ENVIRONMENT AND NATURAL RESOURCES GLOBAL PRACTICE DISCUSSION PAPER #03 CLEAN AIR AND HEALTHY LUNGS Enhancing the World Bank’s Approach to Air Quality Management Yewande Awe, Jostein Nygard, Steinar Larssen, Heejoo Lee, Hari Dulal, and Rahul Kanakia FEBRUARY 2015 WORLD BANK GROUP REPORT NUMBER ACS9035 ENVIRONMENT AND NATURAL RESOURCES GLOBAL PRACTICE DISCUSSION PAPER 03 CLEAN AIR AND HEALTHY LUNGS Enhancing the World Bank’s Approach to Air Quality Management Yewande Awe, Jostein Nygard, Steinar Larssen, Heejoo Lee, Hari Dulal, and Rahul Kanakia © 2015 World Bank Group 1818 H Street NW Washington, DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org Email: feedback@worldbank.org All rights reserved This volume is a product of the staff of the World Bank Group. The findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of the Executive Directors of World Bank Group or the governments they represent. The World Bank Group does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of World Bank Group concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. World Bank Group encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone: 978-750-8400, fax: 978-750-4470, http://www.copyright.com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, World Bank Group, 1818 H Street NW, Washington, DC 20433, USA, fax: 202-522-2422, e-mail: pubrights@worldbank.org. All dollar amounts are U.S. dollars unless otherwise indicated. Cover images starting at the top right: Sura Nualpradid, Ghana; Hung Chung Chih, Beijing; View Apart, South Africa; Andrey Bayda, Cambodia. CONTENTS Foreword vii Acknowledgments ix List of Acronyms xi Executive Summary xiii Chapter One: Introduction 1 Background 1 Objectives of this Report 2 Institutional Context for this Report 2 Drivers of Air Pollution in Developing Countries 4 Sources and Impacts of Air Pollution 5 Air Pollution and Air Quality Management in Developing Countries: Status, Challenges, and Progress 8 Summary 12 Chapter Two: A Review of the World Bank Project Portfolio of Air-Pollution-Relevant Projects 15 Objectives of Portfolio Review 15 Underlying Principles—Linkages between Air Pollution Sources, Air Pollution Concentrations, and Impacts on Human Health 15 Methodology 16 Overview of the Air-Pollution-Relevant Portfolio 17 Potential Public Health Impacts of the Air-Pollution-Relevant Portfolio 22 Observations and Conclusions 23 Chapter Three: A Review of Case Examples 25 Methodology 25 Effectiveness of the Case Study Projects 26 Lessons Learned from the Case Study Projects 29 Chapter Four: A Proposed Approach for Estimating the Air Quality and Health Impacts of World Bank Projects 33 Introduction 33 Integrated Air Quality Management Concept 33 Considerations for Application of Integrated Air Quality Management Concept in World Bank Projects 34 Chapter Five: Conclusions and Recommendations 37 Appendix A: Air Quality Status and Challenges in Various Regions 41 Appendix B: Methodologies for Assessing Health Impacts of Air Pollution Reduction Projects 47 Appendix C: Methodology for Review for the World Bank Project Portfolio that is Relevant to Air Pollution 53 Appendix D: Review of Case Examples 71 Appendix E: Idea Note for Methodology for Integration of SLCP Compounds in Air Quality Management Modeling 89 Appendix F: Project Typologies, Air Pollution Aspects, and Some Relevant Emissions Data Requirements 95 References 103 Enhancing the World Bank's Approach to Air Quality Management iii BOXES Box 1.1: Air Pollutants, Sources, and Health Effects 6 Box C1: Activities under Pollution Management and Environmental Health Theme Code 54 FIGURES Figure ES.1: Comparison of Ambient Air Pollution and Other Environmental Health Risks in Developing Countries (Premature Deaths) xiv Figure ES.2: PM10 Data for Cities in Selected Regions and Countries (Yearly Average in μg/m3) xv Figure ES.3: Economic Burden of Health Damage Associated with Ambient Air Pollution xv Figure ES.4: Trends in Deaths from Ambient Air Pollution (PM), by Region, 1990–2010 xvi Figure ES.5: Air-Pollution-Relevant Single Projects by Product Line xvii Figure 1.1: Global Air Pollution Deaths in 2012 by Disease 7 Figure 1.2: Comparison of Air Pollution (PM) and Other Environmental Health Risks in Developing Countries’ Premature Deaths 7 Figure 1.3: Relationship between Pollution, Urban Labor Productivity, and Urban Livability 8 Figure 1.4: Economic Burden of Health Damage Associated with Ambient Air Pollution as Percentage of GDP 9 Figure 1.5: PM10 Data for Cities in Selected Regions and Countries (Yearly Average in μg/m3) 10 Figure 1.6: Trends in Deaths from Ambient Air Pollution (PM), by Region, 1990–2010 10 Figure 1.7: Development in PM10 Concentrations in Chinese Cities (a) and Huhot City (b), 1997–2012 12 Figure 2.1: Relationship between Air Pollution Sources, Air Pollution Concentrations, and Health Impacts 16 Figure 2.2: Relationship between Emissions, Air Pollution Concentrations, Exposure, and Health Impacts 16 Figure 2.3: Distribution of Air-Pollution-Relevant Projects by Sector Board 18 Figure 2.4: Air-Pollution-Relevant Projects by Product Line 19 Figure 2.5: Air-Pollution-Relevant Projects by Lending Instrument (IBRD/IDA Product Line Only) 19 Figure 2.6: Distribution of Air-Pollution-Relevant Projects by Region 20 Figure 3.1: PM2.5 Concentrations at the Parque O’Higgins Station, Santiago, Chile, 1989–2012 27 Figure 3.2: Annual Ambient PM2.5 Concentrations in Lima-Callao, Peru, 2003–12 (μg/m3; 3-Year Moving Averages) 27 Figure 3.3: Population-Weighted Average Concentration for Different Emission Reduction Scenarios, PM2.5 28 Figure 3.4: Health Impacts of Ambient Air Pollution per Unit of Income in Lima-Callao 28 Figure 3.5: Annual Health Benefits from Five out of Eight Abatement Scenarios in Ulaanbaatar, 2010–13 30 Figure 3.6: Comparing Health Benefits of Two Interventions: Certified Stoves and Relocation into Apartments in Ulaanbaatar 30 Figure 4.1: Framework for Comprehensive Integrated Air Quality Management 34 Figure 4.2: Air Pollution and Health Impact Assessment Process 35 Figure A.1: PM10 Data from Cities in Selected Regions and Countries (Yearly Average in μg/m3) 41 Figure B.1: Relationship between Emissions, Air Pollution Concentrations, Exposure, and Health Impacts 48 Figure B.2: Schematic Representation of AirQUIS 51 Figure D.1: Numbers and Types of Buses in Santiago, Chile, 2006–13 73 iv Clean Air and Healthy Lungs Figure D.2: PM2.5 Concentrations at the Parque O’Higgins Station, Santiago Chile, 1989–2012 74 Figure D.3: Monthly Average PM2.5 Concentrations in Ulaanbaatar, June 2008–May 2009 75 Figure D.4: Population Distribution in Ulaanbaatar, 2008 76 Figure D.5: Modeled Spatial Distribution of PM2.5 in Ulaanbaatar 77 Figure D.6: Annual Health Benefits from 5 Out of 8 Abatement Scenarios in Ulaanbaatar from 2010–23 78 Figure D.7: The Cost of Delaying Short-Term Measures–Comparing Health Benefits between the Certified Stoves and the Relocation into Apartments Scenarios in Ulaanbaatar 80 Figure D.8: Needed Reduction in PM10 and PM2.5 Concentration Levels to Reach Air Quality Standards 81 Figure D.9: Development in PM10 Concentrations at Zuul Ail Station in Ulaanbaatar 2010–13 82 Figure D.10: Cost of Environmental Health Damage in Peru 83 Figure D.11: Health Impacts of Ambient Air Pollution per Unit of Income in Lima-Callao 83 Figure D.12: Marginal Costs and Benefits of Actions to Reduce Particulate Matter Emissions in Peru 84 Figure D.13: Annual Ambient PM2.5 Concentrations in Lima-Callao 2003–12 (μg/m3; 3-Year Moving Averages) 87 Figure E.1: Principal Concept of Analytical Air Quality Management 93 Figure E.2: Population Weighted Average Concentration for Given Emission Reductions, PM2.5 94 TABLES Table 2.1: Distribution of the Reviewed Project Portfolio across World Bank Sectors/Themes and Sector Boards 17 Table 2.2: Distribution of Air-Pollution-Relevant Projects by Typology 21 Table 3.1: Characteristics of the Three Case Study Projects 26 Table 3.2: Estimated Annual Health Impact of Ambient Air Pollution from Particulate Matter in Peru 31 Table A.1: Fifty Cities with the Highest Annual Mean Concentrations of PM10 (in μg/m3) 42 Table C.1: Grouping of Project Activities into Typologies 55 Table C.2: World Bank Air-Pollution-Relevant Projects (Approved FY02–12: Closed Projects) 56 Table C.3: World Bank Air-Pollution-Relevant Projects (Approved FY08–13: Active and Closed Projects) 60 Table C.4: Qualitative Classification of Health Impact Reduction Potential of Abatements in Air-Pollution-Relevant Projects 67 Table D.1: Characteristics of the Three Case Study Projects 72 Table D.2: Emissions from Main Air Pollution Sources in Ulaanbaatar, 2008 (Tons/Year) 76 Table D.3: Population Weighted Exposure to PM in Ulaanbaatar as Calculated by the Air Pollution Model 77 Table D.4: Statistically Significant or Near-Significant Mortality Effect Estimates for PM and NO2, June 2008–May 2009 78 Table D.5: Comparison of Present Value (PV) of Health Benefits (Base Case) with Net Present Value (NPV) of Implementing Costs, and Net Benefit (PV minus NPV) for the Eight Abatement Scenarios, 2010 ($ millions) 79 Table D.6: Average PM10 Concentrations at Zuul Ail Stations 2008–13 82 Table D.7: Estimated Annual Health Impact of Urban Air Pollution from Particulate Matter 83 Table D.8: Peru DPL Program Development Objectives, Key Indicators, Baseline and Program Outcomes 85 Table E.1: Lifetimes, Radiative Efficiencies, and GWPs Relative to CO2 92 Enhancing the World Bank's Approach to Air Quality Management v FOREWORD Paula Caballero Senior Director for Environment and Natural Resources Global Practice World Bank and Ede Iijasz-Vasquez Senior Director for Social, Urban, Rural and Resilience Global Practice World Bank It goes without saying that we all need to breathe to live—but for many people in the world today, breathing can also lead to death. In fact, 3.7 million people died world- wide in 2012 alone from the chronic or the acute effects of breathing in atmospheric pollutants, according to the World Health Organization. Eighty-eight percent of these deaths occurred in low- and middle-income countries. In 2010, ambient air pollution was the fifth leading cause of premature death in the developing world and the third highest environmental health risk. We know that the poor suffer the most because they have fewer options to move to cleaner neighbor- hoods or take protective measures. If left unaddressed, these problems are expected to grow worse over time, as the world continues to urbanize at an unprecedented and challenging speed. Urban, peri-urban, and rural pollution emissions combine in dangerous ways. Deficiencies in transport, industrial practices, workplace safety, and housing expose people to toxic emissions at home, at work, and in-between. In many cities around the world, the lack of adequate air quality control and enforcement puts millions of people at risk of developing debil- itating and often fatal illnesses related to ambient air pollution. The economic burden of air pollution also weighs heavily on developing countries. Premature death, illness, lost earnings, and medical costs can hurt productivity, which is essential for economic growth. This can hamper a growing city’s ability to provide the very opportunities that new urban residents expect, whether services or infrastructure. These risks, in the 21st century, are not acceptable—and the World Bank Group is taking action. The analysis contained in this report is a first step toward a more systematic, cross- sectoral, and evidence-based approach to enhancing air quality management in World Bank lending and technical assistance to developing countries. The report highlights Enhancing the World Bank's Approach to Air Quality Management vii that improving air quality can be achieved in the face of urbanization when proactive leaders are willing to institute the right policies and investments: a nation can have clean air and healthy lungs in addition to the economic benefits of urbanization. It finds that World Bank interventions can be more successful if air quality manage- ment is clearly prioritized rather than treated as an add-on, investments proceed on the basis of solid analytical foundations and are appropriately cross-sectoral to address the diverse sources of ambient air pollution in different country contexts. Report les- sons will help guide the collaboration of different centers of expertise at the World Bank—whether they be staffed by urban, rural, health, transport, energy, or environ- mental specialists—to deliver more optimal pollution management and environmental health results. We also have a good idea of the tools and models that have worked in different parts of the world to turn the corner on decades of rising pollution and put cities on a more livable and economically attractive footing. We now need to step up our game and adopt a more comprehensive approach to fixing air quality, so that people can breathe more easily and enjoy longer, healthier, and more productive lives. At a time when megacities are emerging at great speed in places like South and East Asia and Sub-Saharan Africa, and with much of urban population growth expected to take place in developing countries by 2050, it is imperative that we find collaborative solutions to enhance air quality management and lessen the health burden of pollu- tion. Without further action, air pollution will continue to worsen and become a mas- sive hurdle to achieving a prosperous future for the poor and bottom 40 percent of the world’s population. Effective interventions, on the other hand, promise to deliver mul- tiple wins in terms of productivity, quality of life, and even climate change. Enhanc- ing air quality management is a no-regrets development option. It should become a priority investment. viii Clean Air and Healthy Lungs ACKNOWLEDGMENTS This knowledge product was prepared by a team led by Yewande Awe (World Bank). The core team included Jostein Nygard and Heejoo Lee (World Bank), Steinar Larssen, Hari Bansha Dulal, and Rahul Kanakia (Consultants). The authors would like to acknowledge, with thanks, the valuable advice and inputs of the peer reviewers: Ernesto Sánchez-Triana, Helena Naber, Hocine Chalal, Maged Hamed, and Sameer Akbar (World Bank). The authors are grateful to Marcela Castillo Reyes and Pedro Oyola of the Centro Mario Molina Chile and to Professor Lodoysamba Sereeter of the National University of Mongolia for their valuable technical inputs. Several colleagues also provided useful contributions, including Anca Bogdana Rusu, Craig Meisner, Dora Cudjoe, Flore Martinant de Preneuf, Gary Kleiman, Katelijn Van den Berg, Megha Mukim, Ruma Tavorath, Samuel Kwesi Ewuah Oguah, and Sanjay Srivastava. In addition, Adelaide Barra, James Cantrell, Laura Lewis De Brular, Ruth Maturan Cruz, and Xiaofeng Li are thanked for their various inputs in advancing this work. Editorial services were provided by Linda Starke. Finally, thanks go to Sameh Wahba, Zoubida Allaoua, Valerie Hickey, Paula Caballero, and Ede Ijjasz-Vasquez. Enhancing the World Bank's Approach to Air Quality Management ix LIST OF ACRONYMS AAA Analytical and Advisory Activities FIL financial intermediary loan AFR Africa Region FY fiscal year AirQUIS Air Quality Information System GAINS Greenhouse Gas–Air Pollution Interactions and Synergies AMHIB Ulaanbaatar Air Monitoring and Health Impact Baseline GBD Global burden of disease APC air pollution control GDP Gross domestic product APL adaptable programmatic loan GEF Global environment facility AQG Air quality guideline GHG Greenhouse gas AQM Air quality management GWP Global warming potential BC Black carbon HFCs Hydrofluorocarbons BenMAP Environmental Benefits Mapping and HOB Heat-only boilers Analysis Program IAQM integrated air quality management CEA Country environmental analysis IBRD International Bank for Reconstruction CH4 Methane and Development CMMCh Centro Mario Molina Chile IDA International Development Association CNG Compressed natural gas IIASA International Institute for Applied Systems Analysis CO Carbon monoxide LCR Latin America and the Caribbean CO2 Carbon dioxide Region CoED Cost of environmental degradation MCC Millennium Challenge Corporation CRF Concentration response function MNA Middle East and North Africa Region CSI Core sector indicators N2O Nitrous oxide DALY Disability-adjusted life year NILU Norwegian Institute for Air Research DPL Development policy loan NO2 Nitrogen dioxide EA Eastern Africa NO3 Nitrates EAP East Asia and Pacific Region NOX Nitrous oxides ECA Europe and Central Asia Region NPV Net present value EHS Environment, Health, and Safety O3T Tropospheric ozone EMT Energy and Mining Sector Board OC Organic carbon ENV Environment Sector Board OPCS Operations, Policy and Country Services ENVDPL Environmental Development Policy PAD project appraisal document Loan Program (Peru) Pb Lead ERL emergency recovery loan PDO Project development objective EU European Union Enhancing the World Bank's Approach to Air Quality Management xi PM Particulate matter SLCP Short-lived climate pollutant PM10 Particulate matter with a diameter of SO2 Sulfur dioxide less than 10 microns SO4 Sulfates PM2.5 Particulate matter with a diameter of SPF special financing less than 2.5 microns TAL technical assistance loan PMEH Pollution management and environmental health TRANSANTIAGO Urban Transport Plan for Santiago PV Present value Ulaanbaatar Ulaanbaatar PWE Population-weighted exposure UD Urban Development Sector Board RAINS Regional Air Pollution Information US EPA United States Environmental and Simulation Protection Agency RF Radiative forcing VOC Volatile organic compound SA Southern Africa WAT Water Sector Board SAR South Asia Region WCA Western and Central Africa SIA Secondary inorganic aerosol particles WHO World Health Organization SIL specific investment loan WTP Willingness to pay xii Clean Air and Healthy Lungs EXECUTIVE SUMMARY An estimated 3.7 million people died worldwide in 2012 from either the chronic or the acute effects of breathing atmospheric pollutants, according to the World Health Organization. (When the effects of house- hold air pollution are added, that total rises to 7 million.) The vast majority of these deaths—88 percent—occurred in the developing world. Ambient air pollution was the fifth leading cause of premature death in the developing world in 2010, and the third highest environmental health risk (see figure ES.1). Ambient air pollution causes sub- stantially more illness and death than the use of unimproved sanitation or unimproved water sources. Despite the severity of this problem, the World Bank has had difficulty in adapting its portfolio and instruments to the challenge of addressing ambient air quality concerns. This report aims to examine the Bank’s experi- ence in supporting countries’ efforts to improve ambient air quality and environmental health outcomes and to extract lessons learned and recommendations for enhancing future World Bank support to clients. The specific objectives are to review the problem of ambient air pollution in the developing world, including its health impacts and the current state of air quality management; to understand what types of project activities the World Bank has supported that address air pollution and environmental health; to assess how successful those projects have been in improving air quality and reduc- ing health impacts from air pollution; to draw out lessons and highlight good prac- tices based on case examples; and to recommend ways to enhance future World Bank efforts in reducing air pollution and the associated health burden. The context for this work is that the World Bank has begun to establish a new program on pollution management and environmental health. Its objectives are to help clients reduce pollution levels and associated health risks and thereby improve environmental health outcomes; to generate new knowledge and improve our under- standing of pollution and its health impacts in both urban and rural areas; and to promote increased awareness and understanding of pollution and its health impacts among policy- makers, planners, and other relevant stakeholders in low- and middle-income countries. One of the four components of the program focuses on helping developing countries Enhancing the World Bank's Approach to Air Quality Management xiii FIGURE ES.1. COMPARISON OF AMBIENT industrial firms, and other heavy emitters AIR POLLUTION AND within a geographical area increases and, in the absence of effective interventions, results OTHER ENVIRONMENTAL in a persistent and severe reduction in air qual- HEALTH RISKS IN ity. The World Health Organization’s guideline for par- DEVELOPING COUNTRIES ticulate matter (PM)—the tiny particles that get lodged (PREMATURE DEATHS) in the lungs and generate the greatest health impacts— Tobacco smoking Unimproved sanitation Household air pollution from solid fuels Ambient ozone pollution is that a city’s average annual PM10 levels should be Ambient particulate matter pollution Occupational risks Unimproved water source Residential radon 20 micrograms per cubic meter. This is a value that has Lead exposure been attained or almost attained in cities in developed countries, such as London, New York, and Paris. How- Deaths per 100,000 people 80 70 ever, many large and highly populated cities in develop- 60 ing countries are registering levels that are significantly 50 40 higher. Karachi, Gaborone, and Delhi, for instance, have 30 yearly averages that are above 200 micrograms per cubic 20 10 meter (see figure ES.2). And this problem will only get 0 worse if unaddressed. By the middle of the 21st century, Source: Constructed by authors using global burden of disease (GBD) com- the global urban population is expected to nearly double, pare tool, based on GBD 2010 data. from approximately 3.4 billion in 2009 to 6.4 billion in 2050. Much of that urban growth will take place in small and mid-size cities in developing countries. and cities to improve air quality management. This report is being prepared within the framework of this new Pollution Air pollution also imposes a substantial eco- Management and Environmental Health program and is nomic burden on developing countries. Poor air intended to serve as a preliminary input to inform the devel- quality, resulting in premature death, illness, lost earnings, opment of the program, specifically with respect to enhanc- and increased medical costs, can constrain productivity, ing the Bank’s support to countries in their efforts to control which is essential for economic growth. Estimates of the ambient air pollution. At the outset, the Pollution Manage- economic cost of health damage associated with ambi- ment and Environmental Health program was designed to ent air pollution are typically based on the exposure of focus on urban environmental problems. Hence, although the population to particulate matter and can amount to this report introduces the subject