WHAT THE URBAN HEAT ISLAND EFFECT MEANS FOR EAST ASIA’S CITIES WHAT THE URBAN HEAT ISLAND EFFECT MEANS FOR EAST ASIA’S CITIES UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities II © 2023 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: +1-202-473-1000 Internet: www.worldbank.org SOME RIGHTS RESERVED This work is a product of the staff of The World Bank and the Global Facility for Disaster Reduction and Recovery (GFDRR). The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be construed or considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. RIGHTS AND PERMISSIONS. This work is subject to copyright. THIRD-PARTY CONTENT. The World Bank does not necessarily Because The World Bank encourages dissemination of its own each component of the content contained within the work. knowledge, this work may be reproduced, in whole or in part, for The World Bank therefore does not warrant that the use of any noncommercial purposes as long as full attribution to this work third-party-owned individual component or part contained in is given. the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringements rests solely ATTRIBUTION. Please cite this work as follows: Roberts, with you. If you wish to reuse a component of the work, it is your Mark, Chandan Deuskar, Nicholas Jones, and Jane Park. 2023. responsibility to determine whether permission is needed for Unlivable: What the Urban Heat Island Effect Means for East that reuse and to obtain permission from the copyright owner. Asia’s Cities. Washington DC. The World Bank. Examples of components can include, but are not limited to, tables, figures, or images. TRANSLATIONS. If you create a translation of this work, please add the following disclaimer along with the attribution: This All queries on rights and licenses, including subsidiary rights, translation was not created by The World Bank and should not should be addressed to World Bank Publications, The World Bank be considered an official World Bank translation. The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; e-mail: shall not be liable for any content or error in this translation. pubrights@worldbank.org ADAPTATION. If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Cover design, layout and illustrations: Estudio Relativo UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities III TABLE OF CONTENTS • Foreword......................................................................................................................................... IX • Acknowledgments.........................................................................................................................X • Main Messages............................................................................................................................ XI How Strong Is the Urban Heat Island Effect?................................................................................................XI Why Worry?.................................................................................................................................................................. XII What Can Be Done?................................................................................................................................................ XIII • Abbreviations..............................................................................................................................XIV • Temperature Conversion .......................................................................................................XIV O OVERVIEW 1 • Introduction......................................................................................................................................1 • How Strong Is the Urban Heat Island (UHI) Effect in East Asia?...............................2 UHI Effects Already Strong, with Large Variations across Cities........................................................... 2 Particular Concerns about UHI Effects in Tropical East Asian Cities...................................................5 UHI Effects Often Stronger in Poorer Neighborhoods.................................................................................5 More Severe Heat Waves Likely as UHI Effect Interacts with Climate Change...............................9 • Why Worry about the Urban Heat Island Effect?............................................................11 Threat of Extreme Heat to Cities’ Competitiveness ................................................................................... 11 Impacts of Extreme Heat on Urban Economic Activity ............................................................................ 13 Harmful Impacts of Extreme Heat on Health and Livability................................................................... 14 How Extreme Heat Limits Cities’ Inclusiveness........................................................................................... 15 Air Conditioning: No Panacea............................................................................................................................... 15 • What Can Be Done?...................................................................................................................16 Strategic Option 1: Advance Urban Greening through Strategic Planning....................................... 18 Strategic Option 2: Cool City Spaces through Wind, Shade, and Urban Design........................... 19 Strategic Option 3: Engage Building Owners to Tackle Indoor Heat.................................................. 20 Strategic Option 4: Save Lives through Heat Wave Early Warnings................................................... 21 Strategic Option 5: Protect Heat-Exposed Workers.................................................................................. 22 Strategic Option 6: Mainstream Heat Risk Reduction into City Institutions and Strategies......................................................................................................................... 23 • Conclusion.................................................................................................................................... 25 References................................................................................................................................................................... 25 Notes............................................................................................................................................................................... 28 UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities IV C H A PTE R 1 THE URBAN HEAT ISLAND EFFECT AND THE EVOLUTION OF EXTREME HEAT ACROSS EAST ASIAN CITIES 30 Key Messages.............................................................................................................................................................. 31 • Introduction.................................................................................................................................. 32 • Defining Extreme Urban Heat................................................................................................35 • What Drives Extreme Urban Heat?......................................................................................36 Local Warming Driven by the Urban Heat Island Effect.......................................................................... 36 Global Warming Driven by Climate Change................................................................................................... 36 Interactions between UHI Effects and Climate Change...........................................................................37 • How Is an Extreme Urban Heat Event Measured?......................................................... 37 Local Driver: Intensity of the UHI Effect......................................................................................................... 38 Global Driver: Relative Temperature Anomalies.......................................................................................... 39 • UHI Effects Make East Asian Cities, on Average, 1.6–2.0°C Warmer Than Rural Surroundings.....................................................................41 Strong Average UHI Intensities............................................................................................................................ 41 Diverging UHI Intensities within a City and during the Day................................................................... 44 • While Temperatures Increased Everywhere, Rainfalls Amplified in the Tropics........................................................................................44 Rising Absolute Temperatures............................................................................................................................ 44 Cooler Cities Warming Faster and Hotter Cities Heating Up Further................................................ 45 Increasing Rainfall Totals and Variability in the Tropics...........................................................................47 • Extreme Heat Events Became Increasingly Severe, and More So in the Tropics.....................................................................................................50 Increasing Overall Frequency and Intensity of Extreme Heat Events............................................... 50 Wide Variations across Cities in Evolving Patterns of Extreme Heat Events ............................... 52 Sharper Increases in Extreme Heat Severity in Warmer Cities of Poorer Countries.................. 54 • Urban Greening Mitigates UHI Effects, but Local Conditions Also Matter..........56 • Summary and Conclusion.......................................................................................................60 • Annex 1A Defining Extreme Heat Events............................................................................61 • Annex 1B Determinants of Urban Heat Island Intensity.............................................63 References................................................................................................................................................................... 64 Notes............................................................................................................................................................................... 66 UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities V C H A PTE R 2 THE HEAT ISLAND EFFECT WITHIN EAST ASIAN CITIES: WHO IS MOST EXPOSED & WHAT DETERMINES ITS STRENGTH? 69 Key Messages............................................................................................................................................................. 70 • Introduction................................................................................................................................... 71 • When Cities Are Built, Local Climate Processes Change ...........................................72 • Temperatures Differ Sharply between City Neighborhoods...................................... 75 • Differences in Vegetation and Built Form Drive Heat Differentials.........................77 • Poor Residents Endure Hotter Environments................................................................. 79 • Heat at the Micro Level: Built Form and Thermal Comfort..........................................81 • Urban Heat Will Intensify in Coming Decades................................................................84 • Summary and Conclusion....................................................................................................... 87 References....................................................................................................................................................................87 Notes...............................................................................................................................................................................88 C H A PTE R 3 HOW DOES URBAN HEAT AFFECT CITIES AND RESIDENTS? 89 Key Messages............................................................................................................................................................. 90 • Introduction...................................................................................................................................91 • Extreme Heat Undermines Urban Prosperity...................................................................91 Extreme Heat Makes Cities Less Productive................................................................................................. 91 Reduction in Worker Productivity...................................................................................................................... 93 Damage to Urban Infrastructure........................................................................................................................ 94 Reduction in Global GDP........................................................................................................................................ 94 GDP Losses across the Region........................................................................................................................... 96 Extreme Heat Hurts Educational Outcomes ................................................................................................99 • Extreme Heat Makes Cities Less Livable..........................................................................99 Heat Causes Illness and Death...........................................................................................................................99 Extreme Heat May Exacerbate Crime, Violence, and Conflict............................................................. 101 • Extreme Heat Makes Cities Less Inclusive.................................................................... 103 • Extreme Heat Increases Energy Use for Cooling......................................................... 106 • Summary and Conclusion......................................................................................................107 References................................................................................................................................................................. 108 Notes.............................................................................................................................................................................. 110 UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities VI C H A PTE R 4 MITIGATING AND ADAPTING TO EXTREME URBAN HEAT: POLICY PRINCIPLES AND OPTIONS 111 Key Messages.............................................................................................................................................................112 • Introduction................................................................................................................................. 113 • A Framework for Action: Three Principles, Six Action Areas.................................... 113 • Strategic Option 1: Advance Urban Greening through Strategic Planning.................................................................................................................... 116 • Strategic Option 2: Cool City Spaces through Wind, Shade, and Urban Design .................................................................................................................... 119 • Strategic Option 3: Engage Building Owners to Tackle Indoor Heat ...................122 • Strategic Option 4: Save Lives through Heat Wave Early Warnings.................... 125 • Strategic Option 5: Protect Heat-Exposed Workers ................................................... 131 • Strategic Option 6: Mainstream Heat Risk Reduction into City Institutions and Strategies......................................................................................... 134 • Summary and Conclusion......................................................................................................137 • Annex 4A Mainstreaming Heat Resilience into City Institutions and Strategies........................................................................................................................... 138 References................................................................................................................................................................. 140 Notes..............................................................................................................................................................................142 UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities VII ■ FIGURE O.1 Average Intensity and Trends of Urban Heat Island Effects across LIST OF O East Asian Cities, by Buffer Size .........................................................................................................................................3 FIGURES ■ FIGURE O.2 Ambient Air Temperatures in Bandung, Indonesia, 2022...................................................................6 ■ FIGURE O.3 Relationship between Air Temperature and Ground Cover Characteristics, in Bandung, Indonesia, 2022 .................................................................................................................................................8 ■ FIGURE 0.4 Frequency and Intensity of Extreme Heat Events by Climate Zone and Country Income Level, in East Asian Cities, by Decade (1958-69 to 2010-20)......................................10 ■ FIGURE O.5 Channels of Extreme Heat’s Impacts on a City’s Aggregate Real Output............................... 12 ■ FIGURE O.6 Estimated Impacts of Extreme Heat Events on City Nighttime Light Intensity in East Asia, by Subregion, April 2012–December 2020....................................................................... 14 ■ FIGURE O.7 The “Places, People, Institutions” Framework for Urban Heat Resilience................................ 17 ■ FIGURE O.8 The Heat Wave Risk Management Cycle.............................................................................................. 24 1 ■ FIGURE B1.1.1 Distribution of East Asian Cities, by Country Income Classification and Climate Zone..................................................................................................................................................................... 34 ■ FIGURE 1.1 Trends in the average UHI Intensity and Dispersion across 100 East Asian Cities............................................................................................................................................................. 42 ■ FIGURE 1.2 Evolution of Average Annual Temperature in East Asian Cities, 1958–2020.................................................................................................................................................................................45 ■ FIGURE 1.3 Evolution of Mean Annual Temperature in Selected East Asian Cities, 1958–2020.................................................................................................................................................................................46 ■ FIGURE 1.4 Pace of Temperature Increases in East Asian Cities, 1958–2020................................................................................................................................................................................. 47 ■ FIGURE 1.5 Evolution of Total Annual Precipitation in East Asian Cities, 1958–2020.................................................................................................................................................................................48 ■ FIGURE 1.6 Evolution of Total Annual Precipitation in East Asian Cities, by Climate Zone, 1958–2020..............................................................................................................................................49 ■ FIGURE 1.7 Frequency and Intensity of Extreme Heat in East Asian Cities, by Decade (1958–69 to 2010–20)..................................................................................................................................... 51 ■ FIGURE 1.8 Frequency and Intensity of Extreme Heat Events by Climate Zone and Country Income Level, in East Asian Cities, 2010–20 ....................................................................................55 ■ FIGURE 1.9 Relationship between UHI Intensity and Vegetation, Precipitation, and Population Density in East Asian Cities, Circa 2014........................................................................................ 57 ■ FIGURE 1.10 Relative Effect Sizes of Local Characteristics on UHI Intensity in East Asian Cities......................................................................................................................................................................59 2 ■ FIGURE B2.1.1 The Surface Energy Balance: How Reduced Vegetation and Increased Human-Generated Heat Result in Warmer Urban Temperatures ............................................74 ■ FIGURE 2.1 Ambient Air Temperatures in Bandung, Indonesia, 2022................................................................. 76 ■ FIGURE 2.2 Relationship between Air Temperature and Ground Cover Characteristics, in Bandung, Indonesia, 2022 .............................................................................................................................................. 78 ■ FIGURE 2.3 Relationship between Poverty Rate and Afternoon Temperature, by Subdistrict, in Bandung, Indonesia, 2022 ...............................................................................................................80 ■ FIGURE 2.4 Factors Affecting Heat Stress at the Site Level................................................................................. 82 ■ FIGURE 2.5 The Microcauses of Heat and Cooling in Selected Public Spaces, Bandung, Indonesia................................................................................................................................................................83 3 ■ FIGURE 3.1 Channels of Extreme Heat’s Impacts on a City’s Aggregate Real Output................................93 ■ FIGURE B3.2.1 Estimated Impacts of Extreme Heat Events on City Nighttime Light Intensity in East Asia, by Subregion, April 2012–December 2020.........................................................................................................98 4 ■ FIGURE 4.1 The “Places, People, Institutions” Framework for Urban Heat Resilience............................... 114 ■ FIGURE 4.2 Cooling Effects of Selected UHI Mitigation Strategies.................................................................. 115 ■ FIGURE 4.3 Recommended Site-Level Practices in Hong Kong SAR, China, to Safeguard Air Ventilation and Microclimates Considering Wind Flows..................................................... 120 ■ FIGURE B4.2.1 Heat Health Early Warning System and Public Awareness Campaigns in Japan......... 128 ■ FIGURE 4.4 The Heat Wave Risk Management Cycle............................................................................................. 136 UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities VIII LIST OF BOXES ■ BOX 1.1 The Sample of East Asian Cities: An Overview............................................................................................ 34 ■ BOX 1.2 How Were the 100 Cities Selected?.................................................................................................................38 ■ BOX 1.3 Humidity: A Hidden Threat to Thermal Comfort.......................................................................................... 41 ■ BOX 2.1 How Do a Neighborhood’s Vegetation and Built Surfaces Affect Its Climate?.....................................................................................................................................74 ■ BOX 3.1 The Combined Impacts of the Urban Heat Island Effect and Climate Change................................................................................................................................................................95 ■ BOX 3.2 Impacts of Extreme Heat on Urban Economic Activity across East Asian Cities....................................................................................................................................................... 97 ■ BOX 3.3 Do the Negative Effects of Heat Kick In at a Specific Temperature?............................................. 103 ■ BOX 4.1 Cooling Singapore with Traditional Techniques and Digital Technologies.......................................................................................................................................................121 ■ BOX 4.2 Heat Illness Prevention in Japan................................................................................................................... 126 LIST OF MAPS ■ MAP O.1 Average UHI Intensity in 100 Selected East Asian Cities, 2016–20.....................................................4 ■ MAP 1.1 Average UHI Intensity in 100 Selected East Asian Cities, 2016–20................................................... 43 ■ MAP 1.2 Frequency and Intensity of Extreme Heat Events in East Asian Cities, 2010–20.............................................................................................................................................53 ■ MAP 2.1 Projected Number of Hot Nights Per Year for a Present-Day Reference Period and under Future Climate Scenarios, Phnom Penh, Cambodia .................................................................................................86 LIST OF TABLES ■ TABLE 1B.1 Regression Results for the Potential Determinants of Urban Heat Island Intensity.................................................................................................................................................63 ■ TABLE 2.1 Climate Processes at Different Urban Scales........................................................................................ 73 ■ TABLE 2.2 Projected Heat Wave Days, Tropical Nights, and Energy Needs for Space Cooling, by Climate Change Scenario, in Phnom Penh, Cambodia, 2050......................................................................................................................................85 ■ TABLE 3.1 Elderly Populations in Selected East Asian Countries, 2020 (actual) and 2050 (projected)............................................................................................................................... 105 ■ TABLE 4.1 Actions to Create Cooler City Places, by Scale.................................................................................... 118 ■ TABLE 4.2 Albedo Value for Selected Building Materials .....................................................................................124 ■ TABLE B4.2.1 Japan’s Heat Stroke Alert System: Heat Stress Thresholds and Recommended Actions...............................................................................................127 ■ TABLE 4.3 Do Heat Wave Alerts Save Lives? Evidence from International Studies............................................................................................................................................................ 130 ■ TABLE 4.4 Examples of Actions to Protect Heat-Exposed Workers................................................................. 133 UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities IX FOREWORD East Asia’s cities are in danger of becoming unlivable because of extreme heat. Climate change has contributed to record-high temperatures across East Asia in 2023—including in China, the Lao People’s Democratic Republic, Thailand, and Viet Nam—at a time when many of the region’s countries already have extremely hot and humid baseline climates. However, climate change and location are not the only reasons why so many of East Asia’s cities are so hot. The so-called urban heat island (UHI) effect raises temperatures in cities even further. As with climate change, the causes of this UHI effect are man-made, such as cities’ relative lack of vegetation and higher prevalence of impervious surfaces compared with their surrounding rural areas. While a city’s policy makers cannot hope to solve the climate crisis alone, they can take decisive action to cool their cities by mitigating the UHI effect. This is because, unlike climate change, the UHI effect is a local phenomenon. With the right policies and actions, a city’s leaders can reduce local temperatures and promote better adaptation to extreme heat. By reducing demand for air conditioning and investing in urban greenery, mitigation of the UHI effect at the local level can also carry important climate change co-benefits. Through the analysis of temperature data derived from satellites and on-the-ground recordings, as well as a review of previous research, this report addresses three important questions: How strong is the UHI effect in East Asian cities? Why should policy makers be concerned about the UHI effect and extreme urban temperatures more generally? And what can city leaders do to combat extreme urban heat? The results are sobering but, at the same time, offer cause for hope. On average, for a sample of 100 East Asian cities, cities are 2 degrees Celsius (°C) warmer than their surrounding rural areas. However, this average masks considerable variation across cities both between and within East Asian countries, as well as between neighborhoods within individual cities. Such temperature differences could mean the difference between life and death, especially for vulnerable urban residents such as the poor, the elderly, and outdoor workers. This report also shows that extremely high temperatures due to the combination of the UHI effect and climate change have severe, detrimental impacts not just on the health of East Asia’s urban residents but also on the competitiveness, livability, and inclusiveness of local economies. This report argues that, by applying a “Places, People, Institutions” framework that Anna Wellenstein involves policy interventions to cool the physical spaces of a city (“Places”), to act during heatwaves to prevent death and illness among vulnerable populations East Asia & Pacific (“People”), and mainstream heat resilience into city strategies, operations and Regional Director for budgets (“Institutions”), a city’s leaders, working with other stakeholders, can reduce Sustainable Development local temperatures and better prepare their residents for heat waves. In doing so, city leaders in East Asia can save lives and make their cities more productive, attractive, The World Bank and inclusive places. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities X ACKNOWLEDGMENTS This report was prepared by a team led by Mark Roberts and Chandan Deuskar and was also coauthored by Nicholas Jones and Jane Park with further written inputs from Viviana Mora. The work was conducted mainly under the guidance of Francis Ghesquiere during his time as the World Bank’s East Asia & Pacific II Urban, Disaster Risk Management and Land Practice Manager. Yoonhee Kim and Ming Zhang provided further guidance. The report benefitted greatly from background research by Hiromi Akiyama, 4 Earth Intelligence (4EI), Bandung Institute of Technology (ITB), CAPA Strategies, and VITO. Notably, Nurrohman Wijaya and Diaz Ekaputra from ITB, in collaboration with a team led by Joey Williams at CAPA Strategies, conducted a community heat mapping campaign for Bandung, Indonesia. The team would also like to thank the volunteers in Bandung who participated in the data collection during this campaign. Meanwhile, as the winner of the 2021 Digital Earth Partnership Technology Challenge on Urban Heat, a team led by Jonathan Hendry at 4EI conducted background research on urban heat mapping for a sample of 100 East Asian cities. This challenge was organized jointly by the World Bank and Singapore Space and Technology Ltd., from whom the team would particularly like to thank Adhitya Rajasekaran, Lynette Tan, and Nicolette Yeo. The team was fortunate to receive excellent advice and guidance from the following peer reviewers during the report preparation process: Richard Damania, Joanna Masic, and Paolo Avner. These reviewers are not responsible for any remaining errors, omissions, or misinterpretations. The team also benefitted from extremely helpful discussions with officials from Indonesia’s Ministry of National Development Planning (Bappenas) and the Bandung Environmental Agency on the Bandung community heat mapping work. Excellent feedback on an earlier draft of the report was also provided by Barjor Mehta, and the team further benefitted from very helpful discussions with Ross Eisenberg and Steven Rubinyi. Mary Anderson was the editor, while Pablo Porta and others at Estudio Relativo were responsible for the design and typesetting of the report. This design also benefitted from feedback from Mary Fisk of the World Bank’s formal publishing unit. Last, but not least, the team thanks Kai Xin Nellie Teo for unfailing administrative support. This work received generous financial support from the Global Facility for Disaster Reduction and Recovery (GFDRR) and the City Resilience Program. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities XI MAIN MESSAGES The urban heat island (UHI) effect, especially when considered together with climate change, represents a serious and growing threat to the competitiveness, livability, and inclusiveness of East Asia’s cities. However, because it is a local effect, the leaders of these cities can significantly mitigate its impact on local temperatures while also further promoting greater adaptation to extreme urban temperatures. In this context, this report addresses three critical questions: How strong is the UHI effect in East Asia, and how is it interacting with climate change? Why should policy makers worry about the UHI effect and extreme urban temperatures more generally? What can city leaders do to combat extreme urban heat? HOW STRONG IS THE URBAN HEAT ISLAND EFFECT? On average, East Asian cities are 1.6–2.0 degrees Celsius (°C) warmer than their immediate rural surroundings. Based on analysis of satellite data for 100 East Asian cities, land surface temperatures from 2016 to 2020 were, on average, 1.6°C warmer in these cities than in rural areas within 2 kilometers and 2.0°C warmer than in rural areas within 10 kilometers. Cities in Indonesia, Malaysia, and the Philippines—where average temperatures can be up to 5.9°C warmer—suffer from the strongest UHI effects. UHI effects are much stronger than these averages in some neighborhoods and are strongest at night. The averages mask considerable variation in the strength of the UHI effect within cities. For example, new research for this report shows that in Bandung, Indonesia, there are heat disparities of up to 7.0°C between the hottest and coolest neighborhoods. The UHI effect is typically stronger at night than during the day because the heat stored in the built environment is gradually released after sunset. The UHI effect is often stronger in poorer neighborhoods. In Bandung, a neighborhood’s UHI effect is strongly positively correlated with its poverty rate. This is consistent with evidence for many other cities around the world. Stronger UHI effects in poorer neighborhoods result from a relative lack of urban greenery and a greater prevalence of impervious surface areas. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities XII The UHI effect is a particular concern for tropical East Asian cities where humidity adds to the risk of overheating. These cities already have hot and humid baseline climates that are, furthermore, becoming even hotter and more humid over time because of climate change. In Phnom Penh, Cambodia, the UHI effect already contributes to 23–25 heat wave days a year in the city’s central districts. Fast forward to 2050, and the number of heat wave days a year is projected to more than double for those neighborhoods that have the strongest UHI effects. WHY WORRY? Extreme urban heat damages competitiveness. For the world’s largest cities, literature indicates that productivity losses from the combination of the UHI effect and global warming will decrease real gross domestic product (GDP) by 1.4–1.7 percent for the median city by 2050. For the worst-affected city, the loss could reach almost 11 percent by the end of the century. New evidence presented in this report also shows that large Southeast Asian metropolitan cities and other East Asian cities with warmer baseline climates suffer significant losses in economic activity during extreme heat events. Extreme urban heat hurts health. Already more than 100,000 people in the East Asia and Pacific region die each year owing to causes linked to extreme heat. And projections indicate that future warming during this century will increase heat-related mortality more than it decreases cold- related mortality. Extreme urban heat makes cities less livable. This occurs both directly through the unpleasantness of extreme heat—especially when combined with high humidity—indirectly through other negative impacts, such as potential increases in crime, violence, and car accidents. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities XIII WHAT CAN BE DONE? To address the extreme urban heat challenge, East Asia’s city leaders can adopt a “Places, People, Institutions” policy framework. Heat stress is experienced in the physical spaces of a city. Making streets, plazas, parks, factories, workshops, marketplaces, and homes cooler will reduce heat stress, contributing to a vibrant economy and healthy population. But heat waves are also time-bound emergencies that endanger elderly people, children, exposed workers, and people with health vulnerabilities. Understanding who is at risk and providing them with information, health care resources, and respite from extreme heat is crucial to save lives during heatwaves. Extreme heat is also a challenge that requires coordinated action across city government institutions working in tandem with local stakeholders such as employers and neighborhood associations. Accordingly, this report proposes a policy framework based around three principles: cool the physical spaces of a city (“Places”), act during heatwaves to prevent death and illness among vulnerable populations (“People”), and mainstream heat resilience into city strategies, operations and budgets (“Institutions”). Actions in six priority areas can help city leaders deliver on a “Places, People, Institutions” agenda for heat resilience. These strategic options for action are (a) promote urban greening through strategic planning; (b) cool city spaces through wind, shade, and urban design; (c) engage building owners in tackling indoor heat; (d) save lives through heat wave early warnings; (e) protect heat-exposed workers; and (f) mainstream heat risk reduction throughout city institutions and strategies. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities XIV ABBREVIATIONS AC air conditioner NASA National Aeronautics and Space ASEAN Association of Southeast Asian Nations Administration CO2 carbon dioxide NOAA National Oceanic and Atmospheric °C degrees Celsius Administration °F degrees Fahrenheit SAR special administrative region (of China) 4EI 4 Earth Intelligence TFP total factor productivity GDP gross domestic product UHI urban heat island ILO International Labour Organization URA Urban Redevelopment Authority (Singapore) JMA Japan Meteorological Agency LST land surface temperature(s) TEMPERATURE CONVERSION A temperature in degrees Fahrenheit (°F) can Celsius (°C) FFahrenheit (°F) be obtained by adding 32 to a temperature in degrees Celsius (°C) multiplied by 9/5 (or 1.8). In -5 23 mathematical form: 0 32 F = (C × 9/5) + 32 5 41 10 50 where, F and C denote temperatures in °F and °C, 15 59 respectively. 20 68 25 77 30 86 35 95 40 104 45 113 50 122 UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 1 OVERVIEW INTRODUCTION Over the past 50 years, climate change has caused the earth’s surface to warm by almost 1.2 degrees Celsius (°C) above its preindustrial levels.1 As dangerous as this increase has been, however, it is less than the typical average temperature differential between an East Asian city and its immediate rural outskirts. Travel just 2 kilometers outside of Da Nang in Viet Nam, Mendi in Papua New Guinea, or Wuhan UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 2 in China, and you will find a temperature that is, on average, 1.5–2.5°C cooler than in the city.2 Undertake a similar journey to the rural outskirts of Bandung in Indonesia, George Town in Malaysia, or Manila in the Philippines, and you will feel an even greater temperature drop of between 3.5°C and 5.3°C.3 Just as with global warming, the causes of these large temperature differentials are man-made. An urban heat island (UHI) effect tends to make cities warmer than their rural surroundings because of surfaces and structures, such as roads and buildings, that absorb and retain heat. Together with a relative lack of vegetation, and hence evaporative cooling, this causes a city to be hotter than its rural surroundings. This UHI effect is further compounded by the heat from motorized vehicles and the often widespread use of air conditioning, which transfers heat to the outside of a city’s buildings (Deuskar 2022). Why should an East Asian city’s policy makers worry about the UHI effect? And what can they do to mitigate its impacts and help people and businesses to adapt to higher temperatures, both from the UHI effect and global warming? These are the two key questions addressed by this report. Before getting to these questions, however, the report first addresses another important question—how strong is the UHI effect in East Asia? It further explores the interactions between warming driven by the UHI effect and warming driven by climate change, noting the compounding effects of these twin threats on each other. HOW STRONG IS THE URBAN HEAT ISLAND (UHI) EFFECT IN EAST ASIA? UHI EFFECTS ALREADY STRONG, WITH LARGE VARIATIONS ACROSS CITIES As noted earlier, cities such as Bandung, George Town, and Manila are, on average, 3.5–5.3°C warmer than their rural surroundings within 2 kilometers. More generally, land surface temperatures over 2016–20 were, on average, around 1.6°C warmer in the city than in the nearby countryside within 2 kilometers for a sample of 100 East Asian cities.4 The larger the buffer used to define a city’s rural surroundings, the larger the estimated average UHI effect—from 1.6°C for a 2-kilometer buffer to 2.0°C for UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 3 a 10-kilometer buffer (figure O.1, panel A). Regardless of the exact distance used, however, the estimated intensity of the average UHI effect for East Asian cities has remained broadly stable since 2002 (figure O.1, panel B).5 Figure O.1 Average Intensity and Trends of Urban Heat Island Effects across East Asian Cities, by Buffer Size A AVERAGE INTENSITY, 2016–20 B INTENSITY TRENDS, 2002–20 2.5 2.5 2.0 2.0 UHI intensity (°C) UHI intensity (°C) 1.5 1.5 1.0 1.0 0.5 0.5 0.0 0 08 06 04 10 2 km 5 km 10 km 18 16 14 12 20 20 20 20 20 20 20 20 Buffer size 2 km 5 km 10 km SOURCE: World Bank calculations based on estimates of urban heat island (UHI) intensity produced by 4 Earth Intelligence (https://www.4earthintelligence.com/). NOTE: UHI intensity is estimated by averaging the daytime and nighttime differences in land surface temperatures, in degrees Celsius (°C), between urban and rural areas based on buffer distances of 2, 5, and 10 kilometers (km). Buffers define the location and size of the rural zone around a city. Panel b shows the trends of five-year moving averages. The solid and dotted lines respectively show the 5-year moving averages and their trends. Effects are estimated across a sample of 100 East Asian cities (as further described in chapter 1, box 1.2). Behind the average UHI effect lies considerable variation across East Asian cities (map O.1). The cities with the most severe UHI effects are generally in Indonesia, Malaysia, and the Philippines. Together with Singapore, these countries dominate the list of top 10 cities with the highest estimated UHI intensities: the average of their daytime and nighttime UHI intensities, based on a 5-kilometer buffer, range from 3.5°C to 5.9°C.6 UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 4 By contrast, cities in the continental portion of Southeast Asia, such as those in the Lao People’s Democratic Republic, Myanmar, Thailand, and Viet Nam, generally exhibit less severe UHI effects—in the range of 0.1–2.4°C, based on a 5-kilometer buffer. The same is true for cities at higher latitudes, including most Chinese cities and Seoul, Republic of Korea. With an average UHI intensity of around 3.3°C, Tokyo is an exception. Map O.1 Average UHI Intensity in 100 Selected East Asian Cities, 2016–20 Urban heat island intensity (Average over 2016–20) .0 .0 .0 .0 .9 –1 –2 –3 –4 –5 00 01 01 01 01 0. 