NOAA, 2014b). The map shows the potential evapotranspiration (PE) anomaly for the spring months prior to extremely four hot summers in the modern record: 1980, 1988, 1998, and 2011. Often, unusually low soil moisture values during the spring are followed by unusually hot summers. The high potential evapotranspiration anomaly is tied directly to low soil moisture values (McGregor, 2012). Less solar energy goes to evaporate water and more goes to heat the surface, and the surface temperature rises. This increases the rate of PE. The map shows strong high PE anomalies over the southern Great Plains from north Texas into Oklahoma in the spring season during these years. In conclusion, over the last two decades or so, NOAA has made various models, data bases and analytical tools available on-line. These can be something as simple as a time series graph of an ENSO index to correlations between any climate oscillation index and any Reanalysis variable. Furthermore, this can be done for virtually any region on the Earth. Consequently, the sophistication of atmospheric research and the speed at which it can be done has increased by orders of magnitude compared to what was possible in 1996, and all at the click of a button.

Impact oof ENSO Eventts on U.S. Clim mate Anomalies : Update since 1996

45

Figure 5-4: Potential Evaapotranspiratio on Anomaly April, May, and a June; 1980, 1988, 1998, and 2011

Refereences M Kistler, R., Collins, W.., Deaven, D., Gandin, Kalnay, E., Kanamitsu, M., L., Iredeell, M., Saha, S., White, G.,, Woollen, J., Zhu, Y., Leeetmaa, A., Reynoldds, R., Chelliaah, M., Ebisuzzaki, W., Higggins, W., Jano owiak, J., Mo, K.C C., Ropelewskki, C., Wang, J., Jenne, R.,, and Joseph, D. 1996. The NCE EP/NCAR 40--year reanalyssis project, Buulletin of the American. A Meteorological. Socieety 77, 437–47 70. McGregor, K K.M. 2010. Analysis A of preecipitation pattterns in the Southwest S during thhe 2009–2011 El Niño even nt. Papers of tthe Applied Geography Conferennces 33, 65–74. —. 2012. T The record sum mmer of 2011 1: Heat and ddrought in the southern Plains. P Papers of the Applied A Geogrraphy Confereences 42–51. Mo, K., Scchemm, J., annd Yoo, S. 2009. 2 Influennce of ENSO O and the Atlantic Multidecadall Oscillation on drought oover the Uniteed States. Journal of Climate 222, 5962–5982. NOAA. 2014a. http://ww ww.esrl.noaa.go ov/psd/data/cllimateindices//list/. Last acceessed June 15, 2015. —. 2014b. http://www.eesrl.noaa.gov/p psd/webswitch ch.html. Last accessed June 15, 2015.

CHAPTER SIX NORTHERN HEMISPHERE FLOW ANOMALIES AND U.S. TEMPERATURE AND PRECIPITATION VARIABILITY ANTHONY J. VEGA, C.H. SUI AND K.M. LAU (1997)

1. Introduction The need to understand spatial and temporal variations of atmospheric related phenomena has grown despite continued technological advances. This is especially evident at the regional-scale as phenomena, both surface and atmospheric, are difficult to model, interpret, and link to larger-scale features. One method of linking surface parameter variations to generalized atmospheric flow is through correlations with atmospheric teleconnections. Atmospheric teleconnections are large-scale correlation fields in a variable (pressure or geopotential height) that represent standing long waves in the atmospheric flow. Wallace and Gutzler (1981) identified many such atmospheric teleconnections in the 700 mb geopotential height field. They also devised several simple indices to detect the presence and strength of the most important teleconnections in the Northern Hemisphere at a given time. Use of such indices for climate applications has grown tremendously in recent years (Leathers et al., 1991; Rogers and Rohli, 1991). However, many studies do not fully integrate aspects of many teleconnection indices to define links between the surface environment and the atmosphere. This study uses teleconnection indices determined to be most important to the climate of North America to further define the linkage between the atmosphere and United States regional precipitation and temperature variations.

Northern Hemisphere Flow Anomalies

47

2. Data and Methods The regionalization of U.S. precipitation and temperature is described in Fiscus and Vega (1997). The derived climate regions and the associated climate anomaly time-series from Fiscus and Vega (1997) are used exclusively in this study. Surface data were composed of regionally clustered standardized precipitation and temperature anomalies calculated annually and seasonally (winter=DJF). The clustering procedure utilized rotated principal components analysis (RPCA) of temperature and precipitation data extracted for all 344 United States climate divisions for the 1895–1991 time period. The data were standardized by month (mean=0; std=1) and merged into annual and seasonal arrays. PCA using a Pearson correlation matrix was used to isolate discrete regions of covariation. Varimax rotation techniques were employed to better determine the resulting solutions. Component retention was determined through a combination of the Kaiser-Guttman eigenvalue 1 criterion and a scree plot (Henderson and Vega, 1996). Component loadings were plotted to determine the spatial resolution of the eigenvectors utilizing a 0.45 eigenvector isoline cutoff for the precipitation variable and a 0.5 cutoff for the temperature variable. This procedure was followed in order to define the associated regions precisely with little transitory overlap. Once the final regions were produced spatially, a time-series of the precipitation and temperature anomalies was created to isolate distinct temporal regions of a particular phenomenon. A clustering procedure was used which combined the actual precipitation and temperature anomalies for each climate division included within a particular region. Loadings maps representing the spatial configuration of the components for all annual and seasonal precipitation and temperature regions are presented in Figure 6-1. For the teleconnection analysis, this study utilizes the Pacific-North American (PNA) teleconnection index and the Southern Oscillation Index (SOI) to estimate generalized Northern Hemisphere atmospheric flow. Data were obtained from the Climate Analysis Center. The PNA timeseries used here extends from 1947–1986 and the SOI time-series extends from 1895–1991. The indices were derived from either standardized geopotential heights (PNA) or standardized sea-level pressure variations (SOI). The teleconnection indices are tied to U.S. regional surface parameters (precipitation and temperature) through Pearson correlation coefficients. Because the teleconnection indices have different record lengths, some bias may be introduced into the correlation analysis as higher degrees of

48

Chapter Six

Figure 6-1: United States Annual – Autumn Precipitation (A-E) and Temperature (1-5) Regions

Northern Hemisphere Flow Anomalies

49

freedom are associated with lower amounts of explained variance in significant correlations.

3. Analysis The PNA teleconnection is undoubtedly one of, if not the, most important teleconnection pattern with regard to the North American surface climate. This upper atmospheric pattern has centers of action near the North Pacific subtropical high, the Aleutian low, northwestern North America, and the Florida panhandle. The major characteristic of the PNA flow regime is a ridge, evident in the 700 mb geopotential height field, centered on the Rocky Mountains with adjacent long-wave troughs near the Aleutian and Florida action centers. Changes in the strength and amplitude of the PNA pattern are known to have significant impacts on North American temperature (Rogers and Rohli, 1991) and precipitation (Henderson and Vega, 1996). The pattern has been shown to be temporally persistent on a monthly (Barnston and Livezey, 1987) and intra-monthly time-scale, in addition to being evident at various tropospheric levels (Vega et al., 1995). Wallace and Gutzler (1981) derived a simple index which estimates flow variations in association with the PNA teleconnection pattern. The index denotes meridional (amplified ridgetrough) long-wave flow through positive index values, zonal (deamplified) flow through near-zero values, and reversed flow (meridional troughridge) through negative values. Due to the discontinuous nature of precipitation, few significant relationships are found between the PNA index and the annual time-series. Only the Mid-West (PC1) and Southwest (PC9) regions show a direct relationship with the flow patterns. The weak correlations explain only a small amount of variance and are significant only at the 0.10 alpha level. However, the analysis indicates the relative importance of the long-wave flow regime on the surface climate as a negative correlation exists between the Mid-West region and the index. Zonal flow regimes are therefore responsible for positive precipitation anomalies in this area. Zonal flow ensures that mean wave cyclone trajectories pass over this region. Meridional flow places the mean cyclone trajectories too far to the south during the high positive PNA phase, resulting in negative regional precipitation anomalies. Meridional flow associated with the reversed phase of the PNA places the region under the influence of the mean upperair ridge which also significantly reduces regional precipitation. Identifying physical causes associated with the positively associated

50

Chapter Six

Southwest region is more problematic. The correlation indicates that amplified longwave flow relates significantly to positive precipitation anomalies within the region. This relationship may extend from the dissipation of Pacific induced wave cyclones during positive PNA times. Zonal flow places dissipating cyclones poleward of the region leading to negative precipitation anomalies. Winter correlations fare much better than those of the annual timeseries, validating many previous studies comparing the PNA to the cool season surface environment (Rogers and Rohli, 1991). Significant relationships are found between the index and the Mid-South (PC2), the Mid-West (PC3), the Northwest (PC5), the Upper Plains (PC7), and the Northern Rockies (PC9) precipitation regions. All of the significant correlations are negative, indicating that zonal flow regimes, or reversed flow regimes, relate to enhanced precipitation in these areas. This is climatologically plausible as zonal flow is typically indicative of a poleward displaced mean jet, resulting in wave cyclone trajectories over the northern tier of climate regions in the west and the regions which traverse the upper Mississippi and Ohio River valleys in the east. Of the four significant springtime regions, only the Upper Plains (PC7) shows a weak negative relationship. Positive relationships exist with the Southeast (PC3), the Southern Plains (PC8), and the Central Plains (PC9). This indicates the importance of wave cyclogenesis in the Colorado cyclogenetic region and the resulting trajectories. Positively amplified flow places the mean eastern trough in close proximity to all of the related regions except the Upper Plains (PC7) which lies poleward of the flow axis during these times. Henderson and Vega (1996) showed that the PNA is a reliable flow indicator during the warm season as it relates to the expansion/contraction of pressure cells in close proximity to the PNA action centers. This relationship is demonstrated by the positive correlations between the index and the Southeastern (PC1) and South Central (PC7) regions. The index is influenced by enhanced onshore flow and low level moisture advection associated with the North Atlantic Subtropical High (STH). Such flow induces low level instability in the southeastern U.S. which enhances precipitation. During the autumn season, the PNA index correlates negatively with the Mid-West (PC1), the Northwest (PC4), the Southern Plains (PC5), the Northeastern Plains (PC7), and the Northwestern Plains (PC8). Physical mechanisms responsible for such relationships are similar to the annual

Northern Hemisphere Flow Anomalies

51

time-series as zonal flow significantly affects the northwestern tier regions due to a higher mean latitude of long-wave flow. Surface temperature relationships are easily established with the PNA. Significant correlations agree with previous studies (Leathers et al., 1991) in that negative temperature departures exist in relation to mean troughing while positive relationships relate to prominent ridging. This is true of annual as well as seasonal relationships as many of the temperature regions correlate significantly with the index. The winter and autumn time-series relate especially well to index variations. Many studies provide insight into the relationship between the SOI and the North American surface climate. Douglas and Englehart (1982) provide an in-depth analysis of the links between U.S. precipitation and the SOI, especially for Florida where the SOI explains approximately one half of Florida’s precipitation variability. Further links have been established by Henderson and Vega (1996) for the southern U.S., Vega et al. (forthcoming) for mid-tropospheric variations over the Gulf of Mexico related to extreme SOI phases, and U.S. precipitation patterns and variability (Ropelewski and Halpert, 1989). Although many relationships are apparent, the physical link between atmospheric and SST variations in the equatorial Pacific and the North American surface climate remain ambiguous. Analysis of monthly geopotential height fields reveals a striking correlation between the negative (positive) phase of the SOI (El Niño) and the positive (negative) PNA pattern. It has been surmised that equatorial variations force the PNA into the highly amplified wave pattern through increases in tropic-to-mid-latitude energy and moisture. This is evident through a stronger and more persistent subtropical jet over the western and southern U.S. which induces Gulf of Mexico cyclogenesis (Vega et al., forthcoming), leading to positive precipitation anomalies in much of the United States. This study confirms the relationship between positive precipitation departures in much of the U.S. and the negative phase of the SOI. On an annual basis much of the central U.S. (PCs 5–8) correlates significantly with the SOI. During the El Niño (La Niña) phase of the SOI, positive (negative) precipitation anomalies occur throughout these regions. The Southwest (PC9) region also negatively correlates to the SOI. This is due to increased cool season cyclogenesis off the Pacific Coast during El Niño times caused by an anomalous warm water pool which supports low level

52

Chapter Six

instability. This couples with upper level venting associated with the strengthened subtropical jet, inducing Pacific cyclogenesis. During the winter season, only three precipitation regions correlate negatively to the SOI; the Southeast (PC4), the Lower Plains (PC6), and the Southwest (PC8). During the winter season, the Southeast precipitation region is zonally more inclusive than during the annual time-series. The winter season region includes much of the South Central annual region, implying that much of the association between the Southeast and the SOI resides over the western portion of the region. The Southwest also correlates negatively to the SOI, signifying the link between the warm water anomalies in the southwestern Pacific and cyclogenesis near the west coast of the United States. The impact of more frequent cyclogenesis carries through to the spring season as three precipitation regions correlate negatively to the SOI. All of the associated regions are located in the central to western U.S., implying a significant impact from increased cyclogenetic frequency off the California coast. During summer, two regions correlate significantly with the SOI: the Southeast (PC1) which is positively related and the Southwest (PC8) which is negatively related. The correlations are consistent with those found during other seasons. The autumn season depicts three regional relationships with the SOI. The Lower Plains (PC5), the Northeastern Plains (PC7), and the Southwest (PC9) are all negatively related. Physical links may be tied to increased frequencies of late-season wave cyclones. Regional surface temperature relationships (annual-autumn) show predominantly negative relationships to the SOI. Therefore the physical relationship denotes positive (negative) temperature anomalies during the El Niño (La Niña) SOI phase. This physical link relates to the amplification of the mean longwave flow which produces greater temperature anomalies than during relatively steady zonal flow regimes.

4. Conclusions This study attempts to relate previously derived precipitation and temperature regions to mean atmospheric flow anomalies represented by the Pacific-North American teleconnection pattern and the Southern Oscillation. Results indicate that mean zonal flow across North America relates to broadly defined precipitation increases. This is due to precipitation forcing along the greatest expanse of the conterminous

Northern Hemisphere Flow Anomalies

53

United States. During highly amplified flow regimes, precipitation forcing is confined to areas south and east of the deepened trough. Mean positive precipitation departures are also evident during El Niño events which relate well to increases in energy and moisture into the United States. Temperature relationships are easily defined with areas near mean ridges recording positive anomalies while regions near mean troughs record negative anomalies.

References Barnston, A.G. and Livezey, R.E. 1987. Classification, seasonality and persistence of low-frequency atmospheric circulation patterns. Monthly Weather Review 115, 1083–1126. Douglas, A.V. and Englehart, P.J. 1982. On a statistical relationship between autumn rainfall in the central equatorial Pacific and subsequent winter precipitation in Florida. Monthly Weather Review 190, 2377–2382. Fiscus, D.M. and Vega, A.J. 1997. The derivation of United States temperature and precipitation regions, Proceedings of the National Council for Undergraduate Research (forthcoming). Henderson, K.G. and Vega, A.J. 1996. Regional precipitation variability in the southern United States. Physical Geography 17, 93–112. Leathers, D.J., Yarnal, B., and Palecki, M.A. 1991. The Pacific/North American teleconnection pattern and the United States climate. Part I: Regional temperature and precipitation associations. Journal of Climate 4, 517–528. Rogers, J.C. and Rohli, R.V. 1991. Florida citrus freezes and polar anticyclones in the Great Plains. Journal of Climate 4, 1103–1113. Ropelewski, C.F. and Halpert, M.S. 1989. Precipitation patterns associated with the high index phase of the Southern Oscillation. Journal of Climate 2, 268–284. Vega, A.J., Henderson, K.G., and Rohli, R.V. 1995. Comparison of monthly and intramonthly indices for the Pacific/North American teleconnection pattern. Journal of Climate 8, 2097–2103. Vega, A.J., Rohli, R.V., and Henderson, K.G. The Gulf of Mexico midtropospheric response to El Niño and La Niña forcing. Climate Research (forthcoming). Wallace, J.M. and Gutzler, D.S. 1981. Teleconnections in the geopotential height field during the northern hemisphere winter. Monthly Weather Review 109, 784–812.

CHAPTER SEVEN A CLIMATOLOGICAL ANALYSIS OF LOWER ATMOSPHERIC OZONE TRANSPORT ACROSS PHOENIX, ARIZONA MARK L. HILDEBRANDT (2000)

1. Introduction Each summer, high concentrations of lower atmospheric ozone pose a threat to the respiratory health of the inhabitants of greater metropolitan Phoenix, Arizona. Prolonged exposure to the aerosol may lead to respiratory infection and lung inflammation, with the most drastic result being irreversible changes in lung structure and chronic respiratory illness. As a potential threat to human health, high levels of lower atmospheric ozone across the Phoenix urban area have been a focus of the United States Environmental Protection Agency (EPA) and local government agencies over the past decade (Arizona Department of Environmental Quality (ADEQ), 1988; U.S. EPA, 1998). The Phoenix metropolitan area is particularly susceptible to episodes of high concentrations of lower atmospheric ozone (ADEQ, 1988; U.S. EPA, 1998), and not simply because of the rapidly growing population. The physical geography of the area is as important as the human aspect in producing this particular environmental problem. From a human geography perspective, it is clear that the large population of automobile users within the Phoenix metropolitan area is a significant anthropogenic source of nitrogen dioxide (NO2), a gaseous precursor to ozone (O3). When combined with natural NO2, the concentration of the aerosol in the atmosphere near the Earth’s surface can be high and makes efficient the photochemical process whereby solar radiation disassociates NO2 to ultimately produce ozone (Sillman, 1999). However, the occurrence of

Analysis of Lower Atmospheric Ozone Transport

55

high concentrations of lower atmospheric ozone across Phoenix is critically dependent upon the physical geography of the region. The Phoenix metropolitan area is located at low latitude (33.3°N) in an arid environment. Taken together, the product is a large amount of insolation that drives efficient photochemical production of ozone. In addition, the circulation of the lower atmosphere, and dispersion of ozone, is limited by the topography that surrounds the Phoenix area. The complex topography of the region often dictates the flow of the lower atmosphere (Blumen, 1990), and as the Phoenix metropolitan area (Figure 7-1a) is situated in a valley surrounded by significantly higher terrain (Figure 71b), pollution dispersion is often limited. Furthermore, the location of the region under an area of predominant high pressure results in a consistent lack of a strong background circulation on the synoptic scale. This also precludes significant mixing of the atmosphere and transport of aerosols away from the urban area (Altshuller, 1978). For the very same reasons that efficient pollution dispersion away from the Phoenix area is difficult, it is thought that the movement of lower atmospheric aerosols within the urbanized valley often follows a distinct pattern. It has been demonstrated that the Phoenix Valley area is often subjected to a daytime mountain-valley mesoscale circulation. (Green and Sellers, 1964; Sellers and Hill, 1974; Ellis et al., 2000). The Phoenix Valley serves as a theoretically ideal platform for the development of such a circulation, with large amounts of solar radiation warming the arid mountain-valley environment and no strong background atmospheric circulation to interfere. Capable of transporting lower atmospheric aerosols, the mesoscale circulation is generally thought to result in atmospheric flow up the warmed eastern valley slopes during midday and afternoon hours. As the temperatures of the air overlying the warmed slopes increase, the air becomes less dense, more buoyant, rises, and is replaced with air moving upslope from the valley floor (Arrit and Pielke, 1986), producing an anabatic wind. The result is often the transport of lower atmospheric ozone from areas of high ground traffic near the urban center toward eastern Phoenix suburbs and the higher elevation areas farther east (Ellis et al., 2000). During nighttime hours, the mountainvalley thermal circulation often reverses to a downslope flow or katabatic wind. Rapidly cooling air overlying the mountain slopes becomes denser and consequently “drains” back into the valley below (Simpson, 1997). A significant question is whether or not the cool air drainage plays a role in the diurnal transport of lower atmospheric ozone.