of household air pollu- 0.1–3.2 percent of gross domestic product in selected tion in sections—particularly in more broadly understand- developing countries across six regions (see figure ES.3). ing the magnitude of the problem associated with air pollution—its primary focus in reviewing World Bank proj- Most of the Bank’s attention regarding air pol- ects is ambient air pollution, which is more commonly asso- lution has focused on urbanized regions, notably ciated with cities and urban areas. The significant problem East Asia and the Pacific, where the number and that household air pollution represents in many developing rate of deaths due to air pollution have increased countries is acknowledged and merits additional studies. tremendously over the last 20 years. However, urbanization trends point to other regions that WITHOUT FURTHER ACTION, will soon be at risk of developing air pollution AMBIENT AIR POLLUTION problems. Notably, Africa has leapt ahead of Asia as the most rapidly urbanizing region in the world: it is esti- PROBLEMS WILL GET WORSE mated that its population will increase by about 60 per- OVER TIME cent between 2010 and 2050, with the urban population The primary driver of air pollution is urban- tripling to 1.23 billion, which will be 20 percent of the ization. As cities grow, the number of vehicles, world’s urban population. And by 2033, 20 African cities xiv Clean Air and Healthy Lungs FIGURE ES.2. PM10 DATA FOR CITIES IN SELECTED REGIONS AND COUNTRIES (YEARLY AVERAGE IN μg/m3) 320 300 WHO interim target I 280 260 WHO interim target II 240 EU limit value 220 200 WHO interim target III 180 WHO guideline value 160 140 120 100 80 60 40 20 0 Be u g a La r ab os e D i Ka a tD C i o es Bo a tá W a M aw Lo w N don rk ris h ch a cr ho ijin on k el air oh m co el go Yo ak g ha iti Pa s ra Ac Li D nz or D n ar os C D ew La ta G yp Eg Source: Compiled by authors based on World Health Organization (WHO) ambient air pollution in cities database 2014. FIGURE ES.3. ECONOMIC BURDEN OF HEALTH DAMAGE ASSOCIATED WITH AMBIENT AIR POLLUTION India SAR Bangladesh Pakistan China EAP Indonesia Macedonia Kosovo ECA Kazakhstan Tajikistan Central African Republic Senegal AFR Nigeria Ghana Madagascar Peru Colombia LCR Nicaragua Honduras Guatemala Egypt MNA Jordan Lebanon 0 0.5 1 1.5 2 2.5 3 3.5 Percentage of GDP Source: Authors’ illustration based on various World Bank reports. Enhancing the World Bank's Approach to Air Quality Management xv FIGURE ES.4. TRENDS IN DEATHS standardized monitoring techniques and data collection FROM AMBIENT AIR protocols; the limited role of data in influencing revisions of air quality standards; and a lack of technical capacity POLLUTION (PM), BY for air quality management. REGION, 1990–2010 EAP ECA LCR 1,600 MNA SAR AFR 1,400 THE WORLD BANK NEEDS A COMPREHENSIVE APPROACH 1,200 FOR ADDRESSING AIR POLLUTION Deaths (in 1000s) 1,000 800 Out of 83 air-pollution-relevant projects—that is, projects that included activities that have a 600 potential to reduce air pollution—reviewed, the 400 majority (55 projects, or more than 60 percent) did not include air pollution control among their 200 project development objectives: they focused primarily on sector investments and activities; 1985 1990 1995 2000 2005 2010 2015 air pollution control was an add-on. Because air Year Source: Compiled by Authors using GBD compare tool, based pollution control was not a primary development objec- on data for years shown. http://viz.healthmetricsandevaluation. tive of these projects, they did not include measurement org/gbd-compare/. of baseline data on air quality, make provision for the collection of air quality data over time, or include indi- cators to measure the success of air pollution reduction will be among the largest 100 cities in the world, with interventions that they supported. As such, it was difficult 50 cities having reached 10 million or more. By 2025, to determine project impacts in reducing air pollution Africa will be home to the world’s fastest-growing megac- and associated adverse health outcomes. Instead, only ity, Lagos, with urban population growth rates consider- indicative assessments of project impacts could be made. ably faster than megacities in Asia. All of this, combined Furthermore, almost half of these projects were financed with an underdeveloped air quality monitoring system using concessional funds (global environment facility that makes it difficult to estimate the health impacts of air [GEF] or carbon offset, see figure ES.5). The low prior- pollution in the region, suggests that addressing Africa’s ity accorded to air pollution control within the Bank and air quality issues should be given priority so that they do in client countries is evident in the dearth of collection not worsen over time (see figure ES.4). of baseline data and information during project planning and implementation that would allow for the assessment Despite the health and economic burdens attrib- of impacts of air pollution control components as well as utable to ambient air pollution, and its conse- the limited use of Bank (International Bank for Recon- quences on climate, many cities in developing struction and Development/International Development countries face significant challenges in their Association [IBRD/IDA]) funds for financing projects struggle to improve air quality management. that address air pollution. Some commonly faced challenges to adequate air quality management across regions include inadequate air qual- Most Bank-supported projects with air pollu- ity monitoring infrastructure; lack of national air quality tion control problems are in East Asia and the standards, particularly for pollutants that are important for Pacific, and there were relatively few in South health, notably PM2.5; variable data quality; an absence of Asia and Sub-Saharan Africa. The lack of attention xvi Clean Air and Healthy Lungs FIGURE ES.5. AIR-POLLUTION-RELEVANT WORLD BANK INTERVENTION PROJECTS BY PRODUCT LINE WORKS BEST WHEN IT FY02–12 FY08–13 25 23 IS COMPREHENSIVE AND 20 20 19 BACKED BY UPSTREAM ANALYTICAL WORK Number of projects 15 12 The desk review did reveal some World Bank 10 projects with significant impacts. Three projects were selected as case studies, based on their sound project 5 4 4 design and overall effectiveness. The Peru Environment 1 0 0 Development Policy Loan (DPL) Program and the San- IBRD/IDA GEF Carbon offset SPF tiago Urban Transport Project contributed to significant Note: Data adjusted to eliminate double counting of projects in FY08–FY12. improvements in air quality throughout the Lima-Callao metropolitan area and Santiago metropolitan areas, respectively, while a Bank-supported air quality study in paid to air quality management in the Bank’s portfolio is Mongolia resulted in the development and implemen- of concern, for the South Asia region contains 8 of the tation of new cost-effective interventions for air quality top 10 and 33 of the 50 most-polluted cities (in terms management in Ulaanbaatar. of particulate matter concentrations) in the world (see appendix A for a list of all 50). Despite these issues, air The case examples reviewed demonstrate the pollution control does not appear to be a high priority importance of sound analytical work. The Peru for either the Bank or its clients in the region: South Environment DPL series was the result of a years-long Asia is second in the world in terms of deaths from poor process, which resulted in the Peru Country Environ- air quality. Additionally, the Bank has comparatively mental Analysis (CEA). This CEA identified and quan- few projects in Africa, a region whose rapid urbaniza- tified the key drivers of environmental degradation tion and almost-nonexistent air quality management in Peru and created support for sound environmental (AQM) framework are setting the scene for worsening governance, which resulted in the creation of Peru’s air quality and health outcomes unless there are prompt Ministry of Environment. When the government of interventions. Peru approached the World Bank for support in opera- tionalizing the Ministry of Environment, the eventual Bank projects that address air pollution reduc- loan program incorporated a number of the recom- tion tend to lack a firm analytical underpin- mendations of the CEA. In Ulaanbaatar, on the other ning. Without collecting baseline data on air quality hand, the World Bank wanted to implement a stove- in the region, conducting an emissions inventory to replacement program in the tent villages that surround figure out amounts and sources of air pollution, and the city, but the government was resistant to the notion determining health impacts, it is not possible to appro- and felt that alternative long-term options might be priately target interventions in a cost-effective manner. more cost-effective. Eventually, the government and This analysis can be costly, and it could take at least the Bank embarked on an ambitious study to measure one to two years to complete because it must be con- the sources of emissions in Ulaanbaatar, measure their ducted in all seasons and conditions. However, it is a impacts, and evaluate possible interventions. The study necessity if air pollution control projects are going to allowed the government and donor agencies to come to truly tackle the sources of that pollution in a city or solid conclusions about possible interventions and even- urban area. tually to implement an appropriate program. At the Enhancing the World Bank's Approach to Air Quality Management xvii outset of the Santiago Urban Transport Programmatic DPL, studies were conducted that paved the way for an THE BANK SHOULD environmental focus in the DPL itself targeting air pol- STRENGTHEN UPSTREAM lution control, and throughout the period of the DPL ANALYTICAL WORK, TAKE further studies were undertaken, resulting in the techni- A CROSS-SECTORAL cal assistance loan (TAL) and global environment facil- ity (GEF) projects. APPROACH, AND USE A BROADER ARRAY OF All three case examples demonstrate the INSTRUMENTS FOR importance of engaging in a dialog with client governments. In addition to the data that it yields, REDUCING AIR POLLUTION analytical work provides a mechanism for allowing Air pollution control projects should engage in room for stakeholder engagement. In Santiago, the upfront analytical work. This work is crucial for original Urban Transport Programmatic DPL created determining the most cost-effective interventions the basis for a policy dialog on the application of air for addressing air pollution and should be based quality policies and established the basis for further on solid dialog with clients. The Mongolia and Peru fine-tuned air quality management interventions in case studies both demonstrate the effectiveness of compre- the following TAL and GEF-funded projects. And, as hensive analytical work—notably, an integrated air quality noted, the Peru and Mongolia projects were the out- management process for Mongolia and a country environ- come of years of discussions and negotiations with mental analysis including cost of environmental degradation client governments, which included the use of analyti- for Peru—accompanied by continued dialog with the client cal work to clarify government concerns and assist the and relevant country actors in order to build ownership of client in prioritizing and designing interventions. As a the air pollution control agenda. The results of analytical result, when projects were finally designed and imple- work should be framed in terms that are easily understand- mented, they had more support from the client and in able by policy decision makers: lives lost and economic cost. the end were more successful. Air pollution is a cross-sectoral problem, and A cross-sectoral approach was a key part of the Bank projects should take a cross-sectoral success of these projects. In the Peru and Mongo- approach to addressing it. The approach in many lia cases in particular, the projects evaluated a number of Bank projects that address air pollution has been largely options and attempted to identify air pollution emissions driven by individual sectors. This sector-driven approach sources before deciding on interventions. Furthermore, is reflected in the dominance of the specific investment interventions were subjected to a cost-benefit analysis that lending (SIL) instrument for supporting air-pollution- allowed programs in different sectors to be compared. relevant projects. To better address air pollution, there is For instance, the Mongolia program compared the costs a need to shift from a sector-driven approach to a cross- and benefits of improving stoves in Ger households to the sectoral approach that has air quality management as costs and benefits of relocating the population into apart- its primary focus first and then identifies, based on cost- ments. The Peru DPL compared the costs and benefits benefit analysis, cost-effective abatement options and the of improving fuel quality versus converting vehicles to respective sectors where such interventions should be tar- work on compressed natural gas. In addition, the DPL geted. An integrated air quality management approach involved actions in transport and industry sectors. This is proposed. This approach underscores the need for an comprehensive, bottom-up approach, working outward increased degree of collaboration between different sec- from causes to solutions and engaging various stakehold- tors in the World Bank, likely beyond levels that typically ers across relevant sectors, allows for clearer decision mak- exist, entailing the setting of common targets with respect ing and better outcomes. to air quality management in client countries. xviii Clean Air and Healthy Lungs The Bank should increase synergies between ing, which is needed in many client countries to improve local air quality management and control of air pollution control. short-lived climate pollutants (SLCPs), notably black carbon, that also have health impacts. The Bank should increase support to client gov- Short-lived climate pollutants such as black carbon and ernments, with technical assistance grants and fine particulate matter have significant air quality impli- loans, to promote capacity building for ana- cations. Furthermore, because short-lived climate pollut- lytical work that is required for the design and ants can be removed from the atmosphere on a shorter implementation of effective air quality manage- time scale than long-lived pollutants such as carbon ment interventions. The portfolio review noted that dioxide, the control of SLCPs has the potential to be a technical assistance loans have not often been used in great tool against climate change. These synergies create projects addressing air quality. However, given the consid- an important opportunity, which the Bank should take erable technical and institutional gaps in air quality man- advantage of, to design projects that address air quality agement capacity in client countries, they would seem to management, health, and climate change priorities at the benefit from technical assistance. Furthermore, techni- same time. cal assistance loans can be used to conduct some of the upfront monitoring and analytical work that is needed to Development policy lending (DPL) should be develop air pollution control measures. used to support clients in addressing air pollu- tion control effectively at a sectoral or national The Bank should also promote the use of effec- level. The review of the case studies shows tive and efficient instruments for reducing air that DPLs, based on solid analytical underpin- pollution, notably economic and command- nings, can be an important instrument for help- and-control instruments. The World Bank has often ing countries to achieve air pollution reduction emphasized the use of planning instruments, such as envi- objectives. Unlike traditional investment loans, DPLs ronmental impact assessment, to address air pollution. are able to foster policy and institutional reforms. Because However, this report reviewed air quality success stories air quality is a spatial issue, oftentimes what is needed— from cities in different regions and found that they often- in addition to specific technical interventions—are the times used a variety of instruments—for example, Mexico proper policies and institutional framework. DPLs pro- City removed regressive and inefficient subsidies to fossil vide a vehicle for encouraging governments to take steps fuels and adopted a carbon tax, China instituted pollutant toward better environmental governance. In addition, discharge fees, and Bangkok instituted vehicle emission DPLs are better suited to a cross-sectoral approach than standards. These examples have shown that command SILs. Furthermore, DPLs can support the development of and control (such as fuel quality regulations and emissions a broad menu of instruments for targeting air pollution, standards) and economic instruments (such as pollution ranging from technical interventions to economic instru- charges and the phasing out of regressive fuel subsidies) ments, and they can support institutional capacity build- are effective for reducing air pollution. Enhancing the World Bank's Approach to Air Quality Management xix xx Clean Air and Healthy Lungs CHAPTER ONE INTRODUCTION BACKGROUND In many developing countries, environmental pollution continues to pose serious risks to human health and productivity, making it difficult to achieve economic growth and break out of poverty. Furthermore, the disease and death from environmental degra- dation results in a significant economic burden that is often disproportionately carried by poor people. Effectively addressing environmental health challenges therefore can be a crucial part of a country’s efforts to reduce poverty and increase shared prosperity. This report specifically deals with air pollution, which according to the World Health Organization (WHO) was the single largest environmental health risk globally in 2012 (WHO 2014a).1 Air pollution from outdoor and household sources jointly accounted for more than 7 million deaths (3.7 million from ambient air pollution and 4.3 million from household air pollution).2 Developing countries are the most affected by the del- eterious health effects of air pollution, and they bear a significant economic burden associated with loss of life, illness, and productivity losses. With economic develop- ment, air pollution represents a growing problem, particularly in developing countries, with ambient air pollution mostly affecting cities and urban areas there. Nonethe- less, countries continue to face problems with even the most basic aspects of ensuring adequate air quality management. For decades the World Bank has been supporting developing countries in address- ing issues related to pollution management and environmental health. The current knowledge on the enormous health burden associated with air pollution underscores the need for a comprehensive approach by the Bank to help countries reduce ambi- ent air pollution. This report examines the Bank’s approach to and experience with supporting countries through projects that aim to address ambient air pollution and 1 WHO 2014a. News Release dated 25 March 2014. Available at: http://www.who.int/mediacentre/news/releases/ 2014/air-pollution/en/. 2 According to the WHO, many people are exposed to both indoor and outdoor air pollution. Due to this overlap, mortality attributed to ambient and household sources cannot simply be added together; hence the total estimate of around 7 million deaths. Enhancing the World Bank's Approach to Air Quality Management 1 project activities that could be potentially relevant for » Based on case studies of World Bank projects that reducing air pollution and hence its adverse health out- have incorporated air pollution reduction objectives, comes. It provides recommendations by which the Bank to draw lessons and highlight good practices that could bolster its support to its clients, and, more impor- would be useful for informing further development tant, increase the impact of that support by reducing air of an air quality management agenda in the Bank. pollution concentrations and health impacts in the cities » To recommend ways, going forward, of enhancing and urban areas of developing countries. the ability of the World Bank project portfolio to reduce air pollution in order to reduce the signifi- The following sections present the objectives and key cant associated health burden. aspects of the institutional context for this report, followed by an examination of some of the major drivers of dete- INSTITUTIONAL CONTEXT riorating ambient air quality in developing countries, air pollution sources and impacts, and the status of air qual- FOR THIS REPORT ity management in developing countries. Chapter 2 pre- This section provides a non-exhaustive discussion of sents the results of a desk-based portfolio review of World salient aspects of the internal institutional context for sup- Bank projects that are relevant to the reduction of air pol- porting clients in reducing air pollution and improving lution. This is followed in chapter 3 by an examination associated health outcomes through World Bank–funded of case studies of World Bank projects whose objectives projects. include addressing ambient air pollution, highlighting good practices and lessons for future work in support- Over the years, the World Bank has supported various ing Bank clients. Chapter 4 presents possible approaches countries, in different regions, with projects that address for enhancing future Bank support in helping clients to pollution management and environmental health topics. improve air quality and reduce the associated adverse Most projects that have addressed air quality have done so health outcomes. Chapter 5 presents overall conclusions as a result of a primary focus on investments in a specific and recommendations. sector or thematic area, for example, energy, transport, environment, urban development, or others. While this sector-based approach may affect the contribution of a spe- OBJECTIVES OF THIS REPORT cific sector to air pollution in a country or city, its limitations become evident when dealing with a problem, such as air This report aims to examine the World Bank’s experience pollution, to which there are multiple contributing factors in supporting countries in improving ambient air quality or sources that each require different types of intervention. and environmental health outcomes and to extract les- sons learned and recommendations for enhancing future Under the sector-driven approach, the primary objec- World Bank support to clients. Specifically, the report has tive for many projects has not necessarily been reducing the following objectives: air pollution or improving environmental health. Rather, » To review the problem of ambient air pollution in these benefits are expected as an add-on result of the sec- the developing world, including its health impacts tor development activities that are the primary focus of and the current state of air quality management. the projects. As a result, quantifying or measuring those » To understand, based on a desk review of a selec- benefits is discretionary. A 2002 review of Bank-funded tion of the World Bank’s portfolio of projects with projects in transportation, urban, and water and sani- pollution management and environmental health tation sectors noted that “although health benefits are (PMEH) themes, the types of project activities the expected from the World Bank . . . projects . . . no health World Bank has supported that address air pollu- indicators are included for monitoring.”3 tion and environmental health in client countries. » To assess to what extent those projects have result- ed in improved air quality and health outcomes as- 3 2002 World Bank Consultant report. Evaluation of Environmental Health sociated with air pollution. Indicators in World Bank Projects. 