1. 2. 3. 4. SOURCE: World Bank calculations based on the urban heat island (UHI) intensity data from 4 Earth Intelligence, https://www.4earthintelligence.com/. NOTE: The map presents the five-year (2016–20) averages of UHI intensity, estimated by averaging the daytime and nighttime differences in land surface temperatures between urban and rural areas based on the 5 kilometer buffer. Regarding selection of the 100 cities, see box 1.2. IDN = Indonesia; KHM = Cambodia; MMR = Myanmar; MYS = Malaysia; PHL = Philippines. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 5 PARTICULAR CONCERNS ABOUT UHI EFFECTS IN TROPICAL EAST ASIAN CITIES Strong UHI effects are a particular concern for the nearly 40 percent of East Asia’s urban population living in humid tropical cities such as Hanoi, Viet Nam—where climate change, moreover, is making cities even hotter and more humid over time.7 Tropical East Asian cities have also experienced an upward trend in annual precipitation since 1958.8 This has contributed to increased humidity, compounding the adverse impacts of high temperatures on thermal comfort. Neither human bodies nor cities can cool themselves down as efficiently when extreme heat combines with high humidity. Cities can cool down through convection and the evaporation of water contained in vegetated surfaces. However, neither process works as well under highly humid conditions. Consistent with this, UHI effects and heat waves tend to be stronger in more humid cities (Manoli et al. 2019; Russo, Sillmann, and Sterl 2017). UHI EFFECTS OFTEN STRONGER IN POORER NEIGHBORHOODS The strength of the UHI effect also varies significantly across neighborhoods within individual East Asian cities. A good example of this is provided by Indonesia’s second largest city, Bandung, which has a population of around 8.1 million in its wider metropolitan area (Roberts, Gil Sander, and Tiwari 2019). Research for this report shows that Bandung exhibits marked differences in ambient air temperatures across its neighborhoods, with poorer neighborhoods tending to be hotter (figure O.2).9 UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 6 Figure O.2 Ambient Air Temperatures in Bandung, Indonesia, 2022 A AVERAGE NEAR-GROUND AIR TEMPERATURES B RELATIONSHIP BETWEEN POVERTY RATE AND AIR TEMPERATURE IN SUBDISTRICTS OF BANDUNG, INDONESIA 29.4 °C 12 10 8 % below poverty line R2=0.31 6 24.6 °C 4 N 2 0 1.5 3 0 kilometers -2 29.5 30.0 30.5 31.0 31.5 Afternoon temperature (°C) SOURCE: Jones et al. 2023 and poverty data from Smeru Research Institute’s Poverty and Livelihood Map of Indonesia 2015 (https://smeru.or.id/en/resource/poverty-and-livelihood-map-indonesia-2015). NOTE: In panel A, The ambient air temperature surface map was derived using machine learning techniques whose inputs included field measurements of near-ground air temperature taken on July 31, 2022. The map averages temperatures in degrees Celsius (°C) for three times of day (7–8 a.m., 12–1 p.m., and 5–6 p.m.) to arrive at a citywide average daytime temperature map. In panel B, poverty is defined as the proportion of people with per capita household consumption below the national poverty line (Rp 350,000 a day). Modeled heat data are based on vehicle-based ambient air temperature measurements in degrees Celsius (°C) taken on July 31, 2022. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 7 That poorer neighborhoods tend to experience stronger UHI effects in Bandung and many other cities around the world is likely the result of two processes:10 first, richer households can outbid poorer ones when it comes to renting or buying property in more “pleasant” neighborhoods, which can include neighborhoods with more greenery and fewer impervious surfaces, both of which mitigate the UHI effect. Second, households, businesses, and governments all invest in the built form of city neighborhoods, making decisions about structural types, building materials, plot sizes, spacing between buildings, and vegetation prevalence that are shaped by their budgets and planning regulations. In this context, households and businesses in poorer neighborhoods may be less able to afford investments in, say, urban greenery. Local governments may also target poorer neighborhoods less intensively for such investments, in part because wealthier neighborhoods may be better placed to lobby for the investments. To the extent that poorer households live in more informal or illegal slum neighborhoods, local governments may also be less likely to invest in those neighborhoods. Consistent with these processes, temperatures across Bandung’s neighborhoods are positively correlated with the prevalence of impervious surfaces and negatively correlated with canopy cover (figure O.3). Nevertheless, one must be careful not to overgeneralize—in some cities, including Beijing, Bangkok, and Manila, it is the wealthier neighborhoods that are hotter. More generally, Chakraborty et al. (2019) find that, for 7 out of a global sample of 25 cities, poorer neighborhoods are cooler. Several aspects of urban development, such as a city’s housing market and connectivity, affect whether more affluent residents live in the city centers where UHI intensity is generally higher. This illustrates the importance of studying individual cities in depth before drawing conclusions regarding policy interventions at a local scale. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 8 Figure O.3 Relationship between Air Temperature and Ground Cover Characteristics, in Bandung, Indonesia, 2022 A IMPERVIOUS SURFACE 32 30 Temperature (°C) y = -5E-05x2 + 0.024x + 26.6 28 R 2 = 0.31 26 24 0 20 40 60 80 100 Impervious Surface (%) B TREE CANOPY COVER 32 30 Temperature (°C) y = -0.0002x2 - 0.026x + 28.4 28 R 2 = 0.27 26 24 0 20 40 60 80 100 Canopy Cover (%) SOURCE: Jones et al. 2023, using in situ heat measurements plus European Space Agency’s World Cover 2020 dataset (https://worldcover2020.esa.int/). NOTE: Near-ground air temperature measurements were taken during vehicle traverses of Bandung on July 31, 2022. Panels A and B present temperature measurements in degrees Celsius (°C) against the percentages of land cover that are impervious surface and tree canopy cover, respectively. The equations and R-squared values summarize relationships between air temperature and ground cover characteristics with each dot representing a 100-meter grid cell. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 9 MORE SEVERE HEAT WAVES LIKELY AS UHI EFFECT INTERACTS WITH CLIMATE CHANGE By raising its local baseline temperature, the UHI effect makes a city more prone to heat waves, the frequency and intensity of which have already been increasing in recent decades because of climate change—especially for East Asian cities in tropical zones and in low- and middle-income countries (figure O.4). In the decades ahead, climate change will only further increase the frequency with which East Asian cities experience extreme heat. This is particularly the case for neighborhoods within cities for which the UHI effect is strongest, which, again, often tend to be the poorest neighborhoods. One such example is Phnom Penh, Cambodia’s capital city, which is home to almost 3 million people. According to an urban climate model built for this report, Phnom Penh suffers from a current nighttime UHI effect (measured as the difference between the city center and outlying rural temperatures) of 2.5°C. 11This effect already contributes to 23–25 heat wave days a year in the city’s central neighborhoods, compared with only 2–5 such days a year in surrounding rural areas. Fast-forward to 2050, however, and the number of heat wave days a year is projected to further increase to 40 in those neighborhoods of Phnom Penh most affected by the UHI effect, even under an optimistic climate change scenario.12 Under a pessimistic climate change scenario, these heat wave days rise to 60 days a year.13 By contrast, although they will not remain immune to climate change, leafy neighborhoods with extensive vegetation coverage and more space between dwellings are projected to remain cooler, with around one-third fewer heat wave days in 2050. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 10 Figure 0.4 Frequency and Intensity of Extreme Heat Events by Climate Zone and Country Income Level, in East Asian Cities, by Decade (1958-69 to 2010-20) sd: standard deviations A FREQUENCY BY CLIMATE ZONE B INTENSITY BY CLIMATE ZONE 3.5 2.6 2.6 3.0 3.0 Average anomaly size (sd) Average months per year 2.5 2.5 2.4 2.5 2.4 2.0 2.4 2.4 2.3 2.4 1.6 2.3 1.5 1.4 1.2 2.3 2.3 1.0 0.9 2.2 2.2 0.5 2.2 2.3 0.5 0.3 0.2 0.1 0.1 0.1 0.1 0 2.1 Tropical Nontropical Tropical Nontropical 1958-69 1970s 1980s 1990s 2000s 2010-20 C FREQUENCY BY INCOME LEVEL D INTENSITY BY INCOME LEVEL 2.5 2.6 2.1 Average anomaly size (sd) Average months per year 1.9 2.5 2.0 2.5 2.46 2.5 1.5 1.4 2.4 2.4 2.4 2.4 2.3 1.0 2.3 0.7 0.8 2.2 2.3 0.6 2.2 2.2 0.5 2.2 2.2 0.1 0.1 0.1 0.2 0.1 0.1 0 2.1 Low- and middle-income High-income Low- and middle-income High-income 1958-69 1970s 1980s 1990s 2000s 2010-20 SOURCE: World Bank calculations based on Climatology Lab, TerraClimate, https://www.climatologylab.org/terraclimate.html; European Commission, Global Human Settlement (GHS) Urban Centre Database R2019, https://ghsl.jrc.ec.europa.eu/ghs_stat_ ucdb2015mt_r2019a.php. NOTE: An “extreme hot month” is one in which a city’s temperature for that month is at least 2 standard deviations higher than the month’s city-specific historical norm, as calculated over the period January 1958–December 1999. Frequency is calculated as the number of extreme hot months per year. Intensity is calculated as the average size of the anomaly variable during consecutive extreme hot months. Panels A and C present the average frequency, and panels B and D the average intensity, across the sample of 3,033 East Asian cities. In panels A and B, “tropical” and “nontropical” refer to a city’s latitude in absolute terms—“tropical” being in latitudes 0°–23.5° north and “nontropical” elsewhere. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 11 WHY WORRY ABOUT THE URBAN HEAT ISLAND EFFECT? THREAT OF EXTREME HEAT TO CITIES’ COMPETITIVENESS Effects on GDP. In addition to the more obvious health impacts, the UHI effect, especially when considered together with global warming, represents a serious threat to the competitiveness of East Asian cities. One recent study has estimated that, for a global sample of the world’s 1,692 largest cities, productivity losses from the combination of the UHI effect and global warming will decrease the median city’s real gross domestic product (GDP) by 1.4–1.7 percent by 2050 (Estrada, Wouter Botzen, and Tol 2017). For the worst-affected city, the loss could reach almost 11 percent by the end of the century. Crucially, the projected economic damage from heat in cities during the twenty-first century from both the UHI effect and climate change combined is up to 2.6 times the magnitude of projected losses from climate change alone. Furthermore, the combined economic impact of the UHI effect and global warming is likely to be greater than the sum of their individual impacts. This is because the impact of temperature on a city’s economy is nonlinear: the negative impact of a 1°C increase in temperature—say, from 35°C to 36°C—is greater than that of an increase from 30°C to 31°C (Burke, Hsiang, and Miguel 2015). Given that many East Asian cities, especially tropical ones, are starting off with higher temperatures than in cities in many other parts of the world, these global projections of economic losses probably understate the potential losses that they face in the absence of policy action. Effects on hours worked and labor productivity. Extreme heat associated with both the UHI effect and climate change hurts a city’s economy through many channels (figure O.5). Thus, increased exposure of a city’s workers to extreme heat can result in involuntary reductions in hours worked, and therefore labor input, because of increased work stoppages and absences induced by heat-related illnesses. In addition to reducing the quantity of labor input provided to a city’s economy, exposure to extreme heat also reduces the quality of that input. This is because extreme heat exposure reduces both physical abilities and cognitive performance, meaning that it hinders worker performance in manual and more-skilled occupations UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 12 alike (see, for example, Kahn and Li 2019; Seppänen, Fisk, and Lei 2006). The International Labor Organization (ILO) projects that, by 2030, the equivalent of 3.7 percent of total working hours will be lost in Southeast Asia owing to the combination of the reduced quantity and quality of labor inputs associated with extreme heat exposure. Losses will be particularly large in both the construction and industry sectors, for which it is projected that the equivalent of 8.9 percent and 3.9 percent of working hours, respectively, will be lost (ILO 2019). Cities that are more specialized in these sectors will, therefore, be particularly hard hit. Figure O.5 Channels of Extreme Heat’s Impacts on a City’s Aggregate Real Output Number Physical Human Effectiveness Total factor of hours capital capital of human productivity worked Extreme Extreme heat capital Extreme heat damages heat causes impacts learning infrastructure. Extreme heat Heat exhaustion breakdowns and and makes causes work reduces physical and deterioration labor market stoppages mental abilities. of machinery. less attractive. and absences. Output SOURCE: World Bank elaboration. Effects on human capital. In addition to undermining the effectiveness of a city’s current stock of human capital, increased exposure of the population to extreme heat can, in the longer run, also damage its economic growth by reducing its rate of human capital accumulation. This is because extreme heat exposure hurts both student learning and exam-day performance (Park, Behrer, and Goodman 2021; Park et al. 2020).14 Extreme heat due to the UHI effect and climate change may also undermine a city’s human capital stock in the long run by making it a less attractive place to live for highly skilled workers—both because lower productivity associated with more UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 13 extreme heat leads to lower wages and because workers view an overly hot city as an unpleasant place to live.15 Effects on physical capital. Finally, extreme heat may damage a city’s economy both through its impacts on the physical capital of its firms and through damage to a city’s infrastructure, thereby reducing a city’s total factor productivity. Excessively high temperatures can result in heat-induced breakdowns and accelerated deterioration of machinery, particularly affecting capital-intensive industries. Meanwhile, extreme heat can damage transportation infrastructure by causing roads and rails to buckle because of thermal expansion and by melting or softening paving material. Heat can also reduce the performance of electrical power lines and photovoltaic cells, increase the risk of oil and gas pipeline ruptures, cause geotechnical failures in embankments and earthworks, and damage underground communications infrastructure (Dodman et al. 2022). IMPACTS OF EXTREME HEAT ON URBAN ECONOMIC ACTIVITY Original empirical research for this report for a large sample of East Asian cities (see chapter 3, box 3.3 of the main report) finds that extreme heat events have significant negative impacts on local economic activity in two city groups—large Southeast Asian metros and cities in the rest of East Asia that have warmer baseline climates. A temperature increase of more than 2 standard deviations above a city’s own historical norm for any given month reduces its nighttime light intensity—a commonly used proxy for economic activity—by an estimated 3.9 percent during that month in large Southeast Asian metros (figure O.6, panel A). Elsewhere in East Asia, in cities with warmer baseline climates, such an extreme heat event has an even larger estimated impact on nighttime light intensity: −5.5 percent (figure O.6, panel B). By contrast, extreme heat events have no significant impact on the economic activity of smaller Southeast Asian cities, and the estimated impact on economic activity for East Asian cities with cooler baseline climates is even positive. One potential explanation for this latter result could be that, for cities with cooler climates, a heat event results in more pleasant evenings, which benefit outdoor economic activity, albeit possibly at the expense of daytime economic activity. These results highlight the fact that increases in temperature and heat events driven by the combination of the UHI effect and global warming should be a particular concern for large East Asian cities with tropical climates. Consistent with this, this report also finds that more densely populated cities in East Asia tend to suffer from more severe UHI effects. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 14 Figure O.6 Estimated Impacts of Extreme Heat Events on City Nighttime Light Intensity in East Asia, by Subregion, April 2012–December 2020 A SOUTHEAST ASIA B REST OF EAST ASIA 0.02 0.10 0 0.05 Estimated elasticity Estimated elasticity −0.02 0 −0.04 −0.05 −0.06 −0.10 −0.08 −0.15 Smaller cities Large metros Cooler cities Warmer cities SOURCE: World Bank calculations based on the analysis of Visible Infrared Imaging Radiometer Suite (VIIRS) nighttime lights monthly composites (https://payneinstitute.mines.edu/eog-2/viirs/) and monthly weather data from Climatology Lab, TerraClimate (https://www.climatologylab.org/terraclimate.html). NOTE: Cities are defined as urban centers following the “degree of urbanization” methodology of the Global Human Settlement Layer (GHSL) Urban Centre Database. Each red marker shows the estimated impact of an extreme heat event, obtained by regressing the natural log of sum of lights in city c in month m on a dummy variable which takes the value 1 if temperature of city c in month m deviated from the city’s long-run mean temperature for the month by at least +2 standard deviations; the dummy variable takes the value 0 otherwise. Vertical bars indicate the bounds of the 90 percent confidence interval associated with the corresponding estimated impact. Panel A is based on the analysis of 170 Southeast Asian cities. Panel B is based on the analysis of 536 cities in the rest of East Asia. The analysis of Southeast Asian cities (panel A) covers all 12 months of the year because monthly temperatures remain relatively warm throughout the year for those cities. In contrast, the analysis of cities in the rest of East Asia (panel B) covers only warm months, (June, July, and August) given their more variable climates. HARMFUL IMPACTS OF EXTREME HEAT ON HEALTH AND LIVABILITY Extreme heat makes East Asian cities not only less competitive but also less healthy and livable. More than 100,000 people in East Asia and the Pacific already die each year from causes linked to extreme heat (Zhao et al. 2021). And projections for middle-income countries in East Asia indicate that future warming during this century will increase heat-related mortality more than it decreases cold- related mortality, even after adjusting for increased resilience due to income growth (Bressler et al. 2021).16 Evidence for North American countries furthermore suggests that heat waves can harm mental health, resulting in increased numbers of suicides (Burke et al. 2018), and lead to both more car and workplace accidents (Park, Pankratz, and Behrer 2021; Wu, Zaitchik, and Gohlke 2018). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 15 Extreme heat is also associated with more crime, including, according to some evidence, crime that involves sexual violence (Heilmann and Kahn 2019; Hu et al. 2017; Sanz-Barbero et al. 2018). Hence, extreme heat makes cities not only less healthy but also less livable. As discussed earlier, this may, in turn, damage a city’s ability to attract high-skilled labor. HOW EXTREME HEAT LIMITS CITIES’ INCLUSIVENESS As discussed earlier, in many but not all cities, including in many East Asian cities it is the poor who live in the hottest neighborhoods with the most extreme UHI effects. It follows that, in such cities, the poor are the most likely to bear the brunt of harms from extreme heat—such as increased health risks, heightened risk of crime, and reduced productivity and wages. The poor not only face greater exposure to extreme heat in many cities but also are less financially equipped to adapt. For example, air conditioner (AC) ownership increases with household income in both China and Indonesia (Davis et al. 2021; Pavanello et al. 2021). This lower capacity of the poor to adapt may help explain why— for example, in Hong Kong SAR, China—the relationship between heat and mortality is stronger in poorer districts than in richer ones (Chan et al. 2012). Extreme heat can also affect other dimensions of a city’s inclusiveness, beyond poor versus nonpoor. For instance, it is well known that the elderly, as well as those with disabilities or chronic medical conditions, are most at risk of the health impacts of heat (Deuskar 2022). Given the relatively elderly populations in high-income East Asian countries, extreme heat is already a pressing concern for their cities. It will also increasingly concern cities in low- and middle-income East Asian countries as their populations age. Between 2020 and 2050, the share of the population over the age of 65 is projected to almost quadruple in Mongolia; more than triple in Cambodia, Laos, and Indonesia; and nearly triple in Myanmar and Viet Nam (UN DESA 2017). Varquez et al. (2020) project that, in Jakarta, the combined effects of population aging and rising daytime temperatures will cause the total number of heat-related elderly deaths in August to be 12–15 times higher in the 2050s than it was in the 2010s. AIR CONDITIONING: NO PANACEA When a person faces extreme heat, the most obvious response is to purchase an air conditioner (AC) as soon they can afford it, or if air conditioning is already installed, to turn it up. And given the increasingly hot and humid climates of many low- and middle-income East Asian countries, it is unsurprising that rates of AC ownership UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 16 and use are rapidly growing. Indeed, energy use for air conditioning in Association of Southeast Asian Nations (ASEAN) countries has increased by a factor of 7.5 over the past 30 years and is expected to continue to increase as incomes and temperatures in the region rise (IEA 2019).17 However, although AC use can cool indoor spaces and partially enclosed outdoor spaces (albeit less effectively), it cannot cool open outdoor spaces. Indeed, a local negative externality of AC use is that it exacerbates the UHI effect by transferring heat from indoors to outdoors, creating a potential vicious cycle of intensified UHI effects and increased AC use. Furthermore, AC use is energy intensive and, therefore, so long as energy mixes remain carbon intensive, contributes to climate change. As such, AC represents a serious form of “maladaptation” to extreme heat (Viguié et al. 2020). There is an urgent need for policies that promote other forms of mitigation of, and adaptation to, both the UHI effect and extreme urban heat more generally. WHAT CAN BE DONE? While heat stress presents growing risks to East Asia’s urban populations, the challenge is one that the region’s cities can address decisively and effectively. Working in tandem with national governments and neighborhood-level stakeholders, East Asian cities have at their disposal a wide range of policy and investment options to reduce heat exposure and help residents better anticipate and cope with extreme conditions. However, resilience to extreme heat is no ordinary policy challenge. At least three factors complicate the response, making new and innovative mindsets essential. Making a city resilient to heat requires: Access to an unusually wide variety of skills from across government and the private sector. Contributions are needed not only from urban planners and forestry managers but also from public health specialists, doctors, emergency managers, architects, real estate developers and business owners. Action across physical scales, from changing the material of individual walls and roofs to rethinking areawide land use plans to promote increased greenery and wind flow. New governance structures to ensure both clear lines of authority and coordination of action. Established policy priorities such as transport, health, and education typically have accountability and resources vested in clearly defined agencies, but the authority and capability to reduce heat risks is highly fragmented among government entities. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 17 Leadership is therefore crucial. To rise to the extreme heat challenge, city leaders must establish an evidence base, articulate a vision for change, and build new coalitions for action. To help structure an effective response to the rising heat stress in East Asian cities, this report proposes a policy framework based around three principles: cool the physical spaces of a city (“Places”); act during heat waves to prevent death and illness among vulnerable populations (“People”); and mainstream heat resilience into city strategies, operations, and budgets (“Institutions”). Addressing the three principles in tandem is important given their interlocking nature, and it requires concerted leadership by city authorities working together with central government agencies such as health departments and national meteorological services (figure O.7). Figure O.7 The “Places, People, Institutions” Framework for Urban Heat Resilience Places People Institutions ■ Advance urban greening ■ Save lives through heatwave ■ Mainstream heat mitigation through strategic planning early warnings into city strategies and ■ Cool city spaces through ■ Protect heat-exposed operations wind, shade and design workers ■ Engage building owners to tackle indoor heat SOURCE: World Bank. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 18 Heat stress is experienced in the physical places of a city, both outdoors and in. Making streets, plazas, parks, factories, workshops, marketplaces, and homes cooler will reduce heat exposure, contributing to a competitive economy and healthy population. Heat waves are also time-bound emergencies that endanger people—especially elderly people, children, exposed workers, and those with health vulnerabilities. Identifying the groups that are disproportionately at risk and protecting them during heat waves is critical. Finally, mainstreaming heat resilience across a city’s institutions—its various governmental departments and agencies as well as civil society and professional groups that have essential roles to play—is a prerequisite to effective action. The remainder of this chapter presents six strategic options for action to deliver on the “Places, People, Institutions” agenda for heat resilience. For most cities, all six strategic options are likely to be beneficial and complementary – but the relative importance of each requires careful consideration given existing policy frameworks, and local circumstances such as the prevailing climate and water availability. By considering each option and selecting those that most closely fit local needs, East Asian cities can develop strategic policy and investment frameworks for resilience to extreme heat conditions. STRATEGIC OPTION 1: ADVANCE URBAN GREENING THROUGH STRATEGIC PLANNING Green assets such as trees, grass, and shrubs promote cooler city spaces. However, levels of vegetation tend to be highly uneven between city neighborhoods. Prosperous residential areas tend to have better organized and resourced tree care and may have larger plot sizes, whereas central business districts and poor residential neighborhoods often lack greenery. Expanding the reach of urban nature, especially where trees and shrubs are in short supply, can increase shade and evapotranspiration, resulting in cooler temperatures.18 Urban greening is a key ingredient in heat resilience strategies and benefits diverse groups, including workers and shoppers who frequent dense city centers as well as the poorest residents whose neighborhoods may lack grass, shrubs, and trees. The return on investment from urban greening can be high. A metareview of 220 greening interventions found a reduction in localized temperatures exceeding 2°C in one-third of cases (Santamouris et al. 2017). Planting at “microsites,” (including tree pits and on traffic medians) cools air locally, while planting at extensive sites (such as parks) can generate a “park cool island effect” that benefits surrounding blocks. Pocket parks—small parks of around 1,000 square meters—may confer greater UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 19 cooling benefits than a single large park, especially in high-rise tropical regions (Giridharan and Emmanuel 2018). One study found that expanding tree cover on 10 percent of vacant land could bring temperature reductions of 1–3°C to 647,000 people in Seoul, 772,000 people in Jakarta, and 2.2 million residents in Beijing —at a cost of just US$1–3 per beneficiary (DeJarnett, McDonald, and McDonald 2016). City leaders can consider adopting the so-called 3-30-300 rule: each resident should be able to see at least 3 trees from their home, have 30 percent canopy cover in their neighborhood, and live no farther than 300 meters from a green space (Konijnendijk 2021). Meeting ambitious goals such as the 3-30-300 target across East Asian cities will require resources and long-term planning. Indeed, urban greening is easier said than done. Key challenges in urban forestry expansion include species selection; survival during the crucial first year after planting; tree maintenance and pruning; managing competition between tree roots and the needs of residents and infrastructure; and securing community support for planting and maintenance.19 It further requires strategic planning to expand green assets on large and small sites as well as on publicly owned, commercial, and residential land. Forward planning with a timeline of at least a decade is particularly important for urban forestry given the speed at which trees grow. In this context, adopting dedicated urban forestry master plans is highly advisable. By setting a strategic vision for green assets, such plans provide a powerful tool to ensure that legislative frameworks and budget availability align with a city’s future greening vision and to track progress toward goals. STRATEGIC OPTION 2: COOL CITY SPACES THROUGH WIND, SHADE, AND URBAN DESIGN In addition to “green assets,” wind is a crucial resource, especially for coastal cities. Sea breezes form when air above a city warms at a different rate than air above the sea, providing a cooling effect and dispersing air pollution. In Singapore, air temperatures fall by 2°C when wind speed increases by 1.5 meters per second (Erell, Pearlmutter, and Williamson 2011). However, urban construction can inhibit wind. Tall buildings located directly on the water’s edge can block sea breezes, resulting in significantly warmer and more polluted air inland.20 East Asian cities where rapid growth of tall buildings along the coast is anticipated should learn from examples such as the air ventilation assessment for Hong Kong SAR, China. Just as the city’s urban design guidelines (Hong Kong SAR, UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 20 China, Planning Department. 2015) provide clear guidance on design features and requirements—such as stepped-back buildings, designated breezeways, and arranging buildings to channel wind—designing cities “with the wind” may offer significant benefits (Ng 2009). As well as wind, shade and access to waterfront environments are valuable commodities in hot climates. Urban planning documents and cooling action plans should seek to increase shade in public places. In Freetown, Sierra Leone, the city government introduced awnings in public marketplaces, providing relief for shoppers and street vendors. In Singapore, a distinctive network of covered walkways adds to livability. Restoring and preserving “blue assets” such as rivers, canals and lakes can also improve thermal comfort for city residents as exemplified by the Seoul Urban Greenway: a cool and attractive 10-kilometer urban park created by revitalizing a historic stream in South Korea’s capital (Cho 2010). In promoting cooler urban designs, an important source of inspiration is traditional building practices (“vernacular architecture”), which, in many East Asian cities, offer a rich tapestry of architectural practices that promote ventilation, minimize heat storage in surfaces, and provide shade (Satwiko 1999). STRATEGIC OPTION 3: ENGAGE BUILDING OWNERS TO TACKLE INDOOR HEAT Poorer neighborhoods often have higher outdoor air temperatures, but when indoor conditions are also considered, “thermal inequalities” loom even larger. Addressing indoor heat exposure is essential to reducing heat illness and sustaining labor productivity, since people spend much time indoors. The design and construction of buildings is a key entry point to address heat exposure. City governments’ policy leverage differs depending on building ownership and established planning procedures, and hot buildings cannot be transformed into cool ones overnight. However, city authorities can systematically engage with building owners to promote cooler design and materials while keeping an eye out for “quick wins” that reduce harmful heat exposure of many people at low cost. City authorities can employ a combination of “hard” (mandatory regulations and direct public investment) and “soft” measures (including incentives and information provision) to promote cooler buildings. Among hard measures, city governments, together with relevant national authorities, can review and strengthen building codes and zoning requirements. A case in point is India’s 2016 building code, which specifies reflectivity and emissivity values for roof-covering materials on new or substantially renovated roofs (ESMAP 2020). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 21 City governments can also lead by example through publicly owned facilities or those procured through public-private partnerships. Schools, clinics, government buildings, and sport facilities all provide opportunities to pilot solutions, build construction industry capacity, and create a local evidence base on the effectiveness of cool building measures. Given the limits of public budgets and building code enforcement, a broader set of tools including fiscal and planning incentives and concessional finance for cool building upgrades is worth examining on a case-by-case basis. Hong Kong SAR, China; Singapore; and Tokyo are among the cities offering floor-area ratio bonuses— which permit larger buildings on a given plot size—if developers adhere to green building standards. Property tax rebates for cool roof installations are another option.21 Cool roof programs reduce indoor heat gain through white paint coatings that reflect away the sun’s energy. They represent one of the most cost-effective and scalable heat-reduction measures, as demonstrated by Cool Roofs Indonesia, a program whose interventions have resulted in indoor temperature reductions as high as 10°C (Clean Cooling Collaborative 2022). Integrating cool roofs into public housing programs and incentivizing their adoption by commercial building owners offers major benefits. STRATEGIC OPTION 4: SAVE LIVES THROUGH HEAT WAVE EARLY WARNINGS While cities can take incremental steps to reduce heat exposure across indoor and outdoor settings all year round, the moment of truth for preventing heat-related deaths and illnesses is when a heat wave strikes. To a large degree, deaths during heat waves are both predictable and preventable: they are concentrated in groups that have high exposure to heat—owing to factors such as housing quality or heat- exposed occupations—as well as high vulnerability to heat, whether because of age, occupation, or health status. The most effective measures to prevent a heat wave death are those taken by at-risk individuals or those closest to them: their family, health care providers, or employers. Heat waves cause most deaths when they take populations by surprise by coming early in the season. As a substantial body of evidence shows, heat early warnings can save lives by informing people that their health faces an imminent threat— prompting them to take actions to protect themselves—as well as by activating protocols in health care, education, and workplace settings (Toloo et al. 2013). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 22 Effective early warning systems go beyond simply announcing high temperatures to predicting upcoming heat-related deaths and activating preplanned interventions to safeguard at-risk groups, as several examples illustrate. In the United States, the city of Philadelphia—which in 1995 introduced one of the world’s first population health- focused heat wave warning systems—reduced heat wave mortality by 4.5 percent through the interventions it activates upon declaration of a Heat Health Emergency. These interventions range from activating a “Heatline” telephone service to using libraries and swimming pools as cooling centers and supporting homeless people (Weinberger et al. 2018). And in Hanoi, the Viet Nam Red Cross and Viet Nam Institute of Meteorology, Hydrology and Climate Change jointly implement warnings based on a combined measure of heat and humidity (Jjemba et al. 2020). Japan, a country with a hot summer climate, has developed and refined an extreme heat early warning system centered on the concept of “heat illness reduction.” Alerts are issued by the Japan Meteorological Agency and disseminated via television and radio, private weather companies, email, social media channels, and local government, including via public loudspeakers. Of alert recipients who responded to an evaluation survey, 31 percent changed their behavior to stay indoors, 52 percent hydrated more frequently, and 75 percent checked more regularly on family members (Japan MOE and JMA 2021). Japan’s system has been incrementally improved over many years. Introducing effective heat wave early warnings is not trivial; it requires researching local heat wave–mortality relationships based on public health data, establishing partnerships between the city and national levels, and setting up monitoring and warning issuance mechanisms. However, the benefit-cost ratios of such warning systems are exceptionally favorable. Heat wave early warnings can save lives during heat waves and form the hub of broader heat action plans at the city level. STRATEGIC OPTION 5: PROTECT HEAT-EXPOSED WORKERS Exposure to heat stress varies by occupation. Workers in physically demanding outdoor occupations such as construction, road building and repairs, and waste picking are among the most exposed. However, many indoor jobs also involve great heat stress because of physical intensity, building characteristics, exposure to indoor heat sources (especially in manufacturing), and lack of air conditioning. Many such roles skew toward a female workforce, particularly in industries such as textiles, which are widespread in East Asia. The physiological limits of human heat endurance—established through studies in military, sports, and occupational health and safety—have motivated legislation UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 23 in many countries that limits workplace heat exposure based on Wet Bulb Globe Temperature (Kjellstrom, Holmer, and Lemke 2009). Although the devices required to monitor indoor heat stress are inexpensive, enforcement is often lacking. However, as illustrated through the work of organizations such as the C40 Cool Cities network, a wide range of actions can shield city residents from heat exposure during their working day. These can include increasing access to air conditioning through cooling centers accessible to street vendors and transportation workers. They can also take advantage of heat wave early warnings (see Strategic Option 4). In Ahmedabad, India, the city’s pioneering Heat Action Plan consulted heat-exposed workers and established protocols to be implemented in sectors such as construction during “orange” and “red” alerts. These protocols include distribution of drinking water and work stoppage from 12 p.m. to 4 p.m. (Knowlton et al. 2014). STRATEGIC OPTION 6: MAINSTREAM HEAT RISK REDUCTION INTO CITY INSTITUTIONS AND STRATEGIES Heat mitigation presents some unusual characteristics as a public policy challenge: Managing across physical scales. Making city landscapes cooler requires action across physical scales to change construction and design practices— from the selection of roof and wall materials to the design of buildings and their surroundings as well as incorporating “green” and “blue” elements in areawide plans. Utilizing diverse policy instruments. Heat stress interventions span a wide array of policy instruments—from direct public investments in school buildings or public plazas to “hard” legislative mandates (such as vehicle use limitations) and “soft,” bottom-up measures to incentivize and encourage action by building owners and residents (for example, tree stewardship schemes and public space design guides). Managing across agencies. The challenge of extreme heat in cities requires points of contact with an unusually large number of governmental agencies. From planning and parks departments to transit operators, and from health systems to meteorological agencies, effective action requires that many city and national entities work around the same table. An important step is to elevate extreme heat response to the senior official level. One option is to follow the growing number of cities that have appointed a chief heat officer. Because the actions needed to cool cities are highly fragmented and decentralized, elevating the issues and forging collective action through consultation and public awareness campaigns are key. For cities that are beginning to recognize UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 24 the need to act, a crucial first step might be to establish an extreme heat task force mandated to review evidence of extreme heat, gather and analyze data, consult affected groups, and recommend actions. Hot weather follows an annual seasonality. It follows that heat wave preparedness and response should also follow an annual seasonality—by strengthening preparedness in the run-up to the hot season, activating protocols when a heat wave is imminent, acting to protect the vulnerable during extreme heat events, and conducting after-action reviews before recommencing the cycle (figure O.8). Figure O.8 The Heat Wave Risk Management Cycle During heatwave: ■ Protect vulnerable people Heatwave After imminent: heatwave: ■ Raise awareness ■ Review and improve Before heatwave: ■ Strengthen preparedness SOURCE: World Bank elaboration based on Singh et al. 2020. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 25 CONCLUSION Extreme urban heat arising from the combination of the urban heat island effect and climate change represents a significant and growing drag on the economic competitiveness of many East Asian cities while also contributing to cities that are less healthy, livable, and inclusive. To rise to the urban heat challenge, leaders in East Asian cities should adopt a “Places, People, Institutions” framework that has three key elements—making places cooler, protecting vulnerable people, and mainstreaming heat into institutions. By acting now, East Asian city leaders can help make their cities not only cooler, but more prosperous, inclusive, and livable. REFERENCES Bressler, R. D., F. C. Moore, K. Rennert, and D. Anthoff. 2021. Clean Cooling Collaborative. 2022. “Million Cool Roofs “Estimates of Country Level Temperature-Related Challenge: Local Champions for a Global Movement.” Mortality Damage Functions.” Scientific Reports 11 (1): Clean Cooling Collaborative Blog, March 1. https://www. 20282. cleancoolingcollaborative.org/blog/million-cool-roofs- Burke, M., F. González, P. Baylis, S. Heft-Neal, C. Baysan, S. Basu, challenge-local-champions-for-a-global-movement/. and S. Hsiang. 2018. “Higher Temperatures Increase Davis, L., P. Gertler, S. Jarvis, and C. Wolfram. 2021. “Air Suicide Rates in the United States and Mexico.” Nature Conditioning and Global Inequality.” Global Climate Change 8 (8): 723–29. Environmental Change 69 (1): 102299. Burke, M., S. M. Hsiang, and E. Miguel. 2015. “Global Non-Linear DeJarnett, N., R. McDonald, and R. McDonald. 2016. “Planting Effect of Temperature on Economic Production.” Nature Healthier Air: An Assessment of the Potential for Trees 527 (7577): 235–39. to Reduce Air Pollution and Cool Cities to Enhance Chakraborty, T., A. Hsu, D. Manya, and G. Sheriff. 2019. Human Health.” Presentation at the American Public “Disproportionately Higher Exposure to Urban Heat in Health Association 2016 Annual Meeting & Expo, Denver, Lower-Income Neighborhoods: A Multi-City Perspective.” October 29–November 2. Environmental Research Letters 14 (10): 105003 Deuskar, C. 2022. “Beating the Heat: Measuring and Mitigating Cho, M. 2010. “The politics of urban nature restoration: The Extreme Heat in East Asian Cities.” Technical Working case of Cheonggyecheon restoration in Seoul, Korea.” Paper 1: Literature Review, World Bank, Washington, DC. International Development Planning Review 32, no. 2. Dodman, D., B. Hayward, M. Pelling, V. Castan Broto, W. Chow, Chan, E., Y. Yang, W. B. Goggins, J. J. Kim, and S. M. Griffiths. 2012. E. Chu, R. Dawson et al. 2022. “Cities, Settlements “A Study of Intracity Variation of Temperature-Related and Key Infrastructure.” In Climate Change 2022: Mortality and Socioeconomic Status among the Chinese Impacts, Adaptation, and Vulnerability. Contribution of Population in Hong Kong.” Journal of Epidemiology and Working Group II to the Sixth Assessment Report of the Community Health 66 (4): 322–27. Intergovernmental Panel on Climate Change, Geneva. Erell, E., D. Pearlmutter, and T. Williamson. 2012. Urban Microclimate: Designing the Spaces between Buildings. Oxfordshire, UK: Routledge. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 26 Estrada, F., W. J. Wouter Botzen, and R. S. J. Tol. 2017. “A Global Jones, N., N. Wijaya, J. Williams, Z. Boyce, V. Shandas, and Economic Assessment of City Policies to Reduce Climate C. Deuskar. 2023. “Mapping Urban Heat through a Change Impacts.” Nature Climate Change 7 (6): 403–6. Community-Based Mobile Campaign in Bandung City, ESMAP (Energy Sector Management Assistance Program). 2020. Indonesia.” Background paper for this report. “Primer for Cool Cities: Reducing Excessive Urban Heat— Kahn, M. E., and P. Li. 2019. “The Effect of Pollution and Heat on With a Focus on Passive Measures.” ESMAP Knowledge High Skill Public Sector Worker Productivity in China.” Series 031/20, World Bank, Washington, DC. Working Paper 25594, National Bureau of Economic Giridharan, R., and R. Emmanuel. 2018. “The Impact of Research, Cambridge, MA. Urban Compactness, Comfort Strategies and Energy Kjellstrom, T., I. Holmer, and B. Lemke. 2009. “Workplace Heat Consumption on Tropical Urban Heat Island Intensity: A Stress, Health and Productivity—An Increasing Challenge Review.” Sustainable Cities and Society 40 (2): 677–87. for Low and Middle-Income Countries during Climate Heilmann, K., and M. E. Kahn. 2019. “The Urban Crime and Heat Change.” Global Health Action 2 (1): 2047. Gradient in High and Low Poverty Areas.” Working Knowlton, K., S. P. Kulkarni, G. S. Azhar, D. Mavalankar, A. Jaiswal, Paper 25961, National Bureau of Economic Research, M. Connolly, A. Nori-Sarma et al. 2014. “Development Cambridge, MA. and Implementation of South Asia’s First Heat-Health Hong Kong SAR, China, Planning Department. 2015. “Urban Action Plan in Ahmedabad (Gujarat, India).” International Design Guidelines.” In Hong Kong Planning Standards Journal of Environmental Research and Public Health 11 and Guidelines, Hong Kong SAR, China: Planning (4): 3473–92. Department of Hong Kong. (accessed February 27, 2023), Konijnendijk, C. 2021. “The 3-30-300 Rule for Urban Forestry and https://www.pland.gov.hk/pland_en/tech_doc/hkpsg/ Greener Cities.” Biophilic Cities Journal 4 (2): 2. full/pdf/ch11.pdf. Lindsey, R., and L. Dahlman. 2021. “Climate Change: Global Hu, X., J. Wu, P. Chen, T. Sun, and D. Li. 2017. “Impact of Climate Temperature.” News article, Climate.gov, March 15. Variability and Change on Crime Rates in Tangshan, National Oceanic and Atmospheric Administration, China.” Science of The Total Environment 609 Washington, DC. (December): 1041–48. Manoli, G., S. Fatichi, M. Schläpfer, K. Yu, T. W. Crowther, N. Meili, IEA (International Energy Agency). 2019. “The Future of Cooling P. Burlando, G. G. Katul, and E. Bou-Zeid. 2019. “Magnitude in Southeast Asia.” Report, IEA, Paris. of Urban Heat Islands Largely Explained by Climate and ILO (International Labour Organization). 2019. Working on a Population.” Nature 573 (7772): 55–60. Warmer Planet: The Impact of Heat Stress on Labour Ng, E. 2009. “Policies and Technical Guidelines for Urban Productivity and Decent Work. Geneva: International Planning of High-Density Cities—Air Ventilation Labour Office. Assessment (AVA) of Hong Kong.” Building and Japan, MOE, and JMA (Ministry of the Environment and Japan Environment 44 (7): 1478–88. Meteorological Agency). 2021. “About the National Oke, T. R., G. Mills, A. Christen, and J. A. Voogt. 2017. Urban Launch of the Heat Alert” [in Japanese]. PowerPoint Climates. Cambridge: Cambridge University Press. document, Ministry of the Environment, Tokyo (accessed Park, R. J., A. P. Behrer, and J. Goodman. 2021. “Learning Is February 19, 2023), https://www.wbgt.env.go.jp/pdf/ Inhibited by Heat Exposure, Both Internationally and rma_doc/20210101/doc1.pdf. within the United States.” Nature Human Behaviour 5 (1): Jjemba, E. W., A. Kruczkiewicz, T. N. Quang, H. Nguyen, J. Faucet, 19–27. and R. Singh. 2020. “Applying Forecast Based Actions in Park, R. J., J. Goodman, M. Hurwitz, and J. Smith. 2020. “Heat and Rapidly Urbanizing Cities; The Case of Heatwaves in Hanoi, Learning.” American Economic Journal: Economic Policy Viet Nam.” Abstract no. NH035–07, American Geophysical 12 (2): 306–39. Union (AGU) Fall Meeting Abstracts, AGU, Washington, DC. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 27 Park, R. J., N. Pankratz, and A. P. Behrer. 2021. “Temperature, Seppänen, O., W. J. Fisk, and Q. H. Lei. 2006. “Effect of Workplace Safety, and Labor Market Inequality.” Temperature on Task Performance in Office Environment.” Discussion Paper No. 14560, Institute of Labor Economics Paper No. LBNL-60946, Ernest Orlando Lawrence (IZA), Bonn, Germany. Berkeley National Laboratory, Berkeley, CA. Pavanello, F., E. De Cian, M. Davide, M. Mistry, T. Cruz, P. Bezerra, Singh R., J. Arrighi, E. Jjemba, K. Strachan, M. Spires, and A. D. Jagu et al. 2021. “Air-Conditioning and the Adaptation Kadihasanoglu. 2020. “Heatwave Guide for Cities.” Cooling Deficit in Emerging Economies.” Nature Reference guide, International Federation of Red Cross Communications 12 (1): 6460. Red Crescent Climate Centre, Geneva. Popovich, N., and C. Flavelle. 2019. “Summer in the City Is Toloo, G., G. FitzGerald, P. Aitken, K. Verrall, and S. Tong. 2013. Hot, but Some Neighborhoods Suffer More.” New “Evaluating the Effectiveness of Heat Warning Systems: York Times, August 9. https://www.nytimes.com/ Systematic Review of Epidemiological Evidence.” interactive/2019/08/09/climate/city-heat-islands.html. International Journal of Public Health 58 (5): 667–81. Roberts, M. F. Gil Sander, and S. Tiwari, eds. 2019. Time to ACT: UN DESA (United Nations Department of Economic and Social Realizing Indonesia’s Urban Potential. Washington, DC: Affairs). 2017. World Population Prospects: The 2017 World Bank. Revision. New York: United Nations. Rossitti, G. 2022. “Hot, Wet and Poor? Climate Hazards USAID (United States Agency for International Development). Exposure Bias within Cities.” Unpublished working paper. 2020. “Heat and Urban Environmental Issues: Water Washington, DC: World Bank. Quality.” Quick Reference Guide, USAID, Washington, DC. Roloff, A., ed. 2016. Urban Tree Management: For the Sustainable Varquez, A. C. G., N. S. Darmanto, Y. Honda, T. Ihara, and M. Kanda. Development of Green Cities. Hoboken, NJ: Wiley- 2020. “Future Increase in Elderly Heat-Related Mortality Blackwell. of a Rapidly Growing Asian Megacity.” Scientific Reports Russo, S., J. Sillmann, and A. Sterl. 2017. “Humid Heat Waves at 10 (1): 9304. Different Warming Levels.” Scientific Reports 7 (1): 7477. Viguié, V., A. Lemonsu, S. Hallegatte, A.-L. Beaulant, C. Santamouris, M., L. Ding, F. Fiorito, P. Oldfield, P. Osmond, R. Marchadier, V. Masson, G. Pigeon, and J.-L. Salagnac. Paolini, D. Prasad, and A. Synnefa. 2017. “Passive and 2020. “Early Adaptation to Heat Waves and Future Active Cooling for the Outdoor Built Environment— Reduction of Air-Conditioning Energy Use in Paris.” Analysis and Assessment of the Cooling Potential of Environmental Research Letters 15 (7): 075006. Mitigation Technologies Using Performance Data from Weinberger, K. R., A. Zanobetti, J. Schwartz, and G. A. Wellenius. 220 Large Scale Projects.” Solar Energy 154: 14–33. “Effectiveness of National Weather Service Heat Alerts Sanz-Barbero, B., C. Linares, C. Vives-Cases, J. L. González, J. in Preventing Mortality in 20 US Cities.” Environment J. López-Ossorio, and J. Díaz. 2018. “Heat Wave and the International 116: 30–38. Risk of Intimate Partner Violence.” Science of the Total WHO (World Health Organization). 2018. “Heat and Health.” Environment 644: 413–19. Fact sheet, June 1, WHO, Geneva. https://www.who.int/ Satwiko, P. 1999. “Traditional Javanese Residential Architecture news-room/fact-sheets/detail/climate-change-heat-and- Designs and Thermal Comfort. A Study Using a health. Computational Fluid Dynamics Program to Explore, Wu, C. Y. H., B. F. Zaitchik, and J. M. Gohlke. 2018. “Heat Waves Analyse, and Learn from the Traditional Designs for and Fatal Traffic Crashes in the Continental United Thermal Comfort.” Doctoral dissertation, Te Herenga States.” Accident Analysis and Prevention 119: 195–201. Waka—Victoria University of Wellington, New Zealand. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 28 Zhao, Q., Y. Guo, T. Ye, A. Gasparrini, S. Tong, A. Overcenco, Zivin, J. S. G., Y. Song, Q. Tang, and P. Zhang. 2018. “Temperature A. Urban et al. 2021. “Global, Regional, and National and High-Stakes Cognitive Performance: Evidence from Burden of Mortality Associated with Non-Optimal the National College Entrance Examination in China.” Ambient Temperatures from 2000 to 2019: A Three- Working Paper No. 24821, National Bureau of Economic Stage Modelling Study.” The Lancet Planet Health 5 (7): Research, Cambridge, MA. e415–25. NOTES 1 The surface temperature is averaged increased by 0.9°C, to 24.3°C. reference period (2000–20) for at across land and water. The preindustrial 8 By contrast, average annual least three consecutive days. The period is defined as 1880–1900. Data precipitation has remained broadly optimistic climate change scenario are from Lindsey and Dahlman (2021). stable since 1958 for nontropical corresponds to the Intergovernmental 2 Throughout this report, temperatures East Asian cities overall, although there Panel on Climate Change (IPCC) Shared are generally referred to in degrees is evidence of more severe year-to-year Socioeconomic Pathway (SSP) 1-1.9 Celsius rather than Fahrenheit. variation over 2011–20 than in earlier scenario, under which global carbon A conversion table between Celsius and decades. dioxide emissions are curbed (cut to Fahrenheit can be found in the report’s 9 As discussed in more detail in chapter net zero by about 2050). Meanwhile, front matter. 2 of the main report, this research the pessimistic scenario corresponds 3 The methods underpinning the was undertaken by the World Bank in to the IPCC SSP3-7.0 scenario, under derivation of these temperature collaboration with Bandung Institute which global emissions maintain a differentials are described in more of Technology (Institut Teknologi stronger upward path before stabilizing. detail in the next section and chapter 1 Bandung) and CAPA Strategies using 13 It is also projected that these of the main report. a “citizen science” approach to gather neighborhoods will experience tropical 4 This 100-city sample includes 20 cities field measurements of ambient air nights for 80–90 percent of the year each from China and Indonesia; 13 from temperatures. These temperature in 2050 depending on the climate Viet Nam; 12 from the Philippines; differentials were found to be highest change scenario, where a tropical night 9 from Malaysia, 5 from Myanmar; in the evening, with a 7°C disparity is one for which the Wet Bulb Globe 4 each from Cambodia, Laos, Papua between the hottest and coolest Temperature remains above 25°C. New Guinea, and Thailand; and 1 each neighborhoods as measured between In the absence of cooling, tropical from Japan, the Republic of Korea, 5 p.m. and 6 p.m. Heat-trapping urban nights place the human body under Mongolia, Singapore, and Timor-Leste. materials such as asphalt roads, extreme stress. Particularly for old For more details about how this sample parking lots, walls, and corrugated iron people, young children, and those with was selected, see chapter 1, box 1.2, of roofs absorb solar energy during the health vulnerabilities, such nighttime the main report. day and release it at night. temperatures inhibit cooling, rest, 5 As shown in figure O.1, panel B, the 10 For evidence on other cities around the and recovery and are associated with mean five-year moving average UHI world see, among others, Chakraborty higher death rates. effect was stronger during the years et al. (2019), Popovich and Flavelle 14 Zivin et al. (2018) find that a 1 standard 2010–14. The reasons for this are (2019), and Rossitti (2022). deviation increase in exam-day unclear and would require further 11 The analysis described in this temperature (3.3°C) during the Chinese investigation that is beyond the scope paragraph was carried out for National College Entrance Examination of this report. this report by VITO NV, using its reduces a student’s test score by an 6 In addition to Singapore, the top UrbClim model. estimated 1.1 percent (0.096 standard 10 list includes 1 city in Indonesia, 5 in 12 For this analysis, a heat wave day deviations). This, in turn, reduces the Malaysia, and 3 in the Philippines. is defined as a day for which the probability of a student’s admission 7 Between 1958 and 2020, Hanoi’s temperature exceeds the 90th into a first-tier university by nearly average land surface temperature percentile temperature of a historical 2 percent. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 29 15 A worker may consider an extremely 2018). High temperatures further 18 More generally, in the sample of 100 hot and humid city to be unpleasant reduce water quality by enabling East Asian cities for which this report both because they dislike the heat the development of harmful algal estimates the strength of the UHI itself and/or because extreme heat blooms and toxic bacteria and the effect, a strong negative correlation leads to worse outcomes on various depletion of oxygen in urban water exists between the strength of this other dimensions of a city’s livability, bodies. Spikes in the concentrations effect and a city’s average level as discussed later in the Overview. of waterborne bacteria and viruses of vegetation (see chapter 1 of the An important area of future potential can trigger outbreaks of diseases main report). research would be how extreme heat including cholera, diarrhea, dysentery, 19 For an overview of the key challenges affects the migration decisions of hepatitis A, typhoid, and polio in urban tree management, see Roloff workers of different skill levels. (USAID 2020). (2016). 16 Heat-related illnesses include 17 ASEAN member countries include 20 For an overview of how tall buildings dehydration, heat cramps, and heat Brunei Darussalam, Cambodia, affect airflow, heat, and air pollution, stroke. Overheating also increases the Indonesia, Laos, Malaysia, Myanmar, see Oke et al. (2017) (chapter 4). likelihood of death from preexisting the Philippines, Singapore, Thailand, 21 For a detailed typology of conditions such as respiratory and Viet Nam. interventions for cooler buildings, see and cardiovascular diseases (WHO ESMAP (2020). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 30 CH C H A PT ER 1 THE URBAN HEAT ISLAND EFFECT AND THE EVOLUTION OF EXTREME HEAT ACROSS EAST ASIAN CITIES UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 31 KEY MESSAGES Because of the concentration of artificial surfaces combined with a lack of vegetation, East Asian cities suffer from a strong urban heat island (UHI) effect. From 2016 to 2020, cities were, on average, around 1.6–2.0 degrees Celsius (°C) hotter than their rural surroundings, rising to over 5.0°C hotter in the most extreme case. The UHI effect in East Asia’s cities is compounded by climate change-driven heat stress. Extreme heat events— measured as abnormally high temperatures relative to historical norms—were, on average, 23 times more frequent and 13 percent stronger in 2010–20 than during the 1970s. IIncreased severity of extreme heat varies widely across East Asian cities. The largest increase has occurred in already hot and humid tropical regions, highlighting the challenges that lower-income tropical cities face in heat mitigation and adaptation. The adverse impacts of extreme heat could be equally serious on nontropical cities, notably in the northern and southwestern parts of China, owing to a lack of adaptation. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 32 INTRODUCTION Rising temperature is a growing problem across East Asia1. And nowhere is this problem greater than in the region’s cities, where increasing average temperatures due to climate change are interacting with and compounding the so-called urban heat island (UHI) effect. This effect arises from, among other things, the concentration of impervious surfaces and relative lack of vegetation that are hallmarks of cities, especially those that are poorly planned and characterized by both slums and choking levels of traffic congestion. Indeed, travel from a less built-up part of the Bangkok Metropolitan Region to its central business district, and you will feel a change in atmosphere. The temperature difference used to be 4.4°C in 2000. Because of additional land cover modifications compounded by global warming, the difference had increased to about 5.8°C by 2020 (Pan et al. 2023).2 Extreme urban heat, driven by the UHI effect alongside global warming, is bad for the human body and mind. Those most at risk from heat stress include young children, the elderly, and the unwell, who have less capacity to adapt, as well as low-income populations who tend to live in low-quality housing with less access to air conditioning and other cooling solutions. Between 2000 and 2019, around 20 percent of heat-related global excess deaths occurred in East Asia—a total of just over 101,000 lives lost (Zhao et al. 2021). As later chapters of this report will show, the negative impacts of extreme heat in cities extend well beyond those on health. Extreme heat exposure also undermines the productivity of a city’s workers3 and leads to the failure of critical infrastructure and services,4 thereby acting as a drag on East Asia’s competitiveness at a time when its cities are already struggling with an unfavorable global economic environment. Indeed, it is projected that UHI intensification resulting from urban land expansions between 2015 and 2050 will increase air temperatures by 0.5°C–0.7°C on average globally. This is on top of the average 0.9°C–1.1°C increase that is projected to occur as a result of climate change (Huang et al. 2019).5 The combined impacts could result in a gross domestic product (GDP) loss of 1.7 percent for the global median city (Estrada, Wouter Botzen, and Tol 2017).6 In addition, extreme heat events can increase the risks of climate change-related hazards. In humid conditions, extreme heat increases the risk of flooding by causing more frequent extreme rainfalls. Conversely, in dry conditions, extreme heat enhances evaporation, contributing to water shortages and droughts. Such hot and dry conditions can, in turn, create massive bushfires as in Australia in the summer UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 33 of 2019/20, adding large amounts of carbon dioxide (CO2) to the atmosphere (Robinson et al. 2021). Despite its profound impacts, extreme heat has seldom attracted as much attention from East Asia’s policy makers as have other types of hazards such as floods and tropical cyclones, which tend to cause more obviously apparent damage to people and property. In a warming world, however, rapid urbanization in East Asia means that extreme heat will increasingly threaten city competitiveness as well as human health. The good news is that city officials can undertake concrete actions and policies to promote mitigation and adaptation to extreme urban heat. Although they may not be able to influence global climate change, they can reduce the severity of the UHI effect by, among other things, promoting better urban planning and design. As explored in chapter 4, they can also promote adaptation to extreme heat through a variety of measures. Against this backdrop, this chapter aims to inform the discussions of adverse impacts and policies in later chapters of this report by describing detailed trends of extreme urban heat for East Asian cities. Drawing on a sample of 100 major cities in the region, this chapter provides evidence that the average UHI intensity across the cities has been strong—raising temperatures by around 1.6–2.0°C compared with surrounding rural areas. It also confirms, based on a broader sample of over 3,000 cities (box 1.1), that extremely higher than normal temperatures in East Asia’s cities have become increasingly frequent and intense over the past six decades. The severity of extreme urban heat, however, varies widely across the region. And the UHI intensity depends on a variety of factors, including background climates and a city’s urban form. These findings signpost the importance of designing locally tailored interventions for heat mitigation and adaptation underpinning the policy discussion in chapter 4. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 34 BOX 1.1 THE SAMPLE OF EAST ASIAN CITIES: AN OVERVIEW East Asian countries vary widely in how its entire extent. As such, it identifies cities are in either tropical or subtropical they officially define cities. To overcome 3,033 cities in the East Asia region.a zones between latitudes 35° south and the problems of lack of comparability, this 35° north, of which around 95 percent chapter defines cities using a globally Over 90 percent of identified cities are have somewhat wet climates with annual consistent definition—the degree of from middle-income countries (figure B1.1.1, total precipitation of 750 millimeters or urbanization (Dijkstra et al. 2021). This panel A). China accounts for 61 percent more (figure B1.1.1, panel B). This implies definition classifies a settlement as a city of the cities, followed by Indonesia that many East Asian cities and their if it has a population of at least 50,000 (13 percent), Viet Nam (5.4 percent), residents are exposed to hazardous and a population density of at least 1,500 and Myanmar (4.2 percent). In terms of humid heat. people per square kilometer throughout baseline climates, nearly 75 percent of the Figure B1.1.1 Distribution of East Asian Cities, by Country Income Classification and Climate Zone A BY COUNTRY INCOME CLASSIFICATION B BY CLIMATE ZONE Humid 1,067 High income 174 tropical (35.1%) (5.7%) Humid 1,029 subtropical (33.9%) Upper-middle 1,929 income Subhumid 358 (63.6%) temperate (11.8%) 238 Lower-middle 840 Arid (7.9%) income (27.7%) Humid 219 temperate (7.2%) 90 Low income Subhumid 122 (3.0%) subtropical (4.0%) 0 500 1,000 1,500 2,000 0 300 600 900 1,200 No. of cities No. of cities NOTE: Panel A is based on World Bank country income classifications in fiscal a. The sample excludes Australia, New Zealand, year 2022/23. In panel B, the climate zone of each city is defined based on the and Pacific island countries. To obtain a full picture combination of two parameters: aridity and latitude. Aridity, based on a city’s annual of the overall trend in the region, it includes the total precipitation averaged over 1958–2020 is classified as arid (less than 500 Democratic People’s Republic of Korea even though it is not a World Bank member economy. millimeters), subhumid (500–750 millimeters), and humid (750 millimeters or more). Latitude in absolute terms classifies cities as tropical (0°–23.5° north), subtropical (23.5°–35° north), and temperate (35° and farther north). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 35 DEFINING EXTREME URBAN HEAT The Glossary of Meteorology defines a heat wave as “a period of abnormally and uncomfortably hot and usually humid weather” (Glickman 2000). As this definition implies, various combinations of measures and thresholds that quantify the duration of abnormal and uncomfortable temperature as well as humidity can be applied to identify extreme heat events. To measure temperature thresholds, for example, studies generally use daytime maximum and/or nighttime minimum as well as the daily average temperature (Meehl and Tebaldi 2004). For the duration of the event, the temperature should exceed an agreed-upon threshold for at least one day but generally for at least several consecutive days (Glickman 2000). The abnormality and discomfort thresholds also depend on the normal conditions in the area of interest. In Canada, for example, the threshold for Prince Edward Island with a mild climate is 27°C, whereas a threshold of 35°C is applied to Southeastern British Columbia.7 Nonetheless, three broad definitions of an extreme heat event are apparent in the literature: (a) absolute temperature; (b) extreme heat as an absolute “anomaly” from a city’s long-run average temperature; and (c) extreme heat as a relative “anomaly” from a city’s long-run average (see annex 1A). These definitions capture a person’s perception of an extremely hot day in different ways. Consistent with the absolute temperature definition, someone may perceive a day to be extremely hot because its temperature approaches, say, 40°C. On another day, per the absolute anomaly definition, the same person may feel extremely hot because the temperature is, say, 10°C higher than is historically typical for that time of year. Finally, as implied by the relative anomaly definition, even a smaller temperature anomaly (say, 5°C higher) may feel extremely hot to someone who has lived in a city where the temperature is normally stable throughout the year. By contrast, an absolute anomaly of the same size may have little impact on heat perception in another city with a historically more volatile climate. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 36 WHAT DRIVES EXTREME URBAN HEAT? LOCAL WARMING DRIVEN BY THE URBAN HEAT ISLAND EFFECT Urban heat islands are urbanized areas where temperatures are higher than they otherwise would be if the areas were rural. The “UHI effect” is a local negative environmental externality that may expose urban residents to extreme heat events more often than their observationally equivalent rural counterparts by artificially elevating a city’s long-run average temperature (Houghton 2015). The primary cause is the conversion of natural land to urban land, which reduces a city’s cooling capacity available through evapotranspiration and convection while trapping more heat on artificial surfaces (Zhao et al. 2014).8 Rapid urbanization additionally contributes to the UHI effect because it increases the amount of heat released from human activities such as motor vehicle uses and industrial processes (Li et al. 2020). This highlights the opportunities for local policy makers to target heat mitigation and adaptation through urban planning and local policies. GLOBAL WARMING DRIVEN BY CLIMATE CHANGE Since the 1950s, cities across the globe have experienced large changes in the weather patterns that define their local climates. Global warming, one of the key symptoms of climate change, has been contributing to increased long-run average temperatures in cities and, at least so long as the average temperatures continue to rise, is driving larger absolute and relative temperature deviations from the historical norms to which people have acclimated.9 A small increase in average temperature can have a drastic effect on both the frequency and intensity of extreme heat events. According to the Intergovernmental Panel on Climate Change (IPCC), hot extremes, which used to occur only once in every 50 years during the preindustrial period (1850–1900), now occur every 10 years.10 Furthermore, global warming of 1.5°C and 4.0°C, respectively, are projected to lead hot extremes to occur 8.6 times and 39.2 times in every 50 years and to raise their intensities by 2.0°C and 5.3°C (IPCC 2021). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 37 Global warming is a global negative externality, meaning the burning of fossil fuel somewhere in the world can cause adverse impacts elsewhere. Hence, city leaders cannot mitigate it by acting in isolation as they might mitigate the UHI effect through local planning and policies. But they can take steps to help make their cities and residents more resilient to rising temperatures—both through immediate-term adaptation measures and longer-term planning to reduce future risks. INTERACTIONS BETWEEN UHI EFFECTS AND CLIMATE CHANGE Although the interactions between UHI effects and climate change-induced extreme heat events are not yet fully understood (Souverijns et al. 2022), research shows that, in the face of global warming, city cores are heating up faster than adjacent rural areas because of the UHI effect (Liu et al. 2022). This suggests that the UHI effect can interact with climate change to increase the likelihood of extreme urban heat. Also, it is widely accepted that the surface UHI intensity normally increases with city size, because larger cities tend to be covered by a larger built-up area while emitting a larger amount of anthropogenic heat (Imhoff et al. 2010). Given that East Asian countries are projected to add roughly 141 million inhabitants to their cities by 2050,11 heat stress posed by the UHI effect in combination with global warming will likely increase. HOW IS AN EXTREME URBAN HEAT EVENT MEASURED? This chapter’s analysis draws on two measures of extreme urban heat derived from the local and global drivers discussed in the previous section. Specifically, local and global drivers are measured by the intensity of UHI effect and by the frequency and intensity of relative temperature anomalies, respectively. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 38 LOCAL DRIVER: INTENSITY OF THE UHI EFFECT Traditionally, the intensity of the UHI effect, defined as the difference in mean temperatures between urban and surrounding rural areas, was based on in situ readings of air temperatures at 2 meters above ground level (Liu et al. 2022). With recent advances in space technology, however, it is now feasible to quantify spatially detailed land surface temperatures (LST) for the entire globe through the processing of satellite imagery using remote sensing techniques. The estimates of UHI intensity used in this chapter were derived by 4 Earth Intelligence (4EI) in support of the World Bank12. Bringing LST images together with global land cover images, 4EI classified each pixel in the LST images in and around a sample of 100 East Asian cities (box 1.2) as either urban or nonurban.13 In delineating the location and size of the nonurban zone surrounding a city, three buffer zones with distances of 2, 5, and 10 kilometers were iteratively created around the cluster of urban pixels extracted from the land cover classification. The intensity of a city’s UHI effect was then calculated as the yearly average of daytime and nighttime UHI intensities for each of the years from 2002 to 2020, where UHI intensity is the difference in LSTs averaged across urban and nonurban pixels. BOX 1.2 HOW WERE THE 100 CITIES SELECTED? For satellite mapping of UHI effects 2. The only national capital unselected Seventy-seven cities are in tropical across cities and over time, this chapter in step 1—Naypyidaw, Myanmar—was subregions, while 23 are in temperate focuses on a sample of 100 East Asian manually added to the sample. subregions. These 100 cities have an cities. This sample was constructed in 3. The remaining 54 cities were selected aggregate population of around 417 million three steps: from low- and middle-income countries, people, covering 41.3 percent of East in descending order of population, with Asia’s urban population.a 1. After first excluding Brunei Darussalam, a cap of 20 cities per country. the Democratic People’s Republic of A. Based on population data from the Korea, and Pacific island countries, The final sample includes 20 cities each European Commission’s Global Human the four most populous cities in each from China and Indonesia; 13 from Viet Settlement (GHS) Urban Centre Database low- and middle-income country in Nam; 12 from the Philippines; 9 from R2019 (https://ghsl.jrc.ec.europa.eu/ East Asia were selected alongside the Malaysia; 5 from Myanmar; 4 each from ghs_stat_ucdb2015mt_r2019a.php). single most populous city in each of Cambodia, the Lao People’s Democratic the region’s high-income countries. Republic, Papua New Guinea, and This step identifies 45 cities. Thailand; and 1 each from Japan, the Republic of Korea, Mongolia, Singapore, and Timor-Leste. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 39 GLOBAL DRIVER: RELATIVE TEMPERATURE ANOMALIES In the climate literature, an extreme heat event in a city is often defined as an “absolute” heat anomaly—that is, based on how many degrees Celsius the temperature exceeds its long-run historical average (Dell, Jones, and Olken 2014). This definition may suffice for comparisons of the evolving patterns of extreme heat events across a few cities with similar background climates. Given that this report focuses on the impacts of extreme urban heat with a broad geographic coverage, however, this chapter’s analysis focuses on relative instead of absolute temperature anomalies in measuring extreme urban heat events (a calculation further described in annex 1A). Hence, for any given city, an extreme heat event for any given month is defined in terms of the number of standard deviations of temperature above a city’s own long-run average temperature for the same month.14 The city-specific long-run average and the standard deviation for each month are calculated for the base period January 1958–December 1999.15 When the measure takes on a value of at least +2 standard deviations, a temperature anomaly is considered an extreme heat event because, under normal circumstances, this would be a highly unusual event that occurs less than 2.5 percent of the time and hence comes as a surprise to society at large. The idea behind the choice is that people adapt to local climatic conditions while updating their expectations based on experience over time. Hence, human activities are more likely to be affected by temperatures that are surprisingly higher than long- run averages—for which people are unprepared—rather than by high temperatures per se (Thomson et al. 2022; Tol 2021).16 Research also shows a distinct geographical pattern of future extreme heat events in relation to a specific atmospheric circulation pattern in the background (Meehl and Tebaldi 2004). This implies that, for a cross-city analysis on a broader geographic scale, relative anomalies may be more appropriate to account for diverse local climatic conditions. At the same time, it is important to recognize that, based on the relative anomaly definition, extreme heat may occur across a wide range of absolute background temperatures with a variety of adverse impacts. In the United States, for example, warmer winters, reduced snowfalls, and shorter snow seasons driven by global warming translated into reduced revenues and job losses in winter tourism industries, thereby reducing people’s income and government tax revenues (Burakowski and Magnusson 2012). On the other hand, certain impacts of heat will only manifest themselves at high absolute temperatures, such as melting power transmission lines, railroad tracks, and airport runways. This suggests that there is no one single “correct” definition or threshold of an extreme heat event. For this reason, UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 40 the analysis of relative anomalies will be accompanied by that of absolute temperatures, which will be further complemented by the analysis of relative anomalies across cities in different climate zones.17 In addition to absolute background temperatures, the duration of excessive heat, nighttime temperatures, and the local level of air pollution provide important context for understanding potential adverse impacts of extreme urban heat. Prolonged exposure to anomalous temperatures, driven by a longer extreme heat event or a nighttime heat stress, can trigger sleep deprivation and cardiovascular stress with direct and indirect health impacts (Deschenes and Moretti 2009; He et al. 2022; Obradovich et al. 2017). Moreover, extreme heat and air pollution—which often co-occur—tend to compound the adverse health impacts of each on the other, increasing the risk of mortality, on days hit by both, significantly beyond the sum of their individual risks (Rahman et al. 2022).18 Although the evolving patterns of these dimensions are not analyzed in this regional study, they certainly warrant future research. When it comes to mitigation and adaptation measures, humidity, not just temperature, matters. This is because the ability of both the human body and cities to efficiently dissipate heat depends on the combined level of heat and ambient humidity (Manoli et al. 2019). To highlight the unique challenges posed by humid heat in the tropics (box 1.3),19 we further analyze the evolution of relative temperature anomalies across cities with different levels of humidity.20 UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 41 BOX 1.3 HUMIDITY: A HIDDEN THREAT TO THERMAL COMFORT The urban heat island (UHI) effect and The combined impact of heat and with the evaporation of sweat (UNEP global warming have often been measured humidity is felt most acutely in cities 2021). Exposure to humid heat not only in terms of temperature alone. However, in humid tropical regions, where has adverse effects on human health recent research shows that factoring nearly 40 percent of East Asia’s urban (Raymond, Matthews, and Horton 2020) in humidity alongside heat makes the population reside. Cities can cool down but also leads to significant productivity temperature increase associated with through the processes of convection losses for outdoor workers. Parsons et climate change since 1980 nearly twice (the cycle of heat release to, and cool al. (2022), for example, estimate that as bad as previously calculated (Song et air from, the atmosphere) and the globally, annual labor productivity al. 2022). Based on temperature alone, evapotranspiration of natural surfaces. losses driven by humid heat can amount the 1980–2019 increase in global surface Under humid conditions, neither process to roughly US$2.1 trillion (in constant air temperature appears to be about works as efficiently as under drier 2017 US dollars adjusted for purchasing 0.79 degrees Celsius (°C). If humidity is conditions, however. Indeed, studies power parity).b factored in, surface equivalent potential suggest that humid cities experience temperature, which is an integrated stronger UHI effects and heat waves a. This relationship is based on the measure of surface air temperature and (Manoli et al. 2019; Russo, Sillmann, and increase in the water-holding capacity humidity, increased by about 1.48°C on Sterl 2017). of the atmosphere specified by the average globally and by about 4°C in the Clausius–Clapeyron relation (Robinson tropics. This is because, in a warming Higher humidity affects not only cities. et al. 2021). world, the air can hold more moisture— Human bodies, likewise, cannot cool nearly 7 percent more for every added themselves down efficiently when b. This estimate is based on the 1°C.a When the moisture condenses, this extreme heat is compounded by high construction, as well as the agriculture, then releases heat. Moreover, water humidity. Humans can self-regulate body forestry, and fisheries sectors— vapor itself is a heat-trapping gas that temperatures through sweating. However, that is, both urban and nonurban exacerbates climate change. a high level of ambient humidity interferes economic activities. UHI EFFECTS MAKE EAST ASIAN CITIES, ON AVERAGE, 1.6–2.0°C WARMER THAN RURAL SURROUNDINGS STRONG AVERAGE UHI INTENSITIES During 2002–20, the average intensity of the UHI effect across the sample of 100 East Asian cities increased modestly with minimal fluctuations around the trend. Hence, regardless of the distance used to define the buffer area of each city, the five-year moving average shows that the intensity slightly increased from 2005 UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 42 to circa 2014, then marginally declined afterward (figure 1.1, panel A).21 When rural surroundings are delineated conservatively based on a 2 kilometer buffer, urban areas are, on average, around 1.6°C hotter than surrounding rural areas over 2016–20. Relaxing the buffer sizes to 5 kilometers and 10 kilometers results in stronger average UHI intensities at around 1.8°C and 2.0°C, respectively. During the same period, most cities experienced estimated UHI effects below 3.0°C, as indicated by the third quartile of UHI intensity. Across the 100 cities, however, wide variations exist, with the intensity ranging from around zero to 5.3–6.6°C depending on the choice of buffer distance (figure 1.1, panel B). Figure 1.1 Trends in the average UHI Intensity and Dispersion across 100 East Asian Cities A AVERAGE UHI INTENSITY, 2002–20 B DISPERSION OF UHI INTENSITY, 2016–20 2.5 7 6 2.0 5 UHI intensity (°C) UHI intensity (°C) 1.5 4 3 1.0 2 0.5 1 0 0 2004 2006 2008 2010 2012 2014 2016 2018 −1 2km 5km 10km 2km 5km 10km SOURCE: World Bank calculations based on estimates of urban heat island (UHI) intensity produced by 4 Earth Intelligence, https://www.4earthintelligence.com/. NOTE: In both panels, UHI intensity is estimated by averaging the daytime and nighttime differences in land surface temperatures between urban and rural areas based on buffer distances of 2, 5, and 10 kilometers (km), where each buffer distance defines the location and size of the nonurban zone around the cluster of urban pixels extracted based on the land cover classification. In panel A, the solid and dotted lines show the regional averages of UHI intensity (measured as the five-year moving average of the intensities) across 100 East Asian cities over time and the best fitting straight line of the regional averages, respectively. In panel B, the bottom and top ends of each black line indicate the minimum and maximum values of UHI intensity, respectively. The bottom of the box, the border of two colors, and the top of the box, respectively, depict the first, second (median), and third quartiles of UHI intensity. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 43 Cities with the most severe UHI effects generally belong to Malaysia and the Philippines (map 1.1). These countries, along with Indonesia and Singapore, dominate the list of top 10 cities in terms of estimated UHI intensity—with the average intensity across the 10 cities, based on a 5 kilometer buffer, ranging from 3.5°C to 5.9°C.22 On the other hand, cities in the continental portion of Southeast Asia—those located in Cambodia, Laos, Myanmar, Thailand, and Viet Nam—generally exhibit less severe UHI intensities, in the range of 0.1–2.4°C. The same is true for cities in higher latitudes, including most Chinese cities and Seoul, Republic of Korea. Tokyo is an exception in that its average UHI intensity was around 3.3°C between 2016 and 2020—higher than in other cities around the same latitude. Map 1.1 Average UHI Intensity in 100 Selected East Asian Cities, 2016–20 Urban heat island intensity (Average over 2016–20) .0 .0 .0 .0 .9 –1 –2 –3 –4 –5 00 01 01 01 01 0. 1. 2. 3. 4. SOURCE: World Bank calculations based on the urban heat island (UHI) intensity data from 4 Earth Intelligence, https://www.4earthintelligence.com/. NOTE: The map presents the five-year (2016–20) averages of UHI intensity, estimated by averaging the daytime and nighttime differences in land surface temperatures between urban and rural areas based on the 5 kilometer buffer. Regarding selection of the 100 cities, see box 1.2. IDN = Indonesia; KHM = Cambodia; MMR = Myanmar; MYS = Malaysia; PHL = Philippines. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 44 DIVERGING UHI INTENSITIES WITHIN A CITY AND DURING THE DAY The above averages, however, mask considerable variations in UHI intensities across neighborhoods within individual cities as well as across times of the day. Hence, even within cities with relatively low average UHI intensities, some neighborhoods could be disproportionately exposed to stronger UHI effects. Bangkok provides such an example. During 2000–20, the city was, on average, only 1.4°C warmer than its rural surroundings, where the average temperature was 33.5°C.23 However, residents in the city’s hottest neighborhood had endured the average temperature of 46.3°C during the same period. Relatedly, as chapter 2 shows, the temperature difference between the hottest and the coolest neighborhoods in Bandung, Indonesia, reaches 7°C in the early evening. Also, the UHI intensity is usually stronger at night than during the day because the heat stored in the built environment is gradually released after sunset (Li et al. 2020). Because this chapter’s measure of UHI intensity is constructed by averaging the daytime and nighttime intensities across all pixels in a city throughout the year, it inevitably understates the intensity for some neighborhoods or at night while overstating others or during the day. WHILE TEMPERATURES INCREASED EVERYWHERE, RAINFALLS AMPLIFIED IN THE TROPICS RISING ABSOLUTE TEMPERATURES Previous studies based on satellite measurements or the analysis of observations from weather stations worldwide confirm that, overall, Earth’s temperature has risen by around 1.1°C since the 1880s (Woetzel et al. 2020). An analysis based on the broader sample of 3,033 East Asian cities (box 1.1) shows that, overall, mean annual temperature across the cities increased by around 1.0°C—from 17.8°C to 18.8°C— between 1958 and 2020 (figure 1.2).24 UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 45 Figure 1.2 Evolution of Average Annual Temperature in East Asian Cities, 1958–2020 y = 0.02x + 17.346 R² = 0.8434 19.5 Average annual temperature (°C) 19.0 18.5 18.0 17.5 17.0 0 5 70 75 0 5 10 15 0 5 00 05 20 8 9 6 8 9 6 20 20 20 20 20 19 19 19 19 19 19 19 19 Annual average 5-year moving average Fitted (5-year moving average) SOURCE: World Bank calculations based on Climatology Lab, TerraClimate, https://www.climatologylab.org/ terraclimate.html; European Commission, Global Human Settlement (GHS) Urban Centre Database R2019, https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php. NOTE: The region’s mean annual temperature is calculated by averaging the mean annual temperatures of the 3,033 cities in the full sample (described in box 1.1), by year. The equation and the R-squared value are based on the fitted line of the five-year moving averages. COOLER CITIES WARMING FASTER AND HOTTER CITIES HEATING UP FURTHER Although temperatures have increased everywhere in the region, the magnitude of increase has varied across cities. Whereas Ulaanbaatar, Mongolia, which has a cold arid climate, experienced a sharp increase of 1.7°C in its mean annual temperature (from −1.6°C to +0.1°C) during 1958–2020 (figure 1.3, panel A), the increase in a humid tropical city such as Hanoi, Viet Nam, was below 1.0°C—increasing from 23.4°C to 24.3°C (figure 1.3, panel B). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 46 Figure 1.3 Evolution of Mean Annual Temperature in Selected East Asian Cities, 1958–2020 A ULAANBAATAR, MONGOLIA B HANOI, VIET NAM y =0.031x - 2.076 R² = 0.676 y = 0.015x + 23.188 R² = 0.765 2 26 Mean annual temperature (°C) Mean annual temperature (°C) 1 25 0 24 −1 23 −1 22 −3 21 0 5 0 5 0 5 0 5 0 5 10 15 0 0 5 0 5 0 5 0 5 0 5 10 15 0 200 200 200 200 202 202 198 198 196 199 196 199 198 198 196 199 196 199 197 197 197 197 20 20 20 20 Annual average 5-year moving average Fitted (5-year moving average) SOURCE: World Bank calculations based on Climatology Lab, TerraClimate, https://www.climatologylab.org/ terraclimate.html; European Commission, Global Human Settlement (GHS) Urban Centre Database R2019, https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php. NOTE: The equations and the R-squared values are based on the fitted lines of the five-year moving averages. More generally, increases in mean annual temperature tend to be sharper for cities with arid or semiarid temperate climates (such as Beijing and Ulaanbaatar) than for those with humid tropical climates (such as Hanoi and Manila) (figure 1.4).25 Given the high absolute mean temperatures, however, even relatively modest temperature rises could be stressful for tropical cities. Indeed, the effect of higher temperatures on economic outcomes tends to be nonlinear such that the negative impacts of a given temperature increase are larger for cities with a higher baseline temperature than for those with a lower baseline temperature (see chapter 3). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 47 Figure 1.4 Pace of Temperature Increases in East Asian Cities, 1958–2020 2.5 y = 0.002x² - 0.074x + 1.736 (degrees Celsius, 1958–2020) 2.0 R² = 0.247 Temperature increase 1.5 1.0 0.5 0.0 −5 0 5 10 15 20 25 30 Initial mean annual temperature (degrees Celsius, 1958–62) SOURCE: World Bank calculations based on Climatology Lab, TerraClimate, https://www.climatologylab.org/ terraclimate.html; European Commission, Global Human Settlement (GHS) Urban Centre Database R2019, https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php. NOTE: The initial mean annual temperature (x-axis value) and the temperature increase (y-axis value) are calculated as the five-year average of mean annual temperatures during 1958–62 and the increase from the initial value to the five-year average of mean annual temperatures during 2016–20, respectively, for each of 3,033 East Asian cities in the sample (described in box 1.1). INCREASING RAINFALL TOTALS AND VARIABILITY IN THE TROPICS Although temperature has risen across East Asian cities by an average of 1.0°C since 1958, the average of total annual precipitation has remained relatively stable, albeit with a slight upward trend (figure 1.5, panel A). Over time, however, cities have experienced more severe variations in their levels of total annual precipitation, which affect humidity and thus, in combination with temperature, thermal comfort. Hence, the variation—measured by the absolute values of the residuals from the linear trend of total annual precipitation across cities—has become larger over time (figure 1.5, panel B). This increased variability around the trend may pose challenges for precipitation forecasting and hence for preparedness for weather extremes. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 48 Figure 1.5 Evolution of Total Annual Precipitation in East Asian Cities, 1958–2020 A TOTAL PRECIPITATION B VARIATIONS IN TOTAL PRECIPITATION y = 0.66x + 1,384.36 R² = 0.14 y = 0.9775x - 1303.9 R² = 0.1141 1,600 800 Residual (in absolute terms) 1,500 700 Millimeters 1,400 600 1,300 1,200 500 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 2020 2020 2020 2020 2020 2020 2020 2020 total annual precipitation 5-year moving average Fitted (5-year moving average) SOURCE: World Bank calculations based on Climatology Lab, TerraClimate, https://www.climatologylab.org/terraclimate.html; European Commission, Global Human Settlement (GHS) Urban Centre Database R2019, https://ghsl.jrc.ec.europa.eu/ghs_stat_ ucdb2015mt_r2019a.php. NOTE: In panel A, the equation and the R-squared value are based on the fitted line of the five-year moving averages. In panel B, variations are calculated as the absolute values of the residuals from the linear trend presented in panel A. Once again, however, this regional trend conceals wide disparities across cities with different baseline climates. Overall, cities with the most humid conditions—notably, those in the tropics—have had increasing total annual precipitation, whereas others (except in humid subtropical regions) have exhibited the opposite trend. This can be seen by comparing panel B of figure 1.6 with the other panels in the same figure. This suggests that the already humid climates of East Asia’s tropical cities are becoming even more humid. The trends in figure 1.6 also show considerable interannual variation in total annual precipitation in tropical regions, particularly in the latest decade. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 49 Figure 1.6 Evolution of Total Annual Precipitation in East Asian Cities, by Climate Zone, 1958–2020 A ARID ZONES B HUMID TROPICAL ZONES 800 2,600 y = -0.30x + 359.23 R² = 0.08 y = 1.94x + 2,107.87 R² = 0.22 600 2,400 Millimeters Millimeters 400 2,200 200 2,000 0 1,800 0 5 0 5 0 5 0 5 0 5 10 15 0 0 5 0 5 0 5 0 5 0 5 10 15 0 200 200 200 200 202 202 198 198 196 199 196 199 198 198 196 199 196 199 197 197 197 197 20 20 20 20 C SUBHUMID SUBTROPICAL ZONES D HUMID SUBTROPICAL ZONES 1,100 1,700 y = -0.56x + 679.87 R² = 0.08 y = 0.56x + 1,261.07 R² = 0.06 900 1,500 Millimeters Millimeters 700 1,300 500 1,100 300 900 0 5 0 5 0 5 0 5 0 5 10 15 0 0 5 0 5 0 5 0 5 0 5 10 15 0 200 200 200 200 202 202 198 198 196 199 196 199 198 198 196 199 196 199 197 197 197 197 20 20 20 20 E SUBHUMID TEMPERATE ZONES F HUMID TEMPERATE ZONES 1,100 1,700 y = -0.94x + 628.48 R² = 0.22 y = -0.73x + 1,180.77 R² = 0.04 900 1,500 Millimeters Millimeters 700 1,300 500 1,100 300 900 0 5 0 5 0 5 0 5 0 5 10 15 0 0 5 0 5 0 5 0 5 0 5 10 15 0 200 200 200 200 202 202 198 198 196 199 196 199 198 198 196 199 196 199 197 197 197 197 20 20 20 20 Annual average 5-year moving average Fitted (5-year moving average) SOURCE: World Bank calculations based on Climatology Lab, TerraClimate, https://www.climatologylab.org/terraclimate.html; European Commission, Global Human Settlement (GHS) Urban Centre Database R2019, https://ghsl.jrc.ec.europa.eu/ghs_stat_ ucdb2015mt_r2019a.php. NOTE: The equations and the R-squared values are based on the fitted line of the five-year moving averages. Collectively, the panels represent calculations across the full sample of 3,033 East Asian cities (see box 1.1). For each city, the climate zone is defined based on the combination of two parameters: aridity and latitude. Aridity, as determined by annual total precipitation (averaged over 1958–2020), is classified as either “arid” (less than 500 millimeters); “subhumid” (500–750 millimeters); or “humid” (750 millimeters or more). Latitude in absolute terms classifies cities as “tropical” (0°–23.5° north), “subtropical” (23.5°–35° north), and “temperate” (35° and farther north). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 50 EXTREME HEAT EVENTS BECAME INCREASINGLY SEVERE, AND MORE SO IN THE TROPICS This section highlights the increasing effect of climate change on the frequency and intensity of abnormal heat events, or “temperature anomalies,” over the past six decades for East Asian cities. Again, a temperature anomaly is defined relative to a city’s own long-run mean temperature while considering its historical variability for the second half of the twentieth century (1958–99). INCREASING OVERALL FREQUENCY AND INTENSITY OF EXTREME HEAT EVENTS Overall, East Asian cities experienced an average of only 0.1 extremely hot months per year in the 1970s, which increased to an average of 2.1 months per year during 2010–20 (figure 1.7, panel A). Put differently, extreme heat events increased from under three days per year to over two months per year—a 23-fold increase in four decades. Over the same period, extreme heat events also intensified. The average extreme heat event recorded a temperature that was, on average, around 2.2 standard deviations above a city’s own long-run average in the 1970s. By 2010–20, however, the average intensity had increased to almost 2.5 standard deviations (figure 1.7, panel B). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 51 Figure 1.7 Frequency and Intensity of Extreme Heat in East Asian Cities, by Decade (1958–69 to 2010–20) A FREQUENCY B INTENSITY 2010–20 2.08 2010–20 2.49 2000s 1.33 2000s 2.46 1990s 0.63 1990s 2.37 1980s 0.19 1980s 2.21 1970s 0.09 1970s 2.22 1958–69 0.15 1958–69 2.36 0 0.5 1 1.5 2 2.5 2.1 2.2 2.3 2.4 2.5 Average no. of months per year Average size of anomalies (standard deviations) SOURCE: World Bank calculations based on Climatology Lab, TerraClimate, https://www.climatologylab.org/terraclimate.html; European Commission, Global Human Settlement (GHS) Urban Centre Database R2019, https://ghsl.jrc.ec.europa.eu/ghs_stat_ ucdb2015mt_r2019a.php. NOTE: An “extreme hot month” is one in which a city’s temperature for that month is at least 2 standard deviations higher than the month’s city-specific historical norm, where the historical norm is calculated over January 1958–December 1999. Frequency is calculated as the number of extreme hot months per year and intensity as the average size of the anomaly variable during an extreme heat event—which is defined as a set of consecutive extreme hot months. Panel A presents the average frequency, and panel B the average intensity for each time period, across the full sample of 3,033 East Asian cities (see box 1.1). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 52 WIDE VARIATIONS ACROSS CITIES IN EVOLVING PATTERNS OF EXTREME HEAT EVENTS The spatial distributions of the frequencies and intensities of extreme heat events in the latest decade (2010–20) confirm large differences across East Asian cities. Extreme heat events were frequent throughout Southeast Asia (map 1.2, panel A). Jakarta, Indonesia, is the worst-positioned megacity in the region, but such events are pervasive across all Indonesian cities.26 The same is true for Malaysia, the Philippines, and cities in the southern part of Viet Nam. Unfortunately, these cities also face stronger UHI intensities, exacerbating the challenges they face in adapting to extreme urban heat. In higher latitudes, cities in northeast China (such as those around Shanxi Province) and southern China (such as those around Yunnan Province), in the north of Laos, and in Myanmar also frequently experience extreme heat events. Despite their locations, those cities were, on average, exposed to extreme heat events for over three months per year. And when they did occur, these events were particularly intense for cities in East Java (Indonesia), Thailand, and northern Viet Nam as well as cities in northeast China (map 1.2, panel B). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 53 Map 1.2 Frequency and Intensity of Extreme Heat Events in East Asian Cities, 2010–20 Average extreme hot months A FREQUENCY per year, 2016–20 .0 .0 .0 .0 .2 –1 –2 –3 –4 –7 00 01 01 01 01 0. 1. 2. 3. 4. Phuket Average intensity of extreme B INTENSITY heat 2010–20 .3 .6 .0 .5 .3 –2 –2 –3 –3 –7 10 31 61 01 51 2. 2. 2. 3. 3. Phuket SOURCES: World Bank calculations based on Climatology Lab, TerraClimate https://www.climatologylab.org/terraclimate.html; and European Commission, Global Human Settlement (GHS) Urban Centre Database R2019, https://ghsl.jrc.ec.europa.eu/ghs_stat_ ucdb2015mt_r2019a.php. NOTE: Panels A and B present the average frequency and intensity, respectively, of extreme hot months from January 2010 to December 2020 for each of 3,033 cities in the full regional sample (see box 1.1). An “extreme hot month” is one in which a city’s temperature anomaly variable records a value of at least 2 standard deviations. Frequency is defined as the number of extreme hot months per year and intensity as the average size of the anomaly variable during an extreme heat event—which is defined as a set of consecutive extreme hot months. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 54 SHARPER INCREASES IN EXTREME HEAT SEVERITY IN WARMER CITIES OF POORER COUNTRIES The extent to which the frequency and intensity of extreme heat events have evolved diverges according to background climates. While cities in cooler subregions experienced fewer than two extremely hot months per year in the 2010s, those in the tropics had to endure extreme heat events for over three months per year during the same period (figure 1.8, panel A). Although cities do not diverge much in terms of the intensity (figure 1.8, panel B), extreme heat events in tropical cities can translate into more serious threats to thermal comfort because their higher background temperatures are already amplified by high humidity. Although these evolving patterns confirm increasingly severe extreme heat events as common challenges for many East Asian cities, the heterogeneous patterns highlight the particularly severe struggles faced by tropical cities where the UHI effect also adds stronger heat stress than in nontropical cities. Because low- and middle- income countries tend to be in warmer regions (Diffenbaugh and Burke 2019; Huang et al. 2019), these results by climate zone align with more frequent—and to a lesser extent, more intense—extreme heat events across cities in low- and middle-income countries with lower capacities to cope (figure 1.8, panels C and D). Without adequate strategies for mitigation and adaptation, such challenges alongside these cities’ fast-growing urban populations will likely harm their quality of life and the prospect of economic growth. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 55 Figure 1.8 Frequency and Intensity of Extreme Heat Events by Climate Zone and Country Income Level, in East Asian Cities, 2010–20 sd: standard deviations A FREQUENCY BY CLIMATE ZONE B INTENSITY BY CLIMATE ZONE 3.5 2.6 2.6 3.0 3.0 Average anomaly size (sd) Average months per year 2.5 2.5 2.4 2.5 2.4 2.0 2.4 2.4 2.3 2.4 1.6 2.3 1.5 1.4 1.2 2.3 2.3 1.0 0.9 2.2 2.2 0.5 2.2 2.3 0.5 0.3 0.2 0.1 0.1 0.1 0.1 0 2.1 Tropical Nontropical Tropical Nontropical 1958-69 1970s 1980s 1990s 2000s 2010-20 C FREQUENCY BY INCOME LEVEL D INTENSITY BY INCOME LEVEL 2.5 2.6 2.1 Average anomaly size (sd) 2.5 Average months per year 2.0 1.9 2.5 2.46 2.5 1.5 1.4 2.4 2.4 2.4 2.4 2.3 1.0 2.3 0.7 0.8 2.2 2.3 0.6 2.2 2.2 0.5 2.2 2.2 0.1 0.1 0.1 0.2 0.1 0.1 0 2.1 Low- and middle-income High-income Low- and middle-income High-income 1958-69 1970s 1980s 1990s 2000s 2010-20 SOURCE: World Bank calculations based on Climatology Lab, TerraClimate, https://www.climatologylab.org/terraclimate.html; European Commission, Global Human Settlement (GHS) Urban Centre Database R2019, https://ghsl.jrc.ec.europa.eu/ghs_stat_ ucdb2015mt_r2019a.php. NOTE: An “extreme hot month” is one in which a city’s temperature for that month is at least 2 standard deviations higher than the month’s city-specific historical norm, as calculated over the period January 1958–December 1999. Frequency is calculated as the number of extreme hot months per year. Intensity is calculated as the average size of the anomaly variable during consecutive extreme hot months. Panels A and C present the average frequency, and panels B and D the average intensity, across the sample of 3,033 East Asian cities. In panels A and B, “tropical” and “nontropical” refer to a city’s latitude in absolute terms—“tropical” being in latitudes 0°–23.5° north and “nontropical” elsewhere. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 56 URBAN GREENING MITIGATES UHI EFFECTS, BUT LOCAL CONDITIONS ALSO MATTER Although a city’s local policy makers cannot alone affect the global path of climate change, local policies can significantly help reduce the severity of the UHI effect as a local environmental problem. Given the magnitude of the UHI effect, local action could make a telling reduction in the thermal discomfort that a city’s residents face. To provide insights into potential urban heat mitigation strategies, this section draws on the same sample of 100 cities used for the analysis of UHI intensity (see box 1.2) and evaluates the relative importance of local characteristics—such as a city’s average level of greenness, its population size and density, and its background climatic conditions and topography—in determining its UHI intensity. Consistent with the existing literature (see Bowler et al. 2010 for a review), the results of the analysis highlight the effectiveness of urban greening in mitigating UHI intensity after controlling for other factors that may be correlated with both average greenness and UHI intensity.27 The estimated effect is strong: all else being equal, UHI intensity could be halved if a city’s average level of greenness is doubled (figure 1.9, panel A), or the intensity could decrease by around 40 percent with a 1 standard deviation increase in a city’s average level of greenness (figure 1.10) This confirms a well-established notion that cities with more vegetation and fewer impervious surfaces store less heat while also being able to cool themselves down more effectively through evaporative processes. Vegetation can also enhance pedestrian comfort by offering shade. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 57 Figure 1.9 Relationship between UHI Intensity and Vegetation, Precipitation, and Population Density in East Asian Cities, Circa 2014 A LEVEL OF VEGETATION B LEVEL OF PRECIPITATION 2 Coeff: -1.998; p-value: 0.00 2 Coeff: 0.902; p-value: 0.00 UHI intensity (Log of degrees UHI intensity (Log of degrees 1 1 Celsius, circa 2014) Celsius, circa 2014) 0 0 −1 −1 −2 −2 −0.2 −0.1 0 0.1 0.2 −0.8 −0.4 0 0.4 0.5 Average greenness (Log, 2014) Total annual precipitation (Log, 2014) C POPULATION DENSITY 2 Coeff: 0.466; p-value: 0.03 UHI intensity (Log of degrees 1 Celsius, circa 2014) 0 −1 −2 −0.1 −0.5 0 0.5 0.1 Population density (Log, 2015) SOURCE: World Bank calculations based on the urban heat island intensity (UHI) data from 4 Earth Intelligence, https:// www.4earthintelligence.com/; and the boundaries of 100 cities from the European Commission, Global Human Settlement Urban Centre Database R2019, https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php. NOTE: The figures show the empirical relationships between the intensity of UHI effects—calculated as the averages of the intensities over 2012–16 based on the 5 kilometer buffer—and the level of greenness, population density, and the level of precipitation across 100 East Asian cities in the sample. The regression equation also includes a city’s population, mean annual temperature, mean level of elevation, and the absolute value of latitude, all in the natural log scale, as well as country dummies. Full regression results are presented in Annex 1B. Coeff = Coefficient. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 58 Despite the importance of vegetation for relieving heat stress, it is not alone sufficient to keep cities as cool as their rural surroundings, especially in humid tropical cities. This is not only because the cooling effect of vegetation is limited to a relatively small geographic scale28 but also because background climates matter for the effectiveness of cooling processes.29 In rural areas, natural environments facilitate air movement from the surface to the atmospheric boundary—that is, the efficiency of convection is enhanced. In urban areas, by contrast, the process is suppressed by artificial structures that impede air flow, resulting in larger heat gains during the day. The difference in convection efficiency and evaporative cooling between cities and their rural surroundings is even larger in humid regions where ample precipitation promotes forest cover in rural areas. Thus, under humid environments, the UHI intensity remains high unless the urban morphology is fundamentally changed through, for example, significant citywide increases in green spaces (Manoli et al. 2019; Zhao et al. 2014). Indeed, our analysis of 100 cities finds that UHI intensity tends to increase with the level of precipitation, all else being equal (figure 1.9, panel B). The intensifying effect of precipitation comes out even stronger than the mitigating effect of vegetation (figure 1.10). These pieces of evidence together suggest that, in cities with humid climates, greening must be accompanied by additional heat mitigation strategies, such as increasing the amount of sunlight reflected by artificial surfaces, that is, albedo (Zhao et al. 2014), and improving ventilation. The latter is particularly important for thermal comfort because having more trees can interfere with the body’s cooling mechanism by increasing air humidity while reducing wind speed (Manoli et al. 2019). These various strategies will further be discussed in chapter 4. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 59 Figure 1.10 Relative Effect Sizes of Local Characteristics on UHI Intensity in East Asian Cities Total annual precipitation 0.43*** Population density 0.21** Mean elevation 0.17*** Population 0.04 Latitude -0.08 Mean annual temperature -0.32* Average greenness -0.38*** −0.5 −0.4 −0.3 −0.2 −0.1 0 0.1 0.2 0.3 0.4 0.5 Effect size (buffer distance = 5 kilometers) SOURCE: World Bank calculations based on the urban heat island intensity (UHI) data from 4 Earth Intelligence, https://www.4earthintelligence.com/; and the boundaries of 100 East Asian cities from the European Commission, Global Human Settlement Urban Centre Database R2019, https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php. NOTE: For each variable, the effect size measures the estimated impact on a city’s UHI intensity of a 1 standard deviation increase in the value of the variable based on the sample of 100 East Asian cities in the sample (see box 1.2). Full regression results are presented in annex 1B. Significance level: * = 10 percent, ** = 5 percent, *** = 1 percent. It is also noteworthy that, all else being equal, population density tends to increase the UHI intensity although the magnitude of its impact is smaller than those of greenness and precipitation (figure 1.9, panel C; figure 1.10). High-density, compact urban forms have often been identified as key development strategies for positive agglomeration forces30 with potential co-benefits of climate change mitigation.31 However, the direct and indirect impacts of city densification on the UHI effect is an ongoing debate. On the one hand, a high density associated with well-connected public transportation networks and relatively small property sizes may help ameliorate the UHI effect by reducing human-generated heat from, for instance, private vehicle travels and the energy consumption of buildings (Glaeser 2012; Viguié and Hallegatte 2012). Acting against this, a high density may intensify UHI effects because tall buildings, despite their shading effects, trap heat and pollutants (Li et al. 2020). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 60 The positive correlation between the UHI intensity and population density may also be because of factors that are unaccounted for in this analysis, such as a city’s layout, whereby a grid layout accompanied by tall buildings and narrow streets tends to increase heat storage through radiative exchanges of energy within street canyons (Sobstyl et al. 2018). Although further investigation of these mechanisms is beyond the scope of this report, the above results are suggestive of the importance of urban design and building codes in relieving heat stress especially for cities actively being built out. SUMMARY AND CONCLUSION To help East Asia’s city stakeholders better grasp the threats posed by extreme heat, and to inform the policy discussion of chapter 4, this chapter has presented the evolving patterns of extreme urban heat across East Asian cities. The analysis of 100 major cities in the region reveals that cities are, on average, hotter than their rural surroundings by around 1.6–2.0°C and, in the most extreme case, by over 5°C, presumably owing to inadequate urban planning and policies that contributed to the formation of UHIs. Moreover, these average values mask considerable within-city variation in UHI intensities such that the maximum intensity experienced in some neighborhoods within a city is considerably higher than the average, an issue that chapter 2 explores in depth. A complementary analysis based on a sample of over 3,000 East Asian cities further confirms that global climate change compounds heat stresses posed by the UHI effect. Abnormally high temperatures, relative to the historical norm of each city, have become 23 times more frequent in the past four decades while increasingly deviating from the normal climatic condition over time. Although increasing exposure to extreme heat events is pervasive across the region, more dramatic increases occurred in already hot and humid tropical cities, where the UHI effect is the strongest. Along with the gradually increasing precipitation in the tropics, these evolving patterns highlight the unique struggles that confront rapidly growing cities in Southeast Asian countries. However, cities outside tropical regions, notably those in northern and southwestern China, may also struggle with adverse socioeconomic impacts of extreme heat events because they may not be as well adapted to them. As chapters 2 and 3 will discuss, extreme urban heat affects virtually every aspect of urban life, and the most vulnerable city residents, with limited capacity to adapt, are disproportionately more exposed. Hence, the findings of this chapter suggest that it UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 61 is imperative to implement adequate strategies for heat mitigation and adaptation to safeguard the prosperity, livability, and inclusiveness of East Asian cities and to reduce economic inequality within and across cities. Although local decision-makers, acting in isolation, cannot significantly mitigate global climate change, they can ameliorate the UHI effect through better planning and design of cities in consideration of local climatic conditions. Moreover, they can promote adaptation to extreme urban heat—whether driven by the UHI effect or abnormally high temperatures—through a variety of physical and behavioral measures to be discussed in chapter 4. ANNEX 1A DEFINING EXTREME HEAT EVENTS The first definition, which may be considered an absolute definition, goes back to A. T. Burrows (1901), who defined a heat wave (or, as he termed it, “hot wave”) as a spell of three or more days, on each of which the maximum shade temperature reaches or exceeds 90 degrees Fahrenheit. Based on this definition, a city’s temperature on a given day can be regarded as an extreme heat event if it exceeds a predetermined absolute temperature threshold, say, 35°C, on that day. Formally, it can be written as: Threshold Tc,d > Tc,d ,     (1A.1) where, Tc,dand TThreshold c,d respectively refer to a city’s temperature on a given day and an absolute temperature threshold. The second definition, which associates an extreme heat event with an absolute temperature anomaly, is derived from that of the World Meteorological Organization (WMO). According to this, a heat wave is defined as five or more consecutive days of prolonged heat “during which the daily maximum temperature surpasses the average maximum temperature by 5°C (9°F) or more.”32 More generally, a city is considered as having an extremely hot day if its temperature on that day deviates from a reference temperature by more than a certain absolute threshold—for example, 5°C, as in UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 62 the WMO definition. The reference is often calculated as the city’s long-run average temperature over a base period. This definition may be expressed as follows: Anomalya c,d = (Tc,d - Tc,d) > TThreshold c,d (a) ,     (1A.2) where, Tc,d, Tc,d, and Tc,d respectively denote a city’s temperature on a given day, Threshold (a) its long-run average temperature, and a threshold absolute temperature. The third definition, which is based on the concept of a relative temperature anomaly, is derived by standardizing a city’s absolute anomaly on a given day by the variation in the city’s temperature over the same base period applied to calculate the absolute anomaly. With the absolute anomaly alone, it is inadequate to draw a full picture of extreme heat events because the same degrees of absolute anomaly could be more significant in cities with stable climates than in those with more variable climates. Based on a relative anomaly, a city’s temperature on a given day can be considered an extreme if the standard deviation of the temperature on that day exceeds a predefined threshold, such as +2 standard deviations. Formally, a relative anomaly of a city on a given day can be written as Tc,d - Tc,d Anomalyr = > Tc,d      (1A.3) Threshold (r) c,d SDc where, Tc,d, Tc,d, SDc, and Tc,d respectively refer to a city’s temperature on a given Threshold (r) day, its average temperature over a base period, the standard deviation of the city’s temperatures over the base period, and a threshold standard deviation. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 63 ANNEX 1B DETERMINANTS OF URBAN HEAT ISLAND INTENSITY Table 1B.1 Regression Results for the Potential Determinants of Urban Heat Island Intensity DEPENDENT VARIABLE: UHI INTENSITY (LN, AVERAGE 2012–16) BUFFER DISTANCE 2 KM 5 KM 10 KM Average greenness (ln, 2014) −1.077 −1.998*** −1.755*** [−1.64] [−2.94] [−3.17] Total annual precipitation (ln, 2014) .894*** .902*** .848*** [3.84] [3.88] [4.06] Mean annual temperature (ln, 2014) −.590* −.569* −.458 [−1.81] [−1.86] [−1.45] Latitude (ln of absolute values) −.0601 −.0933 −.0697 [−0.57] [−0.84] [−0.59] Mean elevation (ln, 2014) .056 .122*** .114** [1.29] [2.68] [2.53] Population (ln, 2015) .0245 .0264 .00419 [0.36] [0.34] [0.06] Population density (ln, 2015) .711** .466** .406** [2.63] [2.19] [2.29] Constant −12.5*** −11.5*** −10.6*** [−5.20] [−5.64] [−5.20] No. of observations 99 100 99 R2 .651 .671 .678 Adjusted R2 .556 .582 .590 UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 64 SOURCE: Results based on the analysis of the urban heat island (UHI) intensity data processed by 4 Earth Intelligence, https://www.4earthintelligence.com/, and the rest of the data from European Commission, Global Human Settlement Urban Centre Database (GHS-UCDB) R2019, https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php. NOTE: Discussions in the main text refer to the result in the 5 km column. Numbers in brackets are robust t-statistics. Regressions also include country dummies. The regressions based on the buffer sizes of 2 km and 10 km exclude Bago and Mandalay (both from Myanmar), respectively, because the log transformation of the dependent variable omitted those cities that had negative average UHI intensities during 2012–16. km = kilometers; ln = natural logarithm Significance level: * = 10 percent, ** = 5 percent, *** = 1 percent. REFERENCES Bosker, M., J. Park, and M. Roberts. 2021. “Definition Matters. Dijkstra, L., A. J. Florczyk, S. Freire, T. Kemper, M. Melchiorri, Metropolitan Areas and Agglomeration Economies in a M. Pesaresi, and M. Schiavina. 2021. “Applying the Large-Developing Country.” Journal of Urban Economics Degree of Urbanisation to the Globe: A New Harmonised 125: 103275. Definition Reveals a Different Picture of Global Bowler, D. E., L. Buyung-Ali, T. M. Knight, and A. S. Pullin. 2010. Urbanisation.” Journal of Urban Economics 125 (2): “Urban Greening to Cool Towns and Cities: A Systematic 103312. Review of the Empirical Evidence.” Landscape and Urban Estrada, F., W. J. Wouter Botzen, and R. S. J. Tol. 2017. “A Global Planning 97 (3): 147–55. Economic Assessment of City Policies to Reduce Climate Burakowski, E., and M. Magnusson. 2012. “Climate Impacts on the Change Impacts.” Nature Climate Change 7 (6): 403–6. Winter Tourism Economy in the United States.” Research Glaeser, E. L. 2012. Triumph of the City: How Our Greatest report, Natural Resources Defense Council, New York; and Invention Made Us Richer, Smarter, Greener, Healthier and Protect Our Winters, Pacific Palisades, CA. Happier. New York: Penguin Press. Burrows, A. T. 1901. “Hot Waves: Conditions which Produce Them, Glickman, T. S., ed. 2000. “Heat Wave.” In Glossary of and Their Effect on Agriculture.” Yearbook of the US Meteorology, 2nd ed. Boston: American Meteorological Department of Agriculture 22: 325–36. Society. Cappucci, M. 2023. “Coast-to-coast heat dome sends Houghton, J. 2015. Global Warming. Cambridge: Cambridge temperatures soaring, threatens all-time records.” University Press. Washingtonpost.com, July 13, 2023, He, C., H. Kim, M. Hashizume, W. Lee, Y. Honda, S. E. Kim, P. L. https://www.washingtonpost.com/weather/2023/07/13/ Kinney et al. 2022. “The Effects of Night-Time Warming heatwave-california-arizona-florida-texas-records/. on Mortality Burden under Future Climate Change Corbane, C., P. Martino, P. Panagiotis, F. Aneta, M. Michele, Scenarios: A Modelling Study.” The Lancet Planetary F. Sergio, S. Marcello, et al. 2020. “The Grey-Green Health 6 (8): e648–57. Divide: Multi-Temporal Analysis of Greenness across Huang, K., X. Li, X. Liu, and K. C. Seto. 2019. “Projecting Global 10,000 Urban Centres Derived from the Global Human Urban Land Expansion and Heat Island Intensification Settlement Layer (GHSL).” International Journal of Digital through 2050.” Environmental Research Letters 14 (11): Earth 13 (1): 101–18. 