56

Chapter Seven

Figure 7-1: The Phoenix Metropolitan Area Cities (A) and Surrounding Topography (B) (Black Dot Indicating Falcon Field) (Figure 7-1A Overlays Figure 7-1B)

Analysis of Lower Atmospheric Ozone Transport

57

Aerosol dispersion has been modeled (Berman et al., 1995; Loibi, 1997) and several studies have linked aerosol concentrations to weather patterns of varying scales (Regener, 1957; Niemeyer, 1960; Hosler, 1961; Peterson and Flowers, 1977; Altshuller, 1978; Lalas et al., 1983; Davis and Gay, 1993a; Davis and Gay, 1993b; Schlink et al., 1999; Sillman, 1999). However, as with most aerosols, observational studies of ozone transport are preferred over those that attempt to model the movement as particulate integrity is better maintained without having to simulate the production and destruction of the aerosol. The study that is presented here aimed to provide evidence of the complete diurnal process by which lower atmospheric ozone is transported across the valley within which is situated the Phoenix metropolitan area. The study examined the mesoscale spatial distribution of ozone on high ozone days, while focusing on the mesoscale atmospheric flow responsible for the temporal variation in the spatial patterns. In order to better understand the movement of ozone through the diurnal period, microclimatic data were combined with mesoscale and synoptic scale atmospheric data. The condition of the atmosphere influences the evolution of the mountain-valley thermal circulation, making the study of ozone transport across the Phoenix metropolitan area uniquely appealing to a geographer.

2. Data and Methodology Microclimatic data were gathered from a point location in an area east of the Phoenix urban center that is subject to high afternoon ozone concentrations. Microclimatic instrumentation was installed on the grounds of a regional airport (Falcon Field) in Mesa, Arizona (Figure 71a), a densely populated and burgeoning eastern suburb of Phoenix. It is believed that the microclimate of the site was reasonably representative of the local area within which high ozone levels are common. In order to characterize the lower atmosphere, a tower for the support of microclimatological instruments was erected to a height of 6 meters. Sensors for measuring air temperature (°C), relative humidity (%), and wind speed (m s-1) were fixed at logarithmic heights of 0.6, 1.7, and 5.0 meters. A wind vane fixed atop the tower detected wind direction (degrees). A smaller tower, erected to a height of 1 meter, supported instrumentation for measuring barometric pressure (mb), total solar radiation (W m-2), reflected solar radiation (W m-2), net radiation

58

Chapter Seven

(shortwave and longwave; W m-2), and surface infrared temperature (°C). All tower-supported instruments were leveled and calibrated for accuracy, and also checked daily for high operational performance. Soil thermistors for measuring soil temperature (°C) were buried at depths of 0.01, 0.15, and 0.25 m, while two heat flux transducers for measuring soil heat flux (W m-2) were buried at depths of 0.03 and 0.14 m. For each instrument except the wind vane, measurements were taken on an interval of five seconds, averaged on an interval of five minutes, and recorded to automated dataloggers. Wind direction measurements were simply recorded at the end of each five-minute interval period. Data acquisition began on 11 July and ended on 12 September 1998. Coincident with the timing of the microclimatological measurements, scientists from the ADEQ recorded on-site ozone concentrations using air chemistry analyzers positioned at a height of 6 meters. Data were collected at one-minute intervals and subsequently calculated to one-hour averages centered on each half-hour. The collected ozone data were analyzed on a diurnal basis for the purpose of characterizing a general temporal pattern, particularly on days that ozone concentrations surpassed the EPA standard of an eight-hour average of 0.08 parts per million (ppm). Eleven such days were identified and analyzed in the spirit of determining the influence of any mountainvalley thermal circulation on the transport of ozone across the study site. In expanding the spatial scale of study during the full diurnal period, the spatial patterns of ozone and lower atmospheric meteorological data were analyzed for the 11 days on which the EPA ozone standard was exceeded at the microclimate site. Hourly ozone concentration data for 19 stations across the Phoenix Valley (not shown) were taken as a subset of a larger data set maintained by ADEQ and the Maricopa County Air Quality Division. Surface air temperature, wind speed, and wind direction measured at 16 stations across the Phoenix Valley (not shown) were taken as a subset of the Phoenix Real-time Instrumentation for Surface Meteorology Studies (PRISMS) database. Taken together, the ozone and meteorological data allowed for the construction of hourly ozone distributions and lower atmospheric circulation patterns across the Phoenix Valley for the prescribed 8-hour period of the 11 days of high ozone concentrations. Ozone and air temperature data were interpolated to a 4 x 4 grid covering the Phoenix Valley, with the spatial distribution of each

Analysis of Lower Atmospheric Ozone Transport

59

variable represented through contouring. Average vector winds were simply plotted among the contours of ozone concentration. After considering the microclimatic contributions to the mesoscale transport of ozone, the influence of the broader-scale atmospheric circulation on the evolution of the mountain-valley circulation was analyzed. Synoptic scale patterns of geopotential heights at 500 mb, and geopotential heights, wind directions, and wind speeds at 850 mb were characterized for the 11 days of high ozone. The analysis made use of data and software from the National Climate Diagnostics Center, which makes use of the National Center for Environmental Prediction reanalysis data.

3. Results and Analysis There was no preferred timing within the week or two-month study period for the 11 days of high ozone concentrations at the microclimate site in Mesa. In fact, approximately half (5) of the days occurred on the weekend, when automobile traffic should have been lessened. Each of the days was characterized by an increasing concentration of ozone in the atmosphere generally through 1700 LST (Figure 7-2). The timing of the peak in ozone concentration east of the urban center (4–5 hours after peak insolation) was likely to be at least partially a product of ozone transport eastward from areas of higher ground traffic. After 1700 LST, ozone concentrations steadily declined through sunset. However, on 6 of the 11 days of high ozone, the concentration of lower atmospheric ozone at the microclimate site in Mesa reversed its temporal trend during the nighttime hours. Ozone concentrations increased for 1–2 hours between sunset and midnight before returning to the trend of decreasing ozone in the lower atmosphere. Hourly maps of the spatial distribution of lower atmospheric ozone and vectors of the predominant wind (predominant wind direction, mean wind speed) indicate the transport of ozone over the course of high ozone days as stratified by data from the microclimate site in Mesa. During the hour 1100–1200 LST, the beginning of what appears to be an ozone plume is evident from the composite of ozone concentrations on high ozone days (not shown). The plume coincided with a generally light, east-to-west average wind pattern. The ozone plume became more evident as ozone concentrations increased through 1400 LST, with a maximum centered over Mesa (Figure 7-3A). Wind speeds increased in magnitude and a

60

Chapter Seven S

Figure 7-2: M Mean Ozone Conncentrations on n High Ozone D Days

southwesterly componennt to the win nd direction bbecame sligh htly more frequent on average. Not surprisingly, the ozone pluume extended eastward from the genneral area at which w several major roadwaays intersect in n Tempe, AZ (I-10, U U.S. 60, Rt. 2022, and Rt. 101 1). During tthe three-houur period lead ding up to andd including the t ozone maximum bbetween 1600 and 1700 LS ST (Figure 7--3B), the ozon ne plume expanded innto a large cenntralized areaa of high ozonne concentratiions. The area of maaximum ozonne concentratiion began too flow northeastward, coinciding with strongeer winds posssessing moree of a south hwesterly component. The movem ment of the ozzone was in the direction n of high elevation, but locally it followed f the Salt River. B Beyond 1700 LST, the average ozoone concentraation became diluted in thhe general arrea of its location betw ween 1600 annd 1700 LST (not shown). However, on n 6 of the 11 days of high ozone, the t concentrattion of lowerr atmospheric ozone at the microcllimate site inn Mesa reverrsed its tempporal trend du uring the nighttime hoours. Ozone concentration ns increased ffor 1–2 hours between sunset and m midnight befoore returning to the trend oof decreasing ozone in the lower atm mosphere.

Analysis of Lower Atmospheric Ozone Transport

61

Figure 7-3: Mean Ozone Concentrations (ppm) and Predominant Wind Direction with Mean Wind Speed during the Periods 1300–1359 LST (A), and 1600–1659 (B) LST (for Predominant Wind Vectors, 1.0 cm = 3 m s-1)

62

Chapter Seven

The spatial distribution of ozone on high ozone days may further support what the microclimatic data suggest in terms of an involvement of a mesoscale thermal circulation in the transport of lower atmospheric ozone. The well-defined plume of ozone that moved along the Salt River Valley toward high terrain during the afternoon hours, and subsequently the return flow at night, is indicative of such a circulation. In order for a mesoscale atmospheric flow to transport a relatively coherent concentration of lower atmospheric ozone diurnally back and forth across eastern metropolitan Phoenix, the background or synoptic scale circulation must be conducive. In support of high ozone concentrations and a mountain-valley mesoscale circulation, the synoptic scale atmosphere should be characterized by clear conditions and weak atmospheric circulation. In fact, on high ozone days a dome of high pressure aloft was centered over northeastern Arizona in association with a 500 mb ridge across the western United States (Figure 7-4). The strong area of high pressure aloft was associated with no significant loweratmospheric weather features across Arizona as indicated by the spatial pattern at 850 mb (not shown).

Figure 7-4: Geopotential Heights (dam) at 500 mb

Analysis of Lower Atmospheric Ozone Transport

63

With high pressure aloft and weak lower atmospheric flow, the synoptic atmosphere likely provided an ideal platform for the mesoscale transport of high concentrations of lower atmospheric ozone. Clear sky conditions likely supported the large insolation receipt and extreme surface heating measured at the microclimate site, and also increased photochemical production of lower atmospheric ozone. Likewise, the weak synoptic circulation in the lower atmosphere did not appear to be strong enough to disrupt the mesoscale circulation near the surface. As would be expected, only minor variations from this pattern characterized the synoptic situations associated with the remaining five instances in which nighttime katabatic transport of ozone was evident.

4. Conclusions The concentration of lower atmospheric ozone across metropolitan Phoenix, Arizona, has become of interest to federal, state, and local government agencies due to the potentially dangerous impact of large amounts of ozone on human health. Microclimatic and ozone data were gathered in an area typified by high concentrations of ozone and related to atmospheric characteristics and ozone concentrations on the mesoscale. The microclimate of the high ozone area east of the urban center indicated the evolution of a daytime mesoscale atmospheric flow directed up the slopes of higher terrain to the northeast. Ozone concentrations at the microclimate site suggested that the apparent anabatic wind was capable of producing an influx of ozone from the southwest. Spatial patterns of ozone concentrations on the mesoscale indicated that the anabatic wind did transport high ozone concentrations northeastward across a relatively focused path, from an area of high ground traffic south of the urban center through the eastern suburban areas. Through late afternoon, the mesoscale atmospheric and ozone concentration patterns suggested that the anabatic wind forced the area of high ozone concentrations along the Salt River Valley and up into the area of higher elevations to the northeast. During nighttime hours, evidence from both the microclimate and mesoscale atmosphere indicated the development of a downslope atmospheric flow back into the valley from the higher terrain to the northeast. The apparent katabatic wind was limited spatially to the eastern half of the Phoenix metropolitan area, but appeared to return a portion of the ozone that was advected up the higher terrain during the daytime hours. What the evidence does suggest is that the current practice of monitoring and forecasting dangerous ozone concentrations by using a single composite

64

Chapter Seven

value for the afternoon hours and for the entire Phoenix urban area may not be completely accurate.

References Altshuller, A.P. 1978. Association of oxidant episodes with warm stagnating anticyclones. Journal of the Air Pollution Control Association 28, 152–155. Arizona Department of Environmental Quality (ADEQ). 1998. Air Quality Report (ARS 49-424.10). Phoenix, AZ. Arrit, R. and Pielke, R.A. 1986. Interaction of nocturnal slope flows with ambient winds. Boundary Layer Meteorology 37, 183–195. Berman, N.S., Boyer, D.L., Brazel, A.J., Brazel, S.W., Chen, R.R., Fernando, H.J.S., and Fitch, M.J. 1995. Synoptic classification and physical model experiments to guide field studies in complex terrain. Journal of Applied Meteorology 34, 719–730. Blumen, W. ed. 1990. Atmospheric Processes over Complex Terrain. Meteorological Monographs 23. Boston, MA: American Meteorological Society. Davis, R.E. and Gay, D.A. 1993a. An assessment of air quality variations in the southwestern United States using an upper air synoptic climatology. International Journal of Climatology 13, 755–781. Davis, R.E. and Gay, D.A. 1993b. A synoptic climatological analysis of air quality in the Grand Canyon National Park. Atmospheric Environment 27A, 713–727. Ellis, A.W., Hildebrandt, M.L., Thomas, W.M., and Fernando, H.J.S. 2000. Analysis of the climatic mechanisms contributing to the summertime transport of lower atmospheric ozone across metropolitan Phoenix, Arizona, USA. Climate Research 15, 13–31. Green, C.R. and Sellers, W.D. 1964. Arizona Climate. Tucson, AZ: The University of Arizona. Hosler, C.J. 1961. Low-level inversion frequency in the contiguous United States. Monthly Weather Review 89, 319–339. Lalas, D.P., Asirnakopoulos, D.N., Deligiorgi, D.G., and Helmis, C.G. 1983. Sea-breeze circulation and photochemical pollution in Athens, Greece. Atmospheric Environment 17, 1621–1632. Loibi, W. 1997. Modeling tropospheric ozone distribution considering the spatio-temporal dependencies within complex terrain. In: Kraak, M.J. and Molenaar, M. (eds.), Advances in GIS Research II. London, UK: Taylor and Francis.

Analysis of Lower Atmospheric Ozone Transport

65

Niemeyer, L.E. 1960. Forecasting air pollution potential. Monthly Weather Review 88, 88–96. Peterson, I.T. and Flowers, E.C. 1977. Interactions between air pollution and solar radiation. Solar Energy 19, 23–32. Regener, V.H. 1957. The vertical flux of atmospheric ozone. Journal of Geophysical Research 62, 221–228. Schlink, U., Herbarth, O., Richter, M., Rehwagen, M., Puliafito, J.L., Puliafito, E., Puliafito, C., Guerreiro, P., Quero, J.L., and Behler, J.C. 1999. Ozone-monitoring in Mendoza, Argentina: Initial results. Journal of the Air Waste Management Association 49, 82–87. Sellers, W.D. and Hill, R.H. 1974. Arizona Climate. 1931–1972. Tucson, AZ: The University of Arizona Press. Sillman, S. 1999. The relation between ozone, NOx and hydrocarbons in urban and polluted rural environments. Atmospheric Environment 33, 1821–1845. Simpson, J.E. 1997. Gravity Currents in the Environment and the Laboratory. Cambridge, UK: Cambridge University Press. U.S. Environmental Protection Agency (EPA). 1998. Technical Support Document for the Notice of Final Rule making on Finding of Failure to Attain and Denial of Attainment Extension for Ozone in the Phoenix (Arizona) Metropolitan Area.

CHAPTER EIGHT THE SYNOPTIC CLIMATOLOGY OF LOWER ATMOSPHERIC OZONE EXCEEDANCES IN ST. LOUIS, MISSOURI MARK L. HILDEBRANDT (2001)

1. Introduction It is widely known that anticyclonic activity at the 500 millibar (mb) level of the atmosphere is associated with dry, subsiding air (Altshuller, 1978; Adams and Comrie, 1997). This synoptic condition, as evidenced by high geopotential heights, has been demonstrated to promote high loweratmospheric ozone levels in urban areas, such as Las Vegas, Los Angeles (Davis and Gay, 1993a), Nashville (Banta et al., 1998), and Phoenix (Ellis et al., 1999; Ellis et al., 2000; Hildebrandt, 2000). Anticyclonic activity at the 500 mb levels, coupled with small amounts of moisture advection at the 850 mb levels has been shown to promote dangerous ozone exceedances in metropolitan areas (Davis and Kalkstein, 1990; Davis and Gay, 1993a, 1993b). To date, no studies have addressed the linkages between lower atmospheric ozone concentrations and synoptic conditions in the St. Louis metropolitan area. Lower atmospheric ozone concentrations cannot be understood without studying the linkages between surface characteristics and synoptic circulation. In the 1960s, Niemeyer (1960) and Hosler (1961) linked upper-atmospheric circulation patterns with surface pollution dispersion. Niemeyer (1960) determined that high pollution episodes resulted from low wind speeds, high pressure systems, and subsidence. Altshuller (1978) linked high pollution events with anticyclonic activity. Conversely, low pollution events were linked with other synoptic patterns, such as cold fronts, strong winds, and rain, which may effectively slow down mixing rates of chemical species.

Climatology of Lower Atmospheric Ozone Exceedances in St. Louis

67

Several studies have attempted to identify associations between synoptic scale circulation regimes and general pollutant concentrations and transport within particular urban areas (Davis and Kalkstein, 1990; Davis and Rogers, 1992; Davis and Gay, 1993a, Ellis et al., 1999, 2000; Hildebrandt, 2000). While numerous studies of pollution in the St. Louis metropolitan area were performed in the 1970s (Braham and Wilson, 1978; Changnon, 1978, 1979), no studies addressing the linkages between synoptic climatology and lower atmospheric ozone transport have been performed. Unhealthy levels of lower atmospheric ozone across the St. Louis metropolitan area have been a focus of the United States Environmental Protection Agency (EPA) over the past decade. Traditionally, the EPA has categorized low-level ozone as dangerous to human health if the concentration exceeds 0.12 parts per million (ppm) over the course of one hour on any particular day. St. Louis failed to meet the health-based onehour air quality standard by the Clean Air Act deadline of November 15, 1996. Recent data indicate that St. Louis does not meet the new one-hour standard and federal exceedances are likely to increase (American Lung Association of Eastern Missouri (ALAEM), 1999). The EPA had set a June 29, 2001, deadline for issuing a determination of whether the region had met national ozone standards. Such a determination could have bumped-up St. Louis to the “serious” category, possibly leading to tougher pollution controls and such penalties as the loss of federal highway money. On June 15, 2001, the EPA announced that steps taken by Missouri and Illinois to reduce ozone are sufficient for the region to avoid “serious” nonattainment status, at least for now. The EPA also extended the deadline for meeting ozone standards set by the Clean Air Act to November 15, 2004. While the deadline for possible reclassification has been postponed until 2004, local scientists and agencies will continue their collaborative efforts to combat ozone. The St. Louis Regional Clean Air Partnership (SLRCAP) was established in 1995 as an organization to encourage activities that would reduce air pollution emissions. Due to a wide variety of measures aimed at reducing pollution implemented by SLRCAP, St. Louis’ ozone levels have remained fairly stable over the past few years even with rapid population growth in suburban areas and more vehicular traffic in the metropolitan area (ALAEM, 1999). Despite its measures, current lower atmospheric ozone levels are unhealthy, with the highest

68

Chapter Eight

concentrations consistently located over the eastern portion of the St. Louis urban area during the summer months (ALAEM, 1999). Currently, there is a need to address the climatological conditions associated with ozone exceedances across metropolitan St. Louis. While some studies have attempted to address the theoretical linkages between meteorological phenomena at a synoptic scale and lower atmospheric ozone levels in other cities, no studies have addressed the complex linkages between synoptic scale processes and ozone exceedances in St. Louis. This study will aid regional efforts to improve ozone forecasts as further urban growth in the Midwest may potentially lead to higher pollution levels that are detrimental to the inhabitants of these areas.