2 Clean Air and Healthy Lungs The World Bank classifies projects according to a system models. The guidance however, stops short of assessing of theme and sector codes, which are assigned by the Task health impacts. Team Leader during project preparation. The Pollution Management and Environmental Health theme code is The World Bank Group Environmental Health and relevant to air pollution control. However, some projects Safety (EHS) Guidelines, completed in 2007, provide that have significant air pollution components or activities guidance applicable to facilities or projects that gener- might not be coded to PMEH, making it difficult to accu- ate emissions to air at any stage of the project life cycle.6 rately capture all projects that address air pollution. Fur- They include information about common techniques for thermore, the bundling of the pollution management and emissions management that may be applied to a range environmental health thematic areas together in a single of industrial sectors. The guidelines provide an approach code makes it difficult to identify which projects may be to the management of significant sources of emissions, addressing either thematic area or both. including specific guidance for assessment and monitor- ing of air quality impacts, and additional information on The current approach raises the question of whether most approaches to emissions management in projects located Bank interventions are appropriately positioned to help cli- in areas of poor air quality, where it may be necessary to ent countries make a significant impact on reducing air establish project-specific emissions standards. The EHS pollution and achieving health improvements in the most Guidelines are an important starting point for assessing cost-effective manner. These observations point to the need air pollution reductions associated with a project. Addi- for a more strategic and comprehensive approach in Bank tional steps would, however, be required in order to assess support to clients for reducing air pollution and improving health impacts associated with the project. environmental health outcomes through projects. More recently, the World Bank has begun to establish a The Bank has undertaken institutional initiatives that new program on Pollution Management and Environ- have implications for enhancing its support to clients in mental Health, the objectives of which are to help clients addressing the problem of air pollution. To ensure that to reduce pollution levels and associated health risks and sector and theme codes are accurately applied in lending thereby improve environmental health outcomes; to gener- operations, the Operations, Policy and Country Services ate new knowledge and improve our understanding of pol- (OPCS) Sector/Theme Coding Team reviews all project lution and its health impacts in both urban and rural areas; appraisal documents (PADs).4 Starting in 2009, the World and to promote increased awareness and understanding Bank launched Core Sector Indicators in order to bet- of pollution and its health impacts among policymak- ter measure the impact of its work at the project level.5 ers, planners, and other relevant stakeholders in low- and Notably, in July 2012, several Core Sector Indicators middle-income countries. One of the four components were launched for projects addressing pollution manage- of the program focuses on helping developing countries ment and environmental health themes, including an air and cities improve air quality management. This report is quality–related indicator specified as Particulate matter reduc- being prepared within the framework of this new program tion achieved under the project. The guidance on this indicator on Pollution Management and Environmental Health and notes the requirement for an assessment of the number is intended to serve as a preliminary input to inform the of people with exposure to particulate matter (PM10) in development of the program, specifically with respect to the area of the project: an assessment that would require enhancing the Bank’s support to countries in their efforts mapping areas where PM10 concentrations have been to control ambient air pollution. At the outset, the Pol- reduced by the project, the number of people living in the lution Management and Environmental Health program area, and possible use of air quality management (AQM) was designed to focus on urban environmental problems. Hence, although this report introduces the subject of 4 Further guidance available at http://go.worldbank.org/VGJZN8FJ50 and http://go.worldbank.org/2VBGBMXSK0. 6 In February 2013, the World Bank Group began a three-year process to review 5 Available at: http://go.worldbank.org/979GRAFWO0. the EHS Guidelines. The Guidelines are available at www.ifc.org/ehsguidelines. Enhancing the World Bank's Approach to Air Quality Management 3 household air pollution in sections—particularly in more been made in some countries to control it, the number of broadly understanding the magnitude of the problem highly polluting two- and three-wheelers, mostly powered associated with air pollution—its primary focus in review- by two-stroke engines, has rapidly grown in many cities, ing World Bank projects is ambient air pollution, which is particularly in Asia (PCFV and CAI-Asia 2011). Simi- more commonly associated with cities and urban areas. larly, as shown in the following section, diesel oil demand The significant problem that household air pollution rep- has skyrocketed over the years in developing countries. resents in many developing countries is acknowledged and Despite technological and diesel fuel quality improve- merits additional studies. ments, diesel vehicles are a major source of airborne par- ticulate matter pollution. India’s transport sector has seen DRIVERS OF AIR POLLUTION a surge in demand for diesel models of passenger vehicles due to the lower price of diesel and widening price gap IN DEVELOPING COUNTRIES with petrol (PPAC 2014, p. 19). A number of factors drive the deterioration of ambient air quality in cities in developing countries. Some of these drivers have a reinforcing effect on others, compounding INCREASED ENERGY USE the problem of air pollution. The ongoing rapid growth in motorization in develop- ing countries has already led to a tremendous increase URBANIZATION in energy demand. Globally, the percentage share of The high speed of urbanization is one of the key drivers oil demand by the road transport sector is projected to of air pollution. Today, the majority of the world’s popu- increase from 57 percent in 2010 to 60 percent in 2035. lation lives in urban areas, drawn by the opportunities and Much of this growth is occurring in the developing world. services that cities offer. Unfortunately, cities also intro- In the Asia Pacific, gasoline demand is projected to rise by duce new threats to health and magnify some existing more than 3 million barrels per day (mb/d) between 2012 ones (WHO 2010). By the middle of the 21st century, the and 2035 (OPEC 2013). In India, diesel consumption has global urban population is expected to nearly double from skyrocketed from 3.84 million tons in 1970–71 to 69.08 approximately 3.4 billion in 2009 to 6.4 billion in 2050. million tons in 2012–13 (GOI 2013). This represents about Much of that urban growth will take place in small and 45 percent of India’s total consumption of petroleum mid-size cities in developing countries (UN-Habitat 2009). products (EAI 2013). Significant increases are projected Notably, Africa’s population is estimated to increase by for Africa. After China, gasoline demand is projected to about 60 percent between 2010 and 2050, with the urban increase the most in Africa, resulting in around 2 mb/d of population tripling to 1.23 billion, making Africa the most incremental demand by 2035 (OPEC 2013). And low- and rapidly urbanizing region of the world (followed by Asia) middle-income countries as a whole have seen an overall and home to 20 percent of the world’s urban population. increase of 46 percent in road sector diesel fuel consump- tion per capita between 2000 and 2011. In absolute figures, this amounts to an increase of 69 percent. In Bangladesh, GROWING VEHICLE USE China, and Tanzania, the per capita increase has been 100, Rapid urbanization coupled with growing economic pros- 189, and 107 percent, respectively (World Bank 2014a). perity and lack of public transport infrastructure has led to higher motorization in developing countries, offering improved access to goods, services, and opportunities INADEQUATE MANAGEMENT and providing flexibility and convenience. But motoriza- OF WASTES tion also brings more pollution and more energy use. In Waste production is growing steadily, with stronger trends recent decades, particulate emissions of lead in airborne in developing countries driven by high urbanization rates emissions from vehicles in developing countries have been and economic development (Le Courtois 2012). Global substantially reduced, with almost all countries having production of municipal solid waste has almost doubled successfully taken measures to remove lead from gasoline in the past decade and is projected to reach 2.2 billion (UNEP 2014). At the same time, although efforts have tons a year in 2025. In developing countries, between 4 Clean Air and Healthy Lungs one-third and one-half of municipal solid waste is not » Primary pollutants that are emitted into the atmo- collected (Hoornweg and Bhada-Tata 2012). Low collec- sphere by sources such as fossil fuel combustion tion rates have led to the common practice of disposing of from power plants, vehicle engines, and industrial wastes in open dumps or, particularly in urban and peri- production; by combustion of biomass for agri- urban areas, by open air burning that releases pollutants cultural or land-clearing purposes, and by natural such as black carbon and persistent organic compounds. processes such as windblown dust (SEI 2003). In rural areas, notably in Asia and Africa, burning of crop » Secondary pollutants that are formed within the residues in fields is a recognized source of air pollution atmosphere when primary pollutants react with and is expected to become more widespread with increas- sunlight, oxygen, water, and other chemicals pres- ing cultivation and as other solid fuels become available ent in the air. and affordable (HEI 2010). Another driver of pollution in rural and peri-urban areas is the growth of local indus- Air pollution is addressed by various international and tries that employ polluting processes that use outdated, national air quality guidance documents. The WHO Air polluted machinery and have facilities with inadequate Quality Guidelines provide guidance on thresholds and emission controls. limits for key air pollutants that pose health risks (WHO 2006). The United States Environmental Protection BURNING OF SOLID FUELS Agency considers “criteria” air pollutants as harmful to FOR DOMESTIC PURPOSES public health and the environment. In addition, some In developing countries, about 2.9 billion people live in pollutants that drive climate change—such as short-lived homes where polluting solid fuels such as wood, coal, or climate pollutants (SLCPs), notably black carbon—have dung are the primary fuel burned for cooking (Legros direct health effects; others have indirect health effects et al. 2009). When air pollution generated indoors exits through climate change impacts. This report focuses pri- through windows, chimneys, or gaps in walls and roofs, marily on key air pollutants that are considered harmful it also contributes to ambient air pollution (Smith et al. to public health, such as those addressed by the WHO 1994). Almost three-quarters of the people who rely on guidelines (see box 1.1). solid fuels live in Asia, with India and China accounting for 27 and 25 percent, respectively, of all those using solid fuels for cooking. Much of the health-relevant air pollu- AIR POLLUTION, DISEASE, AND DEATH tion exposure from cooking fuel occurs in the near-house- Health Impacts of Air Pollution—Magnitude hold environment, not just indoors (Smith et al. 2014). of the Problem Furthermore, solid cooking fuels are sufficiently polluting Particulate matter (PM) is the most significant and widespread to appreciably affect the ambient air pol- air pollutant associated with death and disease. lution level and thus cause health impacts far from the According to WHO, short-term exposure to PM is a cause source. Only a limited number of developing countries of respiratory and cardiovascular hospital admissions, have targets in place for meeting the energy needs of poor emergency department visits, and primary care visits; of people, and many are far behind in expanding access to days of restricted activities; and of acute symptoms (cough- modern energy (Legros et al. 2009). ing, phlegm production, wheezing, respiratory infections). Effects attributable to long-term exposure include mortal- ity due to cardiovascular and respiratory disease; chronic SOURCES AND IMPACTS respiratory disease incidence and prevalence (asthma, OF AIR POLLUTION chronic obstructive pulmonary disease, chronic pathologi- AIR POLLUTANTS cal changes); lung cancer; chronic cardiovascular disease; Depending on how they are formed, air pollutants can be and intrauterine growth restriction (for example, low birth classified into two broad categories, both of which may weight) (WHO 2006). According to the results of the 2010 contain chemical compounds in solid, liquid, or vapor Global Burden of Disease (GBD) study, diseases affected phases: by ambient air pollution were the top five causes of the Enhancing the World Bank's Approach to Air Quality Management 5 BOX 1.1. AIR POLLUTANTS, SOURCES, AND HEALTH EFFECTS Particulate matter (PM) refers to small solid and liquid Carbon monoxide (CO) is a gas produced by the incom- particles of various physical dimensions and chemical prop- plete combustion of carbon-based fuels and by some indus- erties. PM may be of natural origin, such as fine soil particles trial and natural processes. The most important outdoor and volcanic ash, or can originate from the combustion of source is emissions from petrol-powered vehicles. It is always fossil fuels, especially coal and diesel fuel, in vehicular and present in the ambient air of cities, often reaching maximum industrial processes; from domestic heating and cooking; concentrations near major highways during peak traffic con- from the burning of waste crop residues and land clearing; ditions. Indoors, it often reaches maximum concentrations and from construction and fire control activities (SEI 2003). near unvented combustion appliances, particularly in poorly Other fine particulates result from slow atmospheric reac- ventilated areas. CO can affect health by reducing oxygen tions among gases such as photochemical smog reactions or delivery to the body’s organs such as the heart and brain and the oxidation of oxides of sulfur or nitrogen. PM is typically tissues; at extremely high levels, it can cause death. characterized by particle size, as either PM10 (aerodynamic Lead (Pb) and other heavy metals, including arsenic, diameter ≤10 microns) or as PM2.5 (aerodynamic diameter cadmium, manganese, mercury, and nickel, that are regularly ≤2.5 microns). The smaller the particles, the deeper they are found in the air can present risks to human health. Pb is the able to penetrate into the lungs, disrupting the exchange of most important heavy metal pollutant for health globally, given oxygen into the blood and causing inflammation and pre- its widespread distribution at concentrations that may damage mature death. Black carbon, a very fine constituent of PM, health. Pb compounds are distributed in the atmosphere, due has gained increased attention in recent times for its health to the combustion of fuels containing alkyl lead additives. As impacts and role in global warming (WHO 2012). PM was countries have eliminated the use of leaded-gasoline in vehi- recently classified as carcinogenic to humans (IARC 2013). cles, Pb is less of an issue from transport and consequently Nitrogen dioxide (NO2) is the most significant form of in urban air quality. Inhalation of Pb is a significant route of nitrogen oxide from a health perspective. The main source of exposure in adults. In children, ingestion of Pb in dust and NO2 is combustion of fuels in motor vehicles and stationary products such as leaded paint is a more important route of sources such as industrial facilities and power plants. NO2 is exposure. Prolonged exposure to Pb is linked to neurological the main source of nitrate aerosols, which form an important and developmental damage in children (World Bank 2008). fraction of PM2.5 and, in the presence of ultraviolet light, of Toxic air pollutants are hazardous air pollutants that are ozone. Long-term exposure to NO2 is associated with reduced known or suspected to cause cancer or other serious health lung function and with bronchitis in asthmatic children. problems, such as damage to the immune system, birth Sulfur dioxide (SO2) is a colorless gas with a sharp odor, defects, and neurological, reproductive, developmental, respi- produced by the combustion of fossil fuels and the industrial ratory, and other health problems. Examples include ben- refining of sulfur-containing ores (SEI 2003; WHO 2006). zene, which is found in gasoline and in chemical and plastic When oxidized, sulfur dioxide becomes sulfate, which is a manufacturing plants; polycyclic aromatic hydrocarbons; particulate. SO2 can affect the respiratory system, as well as polychlorinated biphenyls; volatile organic compounds, diox- irritating the eyes. ins, and furans, which are products of incomplete combustion of carbon-based fuels (SEI 2003). Most air toxics originate Ozone (O3) at ground level is formed by the reaction with from mobile sources (such as cars, trucks, buses) and station- sunlight of mixtures of primary pollutants such as NO2 from ary sources (such as factories, refineries, power plants), as well vehicle and industry emissions and volatile organic com- as indoor sources (such as some building materials and clean- pounds emitted by vehicles, solvents, and industry (WHO ing solvents). Others are released from natural sources such 2006). Excessive ozone in the air can have a pronounced as volcanic eruptions and forest fires. Air toxics may be intro- effect on human health and can cause breathing problems, duced into the body by inhalation and may accumulate over trigger asthma, reduce lung function, and cause lung disease. time in fatty tissue and breast milk. global burden of disease in 2010. More recently, WHO 2014a). The majority of the deaths are associated with estimated that of the 52.8 million deaths globally in 2012, non-communicable diseases (ischaemic heart disease, air pollution from household and ambient sources jointly stroke, lung cancer) that were previously considered to be accounted for 7 million—or one in eight deaths (WHO “diseases of affluence” (see figure 1.1). 