114037. Dell, M., B. F. Jones, and B. A. Olken. 2014. “What Do We Imhoff, M. L., P. Zhang, R. E. Wolfe, and L. Bounoua. 2010. “Remote Learn from the Weather? The New Climate–Economy Sensing of the Urban Heat Island Effect across Biomes Literature.” Journal of Economic Literature 52 (3): in the Continental USA.” Remote Sensing of Environment 740–98. 114 (3): 504–13. Deschenes, O., and E. Moretti. 2009. “Extreme Weather Events, IPCC (Intergovernmental Panel on Climate Change). 2021. Climate Mortality, and Migration.” Review of Economics and Change 2021: The Physical Science Basis. Summary for Statistics 91 (4): 659–81. Policymakers. Working Group I Contribution to the Sixth Diffenbaugh, N. S., and M. Burke. 2019. “Global Warming Has Assessment Report of the Intergovernmental Panel on Increased Global Economic Inequality.” PNAS 116 (20): Climate Change. Cambridge: Cambridge University Press. 9808–13. Li, Y., S. Schubert, J. P. Kropp, and D. Rybski. 2020. “On the Influence of Density and Morphology on the Urban Heat Island Intensity.” Nature Communications 11: 2647. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 65 Liu, Z., W. Zhan, B. Bechtel, J. Voogt, J. Lai, T. Chakraborty, Z.-H. Sargent, T. J. 1972. “Rational Expectations and the Term Wang et al. 2022. “Surface Warming in Global Cities Is Structure of Interest Rate.” Journal of Money, Credit and Substantially More Rapid Than in Rural Background Banking 4 (1): 74–97. Areas.” Communications Earth & Environment 3 (1): 219. Sobstyl, J. M., T. Emig, M. J. Abdolhosseini Qomi, F.-J. Ulm, and Lucas, R. E. Jr. 1972. “Expectations and the Neutrality of Money.” R. J.-M. Pellenq. 2018. “Role of City Texture in Urban Journal of Economic Theory 4 (2): 103–24. Heat Islands at Night Time.” Physical Review Letters 120: Manoli, G., S. Fatichi, M. Schläpfer, K. Yu, T. W. Crowther, N. Meili, 108701. P. Burlando, G. G. Katul, and E. Bou-Zeid. 2019. “Magnitude Song, F., G. J. Zhang, V. Ramanathan, and L. Ruby Leung. 2022. of Urban Heat Islands Largely Explained by Climate and “Trends in Surface Equivalent Potential Temperature: Population.” Nature 573 (7772): 55–60. A More Comprehensive Metric for Global Warming and Maplecroft. 2017. “The Economics of South East Asia’s Rising Weather Extremes.” PNAS 119 (6): e2117832119. Temperatures.” Maplecroft Insights, November 16. Souverijns, N., K. De Ridder, N. Veldeman, F. Lefebre, F. Meehl, G. A., and C. Tebaldi. 2004. “More Intense, More Frequent, Kusambiza-Kiingi, W. Memela, and N. K. W. Jones. 2022. and Longer Lasting Heat Waves in the 21st Century.” “Urban Heat in Johannesburg and Ekurhuleni, South Science 305 (5686): 994–97. Africa: A Meter-Scale Assessment and Vulnerability Mukim, M., and M. Roberts, eds. 2023. Thriving: Making Cities Analysis.” Urban Climate 46: 101331. Green, Resilient, and Inclusive in a Changing Climate. Thompson, V., A. T. Kennedy-Asser, E. Vosper, Y. T. E. Lo, C. Washington, DC: World Bank. Huntingford, O. Andrews, M. Collins, G. C. Hegerland, and Muth, J. F. 1961. “Rational Expectations and the Theory of Price D. Mitchell. 2022. “The 2021 Western North America Heat Movements.” Econometrica 29 (3): 315–35. Wave among the Most Extreme Events Ever Recorded Neo, C. C. 2018. “Urban Warming Worsening Problem in Globally.” Science Advances 8 (18): eabm6860. Singapore, Action Needed while Studies Are Ongoing: Tol, R. S. J. 2021. “The Economic Impact of Weather and Climate.” Researchers.” TODAY, July 12. Working Paper 8946, Social Science Research Network, Obradovich, N., R. Migliorini, S. C. Mednick, and J. H. Fowler. 2017. Center for Economic Studies and Ifo Institute for “Nighttime Temperature and Human Sleep Loss in a Economic Research (CESifo), Munich. Changing Climate.” Science Advances 3 (5): e1601555. UNEP (United Nations Environment Programme). 2021. Beating Pan, L., L. Lu, P. Fu, V. Nitivattananon, H. Guo, and Q. Li. 2023. the Heat: A Sustainable Cooling Handbook for Cities. “Understanding Spatiotemporal Evolution of the Surface Nairobi: UNEP. Urban Heat Island in the Bangkok Metropolitan Region Viguié, V., and S. Hallegatte. 2012. “Trade-Offs and Synergies in from 2000 to 2020 Using Enhanced Land Surface Urban Climate Policies.” Nature Climate Change 2 (5): Temperature.” Geomatics, Natural Hazards and Risk 14 (1): 334–37. 2174904. Woetzel, J., D. Pinner, H. Samandari, H. Engel, M. Krishnan, B. Parsons, L. A., Y. J. Masuda, T. Kroeger, D. Shindell, N. H. Wolff, Boland, and C. Powis. 2020. “Climate Risk and Response: and J. T. Spector. 2022. “Global Labor Loss Due to Humid Physical Hazards and Socioeconomic Impacts.” Research Heat Exposure Underestimated for Outdoor Workers.” report, McKinsey Global Institute, Washington, DC. Environmental Research Letters 17 (1): 014050. World Bank. 2015. East Asia’s Changing Urban Landscape: Rahman, M., R. McConnell, H. Schlaerth, J. Ko, S. Silva, F. Measuring a Decade of Spatial Growth. Washington, DC: W. Lurmann, L. Palinkas et al. 2022. “The Effects of World Bank. Coexposure to Extremes of Heat and Particulate Air Yu, V. 2022. “China Reports ‘Most Severe’ Heatwave and Third Pollution on Mortality in California: Implications for Driest Summer on Record.” The Guardian, September 7. Climate Change.” American Journal of Respiratory and Zhao, L., X. Lee, R. B. Smith, and K. Oleson. 2014. “Strong Critical Care Medicine 206 (9): 1117–27. Contribution of Local Background Climate to Urban Heat Raymond, C., T. Matthews, and R. M. Horton. 2020. “The Islands.” Nature 511 (7508): 216–19. Emergence of Heat and Humidity Too Severe for Human Zhao, Q., Y. Guo, T. Ye, A. Gasparrini, S. Tong, A. Overcenco, Tolerance.” Science Advances 6 (19): eaaw1838. A. Urban et al. 2021. “Global, Regional, and National Robinson, A., J. Lehmann, D. Barriopedro, S. Rahmstorf, and D. Burden of Mortality Associated with Non-Optimal Coumou. 2021. “Increasing Heat and Rainfall Extremes Ambient Temperatures from 2000 to 2019: A Three- Now Far Outside the Historical Climate.” Climate and Stage Modelling Study.” The Lancet Planet Health 5 (7): Atmospheric Science 4 (1): 45. e415–25. Russo, S., J. Sillmann, and A. Sterl. 2017. “Humid Heat Waves at Different Warming Levels.” Scientific Reports 7 (1): 7477. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 66 NOTES 1 This chapter covers 16 World warming under RCP 8.5, the highest days. In the process, these more Bank member countries (Brunei greenhouse gas emissions scenario recent hotter days will also appear Darussalam, Cambodia, China, adopted by the IPCC. Under RCP as larger absolute and relative Indonesia, Japan, the Republic of 4.5, Estrada, Wouter Botzen, and Tol temperature anomalies. Korea, Laos, Malaysia, Mongolia, (2017) project the GDP loss for the 10 In defining hot temperature Myanmar, Papua New Guinea, the median city globally to be 1.4 percent. extremes, IPCC (2021) considers both Philippines, Singapore, Thailand, When the UHI effect is considered, absolute (days when maximum daily Timor-Leste, and Viet Nam) and the the projected total economic damage temperature exceeds, say, 35°C) and Democratic People’s Republic of from heat in cities during the relative (the 90th or higher percentile Korea in East Asia while excluding twenty-first century is 2.6 times the of maximum daily temperature for Australia, New Zealand, and the magnitude of losses when it is not. the calendar day over a base period) Pacific island states. For further discussion, see chapter 3. definitions. 2 Throughout this report, temperatures 7 Some cities and countries even 11 Population projections are are generally referred to in degrees consider the duration of an event based on data from the Shared Celsius rather than Fahrenheit. A to define extreme heat. Adelaide, Socioeconomic Pathway (SSP) conversion table between Celsius Australia, for example, defines a heat Database, version 2.0, on the and Fahrenheit can be found in the wave as five consecutive days at or International Institute for Applied report’s front matter. above 35°C or three consecutive days Systems Analysis (IIASA) website: 3 Labor productivity losses due to at or above 40°C. Chapter 4 discusses https://tntcat.iiasa.ac.at/SspDb/ extreme heat exposure are estimated locally appropriate thresholds in dsd?Action=htmlpage&page=10. to cost the global economy US$300 more detail. 12 4 Earth Intelligence (4EI) (https:// billion every year. The Southeast Asia 8 Evapotranspiration removes heat www.4earthintelligence.com/) is a region is projected to experience the through evaporation from natural global firm with expertise in earth world’s largest labor productivity loss surfaces and transpiration from observation, geographic information from such exposure—16 percent by stomata. Convection releases heat systems (GIS), and remote sensing 2045 (Maplecroft 2017). from the surface back into the analytics. It was competitively 4 When China was grappling with its atmospheric boundary and brings in selected through the Digital Earth summer 2022 heat wave, for example, fresh cool air. This process is more Partnership Technology Challenge many energy-intensive firms had efficient in an environment with rough (https://www.space.org.sg/digital- to reduce their hours of operation surfaces, such as rocks and dense earth-partnership-technology- because of power shortages, while vegetation, than in urban areas with challenge/) jointly organized by the others had to shut down entirely smooth surfaces of buildings and World Bank and Singapore Space (Yu 2022). pavements or in rural areas with low & Technology Ltd during December 5 Based on multi-model ensemble stature, such as shrubs and grasses 2021–March 2022. average projections in Representative (Zhao et al. 2014). 13 The LST images used as input for Concentration Pathway (RCP) 4.5, an 9 As hotter days become more estimating the intensity of the intermediate greenhouse gas emission frequent, this will slowly pull up the UHI effect were produced by the scenario adopted by the IPCC. historical long-run average as more National Aeronautics and Space 6 The projected GDP loss is based on recent hotter days are averaged Administration’s Moderate Resolution temperature increases due to global with relatively more distant cooler Imaging Spectroradiometer onboard UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 67 the Earth Observing System (EOS- expectations and therefore behavior and pollutant dissipation. MODIS) and the National Oceanic on all relevant available information. 19 This is, however, not to downplay the and Atmospheric Administration’s This also implies that they adapt health risk of dry heat that can be Advanced Very High-Resolution to predictable trends and past experienced in, for example, northern Radiometer aboard the Meteorological experiences of events. Under this China. In a dry atmosphere, people Operational satellite (MetOp-AVHRR). theory, for an event to have an can be easily dehydrated because To classify urban and nonurban impact on outcomes, it must come moisture evaporates off their skin too pixels, 4EI leveraged the WorldCover as a “surprise.” The theory of rational quickly, even before they realize they products from European Space expectations was first introduced are sweating (Cappucci 2023). Agency (https://esa-worldcover.org/) into economics by Muth (1961) before 20 A city’s level of humidity is proxied by and land cover maps from Copernicus subsequently being widely integrated its annual total precipitation averaged Global Land Service (https://land. into macroeconomic theory by, over January 1958–December 1999. copernicus.eu/global/products/lc). among others, Lucas (1972) and 21 Visual inspections of various trends— 14 The monthly temperature and Sargent (1972). for example, of other statistics or by precipitation datasets used in this 17 An alternative—and, in principle, very country—suggest that the temporary chapter are drawn from Climatology appealing—approach to measuring increases in the UHI intensities during Lab’s TerraClimate (https://www. extreme heat events would be 2010–14 may not be data artifacts. climatologylab.org/terraclimate. to define a given temperature as Nevertheless, further research html), which covers global terrestrial extreme if there is evidence to show beyond this report is required to surfaces at a spatial resolution of it has significant adverse impacts on establish the reasons for such trends. 1/24th degree (about 4 kilometers) for the outcomes of interest. However, in 22 The list of top 10 cities in estimated the period January 1958–December practice, it is difficult to implement UHI intensity includes 1 city each 2020. such a definition, especially when from Indonesia and Singapore, 5 from 15 The choice of the base period considering a wide range of different Malaysia, and 3 from the Philippines. varies widely across meteorological outcome variables, as is the case 23 The 2000–20 average temperature organizations. For global mean in this report. Thus, for example, as estimate for Bangkok’s rural surface temperature datasets, for the discussion in chapter 3 makes surroundings is based on a 10 example, NOAA uses 1901–2000, clear, there is no unique threshold kilometer buffer. whereas NASA, the Hadley across different outcomes, countries, 24 The analysis in this section Centre/Climatic Research Unit or even time of the year within emphasizes five-year moving Temperature (HadCRUT), and the countries, that can be identified averages. This allows it to take Japan Meteorological Agency use as triggering negative impacts of account of the average intervals of 1951–80, 1961–90, and 1991–2020, high temperatures. Furthermore, recurring climate patterns such as respectively. The choice affects the the level of thermal comfort that a El Niño and La Niña, which occur sizes of anomalies but not the trends given temperature translates into every two to seven years on average over time. also depends on other factors, most (https://oceanservice.noaa.gov/ 16 The relative anomaly definition of notably humidity. facts/ninonina.html). The use of extreme heat events is consistent 18 Extreme heat and air pollution tend five-year moving averages also with the theory that people possess to occur together with the same helps smooth potential temporary “rational expectations.” According underlying meteorological conditions measurement errors in the data. to this theory, people are forward- that reduce the flow of winds, thus 25 The negative relationship between looking agents who base their interfering with convection cooling the increases in mean annual UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 68 temperature and the initial level of greenness across all pixels urban density is associated with a temperatures is even stronger when within its spatial extent, where the 14.2 percent increase in an average the increases are measured as the level of greenness of a pixel is derived worker’s nominal wage. average annual growth rates (AAGRs) based on the highest value of the 31 Evidence based on a global sample of rather than as absolute increases. Normalized Difference Vegetation cities shows that, after controlling for However, we chose to present the Index (NDVI) for the reference year of a city’s size and level of development, pace based on the absolute increases 2014. For more details, see Corbane et city compactness is negatively in figure 1.4 because the calculation al. (2020). associated with the amount of of AAGRs, by definition, omitted 28 For example, Singapore’s Urban production-based CO2 emissions from nine cities that had a negative mean Redevelopment Authority (URA) the residential and transportation annual temperature in the initial year and the Housing and Development sectors. Lower emissions could be (for example, Ulaanbaatar, Mongolia). Board (HDB) find—in their joint study largely attributed to more infill as 26 Megacities are defined as cities on residential areas (Jurong Lake opposed to leapfrog development and with populations over 10 million District and Punggol housing estate) fewer vehicle miles travelled—a result (World Bank 2015). The full sample and a new business area (near Asia achieved especially by discouraging of 3,033 East Asian cities includes Square)—that plants along building personal motorized vehicles (Mukim 11 megacities, in descending order of façades can provide heat stress relief and Roberts 2023). 2015 population: Guangzhou, China; to people within a 4 meter range, 32 Rafferty, John P., “Heat Wave.” Jakarta, Indonesia; Tokyo, Japan; beyond which the green façades have Encyclopedia Britannica, accessed Shanghai, China; Quezon City-Manila, little or no effect (Neo 2018). November 29, 2022, https://www. Philippines; Seoul, Korea; Beijing, 29 It is worth noting that maintaining britannica.com/science/heat-wave- China; Osaka-Kyoto, Japan; Bangkok, more green spaces can be meteorology Thailand; Ho Chi Minh City, Viet Nam; challenging for water-scarce cities. and Jieyang, China. 30 For example, Bosker, Park, and 27 A city’s average greenness is Roberts (2021) estimate that, in calculated as the unweighted mean Java-Bali, Indonesia, a doubling of UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 69 CH A SCENARIO 1 C H A PT ER 2 B SCENARIO 2 THE HEAT ISLAND EFFECT WITHIN EAST ASIAN CITIES: WHO IS MOST EXPOSED & WHAT DETERMINES ITS STRENGTH? UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 70 KEY MESSAGES Within East Asian cities, temperatures can vary sharply between pairs of neighborhoods in close proximity, as evident from in situ measurements in cities such as Bandung, Indonesia. Neighborhoods with densely packed homes, heat-trapping materials, and limited vegetation give up their heat at night more slowly than those with widely spaced buildings and plentiful tree cover. These “thermal inequalities” within cities can be large: heat disparities of up to 7 degrees Celsius (°C) were measured in Bandung, Indonesia, through fieldwork conducted for this report. Climate modeling indicates that East Asian cities will face more hot days and nights per year by midcentury because of the warming background climate, but interventions to reduce heat-trapping surfaces, increase wind flow, and add vegetation could partially counteract this effect. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 71 INTRODUCTION Each day, when city dwellers step outside, they experience the weather: temperature, precipitation, wind, cloud cover, and other atmospheric conditions that (averaged over time) make up the local climate. East Asian cities are experiencing climatic warming due to global processes (see chapter 1). However, urban climates are also shaped by meteorological phenomena that take place on a much finer geographic scale. Urban landscapes, having absorbed the sun’s energy during the day, cool more slowly than do nearby rural areas, owing to the thermal properties of building and paving materials. The density of such materials and their arrangement into built forms such as street canyons differ markedly between neighborhoods. Just as city blocks vary sharply in appearance and character, so too do their thermal properties when interacting with the atmosphere and the sun’s energy—resulting in hyperlocal differences that accentuate the impact of global climate change on residents of heat-prone neighborhoods. Heat in city neighborhoods often tends to reflect and reinforce existing inequalities. Indeed, research across world regions has identified “thermal inequalities” whereby residents of different neighborhoods, despite living very short distances apart, experience sharply different levels of heat stress (Chakraborty et al. 2019). Thermal differences between neighborhoods can be identified and addressed through actions at different spatial scales—from painting an individual building’s roof white to planning urban landscapes for improved wind flow and vegetation coverage. This chapter explores the extent of variation in local temperatures within East Asian cities and the factors that correlate with this variation through two case studies: Bandung (Indonesia) and Phnom Penh (Cambodia). Through local heat stress measurements and climate simulations, the chapter documents variation in heat stress between city neighborhoods, providing insights on a granular level that complement those in chapter 1 on the overall strength of heat island effects in a broad sample of East Asian cities. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 72 WHEN CITIES ARE BUILT, LOCAL CLIMATE PROCESSES CHANGE Urbanization transforms landscapes. When humans build cities, they level terrain, remove vegetation, and introduce new surfaces and structures. They use materials such as asphalt, packed earth, wood, concrete, steel, and glass to introduce buildings and paved surfaces. Such materials are selected for physical properties such as strength and durability, but they also have thermal characteristics such as higher absorption and retention of heat. A city neighborhood, compared with the same site in its undisturbed state, therefore exhibits a radically different interaction with the earth’s atmosphere. In considering neighborhoods from a meteorological perspective, several aspects of urban form are important: Land cover, notably the fraction of land area that is vegetated or sealed with impervious materials such as asphalt Fabric, including the characteristics of the materials used to construct paving and buildings Geometry, including the shape, height, and spacing of buildings. In city landscapes with little vegetation and extensive heat-trapping surfaces, net energy transfers between the earth’s surface and its atmosphere are modified, resulting in higher local air temperatures (box 2.1). The size and orientation of buildings also affect microclimates. Street canyons with a high height-to-width ratio retain thermal energy longer because tall buildings obstruct the release of radiation back into the atmosphere (Johansson 2006; Oke et al. 2017). Cities influence the climate at different scales—from individual parts of buildings to metropolitan regions. Table 2.1 identifies a hierarchy of such scales and the corresponding meteorological phenomena that buildings, neighborhoods, and cities affect. At the level of an individual building, differences in atmospheric conditions can readily be experienced by standing in the shade of a building on a hot day or moving from its wind-exposed side to its sheltered side. At the block and neighborhood levels, pedestrians in East Asian cities routinely adapt their behavior to urban climate effects by walking on sunny or shady sides of the street or seeking out cool, breezy parks for rest and recreation. On the metropolitan scale, differential warming of air above urban areas and their surroundings can induce wind flows that UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 73 affect heat stress and air pollution. At each scale, design and planning interventions can mitigate the heat island effect and promote cooler spaces for the benefit of residents and workers. Table 2.1 Climate Processes at Different Urban Scales TYPICAL HORIZONTAL URBAN UNIT FEATURES CLIMATE PHENOMENA SCALE Facet Roof, wall, road, lawn, pond Shading, storage heat flux, dew, 10 x 10 meters and frost patterns Element Building, tree Wind flow effects (for example, 50 x 50 meters wake behind tall building) Street canyon Row of buildings, line of street Cross-street shading, pedestrian 30 x 200 meters trees microclimate Neighborhood City center, residential Local neighborhood climate, park 2 x 2 kilometers quarter, industrial zone, park, cool island effect, local breezes lake City Built-up area Urban heat island, urban effects 50 x 50 kilometers on humidity and wind SOURCE: Modified based on Oke et al. 2017. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 74 BOX 2.1 HOW DO A NEIGHBORHOOD’S VEGETATION AND BUILT SURFACES AFFECT ITS CLIMATE? Trees and vegetation promote cool urban from liquid to a gas state (generating Urban forestry initiatives can increase environments through their influence on “latent heat”). vegetation cover and change the surface local meteorological processes. Figure energy balance to promote cooler B2.1.1 illustrates their influence by means of The right panel (figure B2.1.1, panel B) environments. When trees are planted the surface energy balance—a framework portrays an urban location. With little across sufficiently large areas, a “park used by meteorologists to characterize the vegetation in this location, a higher share cool island” effect can be observed: transfer of energy between atmosphere of energy is used to generate sensible differential heating of air across green and surface at given locations. heat than latent heat, contributing to versus built-up areas creates localized higher air temperatures. Vehicles and breezes that exert a cooling influence on In the left panel (figure B2.1.1, panel A), factories (“anthropogenic heat” sources) adjacent city blocks. Reduction of areas solar energy meets the earth’s surface raise temperatures further. Materials with under asphalt and similar heat-trapping at a rural location with trees. Energy high thermal inertia such as asphalt retain materials can likewise mitigate urban heat heats the air (generating “sensible heat”). energy during the day and radiate it back islands. Such potential interventions are Through evapotranspiration, energy also at night, contributing to higher nighttime discussed further in chapter 4. transforms the water contained in leaves temperatures. Figure B2.1.1 The Surface Energy Balance: How Reduced Vegetation and Increased Human-Generated Heat Result in Warmer Urban Temperatures A HIGH VEGETATION SITE B DENSE URBAN SITE Q* Q* QA QH QE QH QE QH QE SURFACE ENERGY BALANCE: Q* + QA = QH + QE + ΔQS SOURCE: World Bank elaboration based on Oke et al. 2017. NOTE: The figure presents a simplified representation of the surface energy balance in a densely vegetated site and in a densely built-up urban site. Incoming solar radiation (Q*) + human-generated heat (QA) = sensible heat (QH) + latent heat (QE) + heat storage in substrate and surfaces (ΔQS). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 75 TEMPERATURES DIFFER SHARPLY BETWEEN CITY NEIGHBORHOODS Like many East Asian cities, Bandung, Indonesia, has experienced rapid demographic and territorial growth in recent decades that has transformed its physical character. Located some 90 miles southeast of Jakarta and serving as capital of West Java Province, Bandung’s population grew rapidly in the decades after Indonesia’s independence in 1945, expanding by 2021 to 2.4 million within city limits and around 8.1 million in the Greater Bandung metropolitan area. Rapid growth has resulted in neighborhoods that differ in density, construction type, prevalence of green cover, and building materials. As Indonesia’s second largest metropolitan area, Bandung occupies a position of key economic importance in the country’s urban hierarchy (Roberts, Gil Sander, and Tiwari 2019).1 To gain a fine-grained understanding of heat stress across the city, a “vehicle traverse” study was undertaken for this report in summer 2022 by a team recruited by Bandung Institute of Technology (Institut Teknologi Bandung).2 Thirty-three student “citizen scientists” fitted heat and humidity sensors to motorbikes and drove preplanned routes around the city. More than 51,000 measurements of near-ground air temperature and humidity were conducted in three separate vehicle traverses (conducted at 7–8 a.m., 12–1 p.m., and 5–6 p.m.). By relating the heat measurements to land surface characteristics using a machine learning model, high-resolution heat maps for the entire city area were developed (figure 2.1). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 76 Figure 2.1 Ambient Air Temperatures in Bandung, Indonesia, 2022 A AVERAGE NEAR-GROUND AIR TEMPERATURES 29.4 °C 24.6 °C N 0 1.5 3 kilometers B MEAN NEAR-GROUND AIR TEMPERATURES, BY SUBDISTRICT 25 20 Number of subdistricts 15 10 5 0 26.50 26.75 21.00 27.26 27.50 27.75 28.00 28.25 Temperature (°C) SOURCE: Jones et al. 2023. NOTE: The ambient air temperature surface map was derived using machine learning techniques whose inputs included field measurements of near-ground air temperature taken on July 31, 2022. The map averages temperatures in degrees Celsius (°C) for three times of day (7–8 a.m., 12–1 p.m., and 5–6 p.m.) to arrive at a citywide average daytime temperature map. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 77 DIFFERENCES IN VEGETATION AND BUILT FORM DRIVE HEAT DIFFERENTIALS The resulting datasets show strong spatial variation in heat exposure across Bandung at the time of the heat mapping campaign. Temperature differentials were highest in the early evening, with a 7 degrees Celsius (°C) disparity between hottest and coolest neighborhoods measured via the vehicle traverses conducted at 5–6 p.m. This is consistent with previous urban climate research, which finds the heat island effect is most intense at night, when differential rates of cooling between green and open landscapes versus dense and crowded landscapes drive sharp temperature differences (see, for example, Jáuregui 2000; Oke 1982). Heat island intensity varies systematically with characteristics of the urban surface. Areas where a higher proportion of surface area comprises sealed, impervious surfaces—such as roadways, parking lots, and buildings—show higher temperatures (figure 2.2, panel A). Areas where tree canopy is a higher proportion of surface area are systematically cooler (figure 2.2, panel B). Through hyperlocal measurements of the atmospheric layer that human beings experience daily (that is, ambient air temperatures at near-ground level), the heat mapping campaign provides a more accurate description of spatial variation in heat than is possible through satellite observations or local weather stations (Shandas et al. 2019). In so doing, it draws attention to the pronounced disparity in heat stress experienced between city neighborhoods. Because of the urban heat island (UHI) effect, residents in cities like Bandung already face local temperature disparities that are large enough to influence local health. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 78 Figure 2.2 Relationship between Air Temperature and Ground Cover Characteristics, in Bandung, Indonesia, 2022 A IMPERVIOUS SURFACE 32 30 Temperature (°C) y = -5E-05x2 + 0.024x + 26.6 28 R 2 = 0.31 26 24 0 20 40 60 80 100 Impervious Surface (%) B TREE CANOPY COVER 32 30 Temperature (°C) y = -0.0002x2 - 0.026x + 28.4 28 R 2 = 0.27 26 24 0 20 40 60 80 100 Canopy Cover (%) SOURCE: Jones et al. 2023, using in situ heat measurements plus European Space Agency’s World Cover 2020 dataset (https://worldcover2020.esa.int/). NOTE: Near-ground air temperature measurements were taken during vehicle traverses of Bandung on July 31, 2022. Panels A and B present temperature measurements in degrees Celsius (°C) against the percentages of land cover that are impervious surface and tree canopy cover, respectively. The equations and R-squared values summarize relationships between air temperature and ground cover characteristics with each dot representing a 100-meter grid cell. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 79 POOR RESIDENTS ENDURE HOTTER ENVIRONMENTS Urban climate studies have frequently found a link between the prosperity of neighborhoods and how hot they are. (See, for example, Chan et al. [2012] and Hoffman, Shandas, and Pendleton [2020] as well as the discussion in chapter 3.) Households, businesses, and governments all invest in the built form of city neighborhoods, making decisions about structural type, building materials, plot size, spacing between buildings, and vegetation prevalence that are shaped by their budget constraints and, to varying degrees (depending on the level of enforcement), planning regulations.3 Households also choose residential neighborhoods based, in part, on the “amenities” those neighborhoods offer. Richer households can outbid poorer households when renting or buying in more “pleasant” neighborhoods, which can include those with more greenery and less-impervious surfaces.4 To the extent that a household’s socioeconomic status is correlated with the race or ethnicity of its members, this may also result in racial or ethnic inequalities in exposure to heat island effects. Tree canopy cover and building density are both closely associated with the physical processes that promote heat in urban settings. Poor neighborhoods represent many—but not all—of the city neighborhoods where density is highest and vegetation is lowest. Central business districts and districts where government buildings are concentrated often exhibit high densities of buildings and impervious surfaces and may have low vegetation densities while hosting high-paying residents and workers. In Bandung, similar patterns in the relationship of heat and neighborhood prosperity are evident. Based on poverty map data that estimated the number of people living below Indonesia’s national poverty line in each of the city’s 153 subdistricts, poorer parts of the city systematically have fewer trees. Indeed, the share of area covered by tree canopy in the richest 20 percent of subdistricts is more than double that of the poorest 20 percent of subdistricts (at at 9.2 percent and 3.6 percent of surface area, respectively). Comparing heat stress levels in Bandung with the poverty rates of its subdistricts shows that poorer neighborhoods tend to be hotter than richer neighborhoods (figure 2.3). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 80 Figure 2.3 Relationship between Poverty Rate and Afternoon Temperature, by Subdistrict, in Bandung, Indonesia, 2022 B AFTERNOON AIR TEMPERATURE (°C), BY SUBDISTRICT A RELATIONSHIP BETWEEN POVERTY RATE AND AIR TEMPERATURE IN SUBDISTRICTS OF BANDUNG, INDONESIA 12 Share of population (%) below poverty line 10 8 R2=0.31 6 C SHARE OF POPULATION (%) BELOW POVERTY 4 LINE, BY SUBDISTRICT 2 0 -2 29.5 30.0 30.5 31..0 31.5 Afternoon temperature (°C) SOURCE: World Bank calculations based on temperature data from Jones et al. 2023 and poverty data from Smeru Research Institute’s Poverty and Livelihood Map of Indonesia 2015 (https://smeru.or.id/en/resource/poverty-and-livelihood-map-indonesia-2015). NOTE: Poverty is defined as the proportion of people with per capita household consumption below the national poverty line (Rp 350,000 a day). Modeled heat data are based on vehicle-based ambient air temperature measurements in degrees Celsius (°C) taken on July 31, 2022. In panel (A), a linear regression best fit line is presented with the shaded area denoting the 95% confidence interval. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 81 HEAT AT THE MICRO LEVEL: BUILT FORM AND THERMAL COMFORT Building materials and street layout contribute to city-scale phenomena such as the UHI effect, but they also affect physical processes at the building and block level— including absorption and radiation of heat from surfaces, shading from the sun’s rays, and flow of breezes. Examination of individual sites can yield insights about microfactors that make them hotter or cooler. Whereas temperature measurements are easily relatable, heat stress is necessarily a broader concept than air temperature alone. The human body uses several mechanisms to maintain an internal temperature of 37°C, including sweating and vasodilation (increase of blood flow close to the body’s surface to increase heat shedding). Heat stress results not only from ambient air temperature but also from exposure to radiant heat—such as the sun’s rays (direct or reflected from building materials) or mechanical heat sources such as motor vehicles or the hot air vented outdoors by air conditioners. Humidity is another important contributor to heat stress, because higher water content in the air inhibits heat loss through sweating (see also chapter 1). Finally, wind flow contributes importantly to cooling. When considering heat stress at individual sites, it is important to view these influences holistically (figure 2.4). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 82 Figure 2.4 Factors Affecting Heat Stress at the Site Level Incoming solar radiation (shortwave) Reflected solar radiation Emitted thermal radiation (longwave) Evapotranspiration Shading SOURCE: World Bank elaboration / Estudio Relativo. NOTE: The figure illustrates the effects of solar radiation (both direct and reflected), humidity, shade, and air flow—which, in addition to ambient air temperature, affect levels of heat stress. Local knowledge can be highly applicable to identify and prioritize sites where cooling actions would deliver the greatest benefits. In daily life, heat stress is experienced where daily routines and activities take place in hot environments. Based on a participatory mapping exercise—again involving student “citizen scientists” recruited by Bandung Institute of Technology—figure 2.5 shows sites in Bandung that local residents considered important and representative from a heat mitigation perspective. These include a high school where a lack of shade exposes children to heat; a commercial precinct where dense asphalt adds to a hot experience for shoppers; and a dense settlement where building materials (including iron roofs) and tightly packed buildings contribute to heat stress for primarily low-income residents. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 83 Figure 2.5 The Microcauses of Heat and Cooling in Selected Public Spaces, Bandung, Indonesia 5 3 A SHOPPING CENTER OF "PASAR BARU″ 4 1 2 Heat trapping materials and little shade cause heat 3 exposure for shoppers and merchants. IN FRONT OF THE HIGH SCHOOL BUILDING A DENSE SETTLEMENT IN 1 4 (SMAN 6 BANDUNG) CHAMPELAS ROAD Lack of shade or vegetation. Heat affects school children Dense settlement, lack of trees, metal roofs. and parents. Hot environment affects low-income residents. COOL SPOT NEAR A RECREATION AND SPORT AREAS 2 5 STADIUM ("BABAKAN SILIWANGI″) Dense tree canopy makes for a cool environment. Sports and recreation facilities situated in urban forest provide an attractive amenity. SOURCE: World Bank elaboration / Estudio Relativo. NOTE: The pairs of photographs and thermal images from selected sites illustrate volunteers’ perceptions of why these sites are hotter (red-shaded numbers) or cooler (blue-shaded numbers) than average. The top left map shows these sites’ locations in relation to Bandung city center. The shading of the map indicates near-ground air temperatures derived from field measurements taken on July 31, 2022. The map averages temperatures for three times of day (7–8 a.m., 12–1 p.m., and 5–6 p.m.) to arrive at a citywide average daytime temperature map. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 84 URBAN HEAT WILL INTENSIFY IN COMING DECADES Heat stress disparities are already starkly evident in East Asian cities and often cause low-income communities—which have a lower capacity to adapt through air conditioning and other means—to bear the brunt of heat waves.5 But as climate change intensifies in the coming decades, the combined effects appear certain to place heat-prone neighborhoods under extreme stress, requiring urgent steps to protect vulnerable residents and safeguard labor productivity. Cambodia’s capital city, Phnom Penh, presents one such example. It underwent rapid expansion and densification of its urban footprint between 1995 and 2010, during which its population more than doubled (UN DESA 2015). An urban climate model built for this report estimates the city’s current nighttime UHI effect, measured as the difference between city center and outlying rural temperatures, at 2.5°C.6 Central districts of Phnom Penh already experience strong additional heat stress owing to human influences—some 23–25 heat wave days per year, compared with 2–5 such days in surrounding rural areas. Using temperature datasets for the recent past and modeled atmospheric data for future periods, heat stress variables were modeled for the city. This was done for both (a) an optimistic scenario in which global emissions are curbed (the “SSP1-1.9” scenario); and (b) a pessimistic scenario in which carbon emissions maintain a stronger upward path before stabilizing (“SSP3-7.0”).7 Under the high-emission scenario, the number of heat wave days is projected to increase to 60 days per year (more than a twofold increase) in the most-affected areas by 2050 (table 2.2). Neighborhoods with extensive trees and vegetation, and more space between dwellings, are projected to remain cooler on average, with around one-third fewer days classified as heat waves. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 85 Table 2.2 Projected Heat Wave Days, Tropical Nights, and Energy Needs for Space Cooling, by Climate Change Scenario, in Phnom Penh, Cambodia, 2050 METRIC OPTIMISTIC SCENARIO PESSIMISTIC SCENARIO Annual number of heat wave daysa 40 (73 percent increase) 60 (160 percent increase) Tropical nights (% of year)b 80 percent 90 percent Energy needs for space coolingc 14 percent increase 30 percent increase SOURCE: VITO NV, using its UrbClim model for this report. NOTE: Under the “optimistic scenario,” global carbon emissions are curbed; under the “pessimistic scenario,” emissions maintain a stronger upward path before stabilizing. These scenarios correspond to the selected Shared Socioeconomic Pathways (SSPs) developed for the Intergovernmental Panel on Climate Change—namely, SSP1-1.9 and SSP3-7.0, respectively. a. Heat wave days: the annual number of days where temperatures exceed the 90th percentile temperature of a historical reference period (2000–20) for at least three consecutive days. b. Tropical nights: the annual number of nights where temperature remains above 25 degrees Celsius c. Measured as cooling degree days. Daily cooling needs are calculated by comparing outdoor temperatures with a reference point representing indoor thermal comfort. The cooling requirement for each period is multiplied by the length of the period. This metric captures projected energy demand driven by temperature effects, ignoring the growth driven by increased adoption of air conditioning. Perhaps the most crucial impact of these heat stress disparities, as climate change progresses, will be the health implications of differences in nighttime temperatures. Very hot nights where temperatures remain above 25°C place the human body under extreme stress. Particularly for old people, young children, and those with health vulnerabilities, such temperatures at night inhibit cooling, rest, and recovery and are associated with higher death rates. Even under an optimistic climate scenario, Phnom Penh is projected to experience hot nights for 80 percent of the year by 2050. Under a high-climate-change scenario, the most heat-prone neighborhoods would experience such conditions nearly all year round (map 2.1). Such heat stress levels present a direct threat to human health, particularly where construction quality is poor. In such circumstances, measures to promote cooler physical spaces are essential but not sufficient; city authorities also face a pressing need to intervene in a targeted manner to protect the lives of vulnerable residents during heat waves. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 86 Map 2.1 Projected Number of Hot Nights Per Year for a Present-Day Reference Period and under Future Climate Scenarios, Phnom Penh, Cambodia A PRESENT (2000-2020) 300 250 200 NIGHTS PER YEAR 150 100 50 B MATRIX OF 2030 AND 2050 PROJECTIONS 2030 2050 SSP1-1.9 SSP3-7.0 SOURCE: Analysis conducted using VITO NV’s UrbClim model for this report. NOTE: “Hot nights” are defined as nights where air temperature does not fall below 25 degrees Celsius. In the matrix of models for 2030 and 2050, “SSP1-1.9” represents an optimistic scenario, under which global carbon emissions are curbed. “SSP3-7.0” represents a pessimistic scenario, under which emissions maintain a stronger upward path before stabilizing. These scenarios correspond to the selected Shared Socioeconomic Pathways (SSPs) developed for the Intergovernmental Panel on Climate Change. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 87 SUMMARY AND CONCLUSION Using Bandung in Indonesia and Phnom Penh in Cambodia as case studies, this chapter has shown how neighborhoods within individual East Asian cities are marked by sharp differences in heat stress. These differences result from, among other things, differences in vegetation and built form. And in East Asian cities such as Bandung, the differences also correlate strongly with poverty rates: poorer neighborhoods also tend to be hotter neighborhoods. The chapter also finds that, in future decades, the interaction of the UHI effect with climate change will cause the neighborhoods that are already hottest to become even more uncomfortable, especially at night. The next chapter in this report, chapter 3, further explores how extreme heat affects a city’s competitiveness, livability, and inclusiveness. REFERENCES Chakraborty, T., A. Hsu, D. Manya, and G. Sheriff. 2019. Jones, N., N. Wijaya, J. Williams, Z. Boyce, V. Shandas, and “Disproportionately Higher Exposure to Urban Heat in C. Deuskar. 2023. “Mapping Urban Heat through a Lower-Income Neighborhoods: A Multi-City Perspective.” Community-Based Mobile Campaign in Bandung City, Environmental Research Letters 14 (10): 105003. Indonesia.” Background paper for this report, World Bank, Chan, E., Y. Yang, W. B. Goggins, J. J. Kim, and S. M. Griffiths. 2012. Washington, DC. “A Study of Intracity Variation of Temperature-Related Oke, T. R. 1982. “The Energetic Basis of the Urban Heat Island.” Mortality and Socioeconomic Status among the Chinese Quarterly Journal of the Royal Meteorological Society Population in Hong Kong.” Journal of Epidemiology and 108 (455): 1–24. Community Health 66 (4): 322–27. Oke, T. R., G. Mills, A. Christen, and J. A. Voogt. 2017. Urban De Ridder, K., D. Lauwaet, and B. Maiheu. 2015. “UrbClim – A Fast Climates. Cambridge: Cambridge University Press. Urban Boundary Layer Climate Model.” Urban Climate 12: Roberts, M., F. Gil Sander, and S. Tiwari, eds. 2019. Time to ACT: 21–48. Realizing Indonesia’s Urban Potential. Washington, DC: Hoffman, J. S., V. Shandas, and N. Pendleton. 2020. “The Effects World Bank. of Historical Housing Policies on Resident Exposure Shandas, V., J. Voelkel, J. Williams, and J. Hoffman. 2019. to Intra-Urban Heat: A Study of 108 US Urban Areas.” “Integrating Satellite and Ground Measurements for Climate 8 (1): 12. Predicting Locations of Extreme Urban Heat.” Climate 7 Jáuregui, E. 2000. El Clima de la Ciudad de México. Mexico City: (1): 5. Plaza y Valdés. UN DESA (United Nations Department of Economics and Social Johansson, E. 2006. “Influence of Urban Geometry on Outdoor Affairs). 2015. World Urbanization Prospects: The 2014 Thermal Comfort in a Hot Dry Climate: A Study in Fez, Revision. New York: United Nations. Morocco.” Building and Environment 41 (10): 1326–38. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 88 NOTES 1 Bandung was chosen for this case may also be less likely to invest in study using the UrbClim framework study after consultations with those neighborhoods. (De Ridder et al. 2015). Indonesia’s Ministry of National 5 As discussed in chapter 3, this is 7 The Shared Socioeconomic Pathways Development Planning (Bappenas) not necessarily the case for all East (SSPs) are a set of five scenarios and the city’s local authorities. Asian cities. Hence, out of a sample for global socioeconomic changes 2 The study was also undertaken in of 25 cities globally, Chakraborty projected up to 2100. Developed collaboration with CAPA Strategies. et al. (2019) find that although to inform the Intergovernmental Further details of the study, including temperatures are higher in poorer Panel on Climate Change’s Sixth of the machine learning models neighborhoods in 18 out of 25 cities, Assessment Report, they present applied to derive heat maps for this is not the case in the remaining 7 plausible future states of the world the entire city, can be found in a cities, which include Bangkok, Beijing, under differing social and political background paper for this report and Manila. Several aspects of urban pathways, each with different (Jones et al. 2023). development, such as a city’s housing implications for greenhouse gas 3 For a discussion of Indonesia’s market and connectivity, affect emissions and climate change planning system, see chapter 7 of whether more affluent residents outcomes. The optimistic scenario Roberts, Gil Sander, and Tiwari (2019). live in the city centers where UHI corresponds to “SSP1: Sustainability 4 As the discussion in this paragraph intensity is generally higher. (Taking the Green Road),” which makes clear, the relationship between 6 Differential cooling between urban presents a world that turns toward a neighborhood’s temperature (and, and surrounding rural areas is most sustainability and cooperative more generally, the level of thermal pronounced at night, with urban management of the global commons. comfort it offers) and its prosperity surfaces giving up their heat less Global emissions peak between potentially runs in both directions. quickly because of the thermal 2040 and 2060, and global mean air Hence, wealthier households may inertia of materials and obstruction temperatures rise by 1.0–1.8°C over choose to “sort” toward cooler of radiation by buildings. For these preindustrial levels by 2100. The neighborhoods. They may also, reasons, the nighttime UHI effect is pessimistic scenario corresponds for example, have more success, more pronounced than the daytime to “SSP3: Regional Rivalry (A Rocky or be more active in, lobbying for effect. Climate modeling allows Road),” which depicts a world of investments in urban greenery that estimation of nighttime UHI effects resurgent nationalism, material- help to make a neighborhood cooler that are less readily perceptible intensive consumption, and limited and therefore more pleasant. To through satellite imagery but have cooperation to preserve shared the extent that poorer households major implications for human health resources, resulting in warming of live in more informal or illegal by impeding sleep and recuperation. 2.8–4.6°C over preindustrial levels neighborhoods, local governments Modeling was conducted for this by 2100. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 89 CH C H A PT ER 3 HOW DOES URBAN HEAT AFFECT CITIES AND RESIDENTS? UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 90 KEY MESSAGES Extreme heat—be it from the urban heat island effect, global warming, or the combination of the two negatively affects every aspect of urban life. Extreme heat makes cities less economically productive by reducing the capacity of workers, machinery, and infrastructure. Cities may face average gross domestic product (GDP) losses of over 5 percent by 2100 because of increasing temperatures, with the worst-impacted cities losing over 10 percent of GDP. Extreme heat is responsible for over 100,000 excess deaths per year in the East Asia region, a number that is projected to increase. It also affects livability by increasing rates of illness, accidents, crime, and violence. Extreme heat undermines environmental sustainability by increasing energy consumption and associated carbon emissions because of the increased need for mechanical cooling. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 91 INTRODUCTION Extreme heat—be it from the urban heat island (UHI) effect, climate change, or both combined negatively affects every aspect of urban life. This chapter summarizes the extensive evidence on the negative impacts of extreme heat on urban prosperity, livability, and environmental sustainability. The chapter begins by providing a simple framework to understand the channels through which extreme heat may affect a city’s productive capacity and surveying the evidence that relates to these different channels for the East Asia region. It goes on to discuss the toll of extreme heat exposure on the human body, resulting in over 100,000 excess deaths per year in East Asia as well as higher rates of illness, accidents, suicides, and violent crimes. It also explores the ways in which the negative impacts of extreme heat are felt disproportionately by the urban poor, elderly, and other already disadvantaged groups. Finally, it discusses the impacts of air conditioning, which is being rapidly adopted in the region and is accompanied by higher energy consumption and the outdoor release of heat—further worsening the UHI effect. To the extent possible, this chapter relies on evidence from East Asia but falls back on literature from other countries when the evidence base for East Asian countries is thinner. EXTREME HEAT UNDERMINES URBAN PROSPERITY EXTREME HEAT MAKES CITIES LESS PRODUCTIVE In addition to its more well-known health impacts, which are discussed later in this chapter, extreme heat constrains every input into a city’s productive capabilities. A city’s real output (Y) depends on several factors: the number of hours worked by its workers (L); the amount of physical capital that its firms use in production (K); the supply of human capital (H) and the efficiency with which it is used (e); and a city’s total factor productivity (A), which captures other factors, including technology, that affect a city’s output. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 92 The dependence of a city’s real output on these factors can be captured by its “production function,” which provides a simple and convenient framework for thinking through the different potential channels through which extreme heat might affect a city’s economy. Hence, a city’s production function can be expressed as Y=f(L,K,eH,A). As shown also in figure 3.1, extreme heat—whether due to the UHI effect or climate change—can potentially reduce a city’s real output by Reducing the number of hours worked (L) because of work stoppages and breaks taken during extreme heat as well as absences induced by heat-related illness; Reducing physical capital inputs (K)—for example, owing to heat-induced breakdowns and accelerated deterioration of machinery and physical inputs into production; Reducing the effectiveness (e) of existing human capital inputs in the short run—for example, through the effects of heat exhaustion on physical and mental abilities, which affect both manual labor and cognitive work; Reducing the supply (H) of human capital in the long run by negatively affecting educational outcomes (as further discussed below) and making a city less attractive to high-skilled workers;1 and Reducing total factor productivity (A) owing to, for example, a diminished reliability of infrastructure services because of damage caused by extreme heat to roads, bridges, power plants, power lines, water pipes, and other forms of infrastructure. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 93 Figure 3.1 Channels of Extreme Heat’s Impacts on a City’s Aggregate Real Output L K H e A Number Physical Human Effectiveness Total factor of hours capital capital of human productivity worked Extreme Extreme heat capital Extreme heat damages heat causes impacts learning infrastructure. Extreme heat Heat exhaustion breakdowns and and makes causes work reduces physical and deterioration labor market stoppages mental abilities. of machinery. less attractive. and absences. Y Output SOURCE: World Bank elaboration. REDUCTION IN WORKER PRODUCTIVITY Heat already hurts manufacturing in the region, and the impact will increase with climate change. An extra day with temperature higher than 32 degrees Celsius (°C) decreases the output of Chinese manufacturing firms by 0.45 percent relative to an extra day with temperature of 10–16°C (Zhang et al. 2018). This temperature effect is driven by decreases in total factor productivity (TFP).2 The warming effects of climate change, not accounting for UHI, will result in reductions in output of 12 percent annually by midcentury (2040–59) for the average Chinese firm, assuming no further adaptation by firms (Zhang et al. 2018). If the output share of manufacturing remains at 32 percent of China’s national gross domestic product (GDP), this loss translates to a 3.8 percent reduction in China’s annual GDP. A 14-day heat wave in Nanjing, China, in 2013 reduced productive working time by 2.5 percent across all industries. Losses were highest (4.2–4.5 percent) in outdoor work (such as agriculture, mining, and construction) and lower (about 0.7 percent) in indoor work including manufacturing and services. The impact of the heat wave on health, work productivity, and capacity caused a total loss equivalent to 3.4 percent of the city’s gross value of production in 2013. The largest share of this loss was in manufacturing (Xia et al. 2018).3 UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 94 Heat stress can also reduce performance in cognitive tasks associated with knowledge work, although the mechanisms by which this occurs are not fully understood (Taylor et al. 2016). A review of 24 studies from around the world suggests that performance in office work increases with indoor temperatures up to 21–22°C and decreases when temperatures exceed 23–24°C. Average productivity of office work at an indoor temperature of 30°C is 9 percent below productivity at 22°C (Seppanen, Fisk, and Lei 2006). DAMAGE TO URBAN INFRASTRUCTURE Exposure to extreme heat can also damage urban infrastructure. For example, heat can damage transportation infrastructure by causing roads and rails to buckle owing to thermal expansion and by melting or softening paving material. It can also reduce the performance of electrical power lines and photovoltaic cells. Moreover, heat can increase the risk of rupture of oil and gas pipes, cause geotechnical failures in embankments and earthworks, and damage underground communications infrastructure (Dodman et al. 2022). Infrastructure damage not only results in immediate risks, inconveniences, and costs but also further reduces a city’s real output by reducing TFP (that is, A in the production function). REDUCTION IN GLOBAL GDP Several models suggest that warming due to climate change will also reduce global GDP: Burke and Tanutama (2019) estimate that tropical countries are already more than 5 percent poorer than they would have been had there been no warming since 2000. Burke, Hsiang, and Miguel et al. (2015b) estimate that global warming, if unmitigated, would reduce average global incomes by 23 percent by 2100. Kahn et al. (2019) find that an increase in the average global temperature of 0.04°C per year, in a worst-case scenario, would reduce global real GDP per capita by more than 7 percent by 2100. 4 Crucially, these estimates consider only temperature increases due to climate change. In cities, the total economic damage from heat during the twenty-first century when the UHI effect is taken into account can be 2.6 times the magnitude of losses when it is not (box 3.1). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 95 BOX 3.1 THE COMBINED IMPACTS OF THE URBAN HEAT ISLAND EFFECT AND CLIMATE CHANGE Most discussions of projected 4°C warmer by 2050. One in four could warming due to the UHI effect, these GDP temperature increase in coming decades become more than 7°C warmer by 2100. reductions increase to 1.4 percent and focus exclusively on warming driven 1.7 percent under RCP 4.5 and RCP 8.5, by global climate change. However, the Crucially, the combined economic impact respectively. The worst-off city could face urban heat island (UHI) effect will add of climate change and the UHI effect is GDP losses of up to 10.9 percent in 2100. significantly to the warming in cities. likely to be greater than the sum of their individual impacts. This is because the The total economic damage from heat Estrada, Wouter Botzen, and Tol (2017) impact of temperature on the economy in cities during the twenty-first century estimate future temperature increases is nonlinear: the negative effect of an when the UHI effect is taken into account due to the combined effects of both UHI increase in temperature of 1°C—from, for can be 2.6 times the magnitude of losses and climate change in the 1,692 largest example, 35°C to 36°C—is greater than when it is not.b If ignored, the economic cities in the world, all of which have a from an increase from 30°C to 31°C (Burke, impacts of the UHI effect could undo population of at least 300,000. They Hsiang, and Miguel et al. 2015b). the benefits of efforts focused only estimate that by 2050, the UHI effect on mitigating global climate change. will add 0.84°C to the temperature of Estrada, Wouter Botzen, and Tol (2017) In contrast to global warming, the UHI the median city, and 2.08°C to the most estimate that warming due to climate effect is furthermore a local negative populous 5 percent of cities, over and change alone would reduce the gross environmental externality that can above the warming caused by climate domestic product (GDP) of the median be attenuated through local policy change. The combined effect of UHI city in 2050 by 0.7 percent in an RCP 4.5 actions without the need for global and climate change means that one scenario and by 0.9 percent in an RCP 8.5 collective action. in five cities could become more than scenario.a However, when combined with a. Representative Concentration Pathways (RCP) are scenarios used by the Intergovernmental Panel on Climate Change (IPCC) to represent different possible future levels of global warming. RCP 4.5 represents an intermediate scenario, whereas RCP 8.5 represents a worst-case scenario. b. This is based on the Dynamic Integrated Climate-Economy (DICE) model, which estimates the economic impacts of climate change. It estimates that 2.5°C of warming would lead to a welfare loss equivalent to 1.5 percent of GDP. Economic losses from heat are projected to be greater in poorer countries. Burke, Hsiang, and Miguel (2015b) estimate that unmitigated global warming would widen income inequality between countries. However, this differential impact does not appear to be the result of any greater ability on the part of wealthier countries to adapt to extreme heat, since the differences in countries’ adaptive capacity are insignificant, according to the authors. Instead, the difference occurs mainly because poorer countries tend to have higher baseline temperatures, which means that further warming has a greater impact on their productivity than it does on wealthier, colder countries, where warming could even increase productivity. This differential impact on productivity relates to the importance of nonlinearities in the damage function, which means that the impacts of the UHI effect and climate change-induced warming can potentially compound each other (box 3.1). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 96 GDP LOSSES ACROSS THE REGION In several Southeast Asian countries, reduced labor productivity due to extreme heat induced by climate change is projected to cause GDP losses of 5.5–6 percent by 2030, not even considering the UHI effect. An International Labour Organization (ILO) study estimates that productivity loss due to heat stress cost Southeast Asian countries the equivalent of 3.1 percent of working hours (6.9 million full-time jobs) in 2015, mostly in agriculture and construction (ILO 2019).5 Considering rising temperatures due to climate change and shifts in the economy, the productivity loss due to extreme heat in 2030 is projected to be the equivalent of 3.7 percent of working hours (13 million full-time jobs). Nearly 8.9 percent of working hours in construction and 3.9 percent of working hours in industry are projected to be lost because of extreme heat in Southeast Asia by 2030. In proportional terms, the reduction in working hours due to heat stress in Southeast Asia in 2030 is projected to be highest in Cambodia (7.8 percent), Thailand (6.4 percent), and Viet Nam (5.1 percent). But in absolute terms, the highest projected losses are for Indonesia (equivalent of 4 million full-time jobs) and Viet Nam (3.1 million jobs). By comparison, China (not counted under Southeast Asia) is projected to lose a small proportion (0.8 percent) of working time but a large absolute amount (the equivalent of 5.5 million full-time jobs) by 2030. Kjellstrom (2016), meanwhile, estimates that, in several Southeast Asian countries, reduced labor productivity due to extreme heat will cause GDP losses of 5.5–6 percent by 2030, including 6 percent (US$250 billion) in Indonesia, 5.9 percent (US$85 billion) in the Philippines, and 5.7 percent (US$85 billion) in Viet Nam. Heat-related GDP losses in China in 2030 will be smaller in terms of proportion (0.8 percent) but larger in absolute terms: US$450 billion.6 Again, these projections crucially neglect the UHI effect, which means that they likely underestimate, perhaps substantially, the productivity losses in cities (box 3.1). New analysis conducted for this study suggests that in large Southeast Asian metros as well as in warmer cities in the rest of East Asia during the summer, the negative economic impacts of warmer-than-average temperatures is significant enough to be visible in the intensity of nighttime lights, a widely used proxy for economic activity (box 3.2). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 97 BOX 3.2 IMPACTS OF EXTREME HEAT ON URBAN ECONOMIC ACTIVITY ACROSS EAST ASIAN CITIES News stories about extreme heat The results diverge across cities in In a month hit by an extreme heat event, disrupting economic activity have become different subregions. In Southeast Asia, their nighttime light intensities dropped increasingly common in recent years. where background climatic conditions by an estimated 5.5 percent. By contrast, For example, the British Broadcasting are hot and humid all year round, extreme the estimated impact of extreme heat on Corporation (BBC) reported news of heat had severe negative estimated economic activity for cities with cooler trains being cancelled and rail services impacts on local economic activity background climates is positive, perhaps disrupted in the United Kingdom after the in large metropolitan areas.b Those reflecting the impacts of warmer, more country posted its hottest day on record cities, including quite a few Indonesian pleasant evenings on outdoor economic in July 2022 (Nanji 2022). The Wall Street secondary cities as well as national activity (figure B3.2.1, panel B). Journal carried news of electricity and capitals of the region, darkened by nearly production shortages in China as a result 4 percent on average in a month hit by Given that gross domestic product (GDP) of its worst heat wave in six decades in extreme heat.c By contrast, the estimated and the nighttime lights data do not August 2022 (Xie and Li 2022). impact is insignificant for smaller cities perfectly correspond to each other, any (figure B3.2.1, panel A). Although the estimates of changes in GDP derived To what extent, then, does extreme heat exact mechanism behind this result from lights data must be treated with affect urban economic activity across requires further research, it is in line with caution. That said, the range of the GDP- East Asian cities in the short run? a previous finding that cities with larger to-lights elasticity in the literature (0.3–1) To answer this question, the analysis and denser populations are often under suggests that the estimated percentage in this box employs spatially granular greater heat stresses because of larger- reduction in light intensity associated monthly data on temperatures as well as scale built environments and human with extreme heat could be interpreted as on nighttime lights—a proxy for a city’s activities (Liu et al. 2022). a reduction in local GDP of 1.2–3.9 percent aggregate level of economic activity. for large metropolitan areas in Southeast By measuring extreme heat as a monthly The analysis of 536 other East Asian Asia and 1.7–5.5 percent for cities in temperature anomaly relative to each cities (excluding Southeast Asia) during the rest of East Asia cities with warm city’s historical norm,a this analysis summer months (June–August) reveals baseline climates. estimates how a city’s nighttime light that cities with warmer baseline climates, intensity changed during any month including many manufacturing and when the city experienced extreme heat financial hubs in China were affected between April 2012 and December 2020. by extreme heat quite significantly.d Continued UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 98 Figure B3.2.1 Estimated Impacts of Extreme Heat Events on City Nighttime Light Intensity in East Asia, by Subregion, April 2012–December 2020 A SOUTHEAST ASIA B REST OF EAST ASIA 0.02 0.10 0 0.05 Estimated elasticity Estimated elasticity −0.02 0 −0.04 −0.05 −0.06 −0.10 −0.08 −0.15 Smaller cities Large metros Cooler cities Warmer cities SOURCE: World Bank calculations based on the analysis of Visible Infrared Imaging Radiometer Suite (VIIRS) nighttime lights monthly composites (https://payneinstitute.mines.edu/eog-2/viirs/) and monthly weather data from Climatology Lab, TerraClimate (https://www.climatologylab.org/terraclimate.html). NOTE: Cities are defined as urban centers following the “degree of urbanization” methodology of the Global Human Settlement Layer (GHSL) Urban Centre Database. Each red marker shows the estimated impact of an extreme heat event, obtained by regressing the natural log of sum of lights in city c in month m on a dummy variable which takes the value 1 if temperature of city c in month m deviated from the city’s long-run mean temperature for the month by at least +2 standard deviations; the dummy variable takes the value 0 otherwise. Vertical bars indicate the bounds of the 90 percent confidence interval associated with the corresponding estimated impact. Panel A is based on the analysis of 170 Southeast Asian cities. Panel B is based on the analysis of 536 cities in the rest of East Asia. The analysis of Southeast Asian cities (panel A) covers all 12 months of the year because monthly temperatures remain relatively warm throughout the year for those cities. In contrast, the analysis of cities in the rest of East Asia (panel B) covers only warm months, (June, July, and August) given their more variable climates. a. For a detailed definition of extreme heat events, see chapter 1, annex 1A. b. Southeast Asian countries included in the analysis are Cambodia, Indonesia, Malaysia, Myanmar, the Philippines, Thailand, and Viet Nam. Cities are classified as large metros (smaller cities) if they have a population of at least (below) 1.5 million as classified by the Organisation for Economic Co-operation and Development. c. Indonesian secondary cities classified as large metropolitan areas include Surabaya, Bandung, Yogyakarta, Medan, Semarang, and others. d. These East Asian cities include those in China (including Hong Kong SAR); Japan; the Democratic People’s Republic of Korea; the Republic of Korea; Mongolia; and Taiwan, China. Cities with warmer (cooler) baseline climates are derived based on the top (bottom) half of the global distribution of long-run mean monthly temperature. Warmer cities include the Chinese cities of Guangzhou, Shanghai, Hangzhou, Wuhan, and Chongqing; Hong Kong SAR, China; and some cities from high-income economies including Osaka-Kyoto, Japan. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 99 EXTREME HEAT HURTS EDUCATIONAL OUTCOMES As with office work, heat stress impairs the ability of students to perform cognitive tasks. Heat affects both classroom learning and exam-day performance, with potential long-term impacts on productivity and earnings. Studies show that the number of hot school days is correlated with lower standardized test scores, both in the United States (Park et al. 2020) and across 58 countries (Park, Behrer, and Goodman 2021). These studies show that hot nonschool days were not correlated with lower scores, suggesting a direct impact of heat on learning itself. In the case of the United States, replacing a school day in the 60s on the Fahrenheit scale (roughly 15–21°C) with a hotter school day lowers achievement, with the extent of that damage increasing roughly linearly with temperature above 70 degrees Fahrenheit (21°C). According to a study on China (Zivin et al. 2018), exam-day temperatures affect test scores, with negative consequences for students’ further educational opportunities. An increase of 1 standard deviation in exam-day temperature (3.3°C) during the Chinese National College Entrance Examination reduces the test score by 1.1 percent (0.096 standard deviations), which in turn reduces the probability of admission into a first-tier university by nearly 2 percent. In terms of the production function discussed earlier, the effect of heat on education lowers the supply of skilled labor (H) and the effectiveness of each worker per hour (e). To the extent that skilled labor is an input into innovation, it could also slow the growth of total factor productivity (A). EXTREME HEAT MAKES CITIES LESS LIVABLE HEAT CAUSES ILLNESS AND DEATH Over 100,000 people in the East Asia region die each year because of extreme heat, a number that is likely to increase. Studies of East Asian cities including Bangkok and Hong Kong SAR, China, estimate increases in mortality of 2–6 percent per 1°C increase in temperature above 29°C (Chan et al. 2012). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 100 In a study of global deaths related to nonoptimal temperatures (hot and cold), Zhao et al. (2021) find that the average number of excess deaths related to heat per year was over 21,000 in Southeast Asia and nearly 80,000 in the Rest of East Asia. Surprisingly, however, the equivalent numbers of cold-related deaths were much higher (168,000 and 1.16 million, respectively). Rising temperatures due to climate change (a 0.26°C increase in daily mean temperature per decade from 2000 to 2019 in both regions) means that although heat-related deaths are increasing in these regions, they are far outnumbered by the decrease in cold-related deaths. This is particularly true in Southeast Asia, where heat-related deaths as a proportion of all deaths increased from 0.45 percent in the 2000–03 period to 0.73 percent in 2016–19, while the share of cold-related deaths decreased from 5.38 percent to 3.22 percent.7 By contrast, however, Bressler et al. (2012) project that in all middle-income countries in East Asia (and in the world as a whole), future warming during this century will increase heat-related mortality more than it decreases cold-related mortality in most scenarios, leading to a net increase in mortality, even adjusting for increased resilience due to income growth. Extreme heat affects health in a variety of ways. Heat-related illnesses include dehydration, heat cramps, and heat stroke. Overheating also increases the likelihood of death from preexisting conditions such as respiratory and cardiovascular diseases (WHO 2018). High temperatures further reduce water quality by enabling the development of harmful algal blooms and toxic bacteria and the depletion of oxygen in urban water bodies. Spikes in the concentrations of waterborne bacteria and viruses can trigger outbreaks of diseases including cholera, diarrhea, dysentery, hepatitis A, typhoid, and polio (USAID 2020). Heat also kills and injures people indirectly, by increasing rates of accidents and suicides. The increase in accidents can likely be attributed to the impact of heat stress on cognition and attention (Park, Pankratz, and Behrer 2021). For example, although evidence for East Asian countries is sparse, research shows that in the United States, fatal car crashes increase by a significant amount (3.4 percent) on heat wave days (Wu, Zaitchik, and Gohlke 2018), which is also consistent with earlier cited findings of the impacts of heat on cognitive performance. Furthermore, a recent study using data from California in the United States finds that both indoor and outdoor workplace injuries rise with temperature, including injuries not usually associated with heat—for example, falling, being struck by a crane, or mishandling dangerous machinery (Park, Pankratz, and Behrer 2021). A day with temperatures of 29–32°C has a 5–7 percent higher injury risk than a day with temperatures of 15–21°C, while a day with temperatures above 38°C has a 10–15 percent higher injury risk. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 101 Tragically, heat also correlates positively with suicide rates. Summer peaks in suicide rates have been observed since the nineteenth century, perhaps suggesting that global warming could lead to an increase in suicides. Burke et al. (2018) estimate that suicide rates rise by 0.7 percent in the United States and by 2.1 percent in Mexico per 1°C increase in monthly average temperature. Although the underlying mechanisms linking heat stress to suicide are not fully clear, the authors’ analysis of temperature’s effects on depressive language in social media posts suggests that it is associated with a direct negative impact of heat on mental well-being. Although no studies examine links between heat and suicide in East Asia, Chen et al. (2023) study the relationship between exposure to extreme heat and increased depression in adults over the age of 45 in China. They find that symptoms of depression, such as sleeplessness, distractibility, and lack of motivation, are significantly correlated with the number of days in the preceding year that had extremely high temperatures compared to the local mean. This relationship is only significant among respondents over the age of 60, suggesting that the elderly may be particularly vulnerable to the effects of extreme heat on mental health. Notably, the authors find a stronger correlation between depressive symptoms and exposure to extremely cold days than to hot days, which means that the net impact of increased temperatures on depression may be positive or negative in a given location, depending on how the distribution of extreme hot and cold days changes over time. EXTREME HEAT MAY EXACERBATE CRIME, VIOLENCE, AND CONFLICT Despite the lack of much research on the link between heat and crime in East Asia specifically, international studies suggest that such a link exists. Using highly spatially granular data, Heilmann and Kahn (2019) provide credible causal evidence that in Los Angeles, excessively high temperatures lead to increases in violent crime in particular, with no significant relationship between heat and nonviolent crime. They estimate that violent crime increases by 5.7 percent on days with maximum temperatures above 85°F (29.4°C). The authors argue that the impact of heat on violent crime but not on nonviolent crime strongly supports the notion that the mechanism underlying this relationship is increased aggression rather than greater opportunities for crime on hot days due to reduced policing or increased crowds. Bruederle, Peters, and Roberts (2017) find that in South Africa as a whole, an increase of 1 standard deviation in the daily maximum temperature within an average month of the year (an increase of 2.7°C) increases total crime by 3.7 percent, while an equivalent increase in the daily minimum temperature (1 standard deviation, UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 102 also 2.7°C) increases crime by 5.3 percent. The study also finds that these effects are driven mainly by the impacts of heat on violent crime (particularly sexual violence) and burglary. The authors describe their findings as strongly supporting a “heat- aggression link,” although the impact on burglary is less easily explained. Similarly, a study of Madrid (Sanz-Barbero et al. 2018) found positive correlations between heat waves (temperatures above 34°C) and the incidence of intimate partner violence—a finding the authors describe as consistent with a model that explains aggression as the result of irritability and stress caused by high temperatures. Although further studies are needed to fully document and understand possible links between heat and crime, particularly in East Asia, the literature suggests that impairment of cognitive processes may result in increased crime. A metastudy on the relationship between climate and conflict (defined as being usually but not necessarily violent) finds that a 1 standard deviation increase in temperature increases interpersonal conflict by 2.4 percent and intergroup conflict by 11.3 percent (Burke, Hsiang, and Miguel 2015a). While there appears to be a global relationship between heat and conflict, it is unclear whether this relationship holds in Asia. Mares and Moffett (2016) find that across 57 countries, a 1°C increase in annual mean temperature is correlated with a 5.9 percent increase in national homicide rates, while a 1 standard deviation increase is correlated with a 58 percent increase. However, these results vary by region within their sample, and the correlation between temperature and homicides in Asia is insignificant. In general, the literature does not establish any universal temperature threshold above which the impacts of heat on health, productivity, or crime and violence kick in (box 3.3). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 103 BOX 3.3 DO THE NEGATIVE EFFECTS OF HEAT KICK IN AT A SPECIFIC TEMPERATURE? Adverse physiological impacts occur difficult to answer this question based be the result of varying research design when extreme heat exceeds population- on the existing literature because each choices; means of measuring temperature; specific thresholds. Even small increases study uses different temperature intervals which temperature measure is used in temperatures beyond those for which and thresholds. For example, the studies (for example, air temperature versus a population is prepared increase the discussed in this chapter report the land surface temperature); and data likelihood of illness and death. To some impacts of heat availability. It seems a priori unlikely that extent, the body can acclimatize to all the different impacts of heat that this higher temperatures over time, which is Above 23–24°C on office work chapter discusses—on physical labor, why definitions of heat waves are often (Seppanen, Fisk, and Lei 2006); cognitive tasks, crime, health, and so on— relative to mean local temperatures rather would share a single temperature above than universal temperature thresholds Above 27°C and above 32°C on which their effects are felt, though this (see also the discussion in chapter 1 on manufacturing work (Somanathan et al. intuition would need additional research this point). 2021; Zhang et al. 2018); to be confirmed. Populations can also deacclimatize and Above 12–14°C, above 21°C, and above Additionally, as earlier chapters have reacclimatize to heat over the course 27°C on education (Park et al. 2020; also discussed, other factors besides of a year. This is why heat waves that Park, Behrer, and Goodman 2021; Ziven temperature itself affect the level of take place earlier in the summer, before et al. 2018); and risk from extreme heat. As WHO (2018) the population has had a chance to states, “The precise threshold at which acclimatize, are associated with higher Above 29°C on crime and violence temperature represents a hazardous mortality than later ones (Anderson and (Heilmann and Kahn 2019). condition varies by region, other factors Bell 2011). such as humidity and wind, local levels of These differences in intervals, rather human acclimatization and preparedness For more indirect impacts—on, say, than resulting from different underlying for heat conditions.” productivity, education, and crime—it is thresholds for these impacts, may simply EXTREME HEAT MAKES CITIES LESS INCLUSIVE At the local level, the burden of extreme temperatures on prosperity and livability is often felt disproportionately by the poor. Studies of China and Indonesia find that the adoption of air conditioning is significantly correlated with household income (Davis et al. 2021; Pavanello et al. 2021). This suggests that poorer urban residents are less likely to have access to air conditioning and thus are less able to self-protect against heat. The poor may also live in neighborhoods with less vegetation in the form of parks, street trees, and the like. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 104 In eight large Indonesian metropolitan areas, more affluent neighborhoods (measured using educational attainment as a proxy) face less heat exposure than less affluent ones (Rossitti 2022). Additionally, Chan et al. (2012) find that in Hong Kong SAR, China, the relationship between heat and mortality is stronger in poorer districts. These two studies together suggest that (a) poorer neighborhoods may be more exposed to extreme heat, and (b) when exposed, these neighborhoods’ residents are likelier than wealthier residents to die from that exposure. These results highlight the importance of both heat mitigation (reducing exposure to extreme heat) and adaptation (reducing the damage caused by exposure to extreme heat) measures in cities. However, whether the UHI effect is more intense in poorer or wealthier neighborhoods varies by city. A study of 25 cities across world regions found that in most (18) of them, including Jakarta, Indonesia, poorer neighborhoods did experience higher temperatures (Chakraborty and Lee 2019).8 Conversely, in the remaining 7 cities, including Beijing, Bangkok, and Manila, wealthier neighborhoods experience higher temperatures. The authors note that several aspects of urban development, such as a city’s housing market and connectivity, affect whether more affluent residents live in the city centers where UHI intensity is generally higher. In addition to these direct impacts, exposure to extreme heat could impair the economic mobility of the poor because of the impact on their educational outcomes and economic productivity, as discussed earlier.9 Elderly residents, as well as those with disabilities and chronic medical conditions, are most at risk of the health impacts of heat. Vulnerability to heat-related conditions is significantly affected by respiratory and cardiovascular conditions, which make it harder for the body’s cooling responses to function (see chapter 1, box 1.3). Infants and people over 75 are particularly vulnerable to heat waves. The vulnerability of elderly people is particularly salient in the East Asia region, because the population of the region above the age of 65 is projected to double between 2020 and 2050 (UN DESA 2017). And for many of the region’s countries, it is projected to triple or nearly triple (table 3.1). Varquez et al. (2020) project that in Jakarta, the proportion of the population in the 65+ age group will increase from 3.3 percent in the 2010s to 22 percent by the 2050s. They further project that the total number of heat-related elderly deaths in August will be 12–15 times higher in the 2050s than in the 2010s because of the combined effects of population aging and rising daytime temperatures. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 105 Table 3.1 Elderly Populations in Selected East Asian Countries, 2020 (actual) and 2050 (projected) POPULATION AGE 65+, POPULATION AGE 65+, POPULATION AGE 65+ COUNTRY 2020 (THOUSANDS) 2050 (THOUSANDS, PROJECTED) (RATIO OF 2050 TO 2020) Mongolia 142 534 3.8 The Lao PDR 310 986 3.2 Cambodia 811 2,555 3.2 Indonesia 17,130 52,495 3.1 Malaysia 2,325 6,894 3.0 Viet Nam 7,658 22,414 2.9 Philippines 6,040 17,045 2.8 Singapore 780 2,131 2.7 Myanmar 3,393 8,205 2.4 Timor-Leste 56 131 2.3 Korea, Rep. 8,096 17,831 2.2 Thailand 9,045 19,546 2.2 China 172,263 365,636 2.1 Japan 35,916 39,881 1.1 East Asia totala 271,616 572,498 2.1 SOURCE: United Nations Department of Economic and Social Affairs (UN DESA) Population Division, World Population Prospects 2019 custom online data. a. The East Asia total includes all countries classified by UN DESA in this dataset as belonging to “Eastern Asia” and “South-Eastern Asia.” UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 106 EXTREME HEAT INCREASES ENERGY USE FOR COOLING Air conditioner (AC) ownership is rising rapidly across the region, primarily in cities. It varies across the region, with less than 10 percent of households in Indonesia, the Philippines, and Viet Nam owning ACs. However, energy use for air conditioning in Association of Southeast Asian Nations (ASEAN) countries has increased 7.5 times over the past 30 years and is expected to continue to increase as incomes and temperatures in the region rise (IEA 2019b).10 In China, AC use has grown rapidly since 2000. The country accounted for about one-third of global growth in cooling energy between 2000 and 2017. This was driven primarily by growth in AC ownership in cities. Over 60 percent of households in China now own an AC. This is projected to increase to 85 percent by 2030. Electricity consumption per capita for space cooling in China is still less than half of that in the Republic of Korea or Japan, suggesting that there is considerable room for growth (IEA 2019a). Extreme heat increases energy consumption for mechanical cooling (especially air conditioning). This in turn raises carbon emissions from energy and contributes to global warming (further necessitating AC use). A 1°C rise in temperature increases annual electricity consumption by 3–4 percent in Singapore and 4–5 percent in Hong Kong SAR, China (Ang, Wang, and Ma 2017). However, such annual electricity consumption figures underestimate the spike in electricity demand for cooling during the hottest periods. For example, in Shanghai, China, annual electricity consumption increases by 9 percent per 1°C increase in annual mean surface temperature, but peak electricity use increases by 36 percent per 1°C increase (Li, Pizer, and Wu 2019). This finding is reinforced in other regions. For example, simulations of Paris by Viguié et al. (2020) suggest that if ACs were used to cool all buildings in Paris down to 23°C during a heat wave, the energy required would be equivalent to 81 percent of the current average total daily electricity consumption. Air conditioning further raises outdoor temperatures by contributing to the UHI effect and global warming. Although AC use cools indoor spaces, it further raises outdoor temperatures both directly and indirectly, creating a vicious cycle. ACs directly release heat into streets, raising outside air temperature by 2–4°C during intense heat waves (Viguié et al. 2020), further intensifying the UHI effect. By increasing energy consumption, especially on the scale of East Asian cities, AC use can also further contribute to carbon emissions (depending on the energy source) and UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 107 therefore increase warming due to global climate change. Viguié et al. (2020) describe AC use as a “maladaptation” to extreme heat, because of its negative externalities. SUMMARY AND CONCLUSION Extreme heat negatively affects every aspect of urban life. The research reviewed in this chapter indicates that exposure to such heat has a range of negative impacts. Although the specific thresholds at which these negative impacts come into effect may vary (see box 3.3), increasing temperatures in the region in coming decades are likely to exacerbate several of these impacts. Reduced labor productivity due to extreme heat induced by climate change alone is projected to cause GDP losses in several East Asian countries of 5.5–6 percent by 2030. In cities, the combined impacts of climate change and UHI could more than double the losses to city-level GDP that would be caused by either one on its own. In addition, evidence from East Asia and beyond suggests that exposure to extreme heat worsens educational outcomes; increases the rates of illness, mortality, injury, crime, and violence; necessitates higher energy consumption; and damages urban infrastructure. Cities must act to mitigate the UHI effect and adapt to the increasing frequency and intensity of heat waves, using approaches discussed in the next chapter. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 108 REFERENCES Anderson, G. B., and M. L. Bell. 2011. “Heat Waves in the United Combes, P., and L. Gobillon. 2015. “The Empirics of States: Mortality Risk during Heat Waves and Effect Agglomeration Economies.” In Handbook of Regional Modification by Heat Wave Characteristics in 43 U.S. and Urban Economics, Volume 5, edited by G. Duranton, Communities.” Environmental Health Perspectives 119 (2): J. V. Henderson, and W. Strange, 247–348. Amsterdam: 210–18. Elsevier. Ang, B. W., H. Wang, and X. Ma. 2017. “Climatic Influence on Davis, L., P. Gertler, S. Jarvis, and C. Wolfram. 2021. “Air Electricity Consumption: The Case of Singapore and Conditioning and Global Inequality.” Global Hong Kong.” Energy 127 (May): 534–43. Environmental Change 69 (1): 102299. Bressler, R. D., F. C. Moore, K. Rennert, and D. Anthoff. 2021. Dodman, D., B. Hayward, M. Pelling, V. Castan Broto, W. Chow, “Estimates of Country Level Temperature-Related E. Chu, R. Dawson et al. 2022. “Cities, Settlements Mortality Damage Functions.” Scientific Reports 11 (1): and Key Infrastructure. In Climate Change 2022: 20282. Impacts, Adaptation, and Vulnerability. Contribution of Bruederle, A., J. Peters, and G. Roberts. 2017. “Weather and Crime Working Group II to the Sixth Assessment Report of the in South Africa.” Ruhr Economic Papers No. 739, RWI – Intergovernmental Panel on Climate Change, Geneva. Leibniz Institute for Economic Research, Essen, Germany. Duranton, G. 2015. “Growing through Cities in Developing Burke, M., F. González, P. Baylis, S. Heft-Neal, C. Baysan, S. Basu, Countries.” The World Bank Research Observer 30 (1): and S. Hsiang. 2018. “Higher Temperatures Increase 39–73. Suicide Rates in the United States and Mexico.” Nature Estrada, F., W. J. Wouter Botzen, and R. S. J. Tol. 2017. “A Global Climate Change 8 (8): 723–29. Economic Assessment of City Policies to Reduce Climate Burke, M., S. M. Hsiang, and E. Miguel. 2015a. “Climate and Change Impacts.” Nature Climate Change 7 (6): 403–6. Conflict.” Annual Review of Economics 7 (1): 577–617. Heilmann, K., and M. E. Kahn. 2019. “The Urban Crime and Heat Burke, M., S. M. Hsiang, and E. Miguel. 2015b. “Global Non-Linear Gradient in High and Low Poverty Areas.” Working Paper Effect of Temperature on Economic Production.” Nature No. 25961, National Bureau of Economic Research, 527 (7577): 235–39. Cambridge, MA. Burke, M., and V. Tanutama. 2019. “Climatic Constraints on IEA (International Energy Agency). 2019a. “The Future of Cooling Aggregate Economic Output.” Working Paper No. 25779, in China: Delivering on Action Plans for Sustainable Air National Bureau of Economic Research, Cambridge, MA. Conditioning.” Report, IEA, Paris. Chakraborty, T., and X. Lee. 2019. “A Simplified Urban-Extent IEA (International Energy Agency). 2019b. “The Future of Cooling Algorithm to Characterize Surface Urban Heat Islands in Southeast Asia.” Report, IEA, Paris. on a Global Scale and Examine Vegetation Control on ILO (International Labour Organization). 2019. Working on a Their Spatiotemporal Variability.” International Journal Warmer Planet: The Impact of Heat Stress on Labour of Applied Earth Observation and Geoinformation 74 (1): Productivity and Decent Work. Geneva: International 269–80. Labour Office. Chan, E., Y. Yang, W. B. Goggins, J. J. Kim, and S. M. Griffiths. 2012. Kahn, M. E., K. Mohaddes, R. N. C. Ng, M. H. Pesaran, M. Raissi, “A Study of Intracity Variation of Temperature-Related and J.-C. Yang. 2019. “Long-Term Macroeconomic Effects Mortality and Socioeconomic Status among the Chinese of Climate Change: A Cross-Country Analysis.” Working Population in Hong Kong.” Journal of Epidemiology and Paper No. 2019/215, International Monetary Fund, Community Health 66 (4): 322–27. Washington, DC. Chauvin, J. P., E. Glaeser, Y. Ma, and K. Tobio. 2017. “What Is Kjellstrom, T. 2016. “Impact of Climate Conditions on Different about Urbanization in Rich and Poor Countries? Occupational Health and Related Economic Losses: A Cities in Brazil, China, India, and the United States.” New Feature of Global and Urban Health in the Context of Journal of Urban Economics 98: 17–49. Climate Change.” Asia Pacific Journal of Public Health 28 Chen, Y., R. Sun, X. Chen, and X. Qin. 2023. “Does Extreme (2 Suppl): 28S–37S. Temperature Exposure Take a Toll on Mental Health? Li, Y., W. A. Pizer, and L. Wu. 2019. “Climate Change and Evidence from the China Health and Retirement Residential Electricity Consumption in the Yangtze River Longitudinal Study.” Environment and Development Delta, China.” PNAS 116 (2): 472–77. Economics: 1–25. Published ahead of print, June 5. Liu, Z., W. Zhan, B. Bechtel, https://doi.org/10.1017/S1355770X23000037. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 109 J. Voogt, J. Lai, T. Chakraborty, Z.-H. Wang et al. 2022. “Surface UN DESA (United Nations Department of Economic and Social Warming in Global Cities Is Substantially More Rapid Affairs). 2017. World Population Prospects: The 2017 Than in Rural Background Areas.” Communications Earth Revision. New York: United Nations. & Environment 3 (1): 219. USAID (United States Agency for International Development). Mares, D. M., and K. W. Moffett. 2016. “Climate Change and 2020. “Heat and Urban Environmental Issues: Water Interpersonal Violence: A ‘Global’ Estimate and Regional Quality.” Quick Reference Guide, USAID, Washington, DC. Inequities.” Climatic Change 135 (2): 297–310. Varquez, A. C. G., N. S. Darmanto, Y. Honda, T. Ihara, and M. Kanda. Nanji, N. 2022. “Trains Cancelled and Delayed after Heatwave 2020. “Future Increase in Elderly Heat-Related Mortality Damage.” BBC News, July 20. of a Rapidly Growing Asian Megacity.” Scientific Reports Park, R. J., A. P. Behrer, and J. Goodman. 2021. “Learning Is 10 (1): 9304. Inhibited by Heat Exposure, Both Internationally and Viguié, V., A. Lemonsu, S. Hallegatte, A.-L. Beaulant, C. within the United States.” Nature Human Behaviour 5 (1): Marchadier, V. Masson, G. Pigeon, and J.-L. Salagnac. 19–27. 2020. “Early Adaptation to Heat Waves and Future Park, R. J., J. Goodman, M. Hurwitz, and J. Smith. 2020. “Heat and Reduction of Air-Conditioning Energy Use in Paris.” Learning.” American Economic Journal: Economic Policy Environmental Research Letters 15 (7): 075006. 12 (2): 306–39. WHO (World Health Organization). 2018. “Heat and Health.” Park, R. J., N. Pankratz, and A. P. Behrer. 2021. “Temperature, Fact sheet, June 1, WHO, Geneva. https://www.who.int/ Workplace Safety, and Labor Market Inequality.” news-room/fact-sheets/detail/climate-change-heat-and- Discussion Paper No. 14560, Institute of Labor Economics health. (IZA), Bonn, Germany. Wu, C. Y. H., B. F. Zaitchik, and J. M. Gohlke. 2018. “Heat Waves Pavanello, F., E. De Cian, M. Davide, M. Mistry, T. Cruz, P. Bezerra, and Fatal Traffic Crashes in the Continental United D. Jagu, S. Renner, R. Schaeffer, and A. F. P. Lucena. 2021. States.” Accident Analysis and Prevention 119: 195–201. “Air-Conditioning and the Adaptation Cooling Deficit in Xia, Y., Y. Li, D. Guan, D. M. Tinoco, J. Xia, Z. Yan, J. Yang, Q. Liu, Emerging Economies.” Nature Communications 12 (1): and H. Huo. 2018. “Assessment of the Economic Impacts 6460. of Heat Waves: A Case Study of Nanjing, China.” Journal Rossitti, G. 2022. “Hot, Wet and Poor? Climate Hazards Exposure of Cleaner Production 171: 811–19. Bias within Cities.” Unpublished working paper, World Xie, S. Y., and C. Li. 2022. “China’s Economic Woes Drag On Bank, Washington, DC. with Factory Disruptions, Property Slump.” Wall Street Sanz-Barbero, B., C. Linares, C. Vives-Cases, J. L. González, J. Journal, August 31. J. López-Ossorio, and J. Díaz. 2018. “Heat Wave and the Zhang, P., O. Deschenes, K. Meng, and J. Zhang. 2018. Risk of Intimate Partner Violence.” Science of the Total “Temperature Effects on Productivity and Factor Environment 644: 413–19. Reallocation: Evidence from a Half Million Chinese Seppanen, O., W. J. Fisk, and Q. H. Lei. 2006. “Effect of Manufacturing Plants.” Journal of Environmental Temperature on Task Performance in Office Environment.” Economics and Management 88: 1–17. Paper No. LBNL-60946, Ernest Orlando Lawrence Zhao, Q., Y. Guo, T. Ye, A. Gasparrini, S. Tong, A. Overcenco, A. Berkeley National Laboratory, Berkeley, CA. Urban et al. 2021. “Global, Regional, and National Burden Somanathan, E., R. Somanathan, A. Sudarshan, and M. Tewari. of Mortality Associated with Non-Optimal Ambient 2021. “The Impact of Temperature on Productivity and Temperatures from 2000 to 2019: A Three-Stage Labor Supply: Evidence from Indian Manufacturing.” Modelling Study.” Lancet Planetary Health 5 (7): e415–25. Journal of Political Economy 129 (6): 1797–1827. Zivin, J. S. G., Y. Song, Q. Tang, and P. Zhang. 2018. “Temperature doi:10.1086/713733. and High-Stakes Cognitive Performance: Evidence from Taylor, L., S. L. Watkins, H. Marshall, B. J. Dascombe, and J. the National College Entrance Examination in China.” Foster. 2016. “The Impact of Different Environmental Working Paper No. 24821, National Bureau of Economic Conditions on Cognitive Function: A Focused Review.” Research, Cambridge, MA. Frontiers in Physiology 6: 372. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 110 NOTES 1 Cities potentially become less days reduces output by about 2.7 or warm clothes, resulting in more attractive to high-skilled workers percent because of lower on-the-job deaths during rare cold spells. Testing through two channels: (a) by directly productivity and by an additional this hypothesis could be an important making the city a less pleasant place 1.4 percent because of absenteeism. area for future research. to live because of the associated 4 These figures correspond to 8 Chapter 2 also finds this to be the thermal discomfort; and (b) by the Intergovernmental Panel on case for Bandung, Indonesia. indirectly making the city a less Climate Change’s Representative 9 The impacts of heat on injury and attractive place to live and work in by Concentration Pathways (RCP) 8.5 crime also vary by income group. reducing the marginal productivity (worst case) scenario. For a further The study mentioned earlier of of labor and therefore the real wages explanation of RCP scenarios, see workplace injuries due to heat in that a city’s firms can pay workers in box 3.1, note “a.” the United States (Park, Pankratz, competitive markets. 5 The ILO study calculates productivity and Behrer 2021) finds that the risk 2 TFP, as discussed by Zhang et al. losses as “the equivalent working of injury is five times higher for a (2018), may encompass the concepts hours lost owing to slower work or a worker from the bottom income described by both “A” and “e” in the complete stoppage of work when heat quintile than for someone from the production function introduced above levels are too high for working” (ILO top quintile because of differences (equation 3.1), because the authors do 2019, 91). In terms of the production in the nature and conditions of the not separate out the effectiveness of function discussed earlier, this work performed by these groups. human labor. combines a reduction in effectiveness In turn, injury and illness due to heat 3 Studies for other Asian countries also e (“slower work”) with a reduction in result in lost income, which can throw report negative effects of heat on labor input L (“complete stoppage a financially precarious household manufacturing output. Somanathan et of work”). Kjellstrom (2016) uses the into crisis. The study of heat and al. (2021) find that in garment plants same definition of productivity losses. crime in Los Angeles discussed without climate control (centralized 6 All dollar values in Kjellstrom (2016) earlier (Heilmann and Kahn 2019) air cooling) in and around Delhi are in purchasing power parity (PPP) also shows that the adverse impacts in India, an additional day with a constant 2005 US dollars. of high temperatures on crime are maximum temperature above 35°C 7 Zhao et al. (2021) do not elaborate on concentrated in the poorest quintile reduces average daily output that why warm places would have a higher of city districts, which already have week by as much as 8 percent relative number of cold- than heat-related the most crime. to if the temperature had remained deaths. One potential explanation 10 ASEAN member countries include below 19°C. Among weavers in India, could be that people living in places Brunei Darussalam, Cambodia, the study finds that an additional that are usually warm are not Indonesia, Laos, Malaysia, Myanmar, day with a maximum temperature prepared for cold weather and lack the Philippines, Singapore, Thailand, above 35°C in the six preceding indoor heating, insulated homes, and Viet Nam. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 111 CH C H A PT ER 4 MITIGATING AND ADAPTING TO EXTREME URBAN HEAT: POLICY PRINCIPLES AND OPTIONS UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 112 KEY MESSAGES This report’s analysis points to three key principles—and associated options for action—to equip East Asian cities to combat the urban heat island effect and withstand heat waves: Places: Invest in vegetation, public spaces, and building design for a cooler city People: Protect the health of vulnerable groups during heat waves Institutions: Build a leadership coalition to drive attention to heat hazards across city and national agencies. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 113 INTRODUCTION While heat stress presents growing risks to East Asia’s urban populations, the challenge is one that the region’s cities can address decisively and effectively. Working in tandem with national governments and neighborhood-level stakeholders, East Asian cities have at their disposal a wide range of policy and investment options to reduce heat exposure and help residents better anticipate and cope with extreme conditions. However, resilience to extreme heat is no ordinary policy challenge. At least three factors complicate the policy response, making new and innovative mindsets essential. Making a city resilient to heat requires the following: Access to an unusually wide variety of skills from across government and the private sector. Contributions are needed from built environment disciplines (such as urban planners and forestry managers) but also from public health specialists, doctors, and emergency managers, architects, real estate developers and business owners. Action across physical scales, from changing the material of individual walls and roofs to rethinking areawide land use plans for increased greenery and wind flow. New governance structures to ensure both clear lines of authority and coordination of action. Established policy priorities such as transport, health, and education typically have accountability and resources vested in clearly defined agencies, but the authority and capability to reduce heat risks is highly fragmented among government entities. Leadership is therefore crucial. To rise to the challenge of extreme heat and combat the urban heat island (UHI) effect, city leaders must establish an evidence base, articulate a vision for change, and build new coalitions for action. A FRAMEWORK FOR ACTION: THREE PRINCIPLES, SIX ACTION AREAS To help structure an effective response to rising heat stress in East Asian cities, this report proposes a framework for action based around three principles: cool the physical spaces of a city (“Places”), act during heat waves to prevent death and illness among vulnerable populations (“People”), and mainstream heat resilience UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 114 into city strategies, operations, and budgets (“Institutions”). Addressing the three principles in tandem is important given their interlocking nature and requires concerted leadership by city and national authorities (figure 4.1). Figure 4.1 The “Places, People, Institutions” Framework for Urban Heat Resilience Places People Institutions ■ Advance urban greening ■ Save lives through heatwave ■ Mainstream heat mitigation through strategic planning early warnings into city strategies and ■ Cool city spaces through ■ Protect heat-exposed operations wind, shade and design workers ■ Engage building owners to tackle indoor heat SOURCE: World Bank. Places. Heat stress is experienced in the physical places of a city, both indoors and out. Making streets, plazas, parks, factories, workshops, marketplaces, and homes cooler will reduce heat exposure, contributing to a competitive and inclusive economy and healthy population. Temperature reduction from place-based cooling interventions—including urban greening and cool building materials—can be significant. Indeed, Santamouris et al. (2017) reviewed 220 studies of such measures and found that average daily temperatures at project sites fell by up to 2 degrees Celsius (°C) in approximately two-thirds of cases and by more than 2°C in the remaining cases. These temperature reductions are of roughly the magnitude needed to offset on a localized basis the UHI effects in many East Asian cities. Maximum daily temperatures, i.e. the hottest UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 115 temperature recorded at any point during a 24-hour period, fell by more than 3°C in the proximity of tree and water-based interventions and by more than 5°C in some cases (figure 4.2). Figure 4.2 Cooling Effects of Selected UHI Mitigation Strategies Decrease in max. daily temperature (°C) 0 -2 -4 -6 -8 -10 Water-based Combined green infrastructure Trees and hedges Cool roofs and reflective pavements Reflective pavements Cool roofs Grass SOURCE: Santamouris et al. 2017. Green roofs NOTE: The figure shows temperature reduction in degrees Celsius (°C) from interventions to mitigate urban heat island (UHI) effects, based on a review of performance data from 220 projects. The temperature reduction figures represent decrease in daily maximum temperatures (i.e. the hottest temperature recorded in any 24 hour period) at each project site compared with measurements taken in a control period. For each category of intervention, the yellow shaded box and accompanying whiskers illustrate the range of temperature reduction benefits that were recorded. The top, middle and bottom of the yellow shaded box represents the 25th percentile, 50th percentile, and 75th percentile respectively, while the black lines extend out to the minimum and maximum temperature reduction values that were recorded. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 116 People. But heat waves are also time-bound emergency situations that endanger people—particularly elderly people, children, heat-exposed workers, and those with health vulnerabilities. Identifying the groups that are disproportionately at risk and protecting them during heat waves is imperative. As discussed later in the chapter, experience shows that deaths during heat waves have fallen by double-digit margins after cities introduced heat wave alerts and associated measures to protect the most vulnerable during extreme temperatures. Institutions. Finally, mainstreaming heat resilience across a city’s institutions—its governmental departments and agencies as well as civil society and professional groups that play essential roles—is a necessary response to intensifying urban heat. By establishing coordination structures such as a citywide extreme heat task force and adopting a heat wave risk management. East Asian cities can put themselves on a sound institutional footing to protect citizens, infrastructure, and the economy against future temperature extremes. The following sections present six strategic options for action to deliver on the “Places, People, Institutions” agenda for heat resilience. There is no one-size-fits- all solution given East Asian cities’ differing climates, populations, and institutional starting points. However, by prioritizing needed interventions from across this set of action areas—with consideration of the detailed policy and investment options outlined in Annex 4A at the end of the chapter—each city can design an effective local response to cool its places and protect its population from rising heat stress. STRATEGIC OPTION 1: ADVANCE URBAN GREENING THROUGH STRATEGIC PLANNING Green assets such as trees, grass, and shrubs result in cooler city spaces, and, as the analysis of chapter 1 showed, the intensity of an East Asian city’s average UHI effect is strongly positively correlated with its average level of vegetation. However, as illustrated in chapter 2 for the city of Bandung in Indonesia, vegetation levels also tend to be highly uneven between a city’s neighborhoods. Prosperous residential areas tend to have better organized and resourced tree care and may have larger plot sizes, whereas central business districts and poor residential neighborhoods UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 117 often lack greenery. Expanding the reach of urban nature, especially where trees and shrubs are in short supply, can increase shade and evapotranspiration, mitigating the UHI effect and reducing air temperatures. Urban greening is a key ingredient in heat resilience strategies and benefits diverse groups, including workers and shoppers who frequent dense city centers as well as the poorest residents whose neighborhoods may lack grass, shrubs, and trees. The return on investment from urban greening can be high. Studies estimate that greening interventions can result in a median reduction in localized temperatures of 1.5°C (Santamouris et al. 2017). Planting at “microsites” (such as tree pits and on traffic medians) cools air locally, while planting at extensive sites (such as parks) can also generate a “park cool island effect” benefiting surrounding blocks, particularly in areas downwind of them. This cooling effect may extend up to 2 kilometers, depending on the characteristics of the nearby areas (Santamouris et al. 2017). Multiple pocket parks—small parks of around 1,000 square meters—may confer greater cooling benefits than a single large park of equivalent aggregate size, especially in high-rise tropical regions (Giridharan and Emmanuel 2018). One study found that expanding tree cover on 10 percent of city land could yield a 1.5°C temperature reduction for 647,000 people in Seoul, 770,000 in Jakarta, and 2.2 million in Beijing respectively—at a cost of just US$1–3 per capita for each benefiting resident (McDonald et al. 2016). City leaders can consider using, or adapting, the so-called 3-30-300 rule: each resident should be able to see at least 3 trees from their home, have 30 percent canopy cover in their neighborhood, and live no more than 300 meters from a park or green space (Konijnendijk 2021). Meeting ambitious goals such as the 3-30-300 target across East Asian cities will require resources and long-term planning. Indeed, greening of cities is easier said than done. Key challenges in urban forestry expansion include species selection; survival during the crucial first year after planting; tree maintenance and pruning; managing competition between tree roots and the needs of residents and infrastructure; and securing community support for planting and maintenance. Urban greening requires strategic planning across several physical scales to expand green assets on large and small sites as well as on publicly owned, commercial, and residential land (table 4.1). Forward planning with a timeline of at least a decade is particularly important for urban forestry given the speed at which trees grow.1 In this context, adopting dedicated urban forestry master plans is highly advisable. By setting a strategic vision for green assets in the city, such plans provide a powerful tool to (a) ensure that legislative frameworks and fiscal resources align with a city’s future vision for trees and vegetation, and (b) track progress toward the goals. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 118 Table 4.1 Actions to Create Cooler City Places, by Scale BUILDING NEIGHBORHOOD ASPECT BLOCK SCALE CITY SCALE SCALE SCALE Urban climate effect Heat absorption and Street canyon Local heat island Urban impact on radiation, shading, dynamics: heating intensity, park cool island weather patterns: air ventilation and cooling, pollutant effects circulation, pollutants, circulation fog Relevant actors Building owner, Residents’ association, Neighborhood Urban planner, landlord, tenant real estate developer association, architects’ association, infrastructure or real national government estate developer Example interventions Cool roof, passive Public space design, Urban greening: street Building code (such building design, green arrangement of trees, traffic medians, as cool roof mandate), building elements buildings, urban private land, pocket parks planning incentives greening microsites (such as FAR bonus), design guide, breezeway planning Scale illustration SOURCE: World Bank. PHOTOS: © Nicholas Jones / World Bank; further permission required for reuse. NOTE: The figure shows urban climate effects, relevant actors, and feasible interventions across different physical scales of urban space, arranged from small to large scale. FAR = floor-area ratio. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 119 STRATEGIC OPTION 2: COOL CITY SPACES THROUGH WIND, SHADE, AND URBAN DESIGN Increasing green cover in cities is important, but so too are interventions to preserve beneficial aspects of local climates—such as wind flow generated by the local topography—and promote cooler design of public spaces. Wind effects. For coastal cities in particular, wind is a crucial resource. Sea breezes form because of pressure differentials when air above the city warms at a different rate than air above the sea. The resulting convection cools surfaces and disperses pollution. Indeed, in Singapore, air temperatures fall by 2°C when wind speed is 1.5 meters per second higher (Erell, Pearlmutter, and Williamson 2011). However, urban construction can inhibit wind. Poor building design—such as lines of tall buildings directly on the water’s edge that block sea breezes—have a detrimental effect on thermal conditions and air pollution levels for inland districts (Oke et al. 2017).2 East Asian cities that anticipate rapid growth of further tall buildings along the coast can learn from examples such as the air ventilation assessment in Hong Kong SAR, China (figure 4.3). Just as Hong Kong’s urban design guidelines provide clear guidance on design features and requirements—such as stepped-back buildings, designated breezeways, and building arrangements to channel wind—designing cities “with the wind” may offer significant benefits (Ng 2009). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 120 Figure 4.3 Recommended Site-Level Practices in Hong Kong SAR, China, to Safeguard Air Ventilation and Microclimates Considering Wind Flows 1 STREET ORIENTATION, PATTERN, WIDENING 2 BUILDING HEIGHT AND FORM Air flow Tower setback from podium Downwash wind facing wind direction cannot reach street level street Air flow Tower abutting podium edge Downwash wind facing wind can reach street direction Prevailing wind Prevailing wind Prevailing wind level street 3 PODIUM STRUCTURES 4 WATERFRONT SITES Air flow Podium Podium Air flow Air flow washes Sea or land Sea or land recirculates the street out breeze breeze SOURCE: PlanD 2016. Shade effects. In addition to wind, shade is a valuable commodity in hot climates. Urban planning documents and cooling action plans can promote increased shade in public places. In Freetown, Sierra Leone, an early initiative of the city’s first chief heat officer was to install awnings in three public market places where some 2,300 mainly female market traders sold produce daily in hot conditions (Atlantic Council 2022). In Singapore, a distinctive network of covered walkways has become a hallmark of UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 121 the urban landscape, adding to livability and promoting a vibrant commercial and business landscape that is more comfortably accessible to pedestrians given the city’s hot and humid climate (box 4.1). In promoting cooler urban designs, traditional building practices (“vernacular architecture”)—which, in many East Asian regions, offers a wealth of architectural practices that promote ventilation, minimize heat storage in surfaces, and provide shade (Satwiko 1999)—offer an important potential source of inspiration. BOX 4.1 COOLING SINGAPORE WITH TRADITIONAL TECHNIQUES AND DIGITAL TECHNOLOGIES Planting trees to increase greenery the Walk2Ride program as part of the capacity indicators (share of elderly and coverage has been a key strategy 2013 Land Transport Master Plan, setting infant population, unemployment rate, to control urban temperatures since further ambitions to expand the network employment in construction, access Singapore’s independence in 1965. of covered walkways. to local medical facilities, and others) The government first launched a tree- (Philipp and Chow 2020). planting campaign in 1963 to replicate In 2017, the government of Singapore tree-lined streets from overseas, and launched a major research project In 2019, an interagency working group this was incorporated into the “Garden entitled “Cooling Singapore,” led by led by the Ministry of Sustainability and City” vision (announced in 1967 by Prime a university center with international the Environment (MSE) and the URA Minister Lee Kuan Yew), which accelerated partners. The project has conducted was launched to implement initiatives tree planting. More recently the National research on the urban heat island (UHI) to mitigate UHI effects in Singapore. Parks Board has since 2015 extended the effect, including heat stress mapping, It identified three key strategies for Garden City vision to plant climbers and identifying gaps in knowledge and UHI mitigation: shrubs on bus shelter roofs, on covered technology, and establishing a task force walkways, at railway stations, and on to test intervention options. The project Understanding UHI effects by noise barriers. team tested and simulated alternative expanding the network of climate building designs, conducted surveys of sensors in the country to collect Covered walkways have been a key residents’ experiences and preferences, data and running computational feature since at least the Town Plan of and developed a “digital twin” model to simulation models 1822, which set an essential framework for simulate local climates. Singapore’s spatial development. Today, Simulating the effectiveness of UHI the Urban Redevelopment Authority Cooling Singapore has also created an mitigation strategies through the (URA) mandates that all commercial Urban Heat Vulnerability Index, which Cooling Singapore project and mixed developments “shall provide combines physical exposure with covered walkways along the periphery of socioeconomic indicators to identify Implementing interim UHI mitigation the building facing roads and pedestrian populations most vulnerable to extreme measures by partnering with industry routes,” with design specifications heat. It maps physical exposure indicators and the public. around roof height and minimum width. (measures of temperature and vegetation) The Land Transport Authority launched as well as sensitivity and adaptive UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 122 Urban design effects. With East Asian cities facing a hotter future, architectural and urban design practices have an important role to play in promoting the thermal comfort of urban landscapes. Among them, street depth and orientation are important considerations. The depth of street canyons—the ratio of building height to street width—can affect the amount of the sun’s energy received by building façades during the day as well as influencing air flow. In Bangkok, a study showed that medium-depth street canyons (with a height-width ratio of 0.36 to 0.65) were hottest, providing neither sufficient shade nor unobstructed ventilation (Takkanon and Chantarangul 2019). The effect of street canyon depth varies over a 24-hour period, with more-enclosed spaces being cooler during the day owing to shade but warmer at night because of trapped heat. Orientation of streets affects both shade and ventilation. Streets oriented roughly parallel to the flow of breeze (for example, perpendicular to the shore in a coastal city) have better ventilation and lower temperatures. Water bodies in urban areas can prove highly beneficial to thermal comfort (particularly in dry climates). Where water availability permits, features such as sprinklers, fountains, and misting devices have been implemented with success in cities such as Singapore to cool public spaces such as plazas, train stations, and children’s playgrounds (Lai et al. 2019). Urban design solutions are context specific—differing notably based on local climate, topography, and latitude—but city leaders can incentivize good architectural and design practices by publicizing the urban heat challenge, celebrating good local practices, and building essential considerations into local codes and design guidelines. STRATEGIC OPTION 3: ENGAGE BUILDING OWNERS TO TACKLE INDOOR HEAT Poorer neighborhoods often have higher outdoor air temperatures than wealthier ones, but when indoor conditions are also considered, the “thermal inequalities” loom even larger. Addressing indoor heat exposure is essential to reducing heat illness and sustaining labor productivity, since people spend much time indoors. The design and construction of buildings is a key entry point to address heat exposure. City governments’ policy leverage differs depending on building ownership and established planning procedures, and hot buildings cannot be transformed into cool UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 123 ones overnight. However, city authorities can systematically engage with building owners to promote cooler design and materials while looking out for “quick wins” that reduce harmful heat exposure of at low cost. Passive cooling of buildings refers to architectural elements that reduce indoor temperatures without mechanical cooling. Before the advent of air conditioning, traditional building designs in warm regions featured passive cooling techniques by necessity. Although these may not be sufficient to ensure thermal comfort during the hottest times of the year, they still contribute to cooler temperatures and reduced air conditioning needs. Depending on the climate, passive cooling features may include various elements for shading (overhanging eaves, latticed façades, louvered shades, and others); courtyards; wind towers and other features for ventilation; water for evaporative cooling; the use of earthen surfaces and insulating materials on façades; and the use of materials that reflect the sun’s energy rather than absorbing it (Srivastav and Jones 2009; Yoshino, Hasegawa, and Matsumoto 2007). Building materials differ substantially in the proportion of incoming radiation, from the sun and other sources, that they reflect away. The fraction of incoming energy that a surface reflects is known as its albedo (table 4.2). City authorities can promote the use of high-albedo materials (including the application of white paint coatings to roofs) and other passive cooling interventions through a combination of “hard measures” (mandatory regulations and direct public investment) and “soft measures” (including incentives and information provision). Among hard measures, building codes and zoning requirements can be used to drive adoption of cooler building practices. A good example of this is India’s 2016 building code, which specifies reflectivity and emissivity values required for new or substantially renovated roofs (ESMAP 2020). City governments can also lead by example in their design or renovation of publicly owned facilities including those procured through public-private partnerships. Schools, clinics, government buildings, and sport facilities all provide opportunities to pilot solutions, establish design standards, build construction industry capacity, and create a local evidence base on the effectiveness of cool building measures (table 4.2). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 124 Table 4.2 Albedo Value for Selected Building Materials BUILDING MATERIAL ALBEDO ROADS Asphalt (fresh → weathered) 0.05–0.27 WALLS Concrete 0.10–0.35 Brick (red → white) 0.20–0.60 Limestone 0.40–0.64 ROOFS Tar and gravel, bitumen 0.08–0.18 Tile (old → fresh) 0.10–0.35 PAINTS Black 0.02–0.15 White 0.50–0.90 SOURCE: Adapted from Oke et al. 2017. NOTE: The albedo value of a surface is the fraction of incoming radiation that it reflects. Albedo is scaled from 0 and 1. Surfaces with higher albedo reflect away a higher proportion of incoming solar radiation; for example, a white painted roof with albedo of 0.7 would reflect 70 percent of incoming radiation. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 125 Given the limits of public budgets and building code enforcement, a broader set of tools including fiscal and planning incentives and concessional finance for cool building upgrades may be worth examining on a case-by-case basis. Hong Kong SAR, China; Singapore; and Tokyo are among the cities offering floor-area ratio bonuses— which permit larger buildings on a given plot size—if developers adhere to green building standards. Property tax rebates for cool roof installations are another option. Cool roof programs reduce indoor heat gain through white paint coatings that reflect away the sun’s energy. They represent one of the most cost-effective and scalable heat reduction measures. This is exemplified by Cool Roofs Indonesia, a program whose interventions have resulted in indoor temperature reductions as high as 10°C (Clean Cooling Collaborative 2022). Integrating cool roofs into public housing programs and incentivizing their adoption by commercial building owners can offer major benefits.3 STRATEGIC OPTION 4: SAVE LIVES THROUGH HEAT WAVE EARLY WARNINGS Whatever incremental steps cities take to reduce heat exposure across indoor and outdoor settings year-round, the moment of truth for preventing heat-related deaths and illnesses is when a heat wave strikes. To a significant degree, deaths during heat waves are both predictable and preventable: they are concentrated in groups that have high exposure to heat—owing to factors such as housing quality or heat- exposed occupations—as well as high vulnerability to heat, whether because of age, occupation, or health status. The most effective measures to prevent a heat wave death are those taken by at-risk individuals or those closest to them: their family, health care providers, or employers. Heat waves cause most deaths when they take populations by surprise by coming early in the season. As a substantial body of public health experience has shown, heat-health early warnings can save lives by informing people that their health faces an imminent threat—prompting individuals to take actions that preserve their health— as well as by activating protocols in health care, education, and workplace settings (Toloo et al. 2013). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 126 To maintain credibility and trust, it is essential to issue warnings only when heat stress reaches a level that is known to predict significant adverse health impacts. Design of warning systems therefore requires collaborative work between meteorological agencies, public health specialists, and affected groups. Mortality data should be matched with meteorological information to design warning systems that will be activated when deaths would otherwise occur.4 Effective heat wave early warning systems go beyond simply announcing high air temperatures to activating planned interventions that target the population subgroups known to be most at risk, as several examples illustrate: In the United States, the city of Philadelphia—which introduced one of the world’s first population health-focused heat wave warning systems in 1995—defined a protocol for announcing a “heat health emergency” and accompanied such announcements with a suite of measures, from activating a “Heatline” telephone service to using libraries and swimming pools as cooling centers and increasing outreach to homeless people (Weinberger et al. 2018). In Hanoi, the Viet Nam Red Cross and Viet Nam Institute of Meteorology, Hydrology and Climate Change jointly implement warnings based on a combined measure of heat and humidity and guide vulnerable populations to assistance including mobile cooling centers (Jjemba et al. 2020). In Japan, a country with a hot summer climate, the government has developed and refined an extreme heat early warning system centered on the concept of “heat illness reduction.” Alerts are issued by the Japan Meteorological Agency and disseminated via television and radio, private weather companies, email and social media channels, and local government—including via public loudspeakers (box 4.2). BOX 4.2 HEAT ILLNESS PREVENTION IN JAPAN In Japan, a country known for its Association published “A Guidebook for to develop and refine an alert system. generally mild climate, increased hot the Prevention of Heat Disorder During The Heat Caution Notice system, summers during the 1990s spurred Sports Activities” (JSPO 1994), which, introduced in 2011, provided alerts when recognition of the dangers posed by together with a public information air temperature was projected to exceed extreme heat to human health—and led campaign, helped to reduce cases of heat 35 degrees Celsius (°C). However, by to the development of a robust national illness at sports facilities and among neglecting humidity and radiation—two framework for heat illness reduction. those 14 years old and younger. important aspects of heat stress—the alert system was found to correlate An important early impetus for heat Consecutive hot summers and imperfectly with heat illness cases. health policy was heatstroke among preparations for the 2020 Tokyo Olympic The high frequency of alerts also raised student athletes. In 1994, the Japan Sport Games prompted government action concern for “alert fatigue.” UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 127 An improved system of Heat Stroke A hallmark of Japan’s response to extreme Heat wave alerts have benefited from the Alert (Necchūshō keikai arāto), tested in urban heat has been cooperation across wide array of communication channels, Tokyo then rolled out nationally in 2021, government departments, which was which increasingly include social media. provides citizens with alerts when the initiated through an Urban Heat Island Of alert recipients who responded to an Wet Bulb Globe Temperature is projected (UHI) Ministerial Liaison Council convened evaluation survey, 31 percent changed to exceed 31°C (table B4.2.1). The alerts by the Cabinet Secretariat in the early their behavior to stay indoors, 52 percent are usually issued by 5 p.m. on the 2000s. Where heat illness reduction is hydrated frequently, and 75 percent previous day or, if a sudden temperature concerned, high-level policy coordination checked on family members (Japan MOE rise is expected, at 5 a.m. on the day has meant that alerts issued by the JMA and JMA 2021). Collaborative actions of expected extreme heat. The Japan link with health guidance developed across key heat-exposure settings—from Meteorological Agency (JMA) issues by medical experts (notably the “Heat school sports facilities to elderly care alerts to media, local governments, and Illness Environmental Health Manual,” first homes, construction sites, train stations, private weather forecasting companies to published in 2018) and are supported by and private homes—have helped to avert ensure widespread dissemination. These public awareness measures (figure B4.2.1 heat illness and reduce pressure on the measures reach a large portion of the panel B). health system. population, including through television news flashes and dissemination over local government loudspeaker systems (figure B4.2.1 panel A). Table B4.2.1 Japan’s Heat Stroke Alert System: Heat Stress Thresholds and Recommended Actions HEAT THRESHOLD COLOR CATEGORY MAIN ACTIONS RECOMMENDED (°C WBGT) Red Above 31 Dangerous Exercise should be stopped. Orange 28–31 Warning Refrain from heavy exercise; frequent rest and hydration are strongly encouraged. Yellow 25–28 Watch Frequent rest and hydration are strongly encouraged during heavy exercise. Sky blue 21–25 Caution Hydration is encouraged. Blue Below 21 Generally safe Appropriate hydration is suggested. SOURCE: Japan Meteorological Agency. NOTE: °C WBGT = degrees Celsius (Wet Bulb Globe Temperature). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 128 Figure B4.2.1 Heat Health Early Warning System and Public Awareness Campaigns in Japan A FLOWCHART FOR DISSEMINATION OF HEAT ALERTS JAPAN METEOROLOGICAL AGENCY MOE WEBSITE DISASTER INFORMATION DATA JMA SYSTEM WEBSITE Email list, Website Private sector weather NHK support center PREFECTURE Messaging svstem Email list, Phone apps, THE Website PRIVATE Email list, MEDIA WEATHER Phone apps, COMPANIES Website DATA RELEVANT DATA ORGANIZATIONS LOCAL Email list, GOVERNMENT Phone apps, Website PUBLIC MOBILE LOUDSPEAKERS SPEAKERS CITIZENS Mutual help & reminding UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 129 Figure B4.2.1 Heat Health Early Warning System and Public Awareness Campaigns in Japan B HEAT ILLNESS PREVENTION CAMPAIGN PUBLICITY MATERIALS (ENGLISH LANGUAGE VERSION) SOURCE: SOURCE: Panel (a): World Bank translation and elaboration based on Japan Meteorological Association and Ministry of Environment, 2021. Panel (b): Japan Ministry of Environment (n.d.). A central requirement for heat wave early warning systems is to determine a trigger upon which such warnings will be issued. In Ahmedabad, one of India’s hottest cities, all-cause mortality increased to three times its baseline level on the hottest day of the 2010 heat wave (Azhar et al. 2014). The tripling of citywide daily deaths proved a wake-up call for city leaders, who proceeded to implement South Asia’s first Heat Action Plan, based around a “traffic light” warning system: citizens receive yellow, orange, and red alerts when temperatures exceed 41°C, 43°C, and 45°C, respectively. The thresholds for the alert system were designed by UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 130 examining the correlation of daily deaths from all causes with temperatures from local weather stations. The Heat Action Plan, with the traffic-light alert system at its core, has been credited with helping to avoid around 1,100 deaths per year in the years immediately following its introduction (Hess et al. 2018). Although more evaluations of heat wave early warning systems are needed, existing studies from across world regions already present compelling evidence of the benefits (table 4.3). Heat-related deaths fall into two categories: (a) exertional deaths, which occur when physical effort in hot conditions places body systems under extreme stress; and (b) nonexertional deaths, which occur most frequently at home and among individuals with comorbidities such as cardiovascular disease, reduced lung capacity, or diabetes (Cissé et al. 2022). In both cases, heat wave early warnings based on sound public health guidance can spur vulnerable individuals to hydrate, avoid heat, cool their bodies, and check on neighbors. Table 4.3 Do Heat Wave Alerts Save Lives? Evidence from International Studies CITY STUDIED OBSERVED CHANGE IN MORTALITY STUDY REFERENCE Florence, Italy The odds of mortality during a heat wave among Morabito et al. (2012) the frail elderly population decreased by 9 percent after introduction of alert system compared with a counterfactual scenario. Hong Kong SAR, China Ischemic heart disease deaths among the elderly fell Chau, Chan, and Woo (2009) by 23 percent against expected baseline following the introduction of extreme heat warnings. Paris, France Following the introduction of France’s Heat Health Watch Pascal et al. (2006) Warning System, mortality during the 2006 heat wave was 68 percent lower than expected based on past heat- mortality relationships. Milwaukee, Wisconsin, Heat wave deaths were 49 percent lower than expected Weisskopf et al. (2002) United States levels after the introduction of a suite of heat-health measures, including an alert system. SOURCE: World Bank. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 131 Given East Asian countries’ existing investments in weather forecasting capacities, the direct costs of implementing heat wave early warning systems in countries that currently lack them are likely to be low. They may include the costs of public health studies and community consultations to design warning systems, the cost for weather services to adapt existing forecasting products to locally defined needs, and the cost of city staff and communications specialists to issue warnings that effectively reach the most vulnerable groups. However, the secondary costs may be more significant. An important priority is the strengthening of health care statistics and their use to fine-tune alert mechanisms. Lisbon, Portugal, is one example of a city where heat wave alerts are accompanied by detailed monitoring of disaggregated health statistics. Public health experts at the city’s heat-health observatory (the ÍCARO (ICARUS) surveillance system)5 monitor health outcomes by demographic group, providing insights that have proven valuable for refining the alert system (Leite et al. 2020). Further costs associated with heat wave early warnings may include strengthening the readiness of hospitals, clinics, and ambulance crews to care for heatstroke patients. Ahmedabad’s Heat Action Plan defined health sector protocols to be activated when “orange” or “red” alerts are issued. These include practical measures such as establishing heatstroke treatment stations in hospitals and equipping ambulance crews with ice packs and rehydration fluid. Even considering these secondary costs of establishing an action-oriented heat wave early warning system, the cost-benefit ratios of such systems is likely to be high (Rogers and Tsirkunov. 2011; Williams et al. 2022). STRATEGIC OPTION 5: PROTECT HEAT-EXPOSED WORKERS Exposure to heat stress varies by occupation. Workers in demanding outdoor occupations such as construction as well as road building and repairs are among the most exposed. Physical labor induces high heat production in workers’ bodies that causes a rise in body temperature and subsequent illness. A large share of indoor jobs also entail heat stress due to physical intensity, building characteristics, exposure to indoor heat sources, and lack of ventilation or air conditioning. Many such roles skew toward a female workforce, particularly in industries such as UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 132 garments and textiles, which play a key role in many East Asian economies. Indeed, females make up more than 70 percent of the textile and garment workforce in the Philippines, Thailand, and Viet Nam, and more than 80 percent in Malaysia and Myanmar (ILO 2022). The physiological limits of human heat endurance—established through studies in military, sports, and occupational health and safety research—have motivated legislation in many countries that limits workplace heat exposure as measured by Wet Bulb Globe Temperature, a heat stress measure recognized by the International Organization for Standardization (ISO) for workplace applications (Kjellstrom, Holmer, and Lemke 2009). Although the devices required to monitor indoor heat stress are inexpensive, enforcement is nonetheless often lacking. However, a wide range of actions can reduce heat exposure during the working day, with employers, local government, and workers themselves each having a role to play (table 4.4). Heat-exposed occupations vary greatly in their physical setting, labor market structure, and degree of supervision by employers or government. To identify the most efficient way to reduce heat stress across different occupations, consulting the workers themselves is therefore highly advantageous. In developing its Heat Action Plan, the city of Ahmedabad, India, convened consultations with heat-affected occupational groups including street vendors and traffic police (Knowlton et al, 2014). By hearing directly from heat-exposed workers, the plan’s authors were able to incorporate practical measures into city guidelines while eliciting greater buy-in and awareness. A further benefit of such consultations is to establish sector-specific actions to be activated when alerts are issued by a heat wave early warning system (see Strategic Option 4). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 133 Table 4.4 Examples of Actions to Protect Heat-Exposed Workers CATEGORY ACTION Educational actions Inform heat-exposed workers about symptoms and risks of heat stress Encourage behaviors that mitigate heat stress: drink water; eat less but more frequently; reduce caffeine and alcohol; seek medical attention when needed Train workers to recognize heat illness symptoms and ensure that coworkers receive early intervention when needed (“buddy system”) Employer actions Organize work shifts to avoid exertion during hot afternoon periods Schedule breaks if heat becomes too intense Provide water and shade at work sites Provide workers acclimatization time (for example, initial 50 percent workload) Adapt uniforms and encourage wearing of light, breathable garments Local government actions Consult heat-exposed workers to understand the risks they face and their needs for early warnings and working practice changes Disseminate heat early warnings to employers, labor unions, and heat-exposed workers Review and improve working practices for heat-exposed government staff and contractors Strengthen enforcement of occupational health and safety legislation SOURCE: Adapted from NRDC 2013. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 134 STRATEGIC OPTION 6: MAINSTREAM HEAT RISK REDUCTION INTO CITY INSTITUTIONS AND STRATEGIES Making cities resilient to extreme heat presents a distinct set of public policy challenges that call for an innovative, citywide response: Managing across agencies. The challenge of extreme heat in cities requires points of contact with an unusually large number of governmental agencies. From planning and parks departments to transit operators, and from health systems to meteorological agencies, effective action requires that many city and national entities work around the same table. Managing across physical scales. Making city landscapes cooler requires action across physical scales to change construction and design practices. From selection of roof and wall materials to the design of buildings and their surroundings, interventions are needed to incorporate green elements and cool materials into individual buildings and areawide plans. Utilizing diverse policy instruments. Heat stress interventions span a wide array of policy instruments—from direct public investments in school buildings or public plazas to “hard” legislative mandates (such as building codes) and “soft,” bottom- up measures to incentivize action by building owners and residents (for example, tree stewardship schemes and public space design guides). In contrast to established policy priorities like transportation, health, and education, no single city government agency is generally vested with the mandate and resources required to lead on extreme heat mitigation. Rather, the actions needed for effective heat mitigation are fragmented among many organizations and stakeholders. To make matters more complicated, city governments often lack a clear focal point for leadership on urban heat mitigation. Without an obvious leadership figure or an agreed-upon, long-term vision and strategy, the staffs of city planning, transportation, health, emergency management, or other departments may pay limited attention to the issue despite having taken promising heat mitigation actions within their spheres of influence. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 135 For cities that are beginning to recognize the need to act on this issue, a wise step might be to establish an extreme heat task force mandated to review evidence on extreme heat, gather and analyze data, consult affected groups, and recommend actions. In Tokyo, systematic action to respond to the UHI effect dates back at least to 1998, when the issue was first raised in the Tokyo Metropolitan Assembly, leading to the setting up of a study group within the Tokyo Metropolitan Government’s Bureau of Environment (Yamaguchi 2013). The group raised awareness and secured buy-in from other branches of municipal and national government, leading to an expanding set of legislative initiatives and direct public investments in the following decade. More recently, a growing number of cities worldwide have taken the step of appointing an individual official—sometimes designated as chief heat officer—to lead citywide actions. Whether led by a task force or an individual officer, a key objective should be to increase the visibility of the issue and secure buy-in from stakeholders on whose part action is needed. For example, after appointing its first chief heat officer in 2022, the city of Santiago, Chile, followed a consultative process before introducing the country’s first extreme heat response protocol, which combines color-coded alerts with an information campaign to protect vulnerable groups and heat-exposed workers. For cities that are at an early stage in addressing the UHI effect, taking stock of existing evidence and action on heat mitigation is an important starting point, but strengthening the evidence base through approaches such as heat mapping and public health assessment can help to build momentum. Community-based mapping approaches—such as the vehicle-based “citizen science” campaign conducted in Bandung, Indonesia, for this study (see chapter 2)—can also help to increase visibility and support. Building a partnership with relevant stakeholders inside and outside government, or a “cool city coalition,” can advance heat reduction actions (annex 4A). Because hot weather follows an annual seasonality, heat wave preparedness and response should also follow an annual seasonality—by strengthening preparedness in the run-up to the hot season, activating protocols when a heat wave is imminent, acting to protect the vulnerable during extreme heat events, and conducting after-action reviews before recommencing the cycle (figure 4.4). UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 136 Figure 4.4 The Heat Wave Risk Management Cycle During heatwave: ■ Protect vulnerable people Heatwave After imminent: heatwave: ■ Raise awareness ■ Review and improve Before heatwave: ■ Strengthen preparedness SOURCE: World Bank elaboration based on Singh et al. 2020. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 137 SUMMARY AND CONCLUSION Extreme urban heat—arising from the combination of the UHI effect and climate change—represents a significant and growing drag on the economic competitiveness of many East Asian cities while also contributing to cities that are less healthy, livable, and inclusive. To rise to the urban heat challenge, leaders in East Asian cities can adopt a “Places, People, Institutions” framework corresponding to three key principles: making places cooler; protecting vulnerable populations; and mainstreaming heat mitigation measures into city departments, operations, and budgets. By acting now, East Asian city leaders can help make their cities not only cooler but also more prosperous, inclusive, and livable. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 138 Mainstreaming Heat Resilience into City Institutions and ANNEX 4A Strategies: Suggested Phases for a City Extreme Heat Task Force with Suggested Key Questions and Actions PHASE 2. BUILD THE 3. BUILD A “COOL 4. INVEST, MONITOR, 1. TAKE STOCK EVIDENCE BASE CITY COALITION” MAINSTREAM ■ Review knowledge, ■ Prioritize investments and ■ Strengthen understanding of ■ Review the organizational strategies, and policy actions; integrate Activities heat hazards, vulnerability, and setting for heat action; actions relating to into core budgets; monitor impacts enlist stakeholder support extreme heat and communicate PLACES ■ Have urban heat ■ How does heat intensity differ ■ Which government ■ Which place-based cooling island studies between city neighborhoods? departments and other interventions offer the most been carried out stakeholders can be immediate, high-impact Key questions ■ Do differences in building already? effective champions for benefits? materials and vegetation cooler city spaces? ■ What do city density expose some residents ■ Should city agencies’ design government and to more heat? ■ Which departments and and procurement standards other stakeholders agencies could use their or operational handbooks be ■ What opportunities exist to already do to core operations, staff, revised to integrate cooling increase green cover, make promote cooler and budget to promote considerations? buildings cooler, integrate city spaces? Are greener streets and cooler shade and water into urban ■ Does the city monitor heat- the interventions buildings? design, and improve wind flow? related deaths each year and working? ■ Could residents’ adjust hot-season planning ■ How does heat affect ■ How could heat associations, professional based on the trends? experience of public facilities mitigation actions bodies, employers, trade including schools, hospitals, contribute to unions or botanical buses and trains? existing city gardens support design strategies and and upkeep of green commitments? assets and cool buildings? ■ Map urban greening ■ Conduct urban heat island ■ Engage citizens through ■ Short-list investment options for stakeholders studies using remote sensing, participatory heat cooler and greener city spaces in situ measurements, and/or mapping, focus groups, Supporting measures ■ Map cool building ■ Identify options to integrate climate models and surveys stakeholders passive cooling measures into ■ Pilot cooling interventions ■ Consult forestry building codes and zoning ■ Conduct desk (such as cool roofs on schools stakeholders on options regulations review on policies or public housing) to evaluate to preserve and increase and strategies for ■ Identify options to strengthen potential investment options green cover cool city places markets for cool building ■ Study impacts of buildings on ■ Engage building owners designs and materials through wind to identify options for and construction industry public procurement improved ventilation on indoor heat reduction ■ Identify opportunities to options ■ Conduct a baseline mainstream heat mitigation assessment of urban forestry, into existing city department including tree health, green strategies cover disparities, and species ■ Initiate prefeasibility study for suitability for future climate prioritized investment options UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 139 PHASE 2. BUILD THE 3. BUILD A “COOL 4. INVEST, MONITOR, 1. TAKE STOCK EVIDENCE BASE CITY COALITION” MAINSTREAM PEOPLE ■ Do residents ■ How do deaths and hospital ■ Which groups of workers ■ What actions to protect currently receive admissions vary with heat? have highest heat vulnerable people would prevent information about exposure? deaths and illnesses at the Key questions ■ Which socioeconomic groups upcoming extreme lowest cost? have a higher rate of death, ■ What health and economic heat? illness, or hospital admission losses could be suffered ■ At what heat stress threshold ■ How do doctors and during extreme heat? in the future if workers are should alerts be issued? hospital workers not protected? ■ Do current weather forecasting prepare for the hot products meet the city’s needs ■ Which stakeholders need season? for protecting residents’ lives to disseminate information ■ Does the national during heat waves? for vulnerable groups to meteorological reduce heat exposure and agency provide receive needed support? anticipatory heat stress forecasts for the city? ■ Map public ■ Conduct a study of how daily ■ Consult vulnerable groups ■ Initiate the design and health and early all-cause mortality varies with on how they receive implementation of a heat wave warning system heat stress climate and health early warning system Supporting measures stakeholders information ■ Identify heat stress thresholds ■ Identify response actions to ■ Conduct desk associated with increased ■ Engage doctors to raise accompany heat wave alerts in review of policies mortality and morbidity awareness of heat-related the health sector, in schools, and strategies to illnesses and prioritize and in heat-exposed workplaces ■ Assess the impact of current protect lives during response actions and future heat stress for ■ Plan public information heat waves workers and school children ■ Consult heat-exposed actions with input from health workers on their safety professionals and affected needs groups CROSS-CUTTING ACTIONS 1. Establish leadership: Designate a city official with accountability for delivering heat mitigation outcomes 2. Plan: Establish a multiyear plan with a vision, goals, and targets (integrate into existing strategies or develop a dedicated heat action plan) 3. Coordinate: Convene city departments to coordinate short-run actions (responsibilities during heat emergencies) and long-run actions (investments for a cooler city) 4. Communicate: Drive behavior change through communication ahead of each hot season SOURCE: World Bank. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 140 REFERENCES Atlantic Council. 2022. “Freetown Market Shade Project.” Project Japan, Ministry of Environment (MOE). Heat Illness Prevention description, Atlantic Council, Washington, DC (accessed Information. Website (accessed May 12, 2023). May 11, 2023), https://onebillionresilient.org/project/ https://www.wbgt.env.go.jp/en/sp/heatillness_pr.php freetown-market-shade-cover-project/. Japan, MOE and JMA (Ministry of the Environment and Japan Azhar, G. S., D. Mavalankar, A. Nori-Sarma, A. Rajiva, P. Dutta, A. Meteorological Agency). 2021. “About the National Jaiswal, P. Sheffield, K. Knowlton, and J. J. Hess. 2014. Launch of the Heat Alert” [in Japanese]. PowerPoint “Heat-Related Mortality in India: Excess All-Cause document (accessed May 12, 2023), https://www.wbgt. Mortality Associated with the 2010 Ahmedabad Heat env.go.jp/pdf/rma_doc/20210101/doc1.pdf. Wave.” PloS One 9 (3): e91831. Jjemba, E. W., A. Kruczkiewicz, T. N. Quang, H. Nguyen, J. Faucet, Cissé, G., R. McLeman, P. Aldunce, K. Bowen, D. Campbell- and R. Singh. 2020. “Applying Forecast Based Actions Lendrum, S. Clayton et al. 2022. “Health, Wellbeing and in Rapidly Urbanizing Cities: The Case of Heatwaves the Changing Structure of Communities.” In Climate in Hanoi, Viet Nam.” Abstract No. NH035-07, American Change 2022: Impacts, Adaptation and Vulnerability, Geophysical Union (AGU) Fall Meeting Abstracts 2020 Working Group II Contribution to the Sixth Assessment JSPO (Japan Sport Association). 1994. “A Guidebook for the Report of the Intergovernmental Panel on Climate Prevention of Heat Disorder during Sports Activities.” Change, edited by H.-O. Pörtner, D. C. Adams, M. M. B. JSA, Tokyo. Tignor, E. Poloczanska, K. Mintenbeck, A. Alegría, M. Craig Konijnendijk, C. 2021. “The 3-30-300 Rule for Urban Forestry et al., 1041–1170. Cambridge and New York: Cambridge and Greener Cities.” Preprint, Biophilic Cities University Press. Journal 4 (2). https://static1.squarespace.com/ Clean Cooling Collaborative. 2022. “Million Cool Roofs static/5bbd32d6e66669016a6af7e2/t/6101ce2b17dc Challenge: Local Champions for a Global Movement.” 51553827d644/1627508274716/330300+Rule+Prepri Clean Cooling Collaborative Blog, March 1. https://www. nt_7-29-21.pdf. cleancoolingcollaborative.org/blog/million-cool-roofs- Kjellstrom, T., I. Holmer, and B. Lemke. 2009. “Workplace Heat challenge-local-champions-for-a-global-movement/. Stress, Health and Productivity: An Increasing Challenge Chau, P. H., K. C. Chan, and J. Woo. 2009. “Hot Weather Warning for Low and Middle-Income Countries during Climate Might Help to Reduce Elderly Mortality in Hong Kong.” Change.” Global Health Action 2 (1): 2047. International Journal of Biometeorology 53 (5): 461–68. Knowlton, K., S. P. Kulkarni, G. S. Azhar, D. Mavalankar, A. Jaiswal, Erell, E., D. Pearlmutter, and T. Williamson. 2011. Urban M. Connolly, A. Nori-Sarma et al. 2014. “Development Microclimate: Designing the Spaces between Buildings. and Implementation of South Asia’s First Heat-Health London: Routledge. Action Plan in Ahmedabad (Gujarat, India).” International ESMAP (Energy Sector Management Assistance Program). Journal of Environmental Research and Public Health 11 2020. Primer for Cool Cities: Reducing Excessive Urban (4): 3473–92. Heat–With a Focus on Passive Measures. Knowledge Lai, D., W. Liu, T. Gan, K. Liu, and Q. Chen. 2019. “A Review Series 031/20, ESMAP, World Bank, Washington, DC. of Mitigating Strategies to Improve the Thermal Giridharan, R., and R. Emmanuel. 2018. “The Impact of Environment and Thermal Comfort in Urban Outdoor Urban Compactness, Comfort Strategies and Energy Spaces.” Science of the Total Environment 661: 337–53. Consumption on Tropical Urban Heat Island Intensity: A Leite, A., A. J. Santos, S. Silva, B. Nunes, R. Mexia, and A. P. Review.” Sustainable Cities and Society 40 (2): 677–87. Rodrigues. 2020. “Assessing the use and understanding Hess, J. J., S. Lm, K. Knowlton, S. Saha, P. Dutta, P. Ganguly, A. of the Portuguese heat–health warning system (ÍCARO).” Tiwari et al. 2018. “Building Resilience to Climate Change: Journal of Public Health 42 (2): 395-402. Pilot Evaluation of the Impact of India’s First Heat Action Morabito, M., F. Profili, A. Crisci, P. Francesconi, G. F. Gensini, and Plan on All-Cause Mortality.” Journal of Environmental S. Orlandini. 2012. “Heat-Related Mortality in the Florentine and Public Health 2018 (9): 7973519. Area (Italy) Before and After the Exceptional 2003 Heat ILO (International Labour Organization). 2022. Employment, Wave in Europe: An Improved Public Health Response?” Wages and Productivity Trends in the Asian Garment International Journal of Biometeorology 56 (5): 801–10. Sector. Geneva: ILO. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 141 McDonald, R., N. DeJarnett, T. Kroeger, R. Yeager, and C. Mee. Satwiko, P. 1999. “Traditional Javanese Residential Architecture 2016. “Planting Healthier Air: An Assessment of the Designs and Thermal Comfort: A Study Using a Potential for Trees to Reduce Air Pollution and Cool Cities Computational Fluid Dynamics Program to Explore, to Enhance Human Health.” Panel presentation, American Analyse, and Learn from the Traditional Designs for Public Health Association (APHA) 2016 Annual Meeting Thermal Comfort.” Doctoral dissertation, Te Herenga and Expo, Denver, October 13. Waka – Victoria University of Wellington, New Zealand. McGregor, G. R., P. Bessemoulin, K. Ebi, and B Menne, eds. 2015. Singh, R., J. Arrighi, E. Jjemba, K. Strachan, M. Spires, and Heatwaves and Health: Guidance on Warning-System A. Kadihasanoglu. 2020. “Heatwave Guide for Cities.” Development. Geneva: World Meteorological Organization Reference guide, International Federation of Red Cross and World Health Organization. Red Crescent Climate Centre, Geneva. Ng, E. 2009. “Policies and Technical Guidelines for Urban Srivastav, S., and P. J. Jones. 2009. “Use of Traditional Passive Planning of High-Density Cities: Air Ventilation Strategies to Reduce the Energy Use and Carbon Assessment (AVA) of Hong Kong.” Building and Emissions in Modern Dwellings.” International Journal of Environment 44 (7): 1478–88. Low-Carbon Technologies 4 (3): 141–49. NRDC (National Resources Defense Council). 2013. “Rising Takkanon, P., and P. Chantarangul. 2019. “Effects of Urban Temperatures, Deadly Threat: Recommendations to Geometry and Green Area on Thermal Condition of Urban Prepare Outdoor Workers in Ahmedabad.” Issue brief, Street Canyons in Bangkok.” Architectural Science NRDC, New York. Review 62 (1): 35–46. PlanD (Planning Department, Government of Hong Kong SAR, Toloo, G., G. FitzGerald, P. Aitken, K. Verrall, and S. Tong. China). 2016. “Urban Design Guidelines.” In Hong Kong 2013. “Evaluating the Effectiveness of Heat Warning Planning Standards and Guidelines. Hong Kong SAR, Systems: Systematic Review of Epidemiological China: PlanD. Evidence.” International Journal of Public Health 58 (5): Oke, T. R., G. Mills, A. Christen, and J. A. Voogt. 2017. Urban 667–81. Climates. Cambridge: Cambridge University Press. Williams, S., M. Nitschke, B. Y. Wondmagegn, M. Tong, J. Xiang, A. Pascal, M., K. Laaidi, M. Ledrans, E. Baffert, C. Caserio- Hansen, J. Nairn, J. Karnon, and P. Bi. 2022. “Evaluating Schönemann, A. Le Tertre, J. Manach, S. Medina, J. Cost Benefits from a Heat Health Warning System in Rudant, and P. Empereur-Bissonet. 2006. “France’s Heat Adelaide, South Australia.” Australian and New Zealand Health Watch Warning System.” International Journal of Journal of Public Health 46 (2): 149–54. Biometeorology 50 (3): 144–53. Weisskopf, M. G., H. A. Anderson, S. Foldy, L. P. Hanrahan, K. Blair, Philipp, C. H., and W. T. L. Chow. 2020. “Urban Heat Vulnerability T. J. Török, and P. D. Rumm. 2002. “Heat Wave Morbidity Analysis for Singapore.” Technical report, ETH and Mortality, Milwaukee, Wis, 1999 vs 1995: An Improved Zurich Research Collection. https://doi.org/10.3929/ Response?” American Journal of Public Health 92 (5): ethz-b-000419689. 830–33. Roloff, A., ed. 2016. Urban Tree Management for the Sustainable Weinberger, K. R., A. Zanobetti, J. Schwartz, and G. A. Development of Green Cities. West Sussex, UK: Wellenius. 2018. “Effectiveness of National Weather Wiley Blackwell. Service Heat Alerts in Preventing Mortality in 20 US Rogers, D., and V. Tsirkunov. 2011. “Costs and Benefits of Early Cities.” Environment International 116: 30–38. Warning Systems.” Input Paper to the Global Assessment Yamaguchi, T. 2013. “Heat Island Measures.” Japanese Journal of Report on Disaster Risk Reduction, World Bank, Biometeorology 50 (2): 71–75. Washington, DC. Yoshino, H., K. Hasegawa, and S-I. Matsumoto. 2007. “Passive Ruefenacht, L., and J. A. Acero, eds. 2017. “Strategies for Cooling Effect of Traditional Japanese Building’s Cooling Singapore: A Catalogue of 80+ Measures to Features.” Management of Environmental Quality 18 (5): Mitigate Urban Heat Island and Improve Outdoor Thermal 578–90. Comfort.” Report, Cooling Singapore, Singapore. Santamouris, M., L. Ding, F. Fiorito, P. Oldfield, P. Osmond, R. Paolini, D. Prasad, and A. J. S. E. Synnefa. 2017. “Passive and Active Cooling for the Outdoor Built Environment: Analysis and Assessment of the Cooling Potential of Mitigation Technologies Using Performance Data from 220 Large Scale Projects.” Solar Energy 154: 14–33. UNLIVABLE What the Urban Heat Island Effect Means for East Asia’s Cities CH ToC O 1 2 3 4 142 NOTES 1 Medium- and long-term strategic 2 For further discussion of these issues, 4 The World Meteorological planning for urban forestry is also see chapter 4 (“Airflow”) in Oke et al. Organization has published important important considering two global (2017). guidance on heat-health early environmental threats. First, newly 3 For information on both “hard” and warning system development, introduced pathogens and pests “soft” policy interventions to promote including varieties of heat (for example, Dutch elm disease) cooler cities, see ESMAP (2020) and threshold definition and monitoring have decimated tree populations Ruefenacht and Acero (2017). Based mechanisms (McGregor et al. 2015). in the past. Second, rising global respectively on global experience 5 ÍCARO both translates to Icarus—who, temperatures are causing suitability and Singapore’s local experience, in Greek mythology, perished by zones for individual species to shift, these reports each provide a typology flying too close to the sun—and is an year by year, in a poleward direction. of measures with discussion of acronym for Importância do Calor: Planning on a multidecade timeline their applicability and factors for Repercussão nos Óbitos (“Importance can address these challenges by successful implementation. of Heat: Impact on Deaths”). fostering tree stock genetic diversity and ensuring that newly planted trees are suitable for the future climate. For an overview of key issues in urban tree management, see Roloff (2016). Amid continuing urban growth and the accelerating effects of climate change, East Asian cities suffer from more extreme temperatures than surrounding rural areas—being up to 2 degrees Celsius hotter on average. This urban heat island (UHI) effect is caused by cities’ relative lack of vegetation, the prevalence of impervious surfaces, construction of buildings in locations that block breezes, releases of heat from cars and machinery, and other features of the urban environment. In the decades ahead, the UHI effect will interact with climate change in ways that make cities even more prone to heat waves— already increasing in frequency and intensity—especially among East Asian cities in tropical zones and in low- and middle-income countries. Extreme heat not only lowers the economic competitiveness and livability of cities in the region but also increases the risk of death and illness. Groups such as low-income residents, outdoor workers, and the region’s growing elderly population are the most vulnerable to extreme heat. The poor are also more likely to bear the brunt of these harms: certain urban neighborhoods, particularly poorer ones, may be several degrees hotter than others within the same city. Unlivable: What the Urban Heat Island Effect Means for East Asia’s Cities uses satellite data, on-the-ground data collection, and a review of economic literature to shed new light on the magnitude of the UHI effect and its impacts on East Asian cities. Using a “Places, People, Institutions” framework, the report provides practical suggestions to help policy makers to rise to the extreme urban heat challenge. These actions—such as promoting urban greening, adopting heat-resilient urban design, and implementing heat wave early warning systems—can help to protect East Asia’s urban residents from the impacts of extreme heat, contributing to cities that are more competitive, livable, and inclusive.