2. Data and Methodology The Missouri Department of Natural Resources and the Illinois Environmental Protection Agency, in cooperation with the SLRCAP, have monitored micro-climatological data, including lower atmospheric ozone, at 16 sites within the St. Louis metropolitan area during the period May 1 — September 30 each year since 1992 (Figure 8-1). Data were collected at five-minute intervals and subsequently calculated to one-hour averages (centered on each half-hour) in order to coincide with the basis for EPA standards measurements. These data were provided free-of-charge from SLRCAP, the Illinois EPA, and the Missouri Department of Natural Resources. Once the ozone data were collected for the period of study, the study days were stratified into two categories based on the one-hour average ozone concentrations in order to comply with EPA standards. The first category, high ozone, included days possessing a one-hour average concentration of 0.12 ppm or greater (EPA exceedance level). From the period 1992–2000, one-hour ozone exceedances occurred on 42 different days. The second category, which will not be discussed in this paper, was composed of days possessing a one-hour average ozone concentration of less than 0.12 ppm.

Climatology of Lower Atmospheric Ozone Exceedances in St. Louis

69

Figure 8-1: Map of Ozone Monitoring Sites (Ɣ = Ozone Monitoring Site)

The synoptic climatology associated with the two categories of ozone was then addressed. Specifically, the u (+ westerly, – easterly) and v (+ southerly, – northerly) components of the wind, specific humidity at 850 mb, and geopotential heights at the pressure level of 500 mb were analyzed. Specific humidity (q) is the ratio of the mass of water vapor per unit mass of air, including the water vapor itself (g kg-1). This is a favored measure of atmospheric moisture because it represents the true water vapor content of the atmosphere. Also analyzed were the height fields at 500 mb, for which the base height is 5700 geopotential meters (gpm). Thickness is directly related to the underlying air temperature of an air mass, whereby the warmer the air, the greater the thickness of the atmosphere. Finally, analyzed at 850 mb were the u (zonal) and v (meridional) components of the wind field aloft to indicate differences in the synoptic flow of the lower atmosphere on high and low ozone days. The data and software for the analysis were obtained free-of-charge from the Climate Diagnostic Center (CDC) which makes use of the National Center for Environmental Prediction (NCEP) reanalysis data.

70

Chapter Eight

3. Results an nd Analysis Over thee course of thhe study perio od, 42 days w were classified d as high ozone days (EPA exceedaance). High ozone o days occcurred on all 7 days of the week (F Figure 8-2). However, H dayss at the beginnning of the week w were less likely too have exceeddances. In fact, over half (255) of the 42 hiigh ozone days occurrred between Thursday T and Saturday, whhen automobiile traffic was likely ggreatest. Eachh of the days was characterrized by an in ncreasing concentratioon of ozone in i the atmosp phere throughh 17:00 local standard time (not shown). In orderr for surfacee ozone conccentrations too reach surfa face EPA exceedance levels, the baackground or synoptic scaale circulation n must be conducive too ozone devellopment. In su upport of highh ozone concentrations, the synopticc scale atmospphere should be b characterizzed by clear conditions c and weak attmospheric cirrculation. In fact, f on high oozone days, a dome of high pressurre aloft was centered c over Texas and Okklahoma in asssociation with a 500 mb ridge acrross the centrral United Staates (Figure 8-3). 8 The strong area oof high pressuure aloft was associated a witth no significaant loweratmosphericc weather feattures across th he St. Louis rregion, as ind dicated by the spatial ppattern at 850 mb m (not show wn).

Figure 8-2: V Violations by Daay of the Week

Climatology of Lower Atmospheric Ozone Exceedances in St. Louis

71

Figure 8-3: Geopotential Heights (dam) at 500 mb

With high pressure aloft and a weak lower atmospheric flow, the synoptic atmosphere likely provided an ideal platform for the development of high concentrations of lower atmospheric ozone. Clear sky conditions likely supported large insolation receipt and extreme surface heating, and also increased photochemical production of lower atmospheric ozone. Likewise, the weak synoptic circulation in the lower atmosphere did not appear to be strong enough to disrupt the circulation near the surface. As would be expected, ozone levels were allowed to grow unabated throughout the St. Louis metropolitan region during this period.

4. Conclusions As in other growing metropolitan areas in the central United States, St. Louis ozone levels have been a concern for the past decade. Despite efforts by local organizations to encourage activities that aim to reduce detrimental air pollution, surface ozone levels regularly exceed EPA standards. The St. Louis region will avoid designation as a “serious” ozone polluter and tougher pollution controls and penalties until at least 2004.

72

Chapter Eight

Reclassification to “serious” nonattainment status could bring tougher controls on pollution (e.g., expanded emission tests), monetary penalties, and potential withholding of federal highway money. As such, it is imperative to expand the limited knowledge of climatological conditions that promote ozone exceedances in St. Louis. The results presented in this paper are encouraging. The synoptic climatology associated with high ozone days is logical synoptically speaking. This gives hope to the idea that there is indeed a pattern that coincides with ozone concentrations across St. Louis, making high ozone days foreseeable.

References Adams, D.K. and Comrie, A.C. 1997. The North American Monsoon. Bulletin of the American Meteorological Society 78, 2197–2213. ALAEM. 1999. Ozone Pollution in St. Louis. St. Louis, MO: American Lung Association of Eastern Missouri. Altshuller, A.P. 1978. Association of oxidant episodes with warm stagnating anticyclones. Journal of the Air Pollution Control Association 28, 152–155. Banta, R.M., Senff, C.J., White, A.B., Trainer, M., McNider, R.T., Valente, R.J., Mayor, S.D., Alvarez, R.J., Hardesty, R.M., Parrish, D., and Fehsenfeld, F.C. 1998. Daytime buildup and nighttime transport of urban ozone in the boundary layer during a stagnation episode. Journal of Geophysical Research 103, 22519–22544. Braham, R.R. and Wilson, D. 1978. Effects of St. Louis on convective cloud heights. Journal of Applied Meteorology 17, 587–592. Changnon, S.A. 1978. Urban effects on severe local storms at St. Louis. Journal of Applied Meteorology 17, 578–586. Changnon, S.A. 1979. Rainfall changes in summer caused by St. Louis. Science 205(4404), 402–404. Davis, R.E. and Gay, D.A. 1993a. An assessment of air quality variations in the southwestern United States using an upper air synoptic climatology. International Journal of Climatology 13, 755–781. Davis, R.E. and Gay, D.A. 1993b. A synoptic climatological analysis of air quality in the Grand Canyon National Park. Atmospheric Environment 27A, 713–727. Davis, R.E. and Kalkstein, L.S. 1990. Development of an automated spatial synoptic climatological classification. International Journal of Climatology 10, 769–794.

Climatology of Lower Atmospheric Ozone Exceedances in St. Louis

73

Davis, R.E. and Rogers, R.F. 1992. A synoptic climatology of severe storms in Virginia. Professional Geographer 44, 319–332. Ellis, A.W., Hildebrandt, M.L., and Fernando, H.J.S. 1999. Evidence of lower atmospheric ozone “sloshing” in an urbanized valley. Physical Geography 20, 520–536. Ellis, A.W., Hildebrandt, M.L., Thomas, W.M., and Fernando, H.J.S. 2000. Analysis of the climatic mechanisms contributing to the summertime transport of lower atmospheric ozone across metropolitan Phoenix, Arizona, USA. Climate Research 15, 13–31. Hildebrandt, M.L. 2000. A climatological analysis of lower atmospheric ozone transport across Phoenix, Arizona. Papers and Proceedings of the Applied Geography Conferences 23, 145–153. Hosler, C.J. 1961. Low-level inversion frequency in the contiguous United States. Monthly Weather Review 89, 319–339. Niemeyer, L.E. 1960. Forecasting air pollution potential. Monthly Weather Review 88, 88–96.

CHAPTER NINE SEASONAL TRENDS IN ANTARCTIC TEMPERATURE REANALYSIS JENNIFER M. COLLINS, DAVID R. ROACHE, EDGAR W. KOPP IV AND DOUGLAS LUNSFORD (2012)

1. Introduction The Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report (AR4) (IPCC, 2007) considered that progress in understanding of how climate is changing in space and in time has been gained through improvements and extensions of numerous data sets and data analyses, broader geographical coverage, better understanding of uncertainties, and a wider variety of measurements. However, data coverage remains limited in some regions. This report shows, through the use of station data, that the warming of the climate system is unequivocal, with 11 of the 12 years from 1995 to 2006 ranking among the 11 warmest years in the instrumental record of global surface temperatures since 1850 (Trenberth et al., 2007). It is also noted in the report that the total temperature increase from the period 1850–99 to the period 2001–05 is 0.76 C° (Trenberth et al., 2007). The main conclusion of AR4 is that most of the observed increase in globally averaged temperatures since the midtwentieth century is very likely due to the observed increase in anthropogenic greenhouse gas emissions (Hegerl et al., 2007). Brown et al. (2008) also indicate that extreme daily maximum and minimum temperatures have warmed for most regions of the world since 1950 and that the total area exhibiting positive trends is significantly greater than that which can be attributed to natural variability. Difficulties remain in reliably simulating and attributing observed temperature changes at smaller spatial scales. On such scales, natural climate variability is

Seasonal Trends in Antarctic Temperature Reanalysis

75

relatively larger, making it harder to distinguish changes expected due to external forcings. Uncertainties in local forcings and feedbacks also make it difficult to estimate the contribution of greenhouse gas increases to observed small scale temperature changes (Hegerl et al., 2007). This work focuses on climate change over the southernmost continent, Antarctica. The majority of its land mass falls within the Antarctic Circle, and it is divided into three zones. Two of the zones, East and West Antarctica, are separated by the Transantarctic Mountain chain, while the other zone, the Antarctic Peninsula, includes numerous islands and extends out into the Southern Ocean towards South America (Shirihai, 2002). The majority of the continent is covered in an ice sheet which contains about 2 percent of the Earth’s water (Moss, 1988). It also holds roughly 70 to 85 percent of the world’s fresh water (Shirihai, 2002). As the Earth’s fifth largest continent, it is the only one in which the size of the landmass changes on a regular basis due to the gain and loss of ice (Shirihai, 2002). Antarctica is surrounded by the Southern Ocean, which separates it from the southern tips of Africa, South America, Australia, and New Zealand. Temperatures across the Antarctic continent vary, partly in respect to whether the location is coastal or inland as well as in respect to a location’s elevation above sea level. The Peninsula, as it projects into the Southern Ocean, has a milder climate in comparison to continental areas and is more strongly coupled to the ocean (Monaghan and Bromwich, 2008). The most noticeable coastal changes have occurred in the last few decades along the Antarctic Peninsula (Ferrigno and Williams, 1998). The temperatures in the southern polar regions are sensitive to the ice and snow cover on Antarctica largely due to their relationship with albedo. As Moss (1988, p. 3) notes, “increasing the albedo by extending the permanent ice cover of the Southern Ocean, or its decrease by ice cap melting and ablation would have profound effects on the climate of the Southern Hemisphere.” Krinner et al. (2007) discuss the importance of considering Antarctic temperature trends since they note a general increase of poleward heat transport as global mean temperature increases. There have been few climate change studies focusing on temperature that consider the whole of Antarctica. This may be in part because of the few observing sites due to the harsh conditions there. Of these studies, Doran et al. (2002) found a net cooling of the continental Antarctic between 1966 and 2000, although warming of the Antarctic Peninsula was

76

Chapter Nine

indicated. This work was later challenged by Turner et al. (2002), who argued that the data were inappropriately extrapolated. In 2005, Turner et al. used READER data to survey Antarctic climate change during the past 50 years, revealing a spatially complex pattern with significant peninsular and coastal warming. When looking at temperature trends for the years 1958 to 2002, Chapman and Walsh (2007) found there was considerable sensitivity to study period start and end dates in their analysis. In particular, they found that start dates prior to 1965 produced overall warming trends while start dates between 1966 and 1982 yielded overall cooling trends. They concluded that there was great interannual variability over the continental mass but found no significant trends. However, they noted that warming on the Antarctic Peninsula was significant. On the other hand, Steig et al. (2009) determined that there was a significant warming trend extending beyond the Antarctic Peninsula into the western portion of the continent with the region showing a warming trend of more than 0.1 C° per decade over the last fifty years. This warming was partially offset by cooling in the eastern portion of the continent, though the average near surface temperature increased for the continent as a whole. Monaghan (2009) noted that analysis of ice cores taken from the West Antarctic Ice Sheet shows that West Antarctica’s temperature has increased by about 2 C° since 1950. Warming of the Antarctic Peninsula is the focus of the Vaughan et al. (2003) article examining trends in the context of recent rapid regional warming of high-latitude areas. It concludes that while temperature increases across the rest of the continent have not varied significantly in recent decades in comparison to the global mean temperature increase, warming across the Peninsula was substantially greater than the global rate of warming.

2. Methods In order to determine what, if any, temperature trends have occurred over Antarctica in recent decades, near surface temperature data are examined from the National Centers for Environmental Prediction (NCEP)/National Center for Atmospheric Research (NCAR) Reanalysis (NNR) (Kalnay et al., 1996). Temperatures used in the NNR are at the 2 m level, with T62 Gaussian grid resolution. Temperatures are based on observational data, but the model also has a very strong influence on the reanalysis value (Kalnay et al., 1996). The analysis in this study is conducted for the period 1948 (the start year of the NNR) to 2010, thereby maximizing the length of the data set. Use of reanalysis data in lieu of direct observations allows for assessment over larger areas without gaps as

Seasonal Trends in Antarctic Temperature Reanalysis

77

well as over longer time periods, while at the same time incorporating the increased number of available observations into the model in more recent years. As Hulme et al. (2001) point out, such a historical perspective is essential if the simulated climates of the next century are to be put into their proper context. Comiso (2000) noted, not surprisingly, that trends can fluctuate when a period of ten years or less is used, and it stabilizes significantly when a data set of ten to twenty years is used, with a data set of at least twenty years yielding the most stable trends, concluding that at least for studies of Antarctic temperature trends, any data set used to determine trends must have data for at least twenty years. Trends are considered for each region studied (all of Antarctica, East Antarctica, West Antarctica, and the Peninsula). As shown in Figure 9-1, the West region is bounded by 72°S and spans from 60°W westward to 180°, while the East region is bounded by 65°S and spans from 60°W eastward to 180°. The Peninsula is defined as westerly longitudes north of 72°S, as given by Steig et al. (2009). The trends are determined by examining the slope of the best-fit linear trend from an ordinary least squares (OLS) regression, and the trend is determined to be significant at the 5 percent level (Į=0.05). The Durbin–Watson statistic was calculated to analyze autocorrelation in the yearly averaged time series. No correction was deemed necessary because each yearly averaged time series has a Durbin–Watson test statistic around two, which suggests that the null hypothesis (i.e., that there is no autocorrelation) would not be rejected. The nonparametric Mann– Kendall test (described by Onoz and Bayazit, 2003) has been used for temperature trend detection in other studies. For example, Collins et al. (2009) used it in particular for considering station data in South America. However, for the Antarctica data used here, OLS regression is the appropriate and more robust method because all mean temperatures in Antarctica are considered with residuals that are normally distributed. In addition, the coefficients of skewness and kurtosis are all less than the absolute value of one following the Z distribution for normally distributed data. The means and standard deviations of the NNR temperature are determined for all months of the year for the periods 1948–1979 and 1980–2010 and the results presented in this paper are the average of December–January– February (DJF; summer), March–April–May (MAM; autumn), June–July– August (JJA; winter) and September–October–November (SON; spring). To evaluate the temperature variability in recent decades, the data are further subdivided into 1980–1994 and 1995–2010, and a similar analysis

78

Chapter Nine

Figure 9-1: Antarctica. Scale represents elevation in meters. West and east regions are marked. Peninsula is the land surface outside of these two regions

is conducted. The temperatures of the last 10 years (2001–2010) are also considered since in the recent period there is a marked temperature increase starting at the beginning of this century in Antarctica and globally (Mears et al., 2011). The statistical significance, determined to the 5 percent level (Į=0.05) of the difference between the composites, for the different periods considered in this work, is calculated from the OLS regression by applying the Student’s two-tailed t test.

Seasonal Trends in Antarctic Temperature Reanalysis

79

3. Results Significant increasing temperature trends were found when the annual data set was considered in each region examined: all of Antarctica, East Antarctica, West Antarctica, and the Peninsula (Table 9-1). Analysis by season for each region also shows significant increasing trends for all but the DJF (summer) season. For this season, while the Peninsula shows a significant increasing trend, the other areas examined (all of Antarctica, East Antarctica, and West Antarctica) show significant decreasing trends (Table 9-1, Figure 9-2). Table 9-1: Temperature Trends for Antarctica (All, East, West, Peninsula). Slope (C° per Year) and Significance Obtained from Linear Regression; Time Periods include DJF, MAM, JJA, SON, and the Entire Year (Annual) Time period Annual Annual Annual Annual DJF DJF DJF DJF MAM MAM MAM MAM JJA JJA JJA JJA SON SON SON SON

Region All East West Peninsula All East West Peninsula All East West Peninsula All East West Peninsula All East West Peninsula

Slope 0.034 0.036 0.030 0.034 –0.033 –0.034 –0.022 0.003 0.049 0.047 0.037 0.053 0.083 0.089 0.065 0.055 0.038 0.041 0.039 0.021

Significance

90°F for suummer were shown s in Figu ures 28-2 andd 28-3. The consistent c isopleth pattterns on all maps m revealed d great corresppondence of clo c to the thermal stresss. In winteer (Figure 28--2), both the percent p of tim me with clo > 2.5 and windchill > 800 kcal m-2 hr-1 increased d northward, rrevealing the impact i of latitude. Sim mple correlatioon between laatitude and peercent of timee with clo > 2.5 and windchill > 800 kcal m-2 hr-1 reveealed a high positive relationship (r = 0.94) in both cases. The T maps alsoo showed thatt the lake effect causeed South Bendd to have lessser time with extreme wind dchill and high clo valuues.