6 Clean Air and Healthy Lungs FIGURE 1.1. GLOBAL AIR POLLUTION DEATHS IN 2012 BY DISEASE Household air pollution Ambient air pollution 34% Stroke 40% Lung cancer 6% 6% 26% Ischaemic heart disease 40% Chronic obstructive pulmonary disease 22% 11% Acute lower respiratory disease in children 12% 3% 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% Source: WHO 2014a. The health impacts of air pollution are becom- FIGURE 1.2. COMPARISON OF AIR ing more significant with time. From 1990 to 2010, POLLUTION (PM) AND there was an 8 percent increase in global population- OTHER ENVIRONMENTAL weighted PM2.5, a measure of human exposure (Brauer HEALTH RISKS IN et al. 2012).7 In 2010, 3.2 million deaths or 3.1 percent of the global total of disability adjusted life years (DALYs)— DEVELOPING COUNTRIES’ that is, 76 million DALYs—were attributable to particu- PREMATURE DEATHS Tobacco smoking Unimproved sanitation late matter. In 2012, ambient or outdoor air pollution Household air pollution from solid fuels Ambient ozone pollution caused 3.7 million premature deaths globally, due mainly Ambient particulate matter pollution Occupational risks Unimproved water source Residential radon to exposure to particulate matter (WHO 2014a). This Lead exposure result represents a large increase in burden compared Deaths per 100,000 people 80 with the previous estimate of 1.3 million deaths in 2008, 70 60 attributed to the increased availability of evidence on 50 exposure-health relationships, the more prominent role of 40 non-communicable diseases, and methodological factors 30 20 (WHO 2014a).8 People living in low- and middle-income 10 countries were disproportionately affected by deaths from 0 air pollution: some 88 percent of those premature deaths Source: Constructed by authors using GBD compare tool, based on GBD occurred in low- and middle-income countries that rep- 2010 data. http://viz.healthmetricsandevaluation.org/gbd-compare/. resent 82 percent of the world population. The greatest numbers were in the WHO Western Pacific and South- east Asia regions. Men aged 25 years or older were most Air Pollution Deaths Compared affected by deaths attributable to ambient air pollution, with Deaths from Other Risk Factors accounting for 53 percent, followed by women aged Results based on the GBD 2010 data indicate that house- 25 years or older and children under 5 years old. hold air pollution and outdoor air pollution are the world’s second- and third-largest environmental health risks, respectively (after tobacco smoking) in terms of 7 Clarified during a 2014 personal communication with Michael Brauer, profes- sor at the University of British Columbia. total number of lost healthy life years and number of pre- 8 Reasons include additional evidence that has become more available on the mature deaths caused (Smith et al. 2014). The results in relationship between exposure and health outcomes and the use of integrated developing countries are similar (see figure 1.2). This result exposure-response functions; an increase in non-communicable diseases; the surpasses other well-known environmental health risks, inclusion of the rural population, whereas the previous estimate only covered the urban population; and the use of a lower counterfactual—the baseline namely unimproved water source and unimproved sanita- exposure against which the effect of air pollution is measured (WHO 2014a). tion, which have typically resulted in greater numbers of Enhancing the World Bank's Approach to Air Quality Management 7 FIGURE 1.3. RELATIONSHIP BETWEEN country-level analyses conducted between 2004 and 2014 POLLUTION, URBAN LABOR by the World Bank estimated the economic burden associ- ated with poor environmental health. Cost of environmen- PRODUCTIVITY, AND URBAN tal degradation studies for various countries in different LIVABILITY regions have shown that environment-related risk factors 0.8 (such as inadequate water supply, sanitation, and hygiene, as well as outdoor and indoor air pollution) typically account Urban labor productivity 0.4 for economic losses equivalent to between 2 and 4 percent of GDP (PEP 2008) and can reach up to 9 percent when the 0 cost of environment-related malnutrition and its long-term adverse impacts on cognition and learning are taken into –0.4 account (World Bank 2008). Although people at various levels of economic status may experience high air pollution concentrations, the economic burden of poor environmen- –0.8 Numbeo EIU EIU Liveability UN habitat- tal health is most heavily borne by the poor, who typically pollution index competitiveness index-pollution environment index- index do not have sufficient resources to deal with adverse health environment impacts and are most exposed to environmental risks (PEP Source: Compiled by World Bank Competitive Cities Knowledge Base Team 2008; Sánchez-Triana and Awe 2007; World Bank 2012c). based on various sources. Estimates of the economic cost of health damage associ- premature deaths in developing countries in earlier years. ated with ambient air pollution are typically based on the Ambient and household air pollution jointly represent the exposure of the population to particulate matter and lie world’s greatest environmental health risk. in the cost range that is equivalent to 0.1–3.2 percent of GDP in selected developing countries across six regions Economic Impacts of Air Pollution (see figure 1.4). Cost estimates include the costs from at the City Level premature mortality, morbidity, and cost of illness. The The increased economic activity that accompanies urban- economic costs, expressed in monetary terms, provide an ization often brings increased air pollution. In the long indication of the amount of resources that a country is term, however, air pollution can negatively affect the allocating inefficiently—for example, through medical growth of gross domestic product (GDP) and employ- care expenses and the time lost due to illness—which could ment (the number of available jobs). Air pollution’s instead be allocated to productive uses across the economy impacts on environmental quality could negatively affect if air pollution were to be addressed. In the majority of a city’s livability—defined as the qualities of that city that the cases, premature death represents the greater share of contribute to the quality of life experienced by the city’s the health damage cost associated with air pollution. residents and others—and its productivity. Figure 1.3 shows, based on average scores received by various cities AIR POLLUTION AND AIR on four pollution and livability indices, that high pollu- QUALITY MANAGEMENT IN tion, including air pollution, is negatively correlated with urban labor productivity, while a high score on environ- DEVELOPING COUNTRIES: ment is positively correlated with labor productivity. STATUS, CHALLENGES, AND PROGRESS Economic Burden of Ambient Air AIR QUALITY STATUS AND CHALLENGES Pollution at a Country Level IN VARIOUS REGIONS Poor air quality, resulting in premature death, illness, lost Pollutant levels in many cities in developing earnings, and increased medical costs, can constrain pro- countries far exceed WHO health-based air qual- ductivity, which is essential for economic growth. Several ity guideline values. Particulate matter concentrations 8 Clean Air and Healthy Lungs FIGURE 1.4. ECONOMIC BURDEN OF HEALTH DAMAGE ASSOCIATED WITH AMBIENT AIR POLLUTION AS PERCENTAGE OF GDP India SAR Bangladesh Pakistan China EAP Indonesia Macedonia Kosovo ECA Kazakhstan Tajikistan Central African Republic Senegal AFR Nigeria Ghana Madagascar Peru Colombia LCR Nicaragua Honduras Guatemala Egypt MNA Jordan Lebanon 0 0.5 1 1.5 2 2.5 3 3.5 Percentage of GDP Source: Authors’ illustration based on various World Bank reports. in many cities in low- and middle-income countries are conducted as part of this work indicates that some com- more than twice the levels in high-income countries and monly faced challenges to adequate air quality management several times higher than WHO Air Quality guideline across regions include inadequate air quality monitoring values (see figure 1.5). Particulate matter levels in Asia, infrastructure; lack of national air quality standards, par- Africa, and Latin America are substantially higher than in ticularly for pollutants that are important for health, notably Europe and North America, with several Asian and Afri- PM2.5; variable data quality; the absence of standardized can cities showing the highest levels. monitoring techniques and data collection protocols; the limited role of data in influencing revisions of air quality Despite the significant health and economic bur- standards; and the lack of technical capacity for air quality dens attributable to ambient air pollution and its management (see appendix A for a more detailed review). consequences on climate, many cities in develop- ing countries face significant challenges in their The number and rate of deaths from ambient air efforts to improve air quality management. The pollution point to the urgency for proactive air 2014 version of the WHO database of ambient air pollu- pollution control in rapidly urbanizing regions tion monitoring from 1,600 cities reveals the dearth of air such as Africa. The East Asia and the Pacific (EAP) quality monitoring stations, notably in Sub-Saharan Africa.9 and South Asia (SAR) regions display the fastest rates of A review of air quality management status and challenges increase in the number of deaths from ambient PM pol- lution between 1990 and 2010 (see figure 1.6). However, 9 http://www.who.int/phe/health_topics/outdoorair/databases/AAP_database these observations should be interpreted with caution. _results_2014.pdf ?ua=1 These highly urbanized regions are home to 38 of the Enhancing the World Bank's Approach to Air Quality Management 9 FIGURE 1.5. PM10 DATA FOR CITIES IN SELECTED REGIONS AND COUNTRIES (YEARLY AVERAGE IN μg/m3) 320 300 WHO interim target I 280 260 WHO interim target II 240 EU limit value 220 200 WHO interim target III 180 WHO guideline value 160 140 120 100 80 60 40 20 0 Be u g a La r ab s e D i ka tD C i ta iro es Bo a tá W a M aw Lo w N don rk ris h ch a G go cr ho ijin on oh m co el go Yo ak ha iti Pa a s ra Ac Li D nz or D n ar os C D ew Ka La el yp Eg Source: Compiled by authors based on WHO Ambient air pollution in cities database 2014. FIGURE 1.6. TRENDS IN DEATHS Saharan Africa: by 2033, 20 African cities will be among FROM AMBIENT AIR the largest 100 cities in the world, with 50 cities having reached 10 million or more (UN Habitat 2010). By 2025, POLLUTION (PM), BY Africa will be home to the world’s fastest-growing meg- REGION, 1990–2010 acity, Lagos, with urban population growth rates consid- EAP ECA LCR 1,600 MNA SAR AFR erably faster than megacities in Asia. Coupled with this, air quality monitoring in regions such as Africa is grossly 1,400 inadequate, which does not favor accurate capture of exposure to, and health effects of, ambient air pollution. 1,200 This rapid urbanization suggests that exposure to air pol- Deaths (in 1000s) 1,000 lution and associated premature deaths will increase rap- idly and underscores the urgency of taking advantage of a 800 transient window of opportunity to implement air quality 600 management in a proactive fashion. 400 SOME EXAMPLES OF PROGRESS 200 IN AIR QUALITY MANAGEMENT IN DEVELOPING COUNTRIES 1985 1990 1995 2000 2005 2010 2015 While many developing countries face challenges with Year Source: Compiled by Authors using GBD Compare Tool, based air pollution, some have successfully reduced ambient air on data for years shown. http://viz.healthmetricsandevaluation. pollution concentrations through adoption of technical, org/gbd-compare/. investment, and policy interventions. 50 cities with the highest levels of ambient PM concen- trations globally, based on the latest WHO database (see Mexico City appendix A for listing of the 50 cities). In comparison, Mexico City’s population growth, from fewer than 3 million urbanization is rapidly increasing in regions such as Sub- in 1950 to about 20 million in 2011, created a traffic-choked 10 Clean Air and Healthy Lungs urban sprawl. The increase in emissions, which was exac- covered in thick smog (Fuller 2007). In 1996, average daily erbated by a geography that kept pollutants trapped in the PM10 concentrations in Bangkok were around 100 μg/m3 area, led to the United Nations Environment Programme (compared with the WHO guideline of 50 μg/m3), and declaring, in 1992, that Mexico City had perhaps the annual average visibility at Bangkok’s Don Muang airport world’s worst air pollution problem (Yip and Madl 2001). plunged from almost 16 km in 1964 to less than 8 km in By 2010, however, Mexico had cut ambient concentrations 1996 (World Bank 2002). Within 10 years, however, Bang- of most pollutants in half. Furthermore, ambient concen- kok was being hailed as an air quality management suc- trations of particulate matter dropped by 70 percent, and cess story: average levels of particulate matter were cut airborne lead concentrations dropped by more than 90 per- to 43 μg/m3, even as the number of motor vehicles regis- cent, even as the number of cars in the city doubled, reach- tered in the city rose by 40 percent by 2007 (Fuller 2007; ing more than 4.2 million (O’Connor 2010; Sánchez and Wangwongwatana 2013). Lacy, forthcoming). Much of the improvement was due to interventions that To achieve these results, Mexico City focused on cutting reduced emissions from the city’s vehicle fleet. These emissions from vehicles by mandating the installation of included completely phasing out lead in gasoline; install- catalytic converters in automobiles, removing lead from ing catalytic converters in cars; reformulating gasoline gasoline, reducing the sulfur content of diesel, and strength- to reduce benzene and aromatic compounds; reducing ening a vehicle inspection program that was designed to the sulfur content of diesel from 1 to 0.005 percent by reduce emissions from old, obsolete, and/or poorly main- weight; adopting European Union emissions standards for tained vehicles. The city has also embraced public trans- vehicles; shifting 80 percent of the Thai motorcycle pro- portation: it added hybrid-electric buses to its fleet and is in duction from 2- to 4-stroke engines; establishing routine the process of adding a new line to its Metro system (Par- inspection and maintenance programs for all registered rish et al. 2011; Sánchez and Lacy, forthcoming). vehicles; introducing natural gas for fuel in the transport sector; reducing the number of rickshaws on the streets, Recently, the government has also taken steps to remove since they were the biggest polluters in Thailand (HEI regressive and inefficient fossil fuel subsidies that in 2011 2010; Simachaya 2012; World Bank 2013b); and impos- were equivalent in value to 1.15 percent of GDP. In Octo- ing a ban on import of all cars that did not meet Thai- ber 2013, the Mexican Congress approved a carbon tax land’s new standards. Air quality researchers were heavily that is expected to collect more than Mex$14,500 million involved in shaping these new regulations and AQM prac- or 5.2 percent of total subsidies in 2011 (Sánchez and Lacy, tices, and this analytical mindset has been cited as one of forthcoming). Under the new law, production and import of the reasons for these achievements. fossil fuels will be taxed based on the value of international carbon markets between October 2012 and June 2013; In order to reduce industrial pollution, the city enacted different tax rates apply per unit (liter or ton) and by fuel stringent emissions standards and revoked the operating (gasoline, diesel, coal), based on carbon content. Given that licenses of factories that violated them; established a cap Mexico’s transport sector emits one-third of the greenhouse on the sulfur content of certain fuel oils, such that it cannot gas emissions (CO2) generated from fossil fuels in the coun- exceed 2 percent by weight; and imposed self-monitoring try (World Bank 2013a), it is expected that the reduction on requirements on all industries (HEI 2010; World Bank gasoline subsidies will lead to reduced air pollution, which 2013b). To reduce emissions from open burning of agri- will consequently reduce the health damages associated cultural wastes, a cause of haze pollution, the Cabinet with air pollution that currently cost 1.5 percent of GDP. approved a National Master Plan for Open Burning Con- trol in 2003 and began implementing activities under the Bangkok ASEAN (Association of South East Asian Nations) Agree- During the latter half of the 20th century, Bangkok went ment on Trans-boundary Haze Pollution. Haze mitigation from having clear skies (the Thailand Royal Institute measures under the plan include management of forest didn’t even coin a word for pollution until 1976) to being areas instead of burning; reuse of agriculture wastes as Enhancing the World Bank's Approach to Air Quality Management 11 FIGURE 1.7. DEVELOPMENT IN PM10 CONCENTRATIONS IN CHINESE CITIES (A) AND HUHOT CITY (B), 1997–2012 160 a. b. 140 250 231 microgram per cubic meter 120 237 206 microgram per cubic meter 200 180 100 150 80 113 146 138 109 102 60 97 102 91 100 116 84 76 40 50 70 20 0 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: Compiled based on data received from China National Monitoring Center (CNMC) 2013. bio-energy; improvements in solid waste management to Moving forward, China still faces challenges: its grow- reduce burning of solid waste; and use of awareness cam- ing economy has led to increased industrial and vehicle paigns to promote the prevention, reduction, and mitiga- emissions, and its air quality management framework is tion of air pollution from open burning. probably so extensive that it has already instituted most of the easy fixes. However, since 2012/13, and based upon the new regulation GB 3095–2012, China is now moving Chinese Cities forward by tightening standards (for example, for PM10); Since the late 1990s, China has made significant progress gradually expanding its monitoring network to include in tackling its air quality problems. From 1997 to 2012, PM2.5, CO, and O3 throughout all its cities (an expanded the average PM10 concentrations in Chinese cities went nationwide monitoring system to be in place before the down by around 45 percent, from about 145 μg/m3 to end of 2015); establishing new standards for PM2.5, CO, about 80 μg/m3 (see figure 1.7). By 2009, 83 percent of and O3; instituting strict action plans whose interventions Chinese cities met the intermediate and most stringent are set to be rolled out on fixed timelines; and elaborat- national standards for PM10 compared with 36 percent in ing new action plans in all key regions in China for the 2001 (World Bank 2012a). Furthermore, seven cities com- 2013–17 period (World Bank 2012b; World Bank and the plied with the most stringent standards in 2009, compared Development Research Center of the State Council, P.R. with zero in 2001. China 2014). These achievements were the result of an overhaul of air quality management policy in 2000. China instituted SUMMARY a discharge permit system and total load control in key Poor air quality remains a major challenge for many areas; created pollutant discharge fees that would be developing countries—a challenge that is underscored charged by the type and quantity of pollutant emissions; by the significant health and economic damage that it reinforced pollution control measures on motor vehicles; poses in those countries. And although the gap between and instituted measures to control urban dust pollution. current air quality levels and the values specified by the Furthermore, the government mandated that key cities WHO guidelines appears to be quite wide, it is impor- develop plans for meeting air quality targets on a fixed tant to remember that it took about 50 years for cities in timeline. To measure their progress, the air quality moni- the United States to reach the WHO guideline value for toring system—it measures ambient PM10, NO2, and SO2 PM10 (they went from an average value of 60 μg/m3 in the levels—was also expanded to about 613 cities (World 1960s to around 20 μg/m3 in the 2010s), and Europe had Bank 2012b). similar time frames (World Bank 2012b). Although many 12 Clean Air and Healthy Lungs air quality management lessons can be transferred from perspective. Currently, countries are addressing the prob- the United States and Europe to the developing world, lem with varying levels of success; some of the examples and these lessons could shorten the time frame for achiev- presented in this chapter indicate the promise for achiev- ing reasonable air quality levels in cities of developing ing progress by lowering air pollutant concentrations countries, it is to be expected that substantially reducing and human exposures in developing countries through particulate matter concentrations in the developing world targeted policy instruments, institutional changes, and will be a long-term process that will require a long-term investments to control air pollution. Enhancing the World Bank's Approach to Air Quality Management 13 CHAPTER TWO A REVIEW OF THE WORLD BANK PROJECT PORTFOLIO OF AIR-POLLUTION-RELEVANT PROJECTS OBJECTIVES OF PORTFOLIO REVIEW This portfolio review had the following specific objectives, based on a selection of Bank projects that were assigned the Pollution Management and Environmental Health theme code: to identify projects that include activities with the potential to reduce air pollution, to develop a set of typologies to group similar project activities together in terms of their potential to reduce air pollution, and to assess potential health impacts of the identified projects. UNDERLYING PRINCIPLES—LINKAGES BETWEEN AIR POLLUTION SOURCES, AIR POLLUTION CONCENTRATIONS, AND IMPACTS ON HUMAN HEALTH An understanding of the linkages between sources of air pollution, concentration of pollutants, and pollution impacts on human health provides a guiding context for undertaking the work described in this report. Figures 2.1 and 2.2 summarize these linkages with regard to particulate matter, which is the pollutant with the most delete- rious consequences for health, and appendix B describes the linkages in more detail. A number of terms (air pollution emissions, air pollution concentration, exposure, and health impacts) and relationships (spatial population distribution and concentration- response function) are used in describing these linkages, where terms are data points and relationships are functions that are used to link the terms in order to ultimately arrive at estimates of health impacts. A more detailed description of terms and rela- tionships is provided in appendix B. Enhancing the World Bank's Approach to Air Quality Management 15 FIGURE 2.1. RELATIONSHIP BETWEEN AIR POLLUTION SOURCES, AIR POLLUTION CONCENTRATIONS, AND HEALTH IMPACTS Air pollution Air pollution sources concentration (and exposure) Health impact Traffic Particle size (µm) effect 9.2 to 30 Visible pollution Energy 5.5 to 9.2 Lodges in nose/throat 3.3 to 5.5 Main breathing passages Industry 2.0 to 3.3 Small breathing passages 1.0 to 2.0 Bronchi 0.1 to 1.0 Air sacs Households PM2.5 Human hair Combustion particles, organic Agriculture 50-70µm compounds, metals, etc <2.5 µm (microns) in diameter Spatial distribution of PM2.5, (microns) in diameter annual average (µg/m3) PM2.5 concentrations 2008.06.01 00:00–2009.06.01 01:00 PM10 Construction Less than 10 Dust, pollen, mold, etc <10 µm (microns) in diameter 10–20 20–50 50–100 100–150 150–200 Open burning 200–300 300–400 400–450 90µm (microns) in diameter 450–500 Fine beach sand 500–550 Sewage treatment Greater than 550 (Concentration map can be superimposed with Size of PM2.5 and PM10 relative to population density map to obtain exposure) human hair and beach sand Source: Authors’ illustration and World Bank 2011. FIGURE 2.2. RELATIONSHIP BETWEEN EMISSIONS, AIR POLLUTION CONCENTRATIONS, EXPOSURE, AND HEALTH IMPACTS Air pollution Exposure Health impact Emissions [µg/m3 averaged over a [mortality, morbidity in the [kg/hr] concentration [µg/m3, ppm] period] population over the period] Wind action and Spatial population Concentration-response dispersion distribution function Source: Authors’ illustration based on literature. stages. Project documents reviewed were primarily proj- METHODOLOGY ect appraisal documents, additional project documenta- The methodology had three main elements. A detailed tion, and, where applicable, implementation completion description of these follows brief summaries of the three, reports. The review covered projects financed by IBRD/ including applicable caveats and criteria for portfolio IDA resources (investments, technical assistance loans, selection. Full descriptions of the project typologies devel- and development policy loans) as well as those supported oped under this work and the approach followed for the by concessional finance (such as the Global Environment assessment of potential health impacts of projects are pro- Facility [GEF], carbon offset, and others). vided in appendix C. Projects covered by the portfolio review consisted of the Portfolio review. The portfolio review was a desk- World Bank’s portfolio of 114 closed projects approved based, iterative, expanded process, conducted in two between fiscal years 2002 and 2012 (FY02 and FY12) with 16 Clean Air and Healthy Lungs TABLE 2.1. DISTRIBUTION OF THE REVIEWED PROJECT PORTFOLIO ACROSS WORLD BANK SECTORS/THEMES AND SECTOR BOARDS FY02–12 FY08–13 Number of Number of Projects with Projects with World Bank Sector Managing Number of Activity Relevant Number of Activity Relevant or Theme Sector Board Projects, Total for Air Pollution Projects, Total for