A) Percent of Time T with clo > 2.5; B) Percennt of Time with Windchill Figure 28-2: A > 800 kcal m--2 hr-1

Thee Assessment of o Clo as an Alternative Weathher Stress Index x

289

Figure 28-3: A A) Percent of Time T with clo < 0.5; B) Percennt of Time with HI > 90°F

In summ mer (Figure 28-3), 2 the norrthwest-southeeast running isopleths revealed thaat both the cloo and HI valu ues increased toward the so outhwest. Results of siimple correlattion between latitude l and p ercent of timee with clo < 0.5 and H HI > 90°F displayed stron ng relationshipps (r = –0.81 1 and r = –0.72 respecctively), and so s did elevatio on and percentt of time with clo < 0.5 and HI > 900°F (r = –0.71 and r = –0.76 6 respectively)). Weaker relaationships between lonngitude and peercent of timee with clo < 00.5 and HI > 90°F 9 (r = 0.36 and r = 0.49 respectiively) were fou und. The resuult of the linear regression between hour urly wintertime clo and windchill vvalues of Inddianapolis ind dicated that tthey had a moderate relationship (r = 0.669, r2 = 0.447). Hou urly windchilll values of Ind dianapolis

290

Chapter Twenty-Eight

were calculated using Siple formula and linear regression was performed. The Siple windchill formula is:

‫ ܭ‬ൌ  ሺሺͳͲͲ‫ݒ‬ሻ଴Ǥହ  ൅ ͳͲǤͶͷ െ ‫ݒ‬ሻሺ͵͵ െ  ‫ݐ‬௔ ሻ (Equation 28-6) where K is windchill (kcal m-2 hr-1), v is wind speed (m sec-1), 33 is skin temperature (°C), and ta is air temperature (°C). The result showed a stronger significant relationship between clo and windchill (r = 0.825, r2 = 0.68). The reason for the different results in the linear regression analyses is the variation in the assumptions of these two windchill models. Steadman’s model considered the combined detrimental effect of wind and low temperature on a clothed person while Siple’s model described the cooling power of wind and cold temperature (Dixon and Prior, 1987). In addition, Steadman windchill formula used 30°C as the “skin temperature,” while Siple’s formula used 33°C. This different treatment of the term “skin temperature” in the two equations may lead to different results. The result of linear regression between hourly summertime clo and HI values of Indianapolis revealed that they had a strong relationship (r = –0.804, r2 = 0.646).

4. Conclusion Visual images of the relationships between clo and windchill, and HI provided in Figures 28-2 and 28-3 indicated they had consistent patterns. The higher the percent of time with clo > 2.5 and clo < 0.5, the greater percent of time with windchill > 800 kcal m-2 hr-1 and HI > 90°F respectively. Results from the linear regression analyses revealed that clo had strong relationships with Siple windchill formula and HI. The weaker relationship between clo and Steadman’s windchill model lies on the fact that it considers clothed individuals. Their correspondence suggests that clo could be a potential alternative weather stress index. Further, clo has the merits to be used in all weather and climatic conditions. However, at present, the general public is rarely informed of clo. It would be helpful if the textile manufacturers could attach labels showing clo values of each garment. This would assist the public to recognize clo values of various clothing assemblies, and in time modify the preference of weather stress indices.

The Assessment of Clo as an Alternative Weather Stress Index

291

References Auliciems, A., de Freitas, C.R., and Hare, F.K. 1973. Winter clothing requirements for Canada. Climatological Studies No. 22. Toronto, Canada: Environment Canada. Auliciems, A. and Hare, F.K. 1973. Weather forecasting for personal comfort. Weather 28, 118–121. de Freitas, C.R. 1979. Human climates of northern China. Atmospheric Environment 13, 71–77. Dixon, J.C. and Prior, M.J. 1987. Wind-chill indices - A review. Meteorological Magazine 116, 1–17. Dixon, R.W. 1992. Climatology of Summer Heat Stress Hazard in Texas. Southwest Texas State University. Unpublished Master’s Thesis. Driscoll, D.M. 1992. Thermal comfort indexes: Current uses and abuses. National Weather Digest 17, 33–38. Hay, J.E. 1970. Aspects of the Heat and Moisture Balance of Canada. University of London. Unpublished Ph.D. Dissertation. Kesseler, E. 1993. Wind chill errors. Bulletin of the American Meteorological Society 74, 1743–1744. Landsberg, H.E. 1988. Biometeorology. In: Parker, S.P. (ed.), Meteorology Source Book, New York, NY: McGraw-Hill, 115–123. National Oceanic and Atmospheric Administration. 1980. Heat Stress. Asheville, NC: NOAA/EDIS/NCDC. Steadman, R.G. 1971. Indices of windchill of clothed persons. Journal of Applied Meteorology 10, 674–683. Terjung, W.H. 1966. Physiological climates of conterminous U.S.: A bioclimatic classification based on Man. Annals of the Association of American Geographers 56, 141–179.

CHAPTER TWENTY-NINE IMPACT OF ENSO PHASE ON AUSTRALIAN SHARK ATTACK FREQUENCY RICHARD W. DIXON (2003)

1. Introduction Animal attacks on humans are inherently newsworthy, but subject to excessive media “enthusiasm.” Disproportionate media attention may distort the true vulnerability to the event. Vulnerability to a hazard is based on both risk and exposure (Dixon and Fitzsimons, 2001). In a climatological context, vulnerability to an event depends not only on the characteristics of the event (risk) but also on those human processes (exposure) which may ameliorate or magnify the negative consequences of the event (Pielke and Pielke, 1997). Thus for a study of the vulnerability to shark attacks, consideration must be given not only to where and when the attacks occur but also to those characteristics of the population interacting with the sharks. This paper addresses only the risk aspect of vulnerability by presenting results of an analysis of 51 years of reported shark attacks on humans in the coastal waters of Australia. Quantification of risk of shark attack involves knowledge of both the magnitude and frequency of their occurrence. Many factors such as water temperature, food availability, and coastal currents may influence the presence of sharks in coastal waters. The El Niño / Southern Oscillation (ENSO) phenomenon has been documented to impact both the atmosphere and ocean in many parts of the world with many resultant impacts on human populations and systems (Philander, 1990). This paper attempts to answer the question of a relationship between ENSO phase and shark attacks. Does such a relationship exist? If so, new targeted research can be conducted to elucidate the behavioral (human (Gilbert, 1977) and/or animal) or physical processes involved in elevating the risk.

Impact of ENSO Phase on Australian Shark Attack Frequency

293

The first systematic attempt to record shark attack statistics in Australian waters was conducted by the physician V. M. Coppleson who wrote a series of articles for Medical Journal of Australia and summarized attack statistics from 1919 to 1959 in his book Shark Attack (Butler, 1964). His data compilations included date, location, activity of victim, and outcome. Butler (1964) cited the following conclusions from Coppleson’s work. Attacks were most frequent during the warmer months of October through May and rarely occurred in waters with a climatological mean monthly sea surface temperature of less than 70ºF. Of the 172 attacks in Coppleson’s database 85 (or 49%) were fatal.

2. Methodology 2.1 Shark Attack Data The International Shark Attack file is a worldwide database of reported shark encounters. It had its genesis in an Office of Naval Research funded study started in 1958 (Gilbert et al., 1960; Schultz et al., 1961). The database is now maintained by the Florida Museum of Natural History, a unit of the University of Florida, Gainesville. Information in the database is extracted from a survey conducted for each attack recording such factors as date of attack, location, outcome, and other information related to activity and environmental conditions. Locations are relative, not absolute, which makes a precise mapping problematic. For this study, attack reports, for the years 1950 – 2000 identified as occurring in Australian waters were extracted. This gave a total of 158 reports of which 21 did not contain date information. Those 21 reports were eliminated from the statistical analysis but included in the summary statistics.

2.2 ENSO Phase Data The Climate Prediction Center (CPC) classifies each three-month calendar season according to the departure from normal conditions in equatorial Pacific sea surface temperatures. Departures of +1 C° (–1 C°) or higher (lower) are termed warm, El Niño (cold, La Niña) phase seasons. Departures between these are termed neutral phase seasons. There is considerable variability in the strength and duration of any individual ENSO event. Table 29-1 is a summary of the ENSO phase by season for the 51 years which constitute the time frame of the study as obtained from the CPC.

Chapter Twenty-Nine

294

Table 29-1: Distribution of ENSO Phase by Season Season Summer – JFM Fall – AMJ Winter –JAS Spring – OND

Warm 16 14 17 19

Cold 15 11 12 17

Neutral 20 26 22 15

2.3 Data Analysis We can test the null hypothesis that the average number of shark attacks in a season is invariant of ENSO phase by using traditional inferential statistics. Since for each season there are three phases of ENSO to be evaluated, analysis of variance (ANOVA) would seem to be the preferred technique. However, use of ANOVA with an appropriate posthoc test would indicate that the average number of shark attacks in that season was significantly different for a particular ENSO phase relative to the other phases when each phase is considered individually. In application, it would be most beneficial to know if a particular ENSO phase is riskier relative to not being in that phase. That is, for example, do cold ENSO phase summers have more shark attacks on average than noncold (i.e., warm and neutral phase) summers? This casts the problem as that of two independent samples to be compared, for instance cold phase mean versus “not cold phase” mean. Since this requires the use of three ttests to meet all phases it is necessary to apply a Bonferroni adjustment to the significance level (alpha) to reduce the probability of Type 1 error (Hair et al., 1998). Thus an alpha of 0.017 (0.05 divided by 3) was used for all statistical tests.

3. Results 3.1 Spatial and Seasonal Distribution of Attacks All seven of Australia’s states have marine coastlines. Indeed population tends to concentrate along the southeast (Brisbane to Adelaide) and southwest (Perth) coasts. This is reflected in the attack statistics of Table 29-2. Analysis of exposure to shark attacks which would incorporate the population distribution implied in Table 29-2 is beyond the scope of this paper.

Impact of ENSO Phase on Australian Shark Attack Frequency

295

Table 29-2: Distribution (Total/Fatal) of Shark Attacks by State State Queensland New South Wales Victoria South Australia Western Australia Northern Territory Tasmania Unknown

Total/Fatal 40/13 38/12 16/4 24/11 22/4 0/0 14/5 4/2

Shark attacks have also been recorded in all seasons of the year. Spring (OND) and Summer (JFM) have the highest frequency of occurrence at 48 and 50 attacks respectively, while Fall (AMJ) and Winter (JAS) show less than half the warm season attacks at 20 and 16 attacks. Of the 158 attacks in the database, non-fatal attacks occur about twice as frequently as fatal attacks (107 to 51). This preponderance of warm month attacks confirms the original findings of Coppleson and may be attributed to exposure factors outside the scope of this paper. It is of interest to note the 19502000 fatality rate of 32% is about one-third lower than in Coppleson’s 1919-1959 data. Exposure factors related to improvements in medical response and care are most likely responsible for the decrease.

3.2 Impact of ENSO Phase Table 29-3 displays the results of the independent samples t-test for mean numbers of shark attacks by season and ENSO phase. Statistically significant results are underlined. Only results for the La Niña (cold) phase were statistically significant. The negative value for the mean difference indicates that fewer attacks occur during La Niña fall and winter than during an El Niño or neutral fall and winter. Table 29-3: Shark Attacks by Season (Mean/Difference) during Cold ENSO Phase ENSO Phase JFM AMJ JAS OND

Mean/Difference 1.20/–0.19 0.09/–0.48 0.08/–0.38 1.12/0.00

296

Chapter Twenty-Nine

To interpret this result requires an understanding of the impact of La Niña on the Australian atmosphere. La Niña falls and winters are anomalously dry across most of eastern Australia (Philander, 1990). Thus runoff should be diminished and water clarity should be enhanced. This improved water clarity, perhaps leading to cooler sea surface temperatures, can explain the decrease in shark attack frequency. Note that during the neutral or warm phases, there was no statistically significant seasonal difference in shark attack frequency.

4. Conclusions A statistically significant relationship was found between the seasonal numbers of shark attacks and the La Niña (cold) phase of ENSO. A La Niña fall and winter has a lower number of reported attacks than a neutral or El Niño (warm) phase. This result is most likely due to the impact of anomalous dry conditions on coastal ocean water clarity and possibly sea surface temperature. The finding of a decrease in average number of shark attacks during La Niña falls and winters raises the question of a threshold temperature for shark attacks first alluded to in the earlier work of Coppleson. A follow-on study to examine the impact of ENSO phase on coastal seas surface temperatures is planned. In addition, exposure factors (a function of human behavior) not addressed in this paper may also contribute to this result. This study raises many questions related to the exposure factor in the vulnerability equation which the data as collected in the International Shark Attack File may be unable to address. What activity is riskiest? Are there spatial “hot spots” for shark attacks? Do any of these factors vary spatially or temporally? These and similar questions can be addressed by focused data collection efforts, but the relatively low frequency of shark attacks will always complicate standard statistical analysis.

Acknowledgments The author thanks Dr. George H. Burgess, Director of the International Shark Attack File for access to the shark attack data. In addition, Jessie Morgan of Texas State assisted with the extraction of the Australian subset from the database and E.J. Hanford of Texas State uncovered the reference to the earlier work of Coppleson.

Impact of ENSO Phase on Australian Shark Attack Frequency

297

References Butler, J.C. 1964. Danger – Shark! Boston, MA: Little Brown and Co. Dixon, R.W. and Fitzsimons, D.E. 2001. Toward a quantified hurricane vulnerability assessment for Texas coastal counties. Texas Journal of Science 53, 345–352. Gilbert, P.W. 1977. Two decades of shark research: A review. BioScience 27, 670–673. Gilbert, P.W., Schultz, L.P., and Springer, S. 1960. Shark attacks during 1959. Science 132(3423), 323–326. Hair, J.F., Anderson, R.E., Tatham, R.L., and Black, W.C. 1998. Multivariate Data Analysis. Upper Saddle River, NJ: Prentice Hall. Philander, S.G. 1990. El Niño, La Niña, and the Southern Oscillation. San Diego, CA: Academic Press. Pielke Jr., R.A. and Pielke Sr., R.A. 1997. Hurricanes: Their Nature and Impacts on Society. New York, NY: John Wiley and Sons. Schultz, L.P., Gilbert, P.W., and Springer, S. 1961. Shark attacks. Science 134(3472), 87–88.

CHAPTER THIRTY CLASSIFYING HEAT STRESS EVENTS IN THE CENTRAL UNITED STATES ERIK H. BOWLES (2006)

1. Introduction The central United States experiences a wide variation of climate from year to year, which has a significant influence on living organisms (Rosenberg et al., 1993). Heat stress events produce the potential for harm to people, and hazards researchers are examining phases of the oldest of the traditions within geography: the manner in which humans interact with their environment (Mitchell, 1989). There is a practical aspiration to inform decision makers about the specific environmental conditions that have harmful effects on humans, animals, and natural resources (Harrington and Bowles, 2002). Therefore, it has become necessary to advance our capability to measure the level of exposure people experience during extreme heat events. Geographical contrasts of heat stress are an issue because people’s acclimatization to their regional climate norms would then infer that similar heat events would affect people more severely in areas that do not commonly experience major heat events. This project aims to gain knowledge of an underrepresented climatic hazard. Table 30-1 illustrates that extreme heat events cause more human mortality on an annual basis than almost all other weather-related hazards combined (Robinson, 2001). However, these events have still proven difficult to define universally with no rigorous definition yet in use. The American Meteorological Society (AMS, 2000) defines a heat wave as “a period of abnormally and uncomfortably hot and usually humid weather.” There are several other attempts at adding detail to this definition, which has unfortunately made the problem more complex by not having a collective

Classifying Heat Stress Events in the Central United States

299

Table 30-1: Comparison of Mean Annual Morbidity among Top Warm Season Hazards in the United States from 1979 to 1995 (NOAA, 2006) Heat 350

Lightning 73

Tornado 68

Flooding 135

Hurricane 16

set of parameters (Ward, 1925; Kalkstein et al., 1996; Robinson, 2001; Choi and Meentemeyer, 2002; and Burroughs, 2003). Events like the ones in Chicago in 1995 and Paris in 2003 have proven that heat stress affects tens of thousands of people at a time, even in more developed economic locations (Kalkstein, 1995; Vandentorren et al., 2004). The Chicago event is known to have killed over 500 people, and some estimates for Western Europe from 2003 have surpassed 20,000 deaths. Delworth et al. (1999) and Smoyer et al. (2000) provide information that periods of hot conditions are likely to increase in intensity in the future. This paper focuses on the magnitude of heat events across Kansas and Nebraska from 1980 to 2000. When addressing the problem of excessive heat, one must account for two basic components, temperature and moisture. It has long been known that the stress caused by heat is compounded by the amount of moisture in the air (Lally and Watson, 1960). Water, itself, is a natural energy storage substance, especially with respect to heat (Oke, 1987). As air temperatures rise, the potential vapor content rises, meaning even more energy can be stored in the air. The effect that humidity has on warm temperatures is similar to how cold temperatures are affected by wind to produce ‘windchill’ (Steadman, 1979). As heat and humidity continue to rise, humans and animals become more uncomfortable, and the sultry surroundings can become hazardous to an organism’s health to the point of causing death (Thom, 1959; Steadman, 1984; Kalkstein and Valimont, 1986). For example, temperatures in western Kansas can exceed those in the eastern parts of the state, but higher moisture in the east forces the heat events to be much more intense. Because of this environmental synergism, several indices have been suggested to characterize how the air actually feels to an organism (Houghton and Yaglou, 1923; Thom, 1956; Hevener, 1959; Quayle and Doehring, 1981; Steadman, 1984). Dodd (1965) illustrated where the moisture gradient exists for the United States using dew point averages (Figure 30-1). There is a basic southeast-to-northwest decrease of atmospheric moisture content from the Gulf States to the northern Rockies, indicating that typical heat stress magnitudes would

300

Chapter Thirty

Figure 30-1: Moisture Gradient Decreases from East to West across Nebraska and Kansas (Dodd, 1965)

follow the same pattern. For the study area of Kansas and Nebraska, the gradient is practically east to west; thus, two data points were used to represent conditions across each state. Thom (1956) developed the temperature-humidity index (THI), and this index, as a measure of heat stress on livestock, continues to be used today. One perceived problem with THI is that the index values associated with extreme levels of discomfort do not correspond with uncomfortable dry bulb temperatures. For humans, the most commonly used index for measuring the levels of heat stress in the U.S. is the heat index (HI). Originally termed ‘apparent temperature’, the heat index quantifies heat stress values to depict the combined effect of temperature and humidity on the skin of humans more accurately (Steadman, 1979). This study used the temperature and humidity combination of the heat index to represent heat stress values on an hourly basis and uses an idea first presented by Hubbard et al. (1999) of accumulating (summing over the course of a 24 hour period) the hourly heat stress values above a selected threshold. Two HI thresholds, set by their effects on human health, were provided by Christopherson (2001) as categorizations of heat stress: HI = 90 and HI = 105. Unfortunately, Christopherson does not see fit to accumulate the stressful exposure that occurs from one hour to the next. The HI = 90 and HI = 105 values correspond with temperature-humidity index (THI) thresholds (Table 30-2) used in establishing the Livestock Comfort Index

Classifying Heat Stress Events in the Central United States

301

Table 30-2: Comparison Between Heat Index and THI Values Heat Index 90 105

Temperature-humidity Index 79 84

(1970) and the HI 105 threshold is used by the National Weather Service (NWS) in issuing Heat Advisories. The NWS also uses a nighttime threshold of HI = 80 to document periods when overnight relief from heat stress is minimal. Categorizing hazard magnitudes has become a common practice by researchers. Tornado hazards have been classified for decades employing the Fujita scale from 0–5, and hurricane hazards have been categorized from 1–5 using the Saffir-Simpson scale proposed in 1971. These classification schemes have helped the general public understand the magnitude of the event that has occurred, is occurring, or will occur. Therefore, when designing a classification model for heat waves, following the same 5-category logic was considered appropriate. This paper presents a model has been developed to classify daily heat stress as a category from 1–5, with 5 being the most extreme events. The classification model is implemented for the years 1980–2000 at 4 cities in Kansas and Nebraska spanning across the known moisture gradient from west to east.

2. Data and Methods The National Climatic Data Center (NCDC) is a good source for obtaining long-term weather data for stations across the United States. Previous works with these data have provided assurance that the hourly information needed is available for the period of 1980 to 2000. A goal of this study was to work with hourly data so the Surface Airways data set of the U.S. first-order and airport stations were acquired. Only temperature and humidity were extracted for this research and used to find hourly heat HI values. The four stations used for this study were Topeka in northeast Kansas, Dodge City in southwest Kansas, North Platte in western Nebraska, and Omaha in eastern Nebraska. After the hourly temperature and humidity data were obtained in English units from the Surface Airways dataset, hourly values of HI are to

302

Chapter Thirty

be determined using the Steadman Equation 30-1 which was obtained from Dixon (1997). It was decided that all HI values greater than 70°F would be kept in the analysis. Maintaining the hourly HI values between 70°F and 80°F allows analysis of conditions prior to or just after a heat stress event and analysis of the degree of nighttime relief provided during the early morning hours. Any day that did not have at least one hour of HI greater than 70° F was eliminated from the data set. This procedure leaves out the usually cold periods of winter, early spring, and late fall.

(Equation 30-1) When measuring heat, simply stating that an hour exceeded a parameter is not enough information to express the magnitude of the heat stress. Heat index values that surpass the thresholds of 90°F and/or 105°F must be measured by how much the threshold is surpassed. If Topeka registered a 3:00 pm HI of 109°F, that hour measures 19 F° above the 90°F threshold, and 4 F° above the 105°F threshold. The amount that an hourly observation exceeds either threshold is determined and then the hourly values are summed to find a heat stress magnitude for each day. Parameters for event classification were based on the accumulated amount by which a day exceeded the 90°F and 105°F thresholds. In the above example, results would be termed 19 hours above HI 90 and 4 hours above HI 105 for that hour. Classification of daily heat events was executed by applying the model in Table 30-3. If a daily magnitude exceeded 30 hours above HI = 90, but had no HI = 105 hours, the heat stress for the day would be considered minor, or Category 1. However, the minimum duration for an event to be classified is either three consecutive days of Category 1, or two consecutive days where one is at least a Category 1 and the other is a category between 1 and 5. A third parameter determining daily heat stress magnitudes is recovery, or the number of actual hours that the heat index was measured at or below 75°F. Recovery time attests to how intense an event is throughout its duration. For a day to reach conditions hazardous

Classifying Heat Stress Events in the Central United States

303

Table 30-3: Heat Stress Classification Model Category 1. Minor 2. Moderate 3. Strong 4. Severe 5. Extreme

HI – 90 >30 >80 >120 >160 >210

HI – 105 – >1 >10 >20 >35

Recovery – ”10 ”6 ”2 0

enough to warrant a Category 5 rating, more than 210 hours of HI = 90 must occur with at least 35 hours of HI = 105, as well as having zero actual hours that register a HI less than 75°F. When classes are assigned for a string of consecutive days, this set of days is considered a single heat stress event. Much like tracking changes in hurricane strength, daily heat stress classifications will vary from day to day depending on local conditions. Also like hurricanes, no matter how long the heat event lasts, or how many different classifications are recorded during the event, the highest category recorded is the classification assigned to the event. Summary statistics were compiled to identify the number of times each category occurred during the time period, as well as the average probability that an event could occur each year.

3. Results and Discussion Results for the magnitude and frequency of heat events from the four stations analyzed from the central United States are generally as expected (Figure 30-2). Topeka had the most events (113) with the most major events (Category 4 or Category 5) as well (18). North Platte experienced the least number of events (33) with zero major events over the 20-year period. Although Dodge City totaled more events (91) than Omaha (66), the higher level of moisture availability at Omaha is easily observed with the larger number of major heat stress events at Omaha during this time period (Table 30-4). Geographically, one can observe the difference in heat events across the east-west moisture gradient in both Kansas and Nebraska. Category 1 events are the most dominant out of the events that occur in North Platte and Dodge City. Dodge City in the southwest of the study area averages

304

Chapter Thirty

Figure 30-2: Geographic Variation of Heat Stress Events across the Moisture Gradient and Latitudinally

over three Category 1 events each year, whereas North Platte in the northwest averaged just over one Category 1 event per year. Comparatively, the two eastern cities also exhibit a north-south disparity in the frequency of Category 1 events. Topeka experiences 2.2 Category 1

Classifying Heat Stress Events in the Central United States

305

Table 30-4: Heat Stress Classification Results TOPEKA Frequency Average/21

CAT 1 44 2.20

CAT 2 28 1.40

CAT 3 22 1.10

CAT 4 13 0.65

CAT 5 5 0.25

TOT 112

OMAHA Frequency Average/21

CAT 1 16 0.80

CAT 2 19 0.95

CAT 3 16 0.80

CAT 4 9 0.45

CAT 5 6 0.30

TOT 66

DODGE CITY Frequency Average/21

CAT 1

CAT 2

CAT 3

CAT 4

CAT 5

TOT

75 3.57

14 0.67

1 0.05

0 0.00

1 0.05

91

NORTH PLATTE Frequency Average/21

CAT 1

CAT 2

CAT 3

CAT 4

CAT 5

TOT

27 1.35

4 0.20

2 0.10

0 0.00

0 0.00

33

events on average, and Omaha averages less than one per year, with an 80% chance of one occurring. A significant drop-off in event frequency is observed with Category 2 events in the western cities. This abrupt reduction in heat stress occurrence, 67% chance annually in Dodge City and a mere 20% chance annually in North Platte, is a testament to the lack of moisture needed to produce heat stress conditions greater than the Minor category on a regular basis. Eastward, Topeka has 1.4 Category 2 events per year, and Omaha sees almost one annually. Heat stress events of higher magnitude are extremely rare in the western areas of the study area. Together, both stations only experienced 4 actual events between Category 3 and Category 5 over the 20-year period. North Platte never saw an event higher than a Category 3, but Dodge City did register a Category 5 heat stress event in 1980. The event of 1980 was widespread throughout the central U.S., and these results indicate that the highest order of heat stress can occur in the western Plains during the most extreme of warm season conditions. Topeka still experiences over one Category 3 event per year, and these events are relatively common in Omaha as well.

306

Chapter Thirty

Frequency of the two major categories, 4 and 5, decreases for both Topeka and Omaha, but the probability of a major event occurring is still around 50% annually. Conditions that develop into a Category 5 heat stress hazard are extremely rare regardless of location. Adaptation to such high levels would therefore not be expected for people in any of the areas in this study. However, the knowledge of their occurrence in three of four places in this region warrants a necessity to educate/mitigate for future extreme heat stress hazards. Although a hurricane can reach a Category 5 status on the SaffirSimpson scale, it may be reduced to a Category 3 by the time it makes landfall. The magnitude of an extended period heat stress event can be identified in a similar manner with this model as each day is assigned a value, and prior magnitudes of the event can be accounted for. The difference between heat stress events and the hurricane example is that the whole heat stress event is widely experienced at a given place, whether it is a Category 1 or a Category 5 event. The Category 5 event at Dodge City in 1980 lasted 28 days, but only the third day of the event actually registered a 5 classification. Topeka experienced an event at the same time that was ten days shorter, but eight of the days (four of which were consecutive) registered a Category 5. The nature of these events is notably different, and at present it is still difficult to confidently ascribe the concrete differences in location sensitivities to the hazard. Further analysis will include data from several points around the country, as well as more biometeorological collaboration to continue a progressive understanding of human capacity to adapt during extreme climatic events.

4. Summary Implementation of the new heat stress classification model provides an objective as well as descriptive way to characterize local exposure to extreme heat events. However, objectivity is difficult to preserve when geographic location, local sensitivity, and varying adaptive capacities of the people involved are all things to consider when determining the overall vulnerability associated with heat events. Work in this study did exhibit positive results in the direction of heat stress exposure classification in a manner that could inform people of the hazard levels both during and after an event. Results in this research also provide frequency and probability of specific heat stress occurrences for an area. Future work in this area would allow for better geographic knowledge of heat stress magnitudes, as well

Classifying Heat Stress Events in the Central United States

307

as a refinement of the techniques necessary to describe the effects of certain magnitudes and frequencies of heat stress levels on people.

References American Meteorological Society. 2000. Glossary of Meteorology, 2nd ed. T. Glickman (ed.). Boston, MA: American Meteorological Society. Burroughs, W. 2003. Climate into the 21st Century. New York, NY: Cambridge University Press. Choi, J. and Meentemeyer, V. 2002. Climatology of persistent positive temperature anomalies for the contiguous United States (1950–1995). Physical Geography 23, 175–195. Christopherson, R. 2001. Elemental Geosystems, 3rd Ed. Upper Saddle River, NJ: Prentice Hall. Delworth, T., Mahlman, J., and Knutson, T. 1999. Changes in heat index associated with CO2-induced global warming. Climatic Change 43, 369–386. Dixon, R. 1997. A heat index climatology for the southern United States. National Weather Digest 22, 16–21. Dodd, A. 1965. Dew point distribution in the contiguous United States. Monthly Weather Review 93, 113–122. Harrington, J. and Bowles, E. 2002. Assessing the impact of heat and humidity on livestock: Development of an hourly THI climatology. Papers and Proceedings of the Applied Geography Conferences 25, 311–315. Hevener, O. 1959. All about humiture. Weatherwise 12, 56–85. Houghton, F. and Yaglou, C. 1923. Determining equal comfort lines. Journal of the American Society of Heating and Ventilating Engineers 29, 165–176. Hubbard, K., Stooksbury, D., Hahn, L, and Mader, T. 1999. A climatological perspective on feedlot cattle performance and mortality related to the temperature-humidity index. Journal of Production Agriculture 12, 650–653. Kalkstein, L. 1995. Lessons from a very hot summer. The Lancet 346, 857–859. Kalkstein, L. and Valimont, K. 1986. An evaluation of summer discomfort in the United States using a relative climatological index. Bulletin of the American Meteorological Society 67, 842–848. Kalkstein, L., Jamason, P., Greene, J., Libby, J., and Robinson, L. 1996. The Philadelphia hot weather-health watch/warning system: Development

308

Chapter Thirty

and application, Summer 1995. Bulletin of the American Meteorological Society 77, 1519–1528. Lally, V. and Watson, B. 1960. Humiture revisited. Weatherwise 254–256. Mitchell, J. 1989. Hazards research. In: Geography in America. Gaile, G. and Willmott, C. (eds.). Columbus, OH: Merrill Publishing Company. NOAA. 2006. National Weather Service Weather Forecast Office. National Oceanographic and Atmospheric Administration, St. Louis, Missouri. Page last modified: 9-Jun-2006 3:57 PM UTC. http://www.crh.noaa.gov/lsx/vortex/summer_safety.php Oke, T. 1987. Boundary Layer Climates, 2nd Ed. New York, NY: Routledge. Quayle, R. and Doehring, F. 1981. Heat stress: A comparison of indices. Weatherwise 34, 120–124. Robinson, P. 2001. On the definition of a heat wave. Journal of Applied Meteorology 40, 762–775. Rosenberg, N.J., Crosson, P.R., Frederick, K.D., Easterling III, W.E., McKenney, M.S., Bowes, M.D., Sedjo, R.A., Darmstadter, J., Katz, L.A., and Lemon, K.M. 1993. The MINK methodology: Background and base-line. Climatic Change 24, 7–22. Smoyer, K.E., Kalkstein, L.S., Greene, J.S., and Ye, H.C. 2000. The impacts of weather and pollution on human mortality in Birmingham, Alabama and Philadelphia, Pennsylvania. International Journal of Climatology 20, 881–897. Steadman, R.G. 1979. Assessment of sultriness. Part I: A temperaturehumidity index based on human physiology and clothing science. Journal of Applied Meteorology 18, 861–873. —. 1984. The universal scale of apparent temperature. Journal of Climate and Applied Meteorology 23, 1674–1687. Thom, E. 1956. Measuring the need for air conditioning. Air Conditioning, Heating, and Ventilating 53, 65–70. —. 1959. The discomfort index. Weatherwise 57–60. Vandentorren, S., Suzan, F., Medina, S., Pascal, M., Maulpoix, A., Cohen, J., and Ledrans, M. 2004. Mortality in 13 French cities during the August 2003 heat wave. American Journal of Public Health 94, 1518– 1520. Ward, R. 1925. Climates of the United States. Boston, MA: Ginn and Company.

SECTION VII: COMMUNICATING CLIMATE TO THE PUBLIC

CHAPTER THIRTY-ONE OVERVIEW OF COMMUNICATING CLIMATE TO THE PUBLIC ROBERT V. ROHLI

An important component of any applied research is to optimize the societal benefit of the discovered information. The same holds true in applied climatology, but in the case of applied climatology misconceptions and mistrust must be overcome in some cases because of pre-conceived notions that the public may have about climate information. Persuasion theory that emanated from Yale University in the 1950s holds that new information of any type, whether it be research or simple interpersonal communication, only becomes useful to the consumer of that information after the consumer accepts the truth of the information and overcomes the natural urge to avoid acting on it. Applied climatologists must realize the importance of honest communication of their research to the general public without manipulating the message to push a certain agenda. While the general public may not understand the nuances of the research, clearer methods of communication are important. Four papers from the Applied Geography Conferences are presented here. In the first, Earl Cook (1981) insightfully details some potential pitfalls in forecasting not only climate projections, but projections of any kind. He cautions the forecaster of the difficulties of making forecasts and ensuring that those forecasts are interpreted and acted upon by the public as intended. In the decades since Cook’s essay was written, his words have become more and more familiar to applied climatologists, as they have continually seen their projections ignored, misinterpreted, and misused. The second paper, by John Harrington, Lisa Tabor, and Iris Wilson (2011), uses 42 surveys to identify the extent to which agriculturalists in Kansas are observing the impacts of climate change and how they use climate change information strategically to mitigate negative impacts in their livelihoods. The third paper, also by John Harrington (2012), summarizes

Overview of Communicating Climate to the Public

311

National Science Foundation-supported research on the attitudes of the general public toward climate change, ranging from “alarmist” to “dismissive” and the reasons for those attitudes. The final essay in this monograph, by John Oliver (1996), examines the “future” of applied climatology. Many of Oliver’s statements have indeed come to pass over the last two decades, and it is likely that the events of the future will also support other statements of his.

CHAPTER THIRTY-TWO THE PERILS OF PROJECTION EARL COOK (1981)

1. Introduction The forecaster has an important but unenviable job. He is almost certain to be wrong in the numbers he projects; only rarely does he have sufficient control to be able to forecast in terms of probability. He is more apt to be wrong as the phenomena with which he deals become more complex and as the focus of his interest becomes smaller. The three great systems we live in – the physical, the biological, and the sociocultural – grow more complex in that order. Forecasters of physical phenomena find enough difficulty with weather and earthquakes; how then can we expect to do better, or even as well, in forecasting the effects of human perturbations of the biological and sociocultural environments? Yet, these are the systems or environments of greatest interest to most humans, where we persistently seek foreknowledge of the consequences of our proposed actions. Some of the perils and difficulties of socioeconomic forecasting are dealt with in the papers I have been asked to discuss, one by Branch et al. (1981) (which I shall call the MWR paper), the other by Stenehjem (1981).

2. The Purpose of Assessment In neither paper is it clear for what and for whom socioeconomic impact assessments are made. Surely they are not simply to fulfill a legal requirement; nor is their sole purpose to aid professional planners and project promoters. Indeed, the institutionalization of impact assessments, both environmental and socioeconomic, came about through political pressure exerted by people who feared degradation of their physical, biological, and sociocultural environments, and who wanted either

The Perils of Projection

313

preservation of those environments or the impossible dream of maximum benefit at minimum cost – in other words, to sell out at a high price. The impact statement, although useful in development planning, is in fact a decision instrument, a brief to be used by the opposing forces of development and preservation. In this context, the distributional analysis of costs and benefits, in geographic, demographic, and temporal contexts, is perhaps the most important part of the assessment. Who will get the benefits, and who will pay the costs? Each actor in the decision play will have this question uppermost. Different persons use forecasts in different ways, but almost invariably they are used in a decision mode. The farmer may want rain when you want a sunny day, but both of you will tend to base your behavior tomorrow on the forecast. The realtor in Rifle, Colorado, may want an influx of population, while the retired rancher and his wife fear the resulting rise of taxes that may force them to sell the family home; each will use a forecast to favor or oppose the project that might cause the influx. Distributional analysis gets difficult in the values arena. For instance, how does one weigh the cost to a Sierra Club member living in San Francisco of the building of a coal-fired power plant on the Colorado Plateau? Or the cost to the local residents who see their sons and daughters leaving the Plateau for jobs elsewhere if the plant is not built? Perhaps the assessor should not be expected to go that far in the distributional analysis of the sociocultural impacts of such a power plant, but I agree with the MWR authors that social impact analysis is the logical terminal stage of the assessment procedure. Not many persons in the arena of impact will have all their questions answered by an economic-demographic impact assessment, no matter how good it is. They still will not know, without a social assessment, how their lives or the lives of their children may be changed by the implementation of a proposal to mine, refine, or to build a dam or power plant; and that is precisely what many of them want to know. In addition to answers to questions of income, property values, and taxation, they want to know where their new neighbors will come from, what kind of people they will be, and how they will change the sociocultural environment. Will the present residents be faced with locking their cars while shopping, with keeping their daughters off the streets at night, with increased noise and traffic hazards, with diminution of their

314

Chapter Thirty-Two

preferred means of recreation and with alteration of their own status in the community?

3. Sociocultural Assessment That sociocultural assessment is difficult is shown by the MWR paper, whose authors state that “the attempt to link economic-demographic changes to social-impact assessment” has been “of particular importance,” but they are remarkably unclear about the way or ways in which that attempt has been carried out. It appears from their discussion that the persons concerned with this problem may be nearing “first base,” but have not yet made a “clean hit,” and are very far from “scoring.” We have here a clear and present problem: how to deal with the qualitative extensions of the quantitative assessment. While it is undoubtedly important to attempt to refine economic and demographic assessment procedures by lessons learned from hindsight or post audits of the kind undertaken by Electric Power Research Institute, it may be even more important to attempt similar hindsight studies for social and environmental impacts. Incidentally, I do not believe Stenehjem’s use of the term “socioeconomic” to describe the EPRI-funded studies is appropriate; from the details he gives, they are largely economic and demographic, not social. If socioeconomic impact assessment is to fulfill the broad meaning of that term, it is not sufficient to project only economic benefits and costs, any more than it is sufficient in an environmental impact assessment to project only expectable concentrations of pollutants in air, water, and soil. The bottom line for a community, if that term is used in its human meaning, cannot be a fiscal analysis, but must be a sociocultural assessment.

4. Economic-Demographic Assessment Stress on the social and cultural aspects of assessment in no way demeans the importance of assessment of the economic, demographic and infrastructural impacts; if anything, it means that such assessments should be made with the intellectual rigor so persuasively advocated by Stenehjem (1981), because the utility of the social evaluation depends on the thoroughness and accuracy of the assessments of the economic, demographic, and infrastructural impacts.