Air Pollution Agriculture, fishing ARD 4 2 4 0 and forestry Energy and mining EMT 17 10 14 5 Environment and ENV 46 19 65 16 natural resource management Transportation TR 5 2 10 7 Urban development UD 10 4 33 4 Water, sanitation and WAT 31 9 48 5 flood protection Social development, SDV 1 0 0 0 gender and inclusion Total 114 46 174 37 Note: Number of projects with air-pollution-relevant activities adjusted to eliminate double counting of projects in FY02–12. a pollution management and environmental health the- identified: energy, industry, transport, urban, agriculture, matic content of at least 25 percent as allocated according land administration, and public administration. For each to the World Bank’s system of Sector and Theme codes sector type, relevant Subsector Types were identified and and the Bank’s portfolio of 174 closed and active projects corresponding activity types were grouped under similar approved between FY08 and FY13 with a PMEH the- Subsector Types. matic content of at least 5 percent. Of all the projects reviewed, 83 were found to be air-pollution-relevant— Assessment of health impacts of air-pollution- that is, they included activities that had the potential to relevant projects. Based on the linkages between air reduce air pollution. pollution sources, concentrations, and health impacts, an indicative assessment of the 83 air-pollution-relevant Development of project typologies. The portfo- projects was conducted. lio of 83 air-pollution-relevant projects was mapped to a typology that was developed based on the universe OVERVIEW OF THE AIR- of activities described in the OPCS Guidelines on Sectors and Themes that were deemed relevant to air POLLUTION-RELEVANT pollution.10 The typologies linked project activities to PORTFOLIO Sector and Subsector Types. Seven Sector Types were Projects mapped to the Environment Sector Board account for the largest number of air- pollution-relevant projects. As shown in table 2.1, for 10 The OPCS Guidelines on Sectors and Themes provides guidance on classifi- cation of activities that may be found in World Bank–supported projects (avail- the FY02–12 projects, 46 out of 114 projects were air- able at: http://go.worldbank.org/2VBGBMXSK0). pollution-relevant. Projects mapped to the Environment Enhancing the World Bank's Approach to Air Quality Management 17 FIGURE 2.3. DISTRIBUTION OF AIR-POLLUTION-RELEVANT PROJECTS BY SECTOR BOARD FY02 FY03 FY04 FY05 FY06 FY07 FY08 FY09 FY10 FY11 FY12 6 Number of projects 5 4 3 2 1 0 EMT ENV TR UD WAT ARD Sector board FY08 FY09 FY10 FY11 FY12 FY13 5 Number of projects 4 3 2 1 0 EMT ENV TR UD WAT ARD Sector board Note: Number of projects with air-pollution-relevant activities adjusted to eliminate double counting of projects in FY08–FY12. Sector Board (ENV) accounted for the largest number The distribution of air-pollution-relevant projects by of projects reviewed (46) and 41 percent of all the proj- Sector Board shows that for the FY02–12 portfolio, ects with air-pollution-relevant activities in that period. only EMT and ENV have projects that cover a long After that, the Energy and Mining Sector Board (EMT) enough duration during the specified period (see fig- accounted for the next largest number of projects (10) ure 2.3). For both sets of data, nonetheless, an overall with such activities, followed by the Water (WAT) Sector decreasing trend with time is observed for most Sector Board (9). Boards. For the FY08–FY13 projects, 37 out of 174 projects The majority of air-pollution-relevant proj- were air-pollution-relevant. Again, projects mapped ects do not specify air pollution reduction as a to the ENV accounted for the largest number of proj- development objective. Of the 83 relevant projects, ects reviewed (65) and 43 percent of all the projects 28 included a project development objective related to with air-pollution-relevant activities in that period. air quality, while the remaining 55 did not (see appen- In addition, 16 out of the 65 ENV-mapped proj- dix C). The Sector Types that most frequently stated air ects (25 percent) were air-pollution-relevant. Subse- quality–related project objectives were transport, energy, quently, the Transport Sector Board (TR) accounted policy, and public administration. Sector Types such as for the next largest number of projects (7) with such wastewater/sewage treatment and agriculture typically activities, followed by EMT and WAT, each with 5 did not state air quality–related project objectives. Proj- air-pollution-relevant activities, and Urban Develop- ect activities under the latter Sector Types, however, could ment (UD) with 4. A full listing of all 83 air-pollution- offer significant opportunities for reducing particulate air relevant projects is provided in appendix C, tables C.2 pollution. Manure generation, wastewater treatment in and C.3. open vessels, and the application of manure and fertilizers 18 Clean Air and Healthy Lungs often generate ammonia (NH3) emissions, which interact FIGURE 2.4. AIR-POLLUTION-RELEVANT with sulfur dioxide and nitrogen oxides to form particles PROJECTS BY PRODUCT LINE known as secondary inorganic aerosols. Practices that FY02–12 FY08–13 improve wastewater/sewage treatment and management 25 23 of animals, manure, and fertilizers help to reduce NH3 20 20 19 emissions and consequently production of secondary Number of projects inorganic aerosols. The observed results from the port- 15 folio represent possible missed opportunities for air pol- 12 lution control and may be indicative both of the Bank’s 10 sector-driven approach to air pollution reduction and its 5 4 4 limited attention to, or awareness of, the interactions just 0 1 described (Sun 2012; EEB 2013). 0 IBRD/IDA GEF Carbon offset SPF Note: Data adjusted to eliminate double counting of projects in FY08–FY12. Although IBRD/IDA resources support over 50 percent of the number of air-pollution- relevant projects, concessional finance plays an important role in supporting such projects. Figure FIGURE 2.5. AIR-POLLUTION-RELEVANT 2.4 shows that across both sets of portfolios, the IBRD/ PROJECTS BY LENDING IDA product line accounted for over 50 percent of the INSTRUMENT (IBRD/IDA number of air-pollution-relevant projects. It also shows PRODUCT LINE ONLY) that concessional finance, notably the GEF, played an FY02–12 FY08–13 16 important role in supporting such projects (37 percent, 14 15 14 or 31 out of 83 projects) while carbon finance and spe- Number of projects 12 cial financing were used to a lesser extent. While these 10 projects tend to have the achievement of global objectives 8 such as reducing greenhouse gases as their primary focus, 6 4 they are also relevant for reducing local air pollution (for 4 3 3 example, a project that supports non-motorized transport 2 0 1 0 0 1 1 0 0 as a means to reduce carbon dioxide emissions might also SIL DPL APL TAL FIL ERL reduce particulate matter pollution from vehicles). Note: Data adjusted to eliminate double counting of projects in FY08–FY12. The specific investment lending instrument is the dominant instrument for IBRD/IDA support as East Asia and the Pacific (EAP) and Latin to air-pollution-relevant projects. For IBRD/IDA America and Caribbean Region (LCR), while lending, specific investment loans (SILs) were the most fewer projects have focused on rapidly urbaniz- commonly used instruments to finance projects with air- ing regions such as Africa (AFR). The distribution pollution-relevant activities, followed by development policy of projects (see figure 2.6) with air-pollution-relevant loans. Other instruments, including the adaptable program- activities varies between regions, which might be a reflec- matic loan, technical assistance loan, financial intermediary tion of the development needs or of the level of demand loan, and emergency recovery loan were used to a lesser expressed for these types of activities from countries extent (see figure 2.5). The dominance of SILs underscores in the respective regions. The regional distribution of the sector-based approach that has been used in addressing projects is skewed in favor of East Asia and the Pacific air pollution reduction in Bank-financed projects. Region, where rapid urbanization and industrialization continue to contribute to poor ambient air quality. Out The majority of air-pollution-relevant projects of the total of 83 air-pollution-relevant projects, 27 were have focused on highly urbanized regions such in the EAP Region, followed by 18 projects in LCR, 16 Enhancing the World Bank's Approach to Air Quality Management 19 FIGURE 2.6. DISTRIBUTION OF AIR-POLLUTION-RELEVANT PROJECTS BY REGION FY02 FY03 FY04 FY05 FY06 FY07 FY08 FY09 FY10 FY11 FY12 5 4 Number of projects 3 2 1 0 AFR EAP ECA LCR MNA SAR Region FY08 FY09 FY10 FY11 FY12 FY13 3 Number of projects 2 1 0 AFR EAP ECA LCR MNA SAR Region Note: Data adjusted to eliminate double counting of projects in FY08–FY12. in the Europe and Central Asia Region (ECA), 9 in the wastewater management (13), solid waste management Middle East and North Africa Region (MNA), 6 in AFR, (13), and hazardous chemicals and waste management (8) and 4 in the South Asia Region (SAR). Despite the fact have the largest numbers of projects with air pollution– that several countries in SAR have a large number of cit- reducing activities, followed by public transport (7), ies with very high air pollution concentrations and that policy and institutional improvements (7), and energy the region is experiencing a rapid rate of increase in the efficiency (5). On-farm activities and manure manage- number of air pollution–related deaths (in chapter one, ment, renewables to replace energy production by fossil see figures 1.5 and 1.6), very few AQM-focused proj- fuel, and urban transport fuels also contributed to the ects have been developed in SAR. Furthermore, the low air pollution–reduction portfolio. The activity types that number of air pollution–focused projects in AFR is likely stand out more prominently are wastewater and sewage a reflection of the limited information and knowledge collection and treatment; control and management of about air pollution concentrations and their impacts in municipal dumps and establishment of landfills; man- cities in AFR. agement, storage, and disposal of hazardous chemicals; policy and institutional improvements; and livestock pro- The Urban Sector Type accounts for the major- duction and manure management. Considerable declines ity of air-pollution-relevant projects. As shown by were observed for these activity types: livestock produc- table 2.2, it accounted for 26 such projects, followed by tion and manure management decreased in number from Energy Type (16), Transport Type (15), Industry Type 5 (FY02–12) to 1 (FY08–13), and management, stor- (13), Public Administration Type (7), and Agriculture age and disposal of hazardous chemicals dropped from Type (6). From the perspective of the Subsector Types, 7 (FY02–12) to 1 (FY08–13). 20 Clean Air and Healthy Lungs TABLE 2.2. DISTRIBUTION OF AIR-POLLUTION-RELEVANT PROJECTS BY TYPOLOGY No. of No. of Projects Projects Sector Type Subsector Type Activity Type FY02–12 FY08–13 Agriculture On-farm activities Livestock production and manure management 5 1 Agricultural chemicals Pesticides control Energy Improved management in Rehabilitation, closure of mines 1 mining Cook stove Efficiency improvements, and fuel switch/ 1 improvements renewables Energy efficiency District heating and cooling systems 1 Improving efficiency in boilers and power plants 1 2 Entities providing energy efficiency services 1 Fuel switch in existing LPG, CNG, biogas, and so on to replace 1 installations coal/oil Energy saving/conservation Reducing transmission and distribution losses 1 Retrofitting of buildings: insulation and so on 1 1 Fuel cleaning and Cleaning of coal 1 improvements Renewables to replace Energy production by renewables 3 energy production by fossil fuel Rural electricity to replace local fuel (fossil or 1 biofuel) Industry Production process Cleaner production and eco-efficiency improvements Process improvements and their effects to 1 1 reduce emissions Cleaning of emissions Cleaning of emissions (between the process) 3 and the outlet (end-of-pipe) Hazardous chemicals and Management, storage, disposal of hazardous 7 1 waste chemicals Public Policy and institutional Policy and institutional improvements 3 4 administration improvements Transport Traffic management Urban traffic management and planning 2 systems Vehicles Cleaner transportation technologies 2 Emission inspection, monitoring and 2 maintenance (I&M) Public transport Public transport systems 4 3 Urban transport fuels Cleaning of petrol and diesel fuel Improved fuels 0 Urban roads Repair, construction, maintenance, upgrade of 1 1 urban roads Urban Urban planning/ Urban planning to reduce transport demand upgrading/construction activities Urban upgrading and construction to reduce PM suspension Solid waste Control and management of municipal dumps 3 9 management and establishment of landfills Reduction of small-scale open refuse burning 1 Waste water and sewage Waste water and sewage collection and 9 4 management treatment Enhancing the World Bank's Approach to Air Quality Management 21 Although this listing by number of projects in a given lution and adverse health impacts. Another 24 Sector or Activity Type may give a skewed impression of were categorized as having medium potential, and 11 which types of projects are the most important in terms of were categorized as having high potential impact. Eight their potential to reduce air pollution and adverse health of the 11 projects found to have a potential for signifi- impacts, the discussion in the following sections will give a cant reduction in local air pollution levels and popula- fuller picture of the relative importance of the respective tion exposure in the project-affected areas were under Sector Types in this respect. the Energy and Transport Sector Types: replacement of 40 old boilers in Belgrade (P075343), energy conserva- tion in Sofia/Pernik (P077575), stove improvements or POTENTIAL PUBLIC HEALTH replacements in Ulaanbaatar (P122320), and five pub- IMPACTS OF THE AIR- lic transport system projects—in Santiago (P075343 and P086689), Lima (P074021), Mexico City (P059161), POLLUTION-RELEVANT and Hanoi (P083581). The other three were policy and PORTFOLIO institutional improvement projects in Peru (P101471, This review found that projects were often miss- P116152, and P118713). ing both a description and quantification of the air pollution situation existing before the The 24 projects in the intermediate class for potential intervention of the projects and an analysis of reductions in adverse health impacts included projects the reduction in air pollution and public health addressing energy saving in Belarus (P106719); coal impact resulting from the project implementa- cleaning in Hunan, China (P075730); renewable energy tion. A few projects (for example, the Peru Lima Trans- in Moldova (P084688); power plant emission cleaning in port project and the Mexico Introduction of Climate Shandong, China (P093882); three industry sector proj- Friendly Measures in Transport project) included mea- ects (Zambia copper, P070962; Alshevsk steel, P101615; surements of air pollution concentrations before and after and Bangladesh brick, P098151); implementation of poli- project interventions. However, measurements were not cies to improve urban air quality in Colombia (P081397, sufficient to make adequate assessments of human expo- P095877, and P101301); public transport projects in India sure to air pollution or health impacts. (demonstrations in four cities) (P100589) and São Paulo (P106390); traffic management in Argentina (traffic and In general, the information contained in project land use planning) (P114008) and China (demonstration documentation shows that assessment of reduc- projects on relieving congestion in large cities) (P127036); tions in adverse health impacts as a result of air and 4 improved vehicle technology and inspection main- pollution control activities in projects has not tenance projects—green freight technology demo in been carried out as part of project planning and Guangdong, China (P119654); taxi and microbus replace- implementation. Lack of data in project documenta- ment in Egypt (P119483); and vehicle inspection and tion has not allowed for such assessments as part of this maintenance programs in Xi’an (P092631) and Colombia review. Thus, only a qualitative assessment of the pub- (P115639). lic health impact of the projects could be carried out. In most of the projects, the potential reduction in adverse Of the 48 projects found to have a small but not negligi- public health impact of air pollution was considered to be ble potential impact to reduce health impacts associated associated mainly with reduction in concentrations and with air pollution, 6 projects related to hazardous chemi- exposure to particulate matter. (Appendix C summarizes cals and 32 projects related to wastewater and sewage observations from this exercise.) management, municipal dumps, manure management and livestock, as well as small energy efficiency projects The majority of air-pollution-relevant proj- in Romania (P068062), energy savings by tightening and ects (58 percent or 48 projects) were found to insulating public buildings in Poland (P117333), and wind have only a small potential to reduce air pol- and biomass in Croatia (P071464). 22 Clean Air and Healthy Lungs pollution control is a secondary benefit. To com- OBSERVATIONS AND prehensively address air pollution control in an ef- CONCLUSIONS fective manner, it is best that the process start with Based on the review discussed in this chapter, the follow- a focus first on air pollution sources, after which ing observations and conclusions are highlighted: cost-effective interventions and sectors to be tar- » The criteria for success in addressing air geted are prioritized. pollution are often missing from project » Potential missed opportunities for air pol- planning and design. Although air pollution lution control exist in “non-typical” air concerns cut across projects in various sectors, air- pollution control sectors. The linkages to air pollution-relevant Bank projects do not necessarily pollution control are more evident in Sector Types specify the reduction of air pollution or its adverse such as energy or transport. However, projects that health impacts as a stated objective. Based on the involve wastewater/sewage treatment and man- review, only 28 of the 83 air-pollution-relevant agement of animals, manure, and fertilizers can projects included a project development objec- have a positive effect on reducing air pollution— tive related to air quality, while the remaining 55 specifically particulate matter—by reducing am- did not. A key implication of this point is that the monia emissions, which interact with sulfur dioxide relevant information and data points needed in and nitrogen oxides to form secondary inorganic order to establish baselines and make assessments particles. (Appendix F contains an overview of of health impacts—incorporating an approach how various Sector Types are relevant to air pol- that progresses from sources to concentrations and lution control, including their main pollutants and exposures—are not routinely collected in projects. the kinds of data that must be collected in order to » Almost 60 percent (48 projects) of the air- make assessments of their air pollution reductions.) pollution-relevant projects reviewed have » The limited number of projects in South the potential to achieve only small reduc- Asia and Africa suggest low prioritization tions in the adverse health impacts of air of air pollution control in regions with pollution, based on the qualitative, indicative as- high levels of air pollution and rapidly sessments reported in this chapter. Energy, Trans- urbanizing regions. Although air pollution is port and Public Administration Sector Type proj- a serious problem in regions such as SAR, where ects accounted for the largest number of projects many cities have ambient PM concentrations that with a high potential to reduce adverse health are significantly higher than indicated in WHO impacts of air pollution and were more likely to air quality guidelines and where death rates asso- specify air pollution–control-related project devel- ciated with air pollution are high, there are only a opment objectives. limited number of air-pollution-relevant projects » Air pollution interventions appear to be in the region. Similarly, there are few projects in driven mostly by a sector focus rather than rapidly urbanizing Africa. These observations sug- by an integrated approach to air quality gest that air pollution control has not been a high management informed by upstream ana- priority for both World Bank and client countries lytical work. Only a limited number of the in these regions. reviewed air-pollution-relevant projects seem to » The number of projects supported by con- have supported interventions based on the neces- cessional finance suggests limited prioriti- sary analytical underpinnings, such as full-scale air zation of air pollution control by the World quality management studies to inform the identi- Bank and by client countries. Concessional fi- fication of air pollution control interventions and nance from sources such as GEF or Carbon Finance prioritization of sectors where abatements should plays an important role in supporting air pollution– be targeted. Instead, it appears that the starting reducing activities in Bank projects, accounting for point for many projects is sector-based, and air 48 percent of the air-pollution-relevant projects Enhancing the World Bank's Approach to Air Quality Management 23 covered in this review. In other words, it appears that indicative of the need for support in strengthen- a considerable number of World Bank air pollution ing institutions and appropriate policies for air control projects are funded by external (trust fund) quality management in client countries. However, sources. SILs have been the dominant lending instrument » Despite the significant institutional and in air-pollution-relevant projects. The use of lend- technical challenges to adequate air quality ing instruments such as TALs and DPLs could be management faced by many countries and useful for helping to address these key institutional cities in developing countries, use of tech- and policy challenges up front in the design and nical assistance and policy and institution- planning of air pollution control projects. DPLs al lending instruments is limited. That many in addition provide a vehicle for promoting poli- countries are still facing problems with basic aspects cies that expand the menu of interventions for air of air quality management—such as measuring pollution control, such as the introduction of eco- and monitoring air pollution concentrations—is nomic and other policy instruments. 