The Perils of Projection

315

Stenehjem’s (1981) comments on problems of local labor and capital supply are incisive and appropriate when related to defects in forecasting the demand for labor and capital. Stenehjem (1981) tackles only one part of the assessment problem, but he does it very well. The MWR authors go after bigger game, but do not succeed as well. Stenehjem (1981) is concerned with improving the basement, while the others want to build a house on it. Unfortunately, the house involves questions of the nature of the residents, their living arrangements, and differences in their preferences for wallpaper and furnishings that do not enter into the design of the basement. Or do they? Here is a point on which the two papers appear to differ. Stenehjem (1981) appears to regard the basic inputs to assessment as invariable, whereas the MWR authors suggest that they may be changed by interaction of the potentially affected public with the project planners. This view of the decision process puts the assessor more dynamically into the process, not as a planner, but as a person who evaluates a changing scenario as it is modified, rather than one best plan (from the promoters’ viewpoint) for ultimate acceptance or rejection.

5. The Time Element The time element seems undervalued in these papers. Like an oil spill or a nuclear power plant accident, the immediate consequences of a construction project may differ greatly, in both intensity and extent, from its long-range consequences. It is not a question of short-term economic benefits and long-term social costs, although some critics of traditional benefit-cost analysis seem to have set up such an opposition. It is more a case of temporal diffusion of costs and benefits, both economic and social, with consequent increasing difficulty of distinction and assessment in the extended time frame. A completed project may change all the subsequent history and nature of a community. The time element is of special significance in the following circumstances: (1) high employment and inflow of capital during the construction period, relatively low employment thereafter; (2) high employment and inflow of capital during the period of depletion of a nonrenewable resource, none thereafter; and (3) relatively low social impact during the early years, but basic changes in the community occurring with time by a sort of sociocultural mutagenesis.

316

Chapter Thirty-Two

Western civilization is a boomtown civilization. We have not worried very much about depletion of soils, exhaustion of mines, leveling of forests, or using up nature’s capacity to accept our wastes. We traditionally use a high discount rate in assessing future benefits as well as future costs, whether these are market or nonmarket in nature. Perhaps the most fundamental distinction between economic and demographic assessment, on the one hand, and sociocultural assessment, on the other, is that the former can be done with a discount rate either stated or implied, while the latter should not be. Of course, the former always can be done with greater precision than the latter; but here the words of the great Spanish philosopher Ortega y Gasset (1941) are apt; he pointed out that science has achieved exactness and certainty in prediction “at the cost of remaining on a plane of secondary problems.” And so it is with economics and demography as well. Where science or economics or demography stop, man and society do not.

References Branch, K., Chalmers, J.A., and Pijawka, D. 1981. Toward an Integrated Approach to Socioeconomic Assessment of Energy Resources Development. Tempe, Arizona: Mountain West Research, Inc. Ortega y Gasset, J. 1941. Toward a Philosophy of History. New York, NY: Norton. Stenehjem, E.J. 1981. Forecasting Economic and Demographic Impacts from Energy Development. Denver, CO: Denver Research Institute.

CHAPTER THIRTY-THREE STAKEHOLDER ADAPTATION TO CLIMATIC CHANGE IN KANSAS: WHAT HAVE WE LEARNED? JOHN A. HARRINGTON JR., LISA K. TABOR AND IRIS E. WILSON (2011)

1. Abstract Earth-system variations have combined with transformations induced by an expanding global society to shape a dynamic planet where change has happened and changes will happen in the future. Agricultural stakeholders in Kansas have reacted to a transforming business environment by adopting new strategies to maximize production while experiencing the climatic vagaries of floods and major periods of drought. This paper combines information from 42 stakeholder interviews with ideas from the scholarly literature on past adaptation in the central Great Plains to address the question: are Kansas farmers going to be able to adapt to climate change? There is clear evidence that autonomous adaptation has happened over that last century as new land management practices and technological innovations have been put into play. Interviews in north-central Kansas in 2010 with farmers and ranchers suggest that they are observing a changing environment and that they desire credible information on how future climate change might impact their production practices.

2. Climate and Other Environmental Change in Kansas We live on a rapidly changing planet. Human population is rapidly approaching seven billion and land cover changes dedicated for human

318

Chapter Thirty-Three

food production have exceeded one-third of the terrestrial surface (Foley et al., 2005). The Anthropocene is a time of human-induced environmental change that has been marked by extensive losses in biodiversity; humans are a major driver for the current and sixth major planetary species extinction event. Globally, temperatures have warmed approximately 0.8 C° since 1900 (Karl et al., 2009), an increase that is strongly linked to the burning of fossil fuels and related emission of CO2 into the atmosphere. While warming has been most pronounced in the Arctic, temperature increases and other climate changes will have differing magnitudes and impacts at every scale across the globe. In Kansas, Feddema et al. (2008, 7) projected that there will be “...less frequent, but more intense storms, with longer dry periods between more intense rainfall events. Higher intensity rainfall will mean more runoff and less water storage in the soil during the extended dry periods.”

Wendland (1993) provided an early assessment of potential impacts of climate change on agriculture in Kansas using the CERES-Maize corn yield model and climate projections from the first IPCC report. Corn yields in eastern Kansas were calculated to decline by 19% by the middle of the 21st century. Findings from both Feddema et al. (2008) and Wendland (1993) used the science of climate change to suggest an increase in growing degree days and a greater likelihood for increased storm events with excessive precipitation in Kansas. Stakeholders in agricultural states, like Kansas, have adapted to changing climate and environmental conditions in the past and are likely to face additional challenges in the decades to come. The IPCC defines a stakeholder as: “a person or an organisation that has a legitimate interest in a project or entity, or would be affected by a particular action or policy” (IPCC, 2007, 881). Climate can be difficult to understand because it is an intellectual concept. Climate is not something one can measure with meteorological instruments nor is it something that a human can sense (e.g., hot air, the wind at your back, rain on your face) (Hulme, 2009). One way to characterize the local climate is to generate summary statistics about one or more of the meteorological elements that comprise the climate of a place. Climate change can be even more difficult to understand and detect since it is a transitional process occurring over time. Given that there is considerable natural variability in the atmospheric system, that variability or “noise” makes the signal of climate change (perhaps a slight change in the statistical properties of the system) difficult for stakeholders to

Stakeholder Adaptation to Climatic Change in Kansas

319

perceive. Based on a human tendency to recall only the more significant events, it is a challenge to remember the string of weather events from several decades ago to compare with climate today (Mertz et al., 2009). Kansas has warmed slightly along with the planetary increase (Feddema et al., 2008). But it can be hard to detect climate change in Kansas, especially in areas where there has been a significant expansion of irrigation. Evaporation and transpiration create relatively cooler areas due to increased crop irrigation. The human-induced land cover change of increased irrigation may be “hiding” an ongoing warming trend from Kansas stakeholders. However, much of the groundwater based irrigation is not sustainable in the long term, since much of the practice in western Kansas involves the mining of fossil groundwater. How fast will temperature warm with a shutoff of irrigation in the west? Awareness of the potential challenges that lie ahead is needed for Kansas farmers and ranchers to find realistic ways to balance current constraints and the prospect of a changing playing field. It can be argued that there needs to be a widespread adoption of an improved understanding about climate change among Kansas citizens so that they can construct “...a realistic assessment of the vulnerability” (Feddema et al., 2008, 10) in order to best prepare for the future.

3. Research Goals As part of an NSF-funded research project on climate and energy in Kansas, this paper address aspects of the climate impacts and adaptation component of the overall research effort. Two major components of the effort to-date have been a review of past agricultural adaptation activities within Kansas and interviews of local stakeholders in north-central Kansas (Figure 33-1). Information from these two sources is assembled to address vulnerability and adaptation to climate change in Kansas and the central Great Plains. Generalized findings from interviews with 42 stakeholders in north-central Kansas during the summer of 2010 are discussed in the context of an understanding of past adaptation to climate and environmental change. Specific research goals are: 1. To summarize key themes from the research literature on impacts and adaptation to climate and environmental change in Kansas.

320

Chapter Thirrty-Three

Figure 33-1: Study Countyy Locations within w the Smooky Hill and Flint F Hills Physiographicc Regions within Kansas

2. To reeview the resuults from stak keholder interrviews in nortth-central Kanssas on their perception of o environmenntal change and past modiifications in thheir production n methods duue to these chaanges. 3. To coompare and drraw conclusio ons between w what is being suggested s in thee academic liiterature and what w is actuallly being imp plemented and/oor changed in agriculture. 4. Withh an emphasiss on information sources used by stak keholders, deterrmine what fuurther steps neeed to be takeen to commun nicate the impoortance of adapptation inform mation.

4. Adaptatioon to Clima ate Change in Kansas A review w of literature on agricultural impacts andd adaptation to t climate and environnmental changge in Kansass and the Greeat Plains reg gion was generated too inform an onngoing researcch effort on cliimate change,, impacts, and adaptatiion, and posssible future mitigation m in K Kansas. Based on this review of m more than 40 joournal articles, common adaaptive practicess/methods were identiffied. These themes are: irrigation and irrrigation efficiency, notillage or reeduced tillage managementt practices, crrop rotation, change c in planting dattes, change inn crops planteed, and a shifft to increased d dryland farming. A significant human h characcteristic identiified in this scholarly literature is the general attitude a of ressiliency and aadaptability am mong the farmers and ranchers acrooss the region..

Stakeholder Adaptation to Climatic Change in Kansas

321

A theme regularly mentioned in the climate and environmental change adaptation literature is irrigation since supplemental water can enable a successful crop during periods of below normal rainfall. Unfortunately, irrigation in the central Great Plains is frequently drawn from a primary water source that is utilized at unsustainable rates (Easterling et al., 1993; Harrington, 2005). Higher daily temperatures projected with 21st century climate change will call for a greater need for irrigation than exists today. With irrigation waters already being used in a non-sustainable manner, priorities may need to change to support a decrease in total irrigation water use along with technological improvements to increase irrigation efficiency (Riebsame, 1983; Rosenberg, 1986; Rosenberg, 1992; Easterling et al., 1993; Mendelsohn, 2000; Harrington, 2005; Howden et al., 2007). The need for a decrease in irrigation connects to a shift toward an increase in dryland farming. As temperatures rise, there will be an increase in soil water stress. With less soil water available, dryland cropping with the selection of appropriate drought resistant crop varieties will become very important in order to maintain productivity, a viable regional economy, and food security (Rosenberg, 1986; Easterling et al., 1993; Mendelsohn, 2000). Soil management is critical in agriculture and the practice of reduced tillage or no-till is better than conventional tillage methods, since reduced tillage increases soil organic carbon and water holding capacity (McVay et al., 2006). No-till is frequently suggested as a long-term adaptive planning method for soil conservation in terms of both climate and environmental change (Riebsame, 1983; Rosenberg, 1986; Easterling et al., 1993; McVay et al., 2006; Howden et al., 2007). Crop rotation will also be important to decrease the impacts of soil moisture stress as another long-term adaptation measure (Howden et al., 2007). As temperatures change and seasons warm and cool earlier or later than what is currently considered normal, a change in planting dates will also be necessary to maximize crop yields (Rosenberg, 1992; Easterling et al., 1993; Feddema et al., 2008). Related to the change in planting dates is the idea of a modification in crops planted. If climate is altered and can no longer support the traditional crop being grown, for food security reasons, a more climate appropriate crop or a season of fallow must be considered as a replacement for the traditional crop (Riebsame, 1983; Rosenberg, 1986; Rosenberg, 1992; Easterling et al., 1993; Mendelsohn, 2000; Southworth et al., 2000).

322

Chapter Thirty-Three

There is an attitude of strong cultural resiliency across Kansas and much of the Great Plains. Environmental history and local culture document a strong connection to surviving the Dust Bowl of the 1930s and the severe drought in the 1950s in this region. Both of these extreme climatic events were multi-year and multi-state droughts that had a major impact of economic, social, and environmental conditions in the region. Gutmann (2000) suggests that another Dust Bowl scenario is a possibility as climate changes. However, with the right mitigation and adaptation measures, the impacts could be lessened. For agricultural producers to build on their culture of resilience and deal effectively with potential climate change there is a need to effectively communicate the current science and its implications for climate change across the central Great Plains. Given the mixed messages that are currently available to stakeholders, there is a general mistrust of climate science. Fortunately, there is a strong belief in the ability of the agricultural culture to adopt new technologies and shift production methods if necessary, as has been done historically (Harrington, 2005).

5. Stakeholder Thoughts on Environmental Change Conclusions can be drawn from a farmer’s perception of environmental threats in order to estimate what mitigation and adaptation methods he/she may be willing to utilize (Taylor et al., 1988). This is because people react more strongly to things they have experienced; whether positive or negative and people tend to rely on their personal knowledge for motivation to react. Perception of the need to change is a significant element contributing to farmer willingness and understanding to implement adaptive methods to environmental hazards (Taylor et al., 1988). In the summer of 2010, 42 farmers, ranchers, and local specialists were interviewed in a semi-structured manner across north-central Kansas to gain a better understanding of their perceptions regarding local environmental change. Twenty of those interviewed were crop farmers, ranchers, or were engaged in both activities; the other 22 were local specialists. In an informal interview setting, participants were engaged in a conversation about local environmental and climate change. The topics for the conversation relevant to climate change included:

Stakeholder Adaptation to Climatic Change in Kansas

323

1. 2. 3. 4. 5.

Do you think weather is changing? Do you think another Dust Bowl is likely? Do you use irrigation? Why and with what frequency? How is your soil fertility? Do you have erosion problems? Are you managing differently because of climate and weather events? 6. Where do you get your information about climate and weather?

Other questions dealt with changes in biodiversity, change in water quantity and quality, and whether or not fire was used as a land management technique. While the stakeholders had good knowledge of ongoing changes in species presence (more turkey and deer, fewer quail) and the results of changing land management practices that have reduced soil erosion, the emphasis in this paper is on those aspects of the dialog that dealt with themes relevant to climate change. Out of the 15 farmers who responded to the question about whether they thought climate had changed since they started farming, 12 answered “Yes.” Most noted increased rainfall and greater occurrence of extreme weather events. One farmer said that he has seen longer hot, dry periods and that they are beginning sooner in the year than they used to. Another farmer answered: “Yes. June is now like July; July is like September, who knows about August.” Only one person thought that another Dust Bowl was likely. It was noted by some that, although similar weather conditions might exist, farmers are using new and better methods for managing the soil than were being used in the 1930s, such as no-till practices, so that blowing dust would be less likely in future multi-year droughts. As part of the conversation about land management, it was asked if producers took any precautions and managed differently than they would otherwise due to extreme climate and weather events. Overall, everyone said “No.” A general conclusion based on the stakeholder interviews is that they believe that they had an adaptive capacity and resiliency in terms of land management. Irrigation is used throughout the region by crop farmers and many shared the idea that they used it to increase yields and overall profitability. Most irrigation in the area is usually rain-fed, surface water irrigation stored in stock ponds on the property. It was noted by some ranchers that they occasionally add supplemental water to maintain their pastures during an exceptionally dry year, if they have the water and necessary

324

Chapter Thirty-Three

infrastructure. Irrigation is considered by crop farmers as necessary and a fact of life; however, the significance of the greater need for dryland farming in the future was recognized. Soil fertility was not identified as an issue for farmers or ranchers in the study area. They answered either that there had been no change or that fertility had improved due to an increased adoption of no-till farming. Soil loss through erosion was also thought to have decreased. Some of the reasons noted for the improvement were changes in land management practices such as terraces, installation of grass water ways, and reduced tillage or no-till farming. Stakeholders in Kansas agriculture are accustomed to the local weather, its seasonal cycles, and inter-annual variability. Most do not believe that a Dust Bowl of the magnitude experienced in the 1930s is likely to happen again. They believe that current land management practices will provide them with the ability to make adjustments and that adaptation will protect them from a catastrophe such as occurred in the 1930s. Producers readily acknowledge the benefits that no-till practices have had over the years on reduced soil erosion and improved soil fertility. While they may initially be reluctant to try new methods, once farmers or ranchers see success firsthand through neighbors who are early adopters, they are likely to change land management practices. In the past, agricultural extension services have played a major role in helping communicate new scientific findings through establishment of local research plots. In addition to learning from their neighbors, television stations and the internet were identified as the more common information sources for farmers. Accomplishment can be seen in terms of a variety of adaptation methods and general willingness to adapt to climate variability in Kansas. However, there is still room for improvement in the forms of education and communication about the likely nature of future climate change and the best adaptive practices that can be used by producers and specialists. Hopefully, this increased understanding of possible change and needed response with enable key stakeholders and local specialists to work with policy makers to create a political environment that facilitates adaptation.

Stakeholder Adaptation to Climatic Change in Kansas

325

6. Synthesis of Findings: Emphasis on Information Sources and the Communication of Adaptation Information Interviews with local stakeholders in north-central Kansas provided qualitative information that documents an awareness of environmental change. Farmers, ranchers, and local specialists had an awareness that generally matched on-going changes in species abundance, environmental improvements related to land management practices, and recent variations in weather and climate patterns. An encouraging finding was that the vast majority of farmers and ranchers desired more information about the nature of recent and potential changes in climate. There is concern over the effectiveness of communication pathways between climate science research and local producers in terms of the sharing basic climate science, impacts, and adaptation information. This scholarly information needs to be delivered in a way that is relevant to local farmers and ranchers and in an environment that encourages an open sharing of ideas and concerns. Appropriate word choices that avoid scholarly jargon and connect with local thinking are critical for effective information sharing. Bielders et al. (2001) suggests that a perceived low importance of an environmental hazard tends to make stakeholders less inclined to participate in adaptation measures until activities have attended to other perceived hazards of greater importance. There is frequently an underlying issue with perception studies, where the individual ranks climate change as an overall concern low, but their primary concerns all relate directly to climate change (Mertz et al., 2009). Greater acknowledgment of the methods of communication that work is needed, with an emphasis on strengthening those connections, while looking for possibilities where other practices of communication can be established. As adaptation to climate change in Kansas has been studied, many recommendations for possible future adaptation have been identified. These adaptive approaches include less irrigation and increased irrigation efficiency, no-tillage or reduced tillage management practices, crop rotation, change in planting dates, changes in crops planted, and a shift to increased dryland farming. When comparing the adaptations identified in a survey of the literature to changes made by rural agricultural stakeholders in Kansas, many connections can be made. First, both the scientific literature and the interviews acknowledge that climate is changing.

326

Chapter Thirty-Three

In the case of irrigation, where the adaptation literature advises decreasing its use, irrigation is still considered absolutely necessary for production. However, an increase in dryland farming was suggested in both the literature and by producers as a method to move away from a dependence on irrigation. Soil management, particularly implementing notillage practices, was a strong recommendation for adaptation to climate change and the interviews demonstrated that no-till is becoming an increasingly common practice and is recognized for its many environmental benefits. Producers reported that they did not purposefully manage differently for extreme climate and weather events and the literature did not explicitly encourage different management for specific instances, but overall encouraged crop rotation, shift in planting dates, and shift in crop varieties planted as climate changes. There was little belief that another Dust Bowl is likely but the literature does suggest that there is always a possibility of similar climate conditions returning. Both the literature and the interviews demonstrated a strong sense of regional resiliency and the willingness and belief in the technological capacity to adapt if necessary. Hence there is a good match between suggested future adaptations and past experiences. Overall this comparison shows that stakeholders in Kansas are not only willing to alter their practices as climate changes, but that they acknowledge past effective strategies. Part of the communication issue is that there is little to no direct contact between researchers and stakeholders, which can make translation and information sharing much more difficult (Oppenheimer, 2011). Unfortunately, local specialists, whether they were federal employees with the USDA NRCS or state employees with K-State Extension, expressed reluctance to share information about climate change with local producers; these professionals suggested that the topic was a “non-starter” for effective communication. In the interviews, the most common sources used by stakeholders for climate and weather information are television and the internet. Few of the articles reviewed for this work provided suggestions for readers about locations available to obtain additional information, but when they did, the articles referenced internet resources. This suggests the internet as a main source to communicate climate change adaptation information, though the concern remains on how to direct the attention and interest of agricultural stakeholders to the specific valuable internet websites.