24 Clean Air and Healthy Lungs CHAPTER THREE A REVIEW OF CASE EXAMPLES METHODOLOGY Three cases have been selected for presentation in this report as examples of how operational and analytical activities have addressed air quality, using upfront air pollu- tion assessment approaches (further information about the three projects can be found in appendix D). These three cases provide examples of the three main methodologies for assessing the air pollution and health impacts of abatement actions in projects (see table 3.1): 1. Assessment based on monitoring of air pollution: Santiago Urban Transport Projects (1 development policy loan, 1 technical assistance loan, and 1 GEF grant). 2. Assessment based on a combination of monitoring and modeling: Ulaanbaatar (UB) Air Monitoring and Health Baseline (AMHIB) Study and the Ulaanbaatar Clean Air Project. 3. Assessment based on monitoring of air pollution and assessing health impacts, including on poor groups: Peru Environmental Development Policy Loan program THE SANTIAGO URBAN TRANSPORT PROJECTS These were a set of projects (supported by DPL, TAL and GEF grant) implemented in the 2003 to 2011 period that, in addition to supporting the overall public transportation system in Santiago, also supported air pollution control programs by helping to improve Santiago’s air quality through reducing local air pollutants like SOx, CO, PM, and NOx (particularly through the DPL and TAL projects) and by promoting the reduction of greenhouse gases from ground transportation in Santiago through a long-term modal shift to more-efficient and less-polluting forms of transport (particularly through the GEF project). The program involved interventions to reduce the emissions from public buses, increase the use of bicycles, implement up-to-date emission testing, develop busi- ness schemes for operation of the public transportation system, monitor the bus systems, and implement overall traffic planning. This case is an example of how air quality moni- toring can be used to measure the improvement in air quality that has resulted from project implementation. Enhancing the World Bank's Approach to Air Quality Management 25 TABLE 3.1. CHARACTERISTICS OF THE THREE CASE STUDY PROJECTS Health Type of Set APC Sector Analytical Impact Project project objective Focus Foundation Study Project Cost 1. Santiago urban 1 DPL, 1 TAL, Yes (in Transport Partly referred Preliminary DPL $30.16 million transport projects 1 GEF all 3) to in CAS, health cost TAL $4.8 million 2002 estimate GEF $6.8 million 2. Ulaanbaatar air SIL (Based on Yes Multi-sector, Full-scale Yes, upfront AMHIB: $1 million monitoring and AAA) but mainly AQM plans SIL: $15 million health baseline energy study and Ulaanbaatar clean air project 3. Peru environmental 3 DPLs (based Yes Multi-sector, CEA including Yes, in initial $475 million for three DPL program on CEA) but mainly CoED study CEA DPLs: $330 million, transport $70 million, $75 million ULAANBAATAR AIR MONITORING of how a series of DPL projects were brought together AND HEALTH IMPACT BASELINE under an overall environmental program underpinned (AMHIB) STUDY by upstream analytical work and characterized by dialog In 2007–08, the World Bank started to plan a program for with the government, which was used to gradually build replacing cookstoves in the Ger areas of Mongolia. After ini- a constituency for air pollution control and shape further tial government resistance to its proposals, the World Bank, air pollution control activities. in cooperation with the government of Mongolia, decided to support a full-scale integrated air quality management (IAQM) study11 in order to obtain a complete understand- EFFECTIVENESS OF THE ing of sources, concentration levels, health impacts, and the CASE STUDY PROJECTS most cost-effective abatement options in the short, medium, All case studies were implemented in urban areas and long term. During the portfolio review, this case study that, subsequent to project implementation, saw was selected because it involved a complete air quality and reduced PM concentrations. Particularly in Santiago, health impact assessment methodology, and it created a but also in Lima and initially in Ulaanbaatar, PM moni- baseline for future assessment of improvements in air qual- toring during the project periods showed improved air ity and health conditions. The study formed the basis for quality conditions. Although direct assessment of the proj- the Ulaanbaatar Clean Air Project, which started in 2012.12 ect activities’ impact on improved air quality and health conditions has not been made in any of the projects yet, THE PERU ENVIRONMENTAL indications are that they all had some responsibility for the DEVELOPMENT POLICY LOAN PROGRAM improved air quality situation. Between 2009–11, the Peru Environmental Development Figure 3.1 shows the timing of each of the policy meas- Policy Loan Program (ENVDPL) supported improve- ures for cleaner transportation in Santiago (many of which ments in vehicle emissions, fuel quality, and air quality were applied through implementation of the government monitoring systems in Peru. It was selected as an example of Chile’s 2000–10 Urban Transport Plan for Santiago, the TRANSANTIAGO project) juxtaposed against the 11 For further information on IAQM, see chapter four. reduction of PM concentrations at a monitoring station 12 The stove replacement program that is included in the Ulaanbaatar Clean Air Project is moving in conjunction with a stove replacement program that was close to a major bus route in central Santiago (Parque initiated by the Millennium Challenge Corporation in 2010/2011. O’Higgins Station). 26 Clean Air and Healthy Lungs FIGURE 3.1. PM2.5 CONCENTRATIONS AT THE PARQUE O’HIGGINS STATION, SANTIAGO, CHILE, 1989–2012 First public transport tender Withdrawal of 500 buses Reduction of fuel sulfar 5000 to 35% of the fleet with DPF 3000 ppm Low natural gas sales, (transantiago), LDV standard Sale of fuel increase gasoline and euro IV/V, Sale of gasoline with without Pb Natural gas, diesel use 15 pmm S reduction of fuel sulfar 3000 to 1500 Withdrawal of 3000 buses ppm Transantiago (buses euro 3) 60 31% of the fleet with DPF (transantiago) Reduction of Formal implementation of fuel sulfar 50 1500 to 1000 transantiago ppm Withdrawal of 2200 buses, 5% of the fleet 40 Sale of diesel with 50 ppm with DPF PM2.5 (g/m3) sulfar (transantiago) 43% of the fleet with DPF, 6% euro 30 5 (transantiago) 20 10 0 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: Prepared by M. Castillo and P. Oyola of Centro Mario Molina Chile (CMMCh) based on CMMCh (2008, 2013) and Jhun et al. 2010. FIGURE 3.2. ANNUAL AMBIENT PM2.5 CONCENTRATIONS IN LIMA-CALLAO, PERU, 2003–12 (μg/m3; 3-YEAR MOVING AVERAGES) North East South Callao Center 90 80 70 60 50 40 30 20 10 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: Macizo and Sanchez (forthcoming). In Peru, PM2.5 levels decreased from about 75μg/m3 in health conditions. In Ulaanbaatar, where cookstoves started 2009 to less than 40μg/m3 in 2012, in the metropolitan being replaced during the 2011–13 period,13 it is expected Lima-Callao area (see figure 3.2). that substantive improvements in air quality (and, thereby, health conditions) will be seen from 2013/14 onward. Given these improvements in air quality, it is likely that the case study projects resulted in better 13 141,312 Ger households out of an original 150,000 Ger households had health outcomes. In Santiago and in Lima-Callao, the replaced their stoves by the end of 2013 (at present, the number of Ger house- improvement in air quality indicates substantively improved holds has increased to around 180,000). Enhancing the World Bank's Approach to Air Quality Management 27 FIGURE 3.3. POPULATION-WEIGHTED AVERAGE CONCENTRATION FOR DIFFERENT EMISSION REDUCTION SCENARIOS, PM2.5 300 –30% –50% –80% 250 –18% –15% –24% –65% 200 Gers –48% 60% 150 100 Soils 19% 50 HOBs WHO IT-1 Others Mongolian standard WHO guideline value 0 Baseline Ger stoves HOBs Soils all 3 sectors Source: World Bank 2011. Air quality management studies have been FIGURE 3.4. HEALTH IMPACTS OF critical in setting clear air quality targets and AMBIENT AIR POLLUTION in selecting sector and project activities. Par- PER UNIT OF INCOME IN ticularly in the Ulaanbaatar and Peru projects, Bank- LIMA-CALLAO funded analytical work was instrumental in designing 300 Poor Non-poor specific abatement options and setting both the air quality targets and success criteria for activities. In the 250 Ulaanbaatar study, estimations were made of the con- tributions of each of the main polluting sectors (stoves 200 in Ger households, heat-only boilers [HOBs], soil/dust protection, and other sectors) to high PM concentra- 150 tions, and of the amount of emission reductions that 100 would need to be achieved in each sector in order to reach set air quality standards, that is, WHO and Mon- 50 golian standards. Figure 3.3 shows three different emis- sions reduction scenarios: 30 percent (blue), 50 percent 0 “High-case” (red) and 80 percent (green). As seen from the figure, Base-case “Mid-case” scenario scenario scenario even an 80 percent emission reduction scenario would Source: Larsen and Strukova (2005). not be sufficient to meet the stated standards. In fact, in order to reach Mongolia’s own PM2.5 standard, about a 95 percent emission reduction in the selected sectors would need to be achieved. demonstrated that Ger areas tend to be both poorer Furthermore, air pollution control and health and more polluted than the Ulaanbaatar city center. improvement focused on poor population groups. Figure 3.4 shows that in the Lima-Callao area of Peru, Both in the case of Ulaanbaatar (Ger households) and health impacts of air pollution could be more than three in Peru, the focus was on poor population groups. Ana- times higher for poor people than for non-poor people, lytical work conducted under the UB Clean Air Project relative to income. 28 Clean Air and Healthy Lungs country stakeholders, helped the Ministry of Finance to LESSONS LEARNED FROM understand the air pollution control priorities identified THE CASE STUDY PROJECTS by the study and built trust within the country at various A solid analytical foundation is needed to inform stakeholder levels, which served to reinforce country own- decision making on interventions to help coun- ership of the analysis. The CEA became part of an ongo- tries and cities address air pollution. In Peru, an ing dialog between the World Bank and the government, initial country environmental analysis (CEA), including which formed the basis for the DPL series. As a result, a cost of environmental degradation, provided the ana- the government of Peru implemented actions based on lytical basis for a priority-setting exercise to inform the the CEA, even going beyond the actions contained in government’s decision making and the eventual design the DPL’s policy matrix. In Ulaanbaatar, the underlying of the policy matrix of the DPL program. The major- AMHIB study that evolved over three years formed the ity of the costs associated with air pollution were due to basis for the dialogue with both the national Mongolian health damage costs, specifically morbidity and prema- government and, particularly, the Ulaanbaatar govern- ture death. With the analytical work, Peru was able to ment that resulted in the Clean Air project that included put its resources into cost-effective ways of mitigating the the stove replacement program and so on. damage associated with air pollution. It should be noted that the CEA covered several categories of environmental The best methodological tools should be used to degradation, of which air pollution was a major one. In provide clarity on the costs and benefits of alter- the case of Ulaanbaatar, although later monitoring results nate interventions to reduce air pollution. In the through the AMHIB study showed the highest PM con- case of Peru, once the CEA had identified a priority area— centrations ever measured in any city around the world, health impacts resulting from ambient air pollution—it at the outset of the study the city had very limited knowl- analyzed the costs and benefits of alternative interventions edge about these high PM concentrations, their health to provide clear information to guide the government’s impacts, and the relevant abatement options. The analyti- decision making for reducing air pollution. In the Ulaan- cal foundation developed through the study was invalu- baatar project, calculations of health benefits of alternative able in shaping an overall air pollution control plan and air pollution control interventions were used to convince in designing and implementing interventions within the the authorities in Ulaanbaatar to go for immediate inter- Ulaanbaatar Clean Air Project. Furthermore, at the out- vention options, such as improving stoves, rather than only set of the Santiago Urban Transport DPL, studies were using more costly long-term options like relocating the conducted that paved the way for an environmental focus populace into apartments (see figures 3.5 and 3.6). The in the DPL itself targeting air pollution control. In addi- blue hatched area in figure 3.6 shows the total lost health tion, throughout the period of the DPL, further studies benefit if only the long-term option were implemented. were undertaken, resulting in the TAL and GEF projects. The results of analytical work relating to air pol- The analytical work should be based on a solid lution should be framed in terms that policy mak- dialogue with the country stakeholders, which ers can easily understand. Policy makers often have may take years to evolve. In Peru, the CEA came the responsibility of making decisions on widely disparate from a process of dialog between the Bank and the gov- programs and policies, and they must have the problem and ernment, including a workshop aimed at consensus build- impacts of air pollution framed to them in terms that make ing around the analytical work, and from preparation them easy to compare with competing priorities. The ana- by the Bank of a policy note dedicated to environmen- lytical work on air pollution in both Peru and Ulaanbaatar tal degradation and environmental health policy in Peru was successful in reaching policy makers, notably the Min- (Sánchez-Triana and Awe 2007). That dialogue started istry of Finance in Peru and both the president of Mon- well in advance of the approval of the DPL program. A golia and the mayor of Ulaanbaatar, because they framed step-wise approach and comprehensive scope of analy- the relevant issues in terms of three easy-to-understand ses, combined with the way in which it was shared with metrics: lives lost, economic cost, and the cost of inaction. Enhancing the World Bank's Approach to Air Quality Management 29 FIGURE 3.5. ANNUAL HEALTH An estimation of health impacts from high par- BENEFITS FROM FIVE OUT ticulate matter concentrations is often the most effective way of catching the attention of the OF EIGHT ABATEMENT government and civil society and of providing SCENARIOS IN momentum for air pollution control action. All ULAANBAATAR, 2010–13 three case examples outlined up front the health impacts Scenario 1 startup emissions Scenario 2 certified (lighting techinque) stoves from air pollution in order to draw attention to the pro- Scenario 3 SCC + stoves Scenario 4 electric gram. Health impact data for Peru, estimated in the CEA, 300 Scenario 5 relocation into heating are summarized in table 3.2. apartments 250 Annual health benefit ($ million) Results on the severity and implications of air 200 pollution should be disseminated widely within countries and urban areas. The people who are 150 most affected by a country’s poor ambient air quality are those who live in the polluted areas, and they should be 100 aware of the severity of the problem so that they can raise attention and hold their government accountable for its 50 efforts to remedy the issue. While poor people are dis- proportionately affected by air pollution and other envi- 0 ronmental health problems, they are also the least able to 10 20 1 20 2 20 3 20 4 20 5 20 6 20 7 20 8 20 9 20 0 20 1 20 2 23 1 1 1 1 1 1 1 1 1 2 2 2 afford measures to address its health impacts. Increased 20 20 Source: World Bank 2011. government attention to addressing air pollution is core to a country’s or a city’s poverty reduction efforts. FIGURE 3.6. COMPARING HEALTH BENEFITS OF TWO INTERVENTIONS: CERTIFIED STOVES AND RELOCATION INTO APARTMENTS IN ULAANBAATAR Scenario 2 certified stoves 300 Scenario 5 relocation into apartments 250 Health benefit ($ million) 200 Accumulated health benefit for 150 2010-2023 period: Senario 2 $ 1605 million 100 Senario 5 $ 597.1 million 50 $ 1007.9 million 0 10 12 10 16 18 20 12 20 20 20 20 20 20 20 Source: World Bank 2011. 30 Clean Air and Healthy Lungs TABLE 3.2. ESTIMATED ANNUAL sector but that need to be dealt with in order to effectively HEALTH IMPACT OF AMBIENT reduce air pollution. In the case of Peru, it is not likely that the significant air quality improvements achieved AIR POLLUTION FROM under the Peru ENVDPL would have been possible, for PARTICULATE MATTER IN PERU example, through an investment loan focused solely on the Total Cases/ Total DALYs/ transport sector, which incorporated a component on ret- Health End-Points Year Year rofitting vehicles. The ability of the DPL to support a wide Premature mortality 3,900 29,253 range of policy and institutional actions simultaneously Chronic bronchitis 3,812 8,386 across multiple sectors helps magnify the scope of results Hospital admissions 12,834 205 achievable in addressing a widespread problem such as Emergency room visits/ 251,765 1,133 air pollution. At a sectoral level, policy reform helps to outpatient hospital address the policy and institutional issues that determine visits the effectiveness of interventions or investments to address Restricted activity days 43,347,360 13,004 Lower respiratory 533,457 3,467 air pollution in that sector. Additionally, the original San- illness in children tiago Urban Transport Programmatic DPL created the Respiratory symptoms 137,957,686 10,347 basis for a policy dialogue on the application of air quality Total 65,796 policies and further established the basis for other fine- Source: Larsen and Strukova 2005. tuned AQM interventions in the following TAL and GEF- funded projects. Policy reforms offer a comprehensive way of Approaches to addressing air pollution should addressing air pollution reduction effectively. incorporate a broad set of instruments tailored Providing support for policy reforms and implementation to air pollution. Reducing air pollution requires that of policy reforms at a sectoral or national level are better the appropriate policy instruments are applied. In some methods of achieving broader-ranging results in reducing cases, these could be policy instruments or economic air pollution than interventions or investments in a spe- instruments that are more effective in addressing air cific sector, whose impacts would be primarily localized pollution–related issues that are, for example, associated to the area of the intervention. National and municipal- with market failures or institutional deficiencies than tools level policy reforms have the ability to address policy and (like environmental impact assessment) that seek to miti- institutional issues that lie outside the purview of a specific gate the air pollution impacts of a specific project. Enhancing the World Bank's Approach to Air Quality Management 31 CHAPTER FOUR A PROPOSED APPROACH FOR ESTIMATING THE AIR QUALITY AND HEALTH IMPACTS OF WORLD BANK PROJECTS INTRODUCTION The review of the portfolio of air-pollution-relevant projects and the examination of case examples highlighted, among other aspects, the need for a shift from a sector-driven focus to a comprehensive approach to air quality management in World Bank projects as well as the dearth of baseline information and data that should be included in project planning and preparation in order to assess the impacts of projects on air pollution reduction and health outcomes. These observations are indicative of the need for a strategic, system- atic approach in Bank projects that address air quality management issues. This chap- ter examines how cost-effective air quality management strategies may be applied in the development of World Bank projects that incorporate air pollution abatement measures. INTEGRATED AIR QUALITY MANAGEMENT CONCEPT In order to shift from a sector-driven approach and ensure that a project has an opti- mal economic and technical approach to reducing air pollution and its health impacts, one has to start by applying an air quality management approach that is independent of a specific sector. AQM planning based on an integrated approach can help generate information on health costs related to high air pollution concentrations and the cost of inaction in addressing air pollution, which can help clients make decisions about air pollution control, including identification and prioritization of cost-effective air pollu- tion abatement interventions and determination of the sectors that must be targeted in order to implement those interventions. The integrated air quality management concept is a structured approach to a continu- ous cycle of planning, implementing, evaluating, and adjusting abatement strategies and measures for continual improvements (see figure 4.1). The concept has been used in successful air pollution abatement programs worldwide, and World Bank projects Enhancing the World Bank's Approach to Air Quality Management 33 FIGURE 4.1. FRAMEWORK FOR » Optimizing abatement strategy on the COMPREHENSIVE INTEGRATED most cost-effective interventions—This step is aimed at finding the most effective abatement AIR QUALITY MANAGEMENT options in terms of comparison of control costs Emissions from Monitoring Air pollution with reduced health damage costs. Doing this for pollution sources dispersion modelling concentrations different abatement options helps to rank policies or investments according to their damage reduc- tion per unit of expenditure. Exposure Figure 4.2 illustrates the process for assessment of air pol- lution and health impact assessment in projects. The inte- grated air quality management approach can be extended Optimized to incorporate air pollutants that also have climate Abatement options Impacts and abatement cost-benefit analysis damages impacts, notably short-lived climate pollutants. Some ini- strategy tial work in this respect is presented in appendix E. Source: Authors’ illustration. have incorporated elements of it to varying extents. Ear- lier work conducted by the World Bank has promoted the CONSIDERATIONS FOR IAQM approach in Asian cities (World Bank 1997). APPLICATION OF INTEGRATED AIR QUALITY MANAGEMENT The IAQM approach involves four major steps: » Understanding air pollution sources—This CONCEPT IN WORLD BANK step involves identification of emission sources, PROJECTS including their geographic location, by conduct- In presenting the integrated air quality management ing a detailed inventory and analysis of emission approach, it is important to consider the implications of sources, including stationary and non-stationary adopting and applying such an approach within the insti- (fixed and area) sources. tutional context of the World Bank. » Understanding air quality—This step involves a combination of ground-monitoring data and at- Upfront analytical work over an extended period mospheric dispersion modeling to determine air is required prior to determination of project pollution concentrations and their distribution. concept and target sector for air pollution con- This information can be used to model the ways in trol. Depending on the climate and geographical con- which pollution concentration levels in individual ditions of a city or an urban area, completing IAQM locations change with the introduction of specific that progresses from sources to impacts as outlined in the abatement measures. When monitoring data are steps in the previous section will, based on experience, not available, modeling may be applied. Appendix require at least one or two years (World Bank 2011). The B provides information on state-of-the-art method- upfront analytical work conducted through the IAQM ologies for assessing health impacts as a result of would help to inform the selection and determination air pollution reduction projects. of the correct sector in which cost-effective interven- » Understanding health impacts—In this step, tions for air pollution control should be undertaken. In observed and modeled air pollution concentrations other words, air quality management becomes the initial are translated into impacts by estimating popula- and primary focus, which informs the determination of tion exposure and then applying dose-response or a project concept and the sector in which that project concentration-response functions to link pollution should be. The IAQM approach turns around the cur- levels to health outcomes, specifically morbidity rent World Bank approach, which usually begins with and premature death. the sector. The time requirements of the IAQM process, 34 Clean Air and Healthy Lungs FIGURE 4.2. AIR POLLUTION AND HEALTH IMPACT ASSESSMENT PROCESS AIR POLLUTION ASSESSMMENT MEASUREMENTS MONITORING FOR SOURCE EMISSIONS EMISSSIONS NETWORK APPORTIONMENT INVENTORY DISTRIBUTION PM concentrations at PM elemental Annual emissions per Spatial and temporal station locations concentrations at source type. distribution µg/m3 at stations, station locations Ton/year.source. Ton/km2.hour.source. daily/annnual average. % contribution from sources -Model evaluation- ADDITIONAL DATA DISPERSION MODELED AIR POLLUTION (PM) MODELING DISTRIBUTION Populatio distr. Urban dispersion PM concentration (µg/m3) in each (km2) Topography model in an AQM grid cell each hour throughout the year, Meteorology software system from each source sector. HEALTH EFFECTS ASSESSMENT STUDY OF HEALTH EFFECTS OF PM STUDY OF WILLINGNESS AIR POLLLUTION (PM) Mortality and TO PAY (WTP) TO EXPOSURRE hospitalization data for AVOID HEALTH EFFECTS Combination of PM and individuals in selected OF AIR POLLLUTION population distributions areas Questionnaire study. population weighted average Change in health effect Value of statistical life exposure–PWE (µg/m3) (Relative Risk, RR) as a function of PM concentration RELATIVE RISK AND HEALTH EFFECTS AND THEIR COSTS HEALTH EFFECTS COST DATA Costs in % of GDP. Health cost reduction for selected emission FROM INTERNATIONAL reduction scenarios for the main source LITERATURE sectors. COST BENEFIT CALCULATIONS COST BENEFIT EVALUATION Comparison of control EMISSIONS CONTROL HEALTH BENEFITS FOR option costs with the SCENARIOS AND COSTS EACH CONTROL OPTION monetary value of the Control options for main Avoided health effects and health benefits sources. their monetary value (the reduced health effects as Net present value Net present value a result of the control option) AIR POLLUTION CONTROL STRATEGY Short/medium/long term. To be developed on the basis of this report and other considerations. Source: Adapted from World Bank 2011. and the fact that the correct project concept and sector be taken into account in portfolio planning and in the choice for air pollution cost-effective control interven- preparation cycles of World Bank lending activities that tions emerge as a result of that process, would need to address air pollution. Enhancing the World Bank's Approach to Air Quality Management 35 IAQM requires flexibility and cross-sectoral col- Strengthening support to clients in adopting laboration in order to adapt to the results of IAQM requires a proactive and knowledge-driven IAQM studies. Related to the preceding point, the cor- approach on the part of the World Bank. Often, rect or most cost-effective air pollution abatement options the severity of the problem of air pollution or its health may lie in a different sector than originally envisaged by impacts in a specific urban setting are not known, which the client and/or the World Bank. For example, a client may preclude the client’s expressed demand, to the Bank, may wish to address air pollution through abatement for air pollution control projects. The process of devel- measures in power plants. However, further investigation, oping IAQM plans and the results of the IAQM process as informed by an IAQM study, might reveal that the could help generate information needed for policy mak- main source of air pollution in the target area is a combi- ers to decide on the correct air pollution control interven- nation of agriculture and transport. This implies the need tions for their respective urban settings. In other words, for flexibility by the Bank and the client in adapting to the demand for air pollution control projects would be the results of IAQM studies. It also underscores the need expected to grow as the knowledge of country policy deci- for strengthened collaboration between different sectors sion makers grow. Therefore, implementing the IAQM in the World Bank, likely beyond levels that typically exist, approach implies the need for a proactive stance by the including in the setting of common targets with respect Bank, underpinned by knowledge based on sound analy- to air quality management in client countries. As noted sis, to support the process that guides the client in defin- by Kojima and Lovei (2001), cross-sectoral collaboration ing air pollution control projects, as well as the need to between clients’ sectoral agencies in effectively addressing identify areas where technical and/or other institutional air pollution is also crucial. capacity assistance is required. 36 Clean Air and Healthy Lungs CHAPTER FIVE CONCLUSIONS AND RECOMMENDATIONS This chapter provides some conclusions based on the findings of the portfolio review and the examination of the case studies, along with recommendations to enhance the Bank’s work in supporting countries in reducing air pollution and its associated adverse health outcomes. Air pollution is the world’s largest environmental health issue. 1. Ambient air pollution, notably fine particulate matter, accounted for 3.7 million deaths globally in 2012, and 88 percent of those deaths were in developing countries, according to the World Health Organization. Ambient air pollution imposes a significant economic burden on a country and on individuals through its impacts on public health, reaching up to 3.2 percent of GDP in some devel- oping countries. Poor air quality resulting in illness, lost earnings, and increased medical costs can constrain productivity, which is essential for growth. 2. Household air pollution resulted in another 4.3 million deaths globally in 2012. Smoke from solid fuels burned within households also contributes to ambi- ent air pollution. Together, household and ambient air pollution represent the single largest environmental health risk, responsible for 7 million—or one in eight—deaths in 2012. They each individually cause more deaths than poor hygiene from inadequate water supply and sanitation. Air pollution is a major environmental health risk that is not currently being addressed in a systematic way by the Bank, and one that does not yet appear to be a high priority to either the Bank or its clients. 3. Poor people disproportionately carry the economic burden associated with poor air quality and typically have the fewest resources to deal with its adverse health impacts. With continued economic growth and the rapid urbanization predicted in many developing countries, if the burden of air pollution is not addressed it will continue to grow, making it more difficult for poor people to lead more productive lives and break out of poverty. 4. Ambient air pollution tends to be driven by urbanization. Greater urbaniza- tion means more vehicles and more industrial firms in a smaller space, which means more emissions. However, the examples of Bangkok, Mexico City, and China show that improving air quality can be achieved in the face of Enhancing the World Bank's Approach to Air Quality Management 37 urbanization. With proactive leaders who are will- supported in other regions with very high ing to institute the right policies and investments, levels of air pollution, such as South Asia, a nation can have clean air and healthy lungs, in and rapidly urbanizing regions, such as addition to the economic benefits of urbanization. Africa. Africa is particularly prone to being over- looked on air pollution issues because, ironically, In general, the Bank’s existing approach to air its lack of air quality management infrastructure pollution is piecemeal and inconsistent. makes it very difficult to develop a comprehensive 5. Most World Bank projects that have picture of the health impacts of air pollution in addressed air pollution have done so African cities. However, we do know that air pol- through a sector-driven approach, with lution is driven by urbanization and that Africa is air pollution control activities included as the most rapidly urbanizing region in the world. an add-on. As a result, the majority of air-pol- By 2025, Lagos will be the world’s largest mega- lution-relevant projects do not specify the reduc- city, and Africa will have one-fifth of the world’s tion of air pollution as a development objective. urban population by 2050. Rapid urbanization Consequently, baseline data and information that coupled with lack of institutional capacity to man- are required for making assessments of air qual- age air pollution mean that there is a high like- ity and of the health impacts of air pollution con- lihood that air pollution either already is or will trol activities in projects are not collected during soon be having significant adverse health impacts project planning and preparation. In other words, in the region. most projects do not contain the criteria for assess- ing their success in addressing air pollution and Bank interventions can succeed in addressing air reducing associated adverse health impacts. In pollution if air quality management is given pri- sum, the Bank is willing to build air pollution com- ority and if projects have solid analytical founda- ponents into its projects, but, because of the low tions. priority accorded to air pollution control, projects 8. Upfront analytical work is crucial for do not include the tools that would be necessary to determining the most cost-effective inter- design them properly or to measure their success ventions for addressing air pollution and or failure. Without such assessments, it is not pos- should be based on solid dialogue with cli- sible to see how successful the Bank has been at ents. The Mongolia and Peru case studies both reducing air pollution or its health impacts. demonstrate the effectiveness of comprehensive 6. Bank projects often lack the firm analyti- analytical work—notably an Integrated Air Qual- cal underpinning that they would require ity Management process for the former and a in order to effectively address the issue of Cost of Environmental Analysis, including cost air pollution. Very few of the reviewed projects of environmental degradation, for the latter— had the necessary analytical underpinnings. Proj- accompanied by continued dialog with the client ects did not include full-scale air quality manage- and relevant country actors in order to build own- ment studies to inform the identification of cost- ership of the air pollution control agenda. The effective air pollution control interventions and to results of analytical work should be framed in prioritize the sectors where abatements should be terms that are easily understood by policy decision targeted. Instead, it appears that the starting point makers: lives lost and economic cost. for many projects is sector-based, and air pollution 9. When Bank projects have made air pollu- is seen as a secondary benefit. tion a priority and have been backed by 7. Most Bank-supported air pollution control sound analytical work, they have had suc- projects are in highly urbanized regions, cess in contributing to air pollution reduc- notably East Asia and the Pacific. How- tion in client countries. Significant reductions ever, comparatively few projects have been in air pollution were achieved in the respective 38 Clean Air and Healthy Lungs urban areas covered by the case examples exam- sectoral priorities and its narrow focus renders it ined in this report, although it is difficult to make unable to ask overarching questions about which exclusive attribution to World Bank–supported sectors are the prime drivers of ambient air pol- projects. Nevertheless, the Bank-supported proj- lution and which sectors—or combinations of ects appear to have contributed to the air qual- sectors—should be targeted in order to effectively ity improvements observed. It is expected that reduce ambient air pollution. the reductions in particulate matter concentra- 12. Such a cross-sectoral approach should tions observed had positive impacts on health, al- have air quality management as its prima- though conclusive health impact assessments have ry focus and should then identify, based on not been undertaken. In particular, the Santiago cost-benefit analysis, cost-effective air pol- (Chile) and Lima-Callao (Peru) cases show con- lution abatement interventions and the re- siderable improvements in air quality, while the spective sectors where such interventions Ulaanbaatar project has not been implemented should be targeted. An Integrated Air Quality over a long enough period to draw health conclu- Management approach, as described in chapter sions. All case studies targeted sectors that had a four, underscores the need for an increased de- substantial impact on air quality conditions (trans- gree of collaboration between different sectors portation in Santiago, transportation and industry in the World Bank, likely beyond levels that typi- in Lima, and energy in Ulaanbaatar). cally exist, entailing the setting of common targets 10. Air pollution control is central to reduc- with respect to air quality management in client ing poverty and increasing shared pros- countries. Opportunities for reducing ambient air perity in client countries. Analytical work pollution exist in activities supported in multiple conducted under the Peru case example showed sectors, including those that are not obviously or that poor people are disproportionately affected commonly targeted for air pollution reduction, by the health impacts of poor air quality: in the such as agriculture and wastewater treatment. Lima-Callao area, the health impacts of air pol- Given the cross-sectoral dimensions of air pollu- lution could be more than three times higher for tion, collaboration should be promoted, for exam- poor people than for non-poor people, relative ple, between environment and natural resources; to income. Both the Peru and Ulaanbaatar case urban, rural, and social development; agriculture; examples show that addressing air pollution can water and health; nutrition and population global improve the lives of poor people. Projects that re- practices; and climate change cross-cutting solu- duce air pollution address a problem that most af- tions areas. This requires critical input from op- fects the poor and will most benefit them, and they erational staff to ensure meaningful and effective can therefore be considered as core to the Bank’s solutions for engaging client countries and cities. support to countries’ efforts to reduce poverty and 13. The Bank should increase synergies be- boost shared prosperity. tween managing local air quality and con- trolling the short-lived climate pollutants Recommendations for changing the way that the that also have health impacts, notably black Bank addresses air quality management carbon. Particulate matter is the air pollutant with 11. A shift from a sector-driven approach to a the most harmful effects on human health. Black cross-sectoral approach is needed in order carbon is a very fine constituent of particulate mat- to effectively address a cross-sectoral ter and a cause of climate warming. By building problem such as air pollution. The approach synergies between local air quality management in many Bank projects that address air pollution and the control of SLCPs, climate, air quality, and has been driven by individual sectors. The prob- health objectives can be simultaneously addressed. lem with this approach is that it turns air pollution 14. Development Policy Lending should be into an afterthought by subordinating it to other promoted as an important instrument for Enhancing the World Bank's Approach to Air Quality Management 39 supporting clients on addressing air pol- use of technical assistance loans has been limited lution control effectively at a sectoral or for supporting projects that address air pollution. national level. DPLs are able to promote policy However, given the significant technical and in- and institutional reforms, which provide an en- stitutional capacity gaps in air quality manage- abling framework for air pollution control actions ment in client countries, more technical assistance to be designed and implemented. National-level should be used to support clients. Furthermore, policy reforms have the facility of being able to technical assistance grants could be used to sup- address policy and institutional issues that lie out- port clients in conducting the substantial amount side the purview of a specific sector but that need of analytical work envisaged prior to implementa- to be addressed in order to effectively deal with tion of air pollution control projects. Although not the air pollution problem. In addition, they can be covered by the scope of the current task, in the used to address multiple sectors simultaneously, early 1990s TALs financed by IDA funds were pro- which specific investment loans cannot do. Since vided to environmental authorities, for example in in most urban areas and cities multiple sectors China, which helped to create the regulatory and would typically be contributing to air quality, the monitoring capacity for air quality control. More- DPL could be an effective tool for implementing over, the IAQM approach was also promoted air pollution actions that would address various through several projects and studies at that time. sources/sectors in a comprehensive manner. At a However, it appears that this practice has not been sectoral level, policy reform helps to address the sustained, and it should be reinstated. policy and institutional issues that determine the 16. The Bank should promote the use of effec- effectiveness of interventions or investments to tive and efficient instruments, including address air pollution in that sector. Furthermore, economic and command-and-control in- DPLs can support the development of a broad struments, for reducing air pollution. Ad- menu of instruments for targeting air pollution, dressing air pollution effectively requires a variety ranging from technical interventions to economic of instruments, in addition to technical interven- instruments, and can support institutional capac- tions and investments. This is shown by the cases of ity building, which is needed in many client coun- Mexico City (removal of regressive and inefficient tries to improve air pollution control. A review subsidies to fossil fuels and adoption of a carbon of the case studies shows that development policy tax) and China (institution of pollutant discharge loans, based on solid analytical underpinnings, fees). The Bank has traditionally emphasized the can help countries achieve air pollution reduc- use of planning instruments, such as Environmen- tion objectives. tal Impact Assessment, to address air pollution. 15. Technical Assistance through grants and However, the reviewed cities and the case studies loans should be promoted to support insti- show that other types of instruments—notably, tutional capacity building for air quality command and control (such as fuel quality regula- management by clients and to undertake tions and emissions standards) and economic in- upfront analytical work needed to inform struments (such as pollution charges and phasing decision making for air quality manage- out fuel subsidies)—are effective instruments for ment. As noted from the portfolio review, the reducing air pollution. 