Stakeholder Adaptation to Climatic Change in Kansas

327

This work shows that despite communication barriers, stakeholders in Kansas are trying new methods as conditions change. In addition, agricultural stakeholders recognize that climate change is happening. Given an existing record of successful adaptation, it is likely that Kansas farmers and ranchers will be able to adapt to changes in the future. However, there are still many strides to be made in communicating climate science and adaptation information, particularly how to put the stakeholder in contact with the good information that is available.

Acknowledgments The authors would like to thank Courtney Estes, Brian Nechols, Nathan Owens, and Jordan Waechter for their assistance in conducting stakeholder interviews. We would also like to recognize the Kansas NSF EPSCoR Award (09038060): Phase VI: Climate Change and Energy: Basic Science, Impacts, and Mitigation.

References Bielders, C.L., Alvey, S., and Cronyn, N. 2001. Wind erosion: The perspective of grass-roots communities in the Sahel. Land Degradation & Development 12, 57–70. Easterling III, W., Crosson, P., Rosenberg, N., McKenney, M., Katz, L., and Lemon, K. 1993. Agricultural impacts of and responses to climate change in the Missouri-Iowa-Nebraska-Kansas (MINK) region. Climatic Change 24, 23–61. Feddema, J.J., Brunsell, N.A., Jackson, T.L., and Jones, A.R. 2008. Climate change in Kansas. http://www.climateandenergy.org/LearnMore/InTheNews/ClimateStudy. htm Foley, J.A., DeFries, R., Asner, G.P., Barford, C., Bonan, G., Carpenter, S.R., Chapin, F.S., Coe, M.T., Daily, G.C., Gibbs, H.K., Helkowski, J.H., Holloway, T., Howard, E.A., Kucharik, C.J., Monfreda, C., Patz, J.A., Prentice, I.C., Ramankutty, N., and Snyder, P.K. 2005. Global consequences of land use. Science 309(5734), 570–574. Gutmann, M.P. 2000. Scaling and demographic issues in global change research: The Great Plains, 1880-1990. Climatic Change 44, 377–391. Harrington, L.M.B. 2005. Vulnerability and sustainability concerns for the U.S. High Plains. In: Essex, S.J., Gilg, A.W., and Yarwood, R. (eds.), Rural Change and Sustainability: Agriculture, the Environment and Communities, Cambridge, MA: CABI Publishing, 169–184.

328

Chapter Thirty-Three

Howden, S.M., Soussana, J.-F., Tubiello, F.N., Chhetri, N., Dunlop, M., and Meinke, H. 2007. Adapting agriculture to climate change. Proceedings of the National Academy of Sciences of the United States of America 104, 19691–19696. Hulme, M. 2009. Why We Disagree about Climate Change. Cambridge, UK: Cambridge University Press. IPCC, 2007. Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Parry, M.L., Canziani, O.F., Palutikof, J.P., van der Linden, P.J., and Hanson, C.E. (eds.), Cambridge, UK: Cambridge University Press. Karl, T.R., Melillo, J.M., and Peterson, T.C. (eds.). 2009. Global Climate Change Impacts in the United States. Cambridge, UK: Cambridge University Press. McVay, K.A., Budde, J.A., Fabrizzi, K., Mikha, M.M., Rice, C.W., Schlegel, A.J., Peterson, D.E., Sweeney, D.W., and Thompson, C. 2006. Management effects on soil physical properties in long-term tillage studies in Kansas. Soil Science Society of America 70, 434–438. Mendelsohn, R. 2000. Efficient adaptation to climate change. Climatic Change 45, 583¬600. Mertz, O., Mbow, C., Reenberg, A., and Diouf, A. 2009. Farmers’ perceptions of climate change and agricultural adaptation strategies in rural Sahel. Environmental Management 43, 804–816. Oppenheimer, M. 2011. What roles can scientists play in public discourse? Transactions EOS 92, 133–134. Riebsame, W.E. 1983. Managing drought impacts on agriculture: The Great Plains experience. In: Platt, R. and Macinko, G. (eds.), Beyond the Urban Fringe: Land Use Issues in Non-Metropolitan America, Minneapolis, MN: University of Minnesota Press, 257–270. Rosenberg, N.J. 1986. Adaptations to adversity: Agriculture, climate and the Great Plains of North America. Great Plains Quarterly 6, 202–217. —. 1992. Adaptation of agriculture to climate change. Climatic Change 21, 385–405. Southworth, J., Randolph, J.C., Habeck, M., Doering, O.C., Pfeifer, R.A., Rao, D.G., and Johnston, J.J. 2000. Consequences of future climate change and changing climate variability on maize yields in the midwestern United States. Agriculture, Ecosystems, and Environment 82, 139–158. Taylor, J.G., Downton, M.W., and Stewart, T.R. 1988. Perceptions of drought in the Ogallala Aquifer region. Environment and Behavior 20, 150–175.

Stakeholder Adaptation to Climatic Change in Kansas

329

Wendland, W.M. 1993. Kansas climate with global warming: Agricultural and other economic impacts. Transactions of the Kansas Academy of Science 96, 161–166.

CHAPTER THIRTY-FOUR DEVELOPING STRATEGIES TO CONVEY CLIMATE SCIENCE TO KANSAS STAKEHOLDERS: EVOLUTION AND APPROACH JOHN A. HARRINGTON JR. (2012)

1. Introduction Challenges exist in sharing the knowledge of climate science in a political and social environment that reinforces a message of denial. While the scientific clarity of the anthropogenic climate change message continues to be strengthened, other confusing and “denialist” messages have made many central U.S. citizens uncertain about the science of human-induced climate change. According to the Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report (2007), anthropogenic influences are driving a change in the climate of the planet. Helping civil society better understand the relevant scientific issues, including the nature of anthropogenic climate change and its implications, is confronted with many possible pathways and potential roadblocks (Boykoff, 2007). As with many educational efforts, those who might be helped through attainment of additional information are at different stages in acquiring foundational knowledge and in their ability to understand the relevance of the subject matter (Fink, 2003; Dunlap and McCright, 2008). Leiserowitz et al. (2009) used a survey questionnaire to identify six different audiences, or “Six Americas,” with beliefs related to climate change ranging from alarmed to dismissive. In Kansas and in many of the conservative ‘red states’ in the central part of the U.S., it has been suggested that many citizens respond to certain issues by making decisions that are counter-intuitive (Frank, 2004). Given this awareness, a number of

Strategies to Convey Climate Science to Kansas Stakeholders

331

different reference frames and/or educational approaches may be necessary to assist a variety of possible learners in making a connection with the anthropogenic climate change subject matter (Moser and Dilling, 2004; Nisbett, 2009). Over the course of the last five years, I have had the opportunity and challenge to present information about the nature of climatic change at a number of venues in Kansas (Table 34-1). This time window includes the release of the IPCC Fourth Assessment Report in 2007, the awarding of the Nobel Prize to the IPCC and Al Gore, the lead up to the Copenhagen UNFCCC Conference of the Parties in 2009, and the controversial period known as Climategate related to the hacking of email from several prominent climate scientists. As the scientific information on climate change and its implications has advanced and my experience in connecting with varying audiences has increased, the education approach for the messages I have tried to convey has advanced. This paper provides an overview of major considerations in the evolving presentation and the rationale for some individual components in the overall science message. Multiple presentations and follow-up conversations with Kansas stakeholders have helped identify knowledge frameworks that facilitate scientific communication about anthropogenic climate change. Important qualitative findings support communicating basic physics along with a bit of humor and a design that recognizes a diverse and/or non-scholarly audience.

2. Initial Thinking: Blind Them with Science A common mindset related to science communication is to share a lot of the existing scientific knowledge. Building on what is commonly referred to in the education community as the ‘deficit model,’ the presenter spends much of the available time sharing relevant scientific information on the subject, perhaps highlighting recent findings. According to this ‘loading dock’ model, the scientist selects the most relevant content and delivers it, with the hope that some in the audience end up ‘buying’ some or most of the ideas presented. This approach typically deals more with refining, reminding, and/or reinforcing existing knowledge rather than filling a complete void.

332

Chapter Thirty-Four

Table 34-1: Examples of Venues for Presentations on Anthropogenic and Natural Climate Change Selected to Document the Diversity of Audiences. (More recent events are listed on top) Name of the Group K-State Extension Horticulture Group Haskell Environmental Research Site Flint Hills Discovery Center Planet Under Pressure Students for Environmental Action Kansas Water Issues Forum, Wichita and Hays AAG Annual Meetings Great Plains Rocky Mountain, AAG Prairie Village Community Forum University of Kansas IGERT program Manhattan Science Café K-State Environmental Ethics class K-State Anima Science class Kansas Geography Conference

Type of Group Annual meeting of extension educators University summer class Public forum Professional conference K-State student group State agency – public forum Professional organization Professional organization Local conservation oriented group University class Local science oriented group University class University class Professional organization

In connecting with Kansas citizens, initial thoughts included the idea that I needed to indicate that anthropogenic climate change was a component of a larger issue related to human-driven planetary change and the Anthropocene and that it would be good to include climate change as a component of something larger. Given the sobering nature of the challenges related to ongoing and accelerating global change, humor is inserted to at times lighten the communication. A goal in preparing content for this science-download approach is to find the ‘Goldilocks’ message, the one that has not too much, nor too little, but just the right amount of relevant science content. In searching for a ‘Goldilocks’ message, decisions were made to omit a considerable amount of scientific content and related jargon. Some have used the phrase, ‘less is more,’ to illustrate this point. One topic that is not presented is the global warming potential of various greenhouse gases. Global warming potential is a complex topic and attempts to compare the relative importance of molecules of different greenhouse gases. While the

Strategies to Convey Climate Science to Kansas Stakeholders

333

point is made that certain gases play a key role in how the greenhouse effect works, a comparative assessment of the warming potential of a molecule of CO2 with a molecule of N2O is beyond what is needed for helping introduce the idea that greenhouse gases exist, are increasing, and play an important role in anthropogenic climate change. Another example of a topic that is left out is climate sensitivity, or the amount of change in global mean temperature related to a change in radiative forcing. Part of the reason for not addressing the concept is the frequent use of the term climate sensitivity to refer to a specific change in the system (i.e., the amount of warming related to a doubling in carbon dioxide concentration). It is more important to help people to understand the basic physics of how the global greenhouse effect works, rather than have listeners/learners memorize a number associated with a specific amount of forcing to the system. In trying to connect global climate change with the bigger issue of overall global change, a political cartoon is used to introduce this section. The cartoon shows the Earth in a physician’s office with the doctor suggesting the need for leg amputation due to the planet having a slight fever. It is then pointed out that physicians make numerous observations and perhaps take several measurements to help them understand the complex system(s) they are trying to diagnose. Auto mechanics provide another example of individuals that make numerous observations to diagnose needed adjustments to a complex system. With changes in planetary land cover, losses in biodiversity, changes in atmospheric chemistry, and the rapid use of stored resources, there are a number of planetary scale indicators that the Earth system is under stress. Rockstrom et al. (2009) used this multiple-indicators approach in identifying natural planetary boundaries. A goal for this part of the presentation is to frame the message that anthropogenic climate change is not the only ‘inconvenient truth’ related to human transformations of the planet (Foley, 2009). This scientific framing goes well with the idea that the scientific processes produce relative (rather than absolute) knowledge (Abler et al., 1971). Given the context of natural and anthropogenic global change, a number of pairs of remotely sensed images and examples of repeat photography are presented. These images help document changes in the Earth system with the goal to help people connect the images with ongoing Earth system and climate change. Examples include loss of mass and upslope retreat of alpine glaciers in U.S. national parks, declines in water

334

Chapter Thirty-Four

levels and areal extent of water bodies in arid regions, and recent images of September Arctic sea ice extent.

3. Climate as a Multi-Dimensional System In listening to students in classes and Kansas citizens, two items became clear. First, many did not understand differences among the multiple ways of knowing (Abler et al., 1971; Lee, 1999). Second, most in these audiences did not understand differences between weather and climate. Modifying the message in order to address the nature of science involved preparing content that identified examples of different ways of knowing (e.g., science, theology, common sense, aesthetics, and political ideology) (Abler et al., 1971; Lee, 1999). A relevant cartoon provides another medical message, with a doctor asking a patient if he wants to refuse treatment because the cure involved the use of stem cells. Other humorous examples include cartoons mixing science and religious ways of knowing in modern society. Sometimes it is pointed out that praying for rain has not been shown to have a positive or negative impact on meteorological processes. In addition, the difference among groups in the use of the word ‘theory’ is identified. For many, the word is synonymous with the word guess, or perhaps hypothesis. In the social sciences a theory is often a working model of how either individuals or society functions (such as rational economic decisions), whereas in the natural sciences a theory is a complex system of explanation based on building together a number of scientific laws. Examples of natural scientific theories include atomic theory, plate tectonics, the big bang theory, and the theory of evolution. After making that point, a slide of planetary temperature observations over the last 13 decades is presented. Those data indicate that global warming is an emergent property based on analysis of existing scientific data, and not a theory (however theory is defined). Following some presentations, the conversation addresses the idea that the physics behind the greenhouse effect has characteristics of a scientific theory. Over the last four decades, the importance of using a systems perspective to assist in understanding the world around us has advanced (Steffen et al., 2004). From the perspective of understanding the climate system, the June 1991 eruption of Mt. Pinatubo is an example of a radiative forcing perturbation. The volcanic eruption provided an internal impulse that cooled the Earth system for about three years. Climate

Strategies to Convey Climate Science to Kansas Stakeholders

335

scientists have indicated that correctly modeling the impact of the Pinatubo eruption has helped the modeling community gain confidence in their increasingly sophisticated climate models (Hansen, 2009). To help others understand the difference between weather and climate, a diagram was developed (Figure 34-1) that attempts to illustrate the climate at a given location as a cloud of weather events in threedimensional space. The axes are energy (rather than temperature, since energy fluxes are a big component of a system science perspective), precipitation, and wind. Individual dots within the cloud are designed to illustrate a specific weather event, such as a hot, dry day or a cold, windy, and snowy one. Part of the goal in designing the diagram was to illustrate that the climate, which plays out as a number of weather events, is conceptually much more than 30-year statistical averages of temperature and precipitation data. Another aspect of diagram design was to illustrate the importance of events at the margins of the cloud (the extremes in weather events at a given location).

Figure 34-1: Graphic Designed to Illustrate the Multi-Dimensional Character of the Climate System

336

Chapter Thirty-Four

Modifying the shape and position of the cloud of weather events can be used to illustrate the concept of climatic change (Figure 34-2). In the case illustrated (Figure 34-2), some of the coldest and driest events from the past are no longer a part of the changed collection of weather at the location. New extreme events, such as the one that is high up on the energy axis, would illustrate a change in the intensity of a record high temperature event for the location, whereas an event far to the right on the precipitation axis could illustrate a new record 24-hour rainfall total. The arrow (change vector) suggests that the centroid of weather events at this location has shifted to higher energy, moister, and perhaps less windy conditions. A metric of warming associated with this climate change would be the magnitude of the energy component of the change vector.

Figure 34-2: Modification of Figure 34-1 Designed to Illustrate the MultiDimensional Character of Climate Change

Strategies to Convey Climate Science to Kansas Stakeholders

337

4. Basic Physics and the Global Greenhouse Since the fluxes of energy in the Earth-atmosphere system are a key driver of the climate system, I share the Kiehl and Trenberth (1997) energy budget diagram with audiences (Figure 34-3). Discussion of the energy budget diagram begins with recognition that the values presented are in physical units (Watts per meter squared) rather than as a percentage of the incoming solar energy stream; this diagram was a major step forward provided by Kiehl and Trenberth (1997) for presenting relevant scientific information. Ideas emphasized with regard to shortwave energy flow include: the importance of more energy being absorbed at the surface rather than in the atmosphere, the suggestion that human actions can modify the flows by cutting forests and changing surface reflectance, and the idea that the growth of continental ice sheets would also change both the amount of energy reflected by the surface and the total amount of reflected solar energy lost from the top of the system. Discussion of longwave energy fluxes highlights the recycling of energy within the system by pointing out that the 324 W m-2 of longwave counter or back radiation helping to warm the surface is more than (almost double)

Figure 34-3: An Energy Budget Diagram that Highlights the Importance of Clouds and Greenhouse Gases in the Recycling of Radiation (Source: Kiehl and Trenberth (1997); modified by the author)

338

Chapter Thirty-Four

the 168 W m-2 of shortwave energy absorption. The roles of clouds and greenhouse gases, including the role of water vapor (the most important greenhouse gas), are also identified. Audience reaction indicates that discussion of basic physical science and their personal experiences is effective. A relevant example is the differences in the effectiveness of greenhouse energy recycling during overnight periods with and without cloud cover (with much greater energy loss without cloud cover). Additional connections with individual experience draw on the more rapid cooling that occurs in the evening when the air mass present in a location is much drier. In addition, the basic physics related to the temperature of an object and the wavelengths of the emitted energy are highlighted. Using Wien’s displacement law, the peak wavelength for longwave emissions for Topeka, Kansas, in January and July, was calculated. During July, the warmer temperatures result in a peak wavelength of emissions of 9.7 μm, whereas the cooler winter temperatures produce a longer peak wavelength of 10.7 μm. Using a graph showing energy transmission through the atmosphere by wavelength that also identifies the gases responsible for stopping that transmission for specific wavelength bands, I point out that cooler places on the planet and the period of cooler temperatures should see a greater impact with the increase in CO2 in the atmosphere. Warmer areas on the planet and warmer seasons will be more affected if water vapor concentrations increase. As the basic science would suggest, areas that have warmed the most during the first years of the 21st century are in high latitudes in the Northern Hemisphere (Figure 34-4). Illustration of the fact that more warming has occurred in places that are predictable based on physics is followed by explaining that, in an analysis of warming by specific month, more of the recent warming has occurred in cold season months. Both the global spatial pattern and the seasonal concentration of warming are clearly linked with the physics related to the role of CO2 as a gas that influences energy transmission through the atmosphere.

Strategies to Convey Climate Science to Kansas Stakeholders

339

5. Wrapping Things Up

Figure 34-4: Temperature Anomalies (°C) by Latitude for 2000–2011 Compared with the 1950–1980 Base Period (Source: Created by author at http://data.giss.nasa.gov/gistemp/maps/)

Since a good deal of misinformation exists that can confuse the issue of anthropogenic climate change, it is important to address several of those misconceptions. Use of the multiple skepticism categories identified by Rahmstorf (2004) enables a science-based presentation on attribution, trends in the data, and impacts. A number of ‘fingerprinting’ studies have identified that human-induced releases of greenhouse gases are the primary driver of changes during the last several decades. This conclusion was reported in both the 3rd and 4th IPCC reports (2001, 2007). Several graphics are used to show that the recent upward temperature trend is not related to the approximately 11-year sunspot cycle, and that modeling studies strongly suggest that the rise in planetary temperature since 1980 can be attributed to the increase in greenhouse gases. Trend skepticism is the suggestion that the warming has come to an end; the unusually warm El Niño of 1998 and temperatures since that time have been used by some to suggest a leveling off in the long term trend. In interpreting long term trends in a time series, it is important not to ‘cherry pick’ an outlier in the data series to look at shorter term trends (Figure 345). If looking at only the years 1998 through 2008 (which starts with an El Niño and ends with the 2008 La Niña), observers might deceive themselves

340

Chapter Thirty-Four

into thinking that temperatures are no longer trending upward. However, if they examine the longer term record and compare years with somewhat similar conditions in the tropical east Pacific (i.e., three El Niño years), a different interpretation of the data is warranted (Figure 34-5). Impact skepticism involves the ideas associated with negligible or beneficial impacts from anthropogenic climate change. Frequently questions from audience members involve the general idea that more CO2 will help plants grow better. While it is correct that additional CO2 generally increases photosynthesis, not all plants respond in the same way. Poison ivy and kudzu are plants which respond favorably to increased CO2 that audience members seem to not want to have enhanced growth. Delay skepticism, a fourth type not originally identified by Rahmstorf (2004), involves the idea that humanity has considerable time to make adjustments to current practices. In discussing delay skepticism, I frequently point out that many of the decisions we make today have builtin lifetimes of decades (e.g., the decision to build a new power plant is based on a certain number of years for a return on investment costs). In addition, the costs for making adjustments in lifestyles are forecast to

Figure 34-5: Global Temperatures (°C from NASA GISS) for 1981–2010 with Three Major El Niño Events Indicated

Strategies to Convey Climate Science to Kansas Stakeholders

341

increase as the planet gets warmer and the global population continues to increase (IPCC, 2007). Prior to ending the formal presentation and opening the session up for questions and discussion, ideas of ethical considerations, planetary stewardship, and the burdens that humanity today may be leaving for future generations are presented. Concepts presented in this last part of the overall presentation suggest that when we first started the atmospheric chemistry experiment with humanity increasing the concentration of greenhouse gases, there were not clear scientific understandings of the implications. But, with four IPCC reports and decades of scientific study, we now know that we are violating the precautionary principle that ‘we should first do no harm.’ Quotes from Buckminster Fuller regarding system change and building a better model and from Margaret Meade related to the importance of a small number of people being responsible for big changes are presented to help the audience recognize that change is possible.

6. Summary During the past few years, I’ve had many opportunities to communicate with students and members of civil society about natural and anthropogenic climate change. In thinking about the questions asked and the responses received in various aspects of the message, adjustments have been made to hopefully improve communication effectiveness. Major ideas addressed in this process include: • the deficit model • searching for the Goldilocks message • deciding which topics to exclude, such as global warming potential • climate as a complex multi-dimensional system • contrasting science with other ways of knowing • that warming is an emergent property based on analysis of scientific observations • that warming and other changes are happening locally • presenting the basic physics associated with how the greenhouse effect works • using Wien’s displacement law to illustrate reasons that colder places should be warming faster with increased CO2 • addressing the multiple forms of skepticism • ethical considerations • the idea that human-induced system change is possible

342

Chapter Thirty-Four

Communicating scientific content to a diverse population is a challenge. That difficulty is more acute when political ideology muddies understandings. Through an iterative process of listening to audience responses and adjusting the messages being communicated, it is believed that greater effectiveness and audience appreciation is being achieved.

Acknowledgments Two NSF grants have helped fund aspects of the ideas presented in this paper: Award No. EPS-0903806, PHASE VI: Climate Change and Energy: Basic Science, Impacts, and Mitigation and Award No. CCEP1043393, Central Great Plains Climate Change Education Partnership.

References Abler, R., Adams, J., and Gould, P. 1971. Spatial Organization: The Geographer’s View of the World. Englewood Cliffs, NJ: Prentice-Hall, Inc. Boykoff, M. 2007. From convergence to contention: United States mass media representations of anthropogenic climate science. Transactions of the Institute of British Geographers 32, 477–489. Dunlap, R.E. and McCright, A.M. 2008. A widening gap: Republican and Democratic views on climate change. Environment 50, 26–35. Fink, D. 2003. Designing Significant Learning Experiences. San Francisco, CA: Jossey-Bass. Foley, J. 2009. The Other Inconvenient Truth: The Crisis in Global Land Use, Yale Environment 360. http://e360.yale.edu/feature/the_othrer_inconvenient_truth_the_ crisis_in_global_land_use/2196/ Frank, T. 2004. What’s the Matter with Kansas? How Conservatives Won the Heart of America. New York, NY: Metropolitan Books. Hansen, J. 2009. Storms of My Grandchildren: The Truth about the Coming Climate Catastrophe and Our Last Chance to Save Humanity. New York, NY: Bloomsbury USA. IPCC 2001. Climate Change 2001: The Scientific Basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Houghton, J.T.; Ding, Y.; Griggs, D.J.; Noguer, M.; van der Linden, P.J.; Dai, X.; Maskell, K.; and Johnson, C.A., (eds.), Cambridge, UK and New York, NY: Cambridge University Press.

Strategies to Convey Climate Science to Kansas Stakeholders

343

—. 2007. Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Solomon, S., Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt, K.B., Tignor, M., and Miller, H.L. (eds.), Cambridge, United Kingdom and New York, NY, USA: Cambridge University Press. Kiehl, J. and Trenberth, K. 1997. Earth’s annual global mean energy budget. Bulletin of the American Meteorological Society 78, 197–208. Lee, J. 1999. The Scientific Endeavor: A Primer on Scientific Principles and Practice. San Francisco, CA: Benjamin Cummings. Leiserowitz, A., Maibach, E., and Roser-Renouf, C. 2009. Climate Change in the American Mind: Americans’ Climate Change Beliefs, Attitudes, Policy Preferences, and Actions. New Haven, CT: Yale University. Moser, S. and Dilling, L. 2004. Making climate hot: Communicating the urgency and challenge of global climate change. Environment 46, 32– 46. Nisbett, M. 2009. Communicating climate change: Why frames matter for public engagement. Environment 51, 12–23. Rahmstorf, S. 2004. The climate skeptics. In: Weather Catastrophes and Climate Change - Is There Still Hope for Us? Munich Re, (ed.). PgVerlag: Munich, 76–83. Rockstrom, J., Steffen, W., Noone, K., Persson, A., Chapin, F.S., Lambin, E.F., Lenton, T.M., Scheffer, M., Folke, C., Schellnhuber, H.J., Nykvist, B., de Wit, C.A., Hughes, T., van der Leeuw, S., Rodhe, H., Sorlin, S., Snyder, P.K., Costanza, R., Svedin, U., Falkenmark, M., Karlberg, L., Corell, R.W., Fabry, V.J., Hansen, J., Walker, B., Liverman, D., Richardson, K., Crutzen, P., and Foley, J.A. 2009. A safe operating space for humanity. Nature 461(7263), 472–475. Steffen, W., Sanderson, A., Tyson, P., Jager, J., Matson, P., Moore III, B., Oldfield, F., Richardson, K., Schnellnhuber, H., Turner II, B., and Wasson, R. 2004. Global Change and the Earth System: A Planet under Pressure. New York, NY: Springer-Verlag.

CHAPTER THIRTY-FIVE THE PROSPECT FOR APPLIED CLIMATOLOGY JOHN E. OLIVER (1996)

1. Introduction In a recent publication outlining the nature and rationale of the Applied Geography Conferences, Frazier et al. (1995) state that, “applied geographic research seeks a solution to a particular spatial or environmental problem by extending the scientific method to evaluation and implementation stages for a client or an audience who wishes its resolution.” Given such a clear and focused definition, it is interesting to consider how the goals of given subdivisions of geography fit within this framework. Within the field of climatology, an organized subfield identified as Applied Climatology has developed (Miller, 1977). The purpose of this paper is to: (1) examine what is encompassed by the title Applied Climatology, and (2) relate applied climatic research to the definition of applied geography as used by Frazier et al. (1995). In attempting to answer these questions, there is the inevitable danger of getting bogged down in definitions and boundaries. This hazard can be effectively diminished by showing how the very nature of climatology has such broad applications, and that the discussion must go beyond mere semantics.

The Prospect for Applied Climatology

345

2. The Background History plays a critical role in the present-day view of applied climatology. Climatology began when fishermen and farmers began to realize that daily and seasonal cycles of atmospheric conditions could be identified. Applied climatology originated shortly thereafter when those same people applied their knowledge of the atmosphere to such things as planting activities and sailing schedules. Many scholars cited in the history of climatology actually were more concerned with the effects and impacts of climate than the actual mechanisms of climate. Just two examples illustrate this point. In the much cited On Airs Waters and Places, Hippocrates (ca 420 B.C.) contrasted easy-going Asiatics with penurious Europeans and found the difference climatically induced; the great Arab historian-geographer Ibn Khaldun (1332–1406) identified world areas and noted that areas of this middle (temperate) zone, “provided the climate in which people were neither too stolid nor excessively passionate.” It is interesting to note that the original purpose of scientific weather observation in the United States, urged by Thomas Jefferson and Surgeon General Joseph Lovell, was for practical applications of land use and health. In fact, one of the earliest climatologies of the United States was written by Samuel Forry (1842). An M.D., Forry called his text, The Climate of the United States and Its Endemic Influences, and much of its content relates the health of military personnel to climatic conditions. Texts such as this placed both climatology and its applied aspects in good standing in the academic community. However, this was to change. Perhaps the most influential factor in the growth of applied climatic research was the rise and fall of climatic determinism. The extreme stance of determinists of the mid-20th century led to a backlash effect in which climate and environment, when considered in any aspect of human activities, became an area of non-research. Voluntarism, at the other end of the spectrum of thought, became viable. The adverse reaction to the overstated tenets of determinism was clearly a deterrent to research in many aspects of applied climatology. Another factor compounding its advancement was the rapid development of meteorology in the 1930s and 1940s. The mathematical/applied physics of the new meteorology caused geographical and applied climatology to become somewhat slighted disciplines of the atmospheric sciences. Fortunately, there were a number

346

Chapter Thirty-Five

of outstanding scholars, such as C.W. Thornthwaite and Helmet Landsberg, whose research maintained the applied climatologic approach. It was in fact Landsberg who provided an early meaningful definition of applied climatology.

3. Evolving Definitions In the 1951 Compendium of Meteorology, Landsberg and Jacobs (1951) write, “If we consider climate as the statistical collective of individual conditions of weather, we can define applied climatology as the scientific analysis of this collective in the light of a useful application for an operational purpose.” This definition relates applied climate to technology which, for convenience, may be described under the head of technoclimatology. Technoclimatology is the utilization of climatic data as input into a given technical problem and, not surprisingly, the academic literature on the topic is quite limited. Much of the work completed is utilized on a local basis for solving a particular problem. Often, the applied data are used in legal situations in suits ranging from flooding problems to roof collapse resulting from snowfall. The need in such areas is considerable and has given rise to specialists who label themselves forensic climatologists. Depending upon their nature, air pollution studies fall within the rubric of technoclimatology. But this brings us to a problem in determining the nature of applied geography, one of increasing specialization in the academic world. Some areas of study that might have been considered technoclimatology have become so specialized that they are now identified fields of endeavor. Air pollution meteorology/climatology is now as much in the fields of chemistry and computer modeling as it is climatology. Many analogous cases exist with urban climatology providing another example. Oliver (1973) suggested a broader scope for the study of applied climatology stating, “It might be considered as the application of principles and concepts of climatology to spheres of endeavor that concern man and his past and present environments.” This definition suggests the inclusion of relationships between climate, the physical geographic environment, the human-economic environment, and aspects of technoclimatology. Oliver also suggests that the applied climatologist should generate research in the specialty fields of geomorphology, soils, and vegetation studies in an organized study of the biophysical realm.

The Prospect for Applied Climatology

347

The broad base of Oliver’s definition contrasts with that of Landsberg and Jacobs. Perhaps one is too broad and the other too limited in seeking out important research areas. An excellent alternate was given by Smith (1987) who suggested that applied climatology concerns “the use of both archived and real-time climatic information to solve a variety of social, economic, and environmental problems.” This definition corresponds closely to one of the basic categories encompassed by the National Climate Program Act (U.S. Congress, 1978), the goal of which is to “assist the nation and the world to understand and respond to natural and man induced climate processes and their implications.” Interestingly, it also corresponds to the already cited definition of the nature of applied geography provided by Frazier et al. (1995). It is Smith’s definition of applied climatology that provides a framework for identifying exciting research problems.

4. Areas of Research The discussion on the content of applied climatology is not an exercise in definitions, but rather a method to identify research themes and topics. Given what has been stated, I suggest that there are three classes of applied climatology that afford enormous possibilities for research.

4.1 Technoclimatology As already noted, this is often the realm of research completed as an identified funded report for a specific problem. However, in their outline of applied climatology, Landsberg and Jacobs (1951) provided a listing of appropriate topics. This is reproduced in Table 35-1. While dated, there is much here that has not been the objective of applied geographic research for a number of years but which still offers opportunities. Of significance in this contribution is the identification of the class of problem. Here the authors provide a guide to the nature of the research necessary to solve the applied problem.

4.2 The Socioeconomic Focus In recent years, researchers have been increasingly concerned with the impacts of climatic extremes upon the socioeconomic environment. What may be considered a model for this type of research is the impacts of severe winters as interpreted by Changnon (1979). Other avenues of research include such studies as climate assessment in relation to tourism

348

Chapter Thirty-Five

Table 35-1: List of Topics (After Jacobs and Landsberg 1951)

Field of Application Advertising and marketing Aerial photography Airfield construction Airline operation Air pollution control Architecture, housing construction City planning Clothing Crop protection and planning District heating Forest fires Gas and oil dispatching Health resorts Heating and cooling plants Highway construction Hydroelectric power Insurance Land utilization Lubrication Military operations Open air theater operation Power dispatching Power lines Shipping and transportation Storage Windpower

Problem Class* C C A C B,C A,B,C B A C C C C D A B B A D A,C C B,C C A C A,B A,B

* Class of problem comprises A. Design and specification B. Location and operation C. Planning of operation D. Relation of climate to biological problem

(McGregor, 1985), relationships of seasonality to crime (Harries et al., 1985), consideration of various bioclimatological indices (Warrick and Riebsame, 1983), and human responses to past and future climates.

The Prospect for Applied Climatology

349

The great surge of perception studies in geography of the 1970s did not impinge greatly upon work in applied climatology. While there are exceptions, they are surprisingly few in relation to the potential contributions of research in this area. Another fruitful line of research concerns studies of climatic impacts upon social conditions, which mostly deal with the effects of climatic variations in both developed and underdeveloped areas (Warrick and Riebsame, 1983).

4.3 Integrated Studies One of the most marked features of published research in recent years is the increasing interest in what may be termed integrated studies. In these, interconnections are sought between identified events in terms of their potential causation and eventual impact. Such research often requires the construction of models or scenarios. Examples include potential impacts resulting from El Niño-Southern Oscillation events and reconstruction and assessment of historic eras. Climatic change and variation provide the basis for a number of integrated assessments. Clearly, the scenario, “What would happen if climate were to change. . .” presents many potential avenues for research, as seen in the wide variety of papers. It is worth noting, however, that integrated studies do not need to involve components of major world events. Applied geographers can make and have made important contributions to multifaceted research problems. Acid rain distribution and perceptions of ozone depletion provide apt examples. Resource development is also a fertile area for interdisciplinary work. As a case in point, climatic aspects of water resources may consider almost all atmospheric components of the hydrologic cycle including snowmelt, changing precipitation patterns, extreme events, and hydroclimatic modeling. Water balance studies continue to provide significant information as shown by Mather (1978). As part of the thrust toward a better understanding of drought, studies on desertification and land degradation also appear frequently in the literature (Vanypersele and Verstraete, 1986).

5. Conclusion By considering applied climatology as a way to use climatic information to solve a variety of social, economic, and environmental problems, an optimum example of the goals envisaged for the Applied

350

Chapter Thirty-Five

Geography Conferences is provided. Through carefully examining the entire gamut of climatic studies, those most applicable to these goals may be identified. It appears that research in technoclimatology, the socioeconomic/climate interface, and in multifaceted integrated studies provide guides to potential topics. The charge of the National Climate Program Act (U.S. Congress, 1978) has resulted in many symposia, meetings and task forces, most of these are held under titles other than geography. Like the integrated approach to climate impact, applied climatic studies require an umbrella organization to bring together researchers with seemingly very different interests. This trend certainly will continue and it is anticipated that participants of the Applied Geography Conference will be active in such research. There is much to be completed.

References Changnon, S.A. 1979. How a severe winter impacts on individuals. Bulletin of the American Meteorological Society 60, 110–114. Forry, S. 1842. The Climate of the United States and its Endemic Influences. Boston, MA: Little Brown. (Reprinted by AMS Press, New York) Frazier, J.W., Epstein, B.J., and Schoolmaster, F.A. 1995. Contributions to applied geography: The 1978–1994 Applied Geography Conferences. In: Schoolmaster, F.A. (ed.), Papers and Proceedings of Applied Geography Conferences 18, 1–13. Arlington, VA: George Mason University. Harries, K.D., Stadler, S., and Zdorkowski, T. 1985. Seasonality and assault: Explorations in interneighborhood variations, Dallas, 1980. Annals of the Association of American Geographers 74, 590–604. Landsberg, H. and Jacobs, W.C. 1951. Applied climatology. In: Malone, T.F. (ed.), Compendium of Meteorology, Boston, MA: American Meteorological Society, 976–992. Mather, J.R. 1978. The Climatic Water Balance in Environmental Analysis. Lexington, MA: D.C. Heath. McGregor, K.M. 1985. The tourism climatic index: A method for evaluating world climates for tourism. Canadian Geographer 29, 220– 233. Miller, D.H. 1977. Water at the Surface of the Earth: An Introduction to Ecosystem Hydrodynamics. New York, NY: Academic Press.

The Prospect for Applied Climatology

351

Oliver, J.E. 1973. Climate and Man’s Environment: An Introduction to Applied Climatology. New York, NY: John Wiley. Smith, K. 1987. Applied climatology. In: Oliver, J.E. and Fairbridge, R.W. (eds.), The Encyclopedia of Climatology, New York, NY: Van Nostrand Reinhold, 64–68. United States Congress. 1978. The National Climate Program Act. Conference report to accompany HR6669, Washington D.C.: U.S. Government Printing Office. Vanypersele, J.P. and Verstraete, M.M. 1986. Climate and desertification – editorial. Climatic Change 9, 1-4. Warrick, R.A. and Riebsame, W.E. 1983. Societal response to CO2 induced climatic change: Opportunities for research. In: Chan, R.S., Boulding, E.M., and Schneider, S.H. (eds.), Social Science Research and Climatic Change: An Interdisciplinary Appraisal. Dordrecht, Holland: Reidel.