40 Clean Air and Healthy Lungs APPENDIX A AIR QUALITY STATUS AND CHALLENGES IN VARIOUS REGIONS Pollutant levels in many developing countries far exceed than the WHO Air Quality Guideline value of 20 μg/m3. World Health Organization (WHO) Air Quality Guide- Most national air quality standards include PM10 (CAI- line values. Particulate matter concentrations in many cit- Asia 2010; Patdu 2012). Despite its links to cardiovascular ies in lower- and middle-income countries are more than and respiratory diseases, PM2.5 is not yet part of many regu- twice the levels in high-income countries, and several latory ambient air quality monitoring networks. Countries times higher than WHO health-based air quality guide- are slowly moving toward development of PM2.5 standards. lines (see figure A.1 and table A.1). Particulate matter lev- Some of the countries that have progressed on this front els in Asia, Africa, and Latin America are substantially include Thailand, Mongolia, Pakistan, China, India, Bang- higher than in Europe and North America, with several ladesh, and Sri Lanka, although PM2.5 standards are higher Asian and African cities showing the highest levels. than WHO Air Quality Guideline values. A number of countries, such as Afghanistan and Myanmar, still do not In Asian cities, air quality levels remain well above the have national air quality standards. Challenges associated maximum levels set by national and international stand- with air quality management in Asian cities and countries ards, posing a challenge with increasing population. A include limited coverage by air quality monitoring systems, survey of over 200 Asian cities showed that only two cit- variable quality of data, technical challenges that hinder ies had an annual average PM10 concentration within the sustained monitoring efforts, and a limited role of data in WHO Air Quality Guidelines, while about 58 percent of influencing review and revisions of standards (Patdu 2012). the cities had annual PM10 levels exceeding the most leni- ent WHO interim target of 70 μg/m3 (CAI-Asia 2010). On In Latin America and the Caribbean, air pollution average, PM10 concentrations were about 4.5 times higher remains a challenge in growing urban areas, as evidenced FIGURE A.1. PM10 DATA FROM CITIES IN SELECTED REGIONS AND COUNTRIES (YEARLY AVERAGE IN μg/m3) 320 300 WHO interim target I 280 260 WHO interim target II 240 EU limit value 220 200 WHO interim target III 180 WHO guideline value 160 140 120 100 80 60 40 20 0 Be u g a La r ab os e D i Ka a tD C i o es Bo a tá W a M aw Lo w N don rk ris h ch a cr ho ijin on k el air oh m co el go Yo ak g ha iti Pa s ra Ac Li D nz or D n ar os C D ew La ta G yp Eg Source: Compiled by authors based on WHO Ambient air pollution in cities database 2014. Enhancing the World Bank's Approach to Air Quality Management 41 TABLE A.1. FIFTY CITIES WITH THE HIGHEST ANNUAL MEAN CONCENTRATIONS OF PM10 (IN μg/m3) Annual Mean PM10 City Country Levels (μg/m3) Year of Measurement 1 Peshawar Pakistan 540 2010 2 Rawalpindi Pakistan 448 2010 3 Mazar-e Sharif—Camp Northern Lights Afghanistan 334 2009 4 Gwalior India 329 2012 5 Ahvaz Iran 320 2010 6 Hamad Town Bahrain 318 2012 7 Raipur India 305 2012 8 Delhi India 286 2010 9 Karachi Pakistan 273 2010 10 Kabul—ISAF HQ Afghanistan 260 2009 11 Ma’ameer Bahrain 257 2012 12 Ras Hayan Bahrain 250 2012 13 Nabih Saleh Bahrain 244 2012 14 Lucknow India 219 2010 15 Firozabad India 219 2010 16 Kanpur India 212 2010 17 Amritsar India 210 2012 18 Ludhiana India 207 2012 19 Allahabad India 202 2010 20 Agra India 200 2010 21 Khanna India 200 2012 22 Lahore—Johar Town Pakistan 198 2010 23 Jodhpur India 196 2012 24 Rajshahi Bangladesh 195 2013 25 Narayonganj Bangladesh 181 2013 26 Dhaka Bangladesh 180 2013 27 Dakar Senegal 179 2012 28 Hidd Bahrain 178 2012 29 Dehradun India 175 2011 30 Darkhan Mongolia 174 2009 31 Chandrapur India 174 2010 32 Al Gharbia—Biya Zayed UAE 171 2011 33 Bhopal India 171 2012 34 Abu Dhabi UAE 170 2011 35 Doha Qatar 168 2012 36 Gazipur Bangladesh 166 2013 37 Patna India 164 2011 38 Barisal Bangladesh 160 2013 39 Gobindgarh India 159 2012 40 Jaipur India 155 2012 41 Lanzhou China 155 2010 42 Khurja India 154 2011 43 Al Wakrah Qatar 152 2012 44 Ulaanbaatar Mongolia 148 2010 45 Kota India 146 2012 46 Udaipur India 143 2011 47 Delta cities Egypt 140 2011 48 Al Ain—Urban/Residential UAE 138 2011 49 Mumbai India 136 2010 50 Tsetserleg Mongolia 136 2009 Source: Compiled based on WHO 2014b. 42 Clean Air and Healthy Lungs by increasing mortality attributable to air pollution. It limited, as evidenced by low respondent rates for the few estimated that at least 100 million people in the Region surveys that have examined the issue, indicative of a lack are exposed to air pollution above WHO recommended of air quality standards and published information on levels (Bell et al. 2006). Although there have been great them (World Bank 2007a; Vahlsing and Smith 2012).14 strides made with reducing air pollution in several cities, Only a few countries have established ambient air qual- notably Mexico City, Bogotá, São Paulo, and Santiago, ity standards. World Bank (2007a) found that air quality a review of monitoring data for some 25 countries dur- monitoring is sparse in most cities, and that monitoring ing the period 1997–2011 found that of the 11 cities that networks are characterized by brief periods of opera- recorded concentrations of PM2.5 in 2011, 10 exceeded tion. By 2007, only seven countries had operational the WHO and U.S. Environmental Protection Agency routine monitoring systems, and nine countries had (USEPA) annual guidelines and 8 exceeded the European established or proposed ambient air quality standards. Union (EU) annual guideline (CAI 2012). Of the 16 cities Furthermore, emissions inventories of key pollutants are that measured PM10 concentrations in 2011, all exceeded lacking, as are studies on the adverse impacts of air pol- the WHO guideline (20 μg/m3) and 9 exceeded the EU lution on public health. annual guideline (40 μg/m3). Reported challenges associ- ated with air quality management relate to the absence Much work on addressing air pollution in Sub-Saharan of national standards and data gaps. The study reported Africa has focused on vehicular sources, which are esti- that of the 25 countries reviewed, approximately one- mated to contribute over 90 percent of air pollution in third, including Honduras, Belize, Haiti, Cuba, Para- African cities (UNEP 2014). In this regard, initiatives guay, Guatemala, and Uruguay, either had no standards such as the Partnership for Clean Fuels and Vehicles in place or had no information on them (CAI 2012). A have been instrumental in helping to reduce vehicular more recent survey of air quality monitoring indicated air pollution in developing countries through the promo- that of the 42 largest cities in the region, data were col- tion of clean fuels and vehicles. To date, all countries in lected in only 20 countries (Sánchez-Triana and Sanchez Sub-Saharan Africa have completely phased out lead in forthcoming). Both surveys found that approximately half gasoline (UNEP 2014). In addition, a few countries have of the countries did not have standards for PM2.5, which is specified fuel standards. Between 2008 and 2009, most more important for health. Instead, where national stand- African countries adopted Regional Framework Agree- ards existed, they were for PM10, and monitoring data ments on Air Pollution aimed at regional cooperation were considered to be of variable quality, a reflection of on harmonizing national air quality management leg- the absence of standardized monitoring techniques and islation, standards, monitoring, and data management data collection protocols (CAI 2012; Sánchez-Triana and procedures.15 Furthermore, African ministers respon- Sanchez forthcoming). sible for health and environment signed the Libreville Declaration on Health and Environment that commits Some of the few countries in Africa where air quality is 53 participating countries to effectively address environ- being monitored exhibit some of the highest particulate mental impacts on health, including of air pollution. matter concentrations measured in any city in the world. Cities, for example, in Nigeria, Botswana, Ghana, Sen- 14 In the review by World Bank 2007a, information was gathered primarily egal, and Madagascar have PM concentrations that through surveys sent to about 40 African countries, of which 25 responded. Addi- greatly exceed both the most relaxed WHO interim tar- tional information sources were incorporated for 2 more countries, bringing the total number of countries to a total of 27. In Vahlsing and Smith 2012, surveys get standards (70 μg/m3) and their own national stand- were sent to 49 African countries, as part of a global survey, of which 6 responded. ards. World Bank (2007a) reported that the main causes 15 Regional air pollution framework agreements have been developed for West of urban air pollution are the use of fossil fuels in trans- and Central Africa/WCA (“Abidjan Agreement” from July 2009 adopted by port, power generation, industry, and domestic sources, 21 WCA countries), Eastern Africa/EA (“Nairobi Agreement” adopted by 11 NA countries in October 2008), and Southern Africa/SA (“Lusaka Agreement” in addition to the burning of firewood and agricultural adopted by 14 SA countries in March 2008), while a framework agreement and animal waste. Up-to-date information on air qual- for North Africa/NA (presented and reviewed by 6 NA countries in Tunis in ity in Sub-Saharan African countries and cities is very November 2009) is still to be formally adopted. Enhancing the World Bank's Approach to Air Quality Management 43 Notwithstanding, there remain considerable challenges production activities, road transport, industrial installa- to be addressed to more comprehensively deal with air tions, burning of wood and lignite, and agriculture. Air pollution from all sources—stationary and non-station- pollution continues to pose a challenge in urban areas ary—in the region. Key areas for improvement include in many countries. Air quality monitoring is evolving establishing and operationalizing air quality monitoring in countries. Many countries have legal obligations to networks, improving understanding of the source struc- monitor air quality in accordance with national law and ture of air pollution, understanding air pollution health requirements for EU accession. Some of the challenges impacts better, strengthening and enforcing existing legis- include limited continuous or automated monitoring lation, and developing emission and air quality standards capacity; limited numbers of, and coverage by, monitor- and emission inventories to enable implementation of ing stations, many of which measure total suspended control measures in cities. particles rather than PM2.5 and PM10; and variable moni- toring data quality. In the Middle East and North Africa Region, major sources in many countries include vehicles and In Kosovo, for example, according to 2012 State of the stationary sources such as power generation, refineries, Air Report, the first automatic monitoring station began metal smelters, desalination plants and other industrial to operate in 2009, although this number was expected to activities, and open-air burning of municipal and agri- increase to 9 covering 8 municipalities in Kosovo by 2012 cultural wastes (World Bank 2008). Seasonal sand and (KEPA 2013). According to the same report, from 2010 dust storms often aggravate air quality, carrying pollu- to 2011 the number of days during which PM10 levels tion great distances. Many countries have established exceeded the EU daily limit value of 50 μg/m3 in Pristina national air quality standards and monitoring networks. ranged from 41 to 99, compared with the EU require- For example, Egypt has set up ambient air quality stand- ment of 35 days. ards and established countrywide monitoring networks for ambient air quality, in addition to industrial emis- In Kazakhstan, the mining industry and industrial manu- sion monitoring networks that target air pollution from facturing are sources of a significant amount of particulate industrial facilities such as cement and fertilizer factories matter pollution. A 2013 report that examined air quality in (World Bank 2013b). Despite such advances, air pollu- four industrial oblasts showed that EU air quality standards tion remains a challenge in many cities and countries in were exceeded in 10 out of 11 cities, and concentrations the region, with ambient levels of air pollutants exceed- were often many times higher than EU annual limit val- ing WHO air quality guidelines. In Egypt, notwithstand- ues (JERP/World Bank 2013). Of 78 monitoring stations ing a decreasing trend in annual mean concentrations in 28 urban or industrial areas, 56 stations were manual. between 2000 and 2009, levels of pollutants such as The report noted that the existing air quality monitoring PM2.5, PM10, and lead are still above WHO guidelines, network was outdated compared with current international as well as national standards in the cases of PM10, NO2, criteria in a number of ways: too few monitoring sites, given and lead. In other countries, technical and institutional the country size and presence of heavy industry; a selection capacities need to be strengthened to more effectively of monitored pollutants that do not incorporate priority address increasing air pollution, including through pollutants that threaten public health; monitoring based enhancing air quality monitoring networks and ensur- on slow, laborious manual analyses and unrepresentative ing adequate staffing of entities responsible for air qual- sampling; and air quality standards and related practices ity monitoring with trained personnel, as initial steps that are not based on current knowledge on health impacts (World Bank 2008). of air pollution. In addition to modernizing the air quality monitoring system, addressing regulatory gaps that did not In Europe and Central Asia, key anthropogenic foster environmental compliance by industry—a key con- sources of air pollution include energy and mining tributor to air quality—was found to need attention. 44 Clean Air and Healthy Lungs A 2005 study found that in all of Europe, Macedonia indicate that annual average PM10 concentrations regu- ranked the highest in terms of the population’s expo- larly exceed the EU standard of 40 μg/m3. Although sure to PM pollution, a situation that changed little by PM10 levels have been on the decline since 2005—reach- 2011 (World Bank 2013c). Most of this exposure is in ing its minimum in the aftermath of the economic cri- Skopje—where most people live—whose geography lim- sis—the 2011 and 2012 readings suggest it is increasing its air circulation in the winter months. Observations again, reaching more than 100μg/m3 in several cities from Macedonia’s monitoring network for eight cities (World Bank 2013c). Enhancing the World Bank's Approach to Air Quality Management 45 APPENDIX B METHODOLOGIES FOR ASSESSING HEALTH IMPACTS OF AIR POLLUTION REDUCTION PROJECTS INTRODUCTION EMISSIONS, CONCENTRATIONS, EXPOSURE, HEALTH IMPACT This appendix reviews the basic components and meth- Air pollution emission is the release of pollutant odologies in assessing the health impacts of projects compounds into the air from point sources (for example, intended to reduce air pollution. stacks, chimneys), non-point sources (for example, line sources such as traffic or area sources such as a number The measurement of World Bank Core Sector Indica- of small individual sources distributed over an area), and tors (CSI) is mandatory in the World Bank project results natural sources. The pollutant emissions mixing with the frameworks. The CSIs related to the assessment of air surrounding air due to wind action and dispersion deter- quality aspects of projects that have a pollution manage- mines the air pollution concentration. ment and environmental health (PMEH) theme include “the Particulate Matter reduction achieved under the Air pollution concentration is the resulting quantity project, in μg/m3” and require as mandatory supple- of pollutant compounds in the air after the release (emis- mental data the “the number of people with exposure to sion) and mixing of pollutants. The pollution concentra- PM10 in the area of the project.” As such, the guidelines tion can be monitored—that is, measured by monitoring require population exposure but do not require the health equipment—or modeled, that is, calculated using disper- impact. The purpose of this appendix is to describe the sion and chemical transport models with input of neces- methodologies used for assessing the PM concentrations sary data. The pollutant concentration will always vary and exposure in projects that reduce pollution, as well as considerably with location and time since emissions as for the health impact assessment. well as dispersion vary with time, and dispersion is also The first part of this appendix provides a general over- dependent upon meteorological conditions such as wind view of the steps in the health impact assessment meth- and temperature. When calculated with models, the con- odology and the different data and information that are centrations are presented as a concentration “field” on a necessary. Also provided are examples of available soft- map of the area surrounding the source of emissions, with ware tools that are commonly used for supporting the sci- a suitable spatial resolution (usually 1 km or less) as well as entific assessment. an indication of how the field varies with time (hour-by- hour, or day-by-day, dependent upon the chemical com- BASIC TERMS AND pounds of interest and its health impact characteristics). RELATIONS FOR HEALTH Exposure is assessed by combining the air pollution con- IMPACT PREDICTION centration with the number of people in the affected area. A number of terms (air pollution emissions, air pollution The extent of exposure is influenced by the population’s concentration, exposure, and health impacts) and relation- mobility: the way that inhabitants stay and move through ships (spatial population distribution and concentration- the area and are exposed to pollutants. In practice, a map response function) are used in describing these linkages, of the concentrations is overlaid with a map of the popu- where terms are data points and relationships are functions lation distribution. The exposure is measured in relation that are used to link the terms in order to ultimately arrive to a certain time interval, such as an hour or day in the at estimates of health impacts. A more detailed descrip- case that the health impact results from short-term expo- tion of terms and relationships is provided in figure B.1. sure, or a year in the case of long-term exposure. Enhancing the World Bank's Approach to Air Quality Management 47 FIGURE B.1. RELATIONSHIP BETWEEN EMISSIONS, AIR POLLUTION CONCENTRATIONS, EXPOSURE, AND HEALTH IMPACTS Air pollution Exposure Health impact Emissions [µg/m3 averaged over a [mortality, morbidity in the [kg/hr] concentration [µg/m3, ppm] period] population over the period] Wind action and Spatial population Concentration-response dispersion distribution function Source: Authors’ illustration based on literature. Health impact is the result of the exposure to com- odologies for assessing air pollutant concentrations are pounds that produce a health-damaging effect, generally reviewed in this section. expressed as cases per year (or per day) attributed to the pollutant. The specific nature of the health impact (death, Concentrations can be measured using a monitoring sys- type of disease, severity, and so on) depends on the char- tem (one or more monitoring stations), or the concentra- acteristics of the air pollutant compound and is found tion variation can be assessed by the use of models. Models through research and clinical studies. need rather detailed and, usually, complex input data: mainly information on emissions, meteorology, and topog- raphy, and their spatial and temporal distribution within SPATIAL POPULATION DISTRIBUTION, the affected area. The complexity of the methodology to CONCENTRATION-RESPONSE assess air pollution concentration depends on the following FUNCTION factors. The spatial population distribution describes how » Nature of polluting compounds: Some com- population density varies within a given area and is used pounds cause damage as a result of short-term to calculating the exposure. Depending on the characteris- peaks or high concentrations over a fairly short tics of the air pollutant compound, additional characteris- period (such as 8–24 hours for ozone, and 24 hours tics of the population distribution, such as age distribution for PM and NO2). Other compounds cause dam- and vulnerable population groups, may be needed to pro- age as a result of long-term exposure (such as a vide the needed basis for impact assessment. year or longer for PM, benzene, heavy metals). Such damage potentials are reflected by air quality Concentration-response function (CRF), or dose- standards that specify the time averaging period of response relationship, describes how the likelihood and the concentration assessment. severity of adverse health effects (response) are related to » Spatial variability: The complexity is also deter- the condition of exposure to a pollutant (dose). mined by the spatial variability of the concentrations within the area. In a situation where concentrations METHODOLOGIES do not show a strong variation (such as when the emis- FOR ASSESSMENT sions are distributed evenly across the area), monitor- OF AIR POLLUTANT ing at one station could be sufficient to characterize the concentration level. But the more frequent situ- CONCENTRATIONS ation is that the concentration varies considerably The methodologies typically applied to assess the health within the area, and that monitoring at a number of impact or its reduction due to mitigation16 projects vary stations, or a modeling approach, is necessary. according to how the concentrations are assessed. Meth- » Time variations: The typical situation is that concentrations vary considerably with time (from 16 Mitigation here refers to air pollutants, not greenhouse gases. hour-to-hour, day-to-day, and so on), as a result of 48 Clean Air and Healthy Lungs time variations in emissions from various pollut- the area, meteorology, and topography) should be ing sources in the area. This is the typical situation assessed and made available. in a city where emissions from traffic and heating 3. The model should be capable of calculating short- sources, as well as industrial sources, vary by time and long-term averages in a grid system cover- of the day. Another equally important factor in ing the area. The grid cell size depends upon the time variation is the weather (wind and tempera- variability of emissions and the type of mitiga- ture), which determines the mixing and transport tion. Examples: power plant/industrial stacks: of the pollutant emissions in different directions 1 × 1 km (or possibly even smaller); traffic mitiga- and at varying speeds. tion in urban areas: 0.5 × 0.5 km or 1 × 1 km. Such factors of variability make it necessary that measure- ments or modeling of concentrations cover a time period CASES WHERE SIMPLIFIED METHODOLOGY OF MEASUREMENT long enough to capture the resulting variations, including MAY BE USED peaks, and long enough to enable a representative average A simpler assessment methodology may be used in many concentration to be determined (for the cases where the of the projects: measuring the concentration instead of impact is result of long-term exposure). using models. In terms of costs, however, the measure- ment method is not necessarily less expensive. The actual With regards to assessing the concentration reduction number and location of such stations should, for each effect of a mitigation project, a baseline of the air pol- given project, be guided by air pollution experts. lution situation before the mitigation is required. In a situation where monitoring can be used, the monitoring The practicality of using the measurement method also data from the relevant stations should be available from requires checking whether monitoring stations are oper- a period before the mitigation (at least several months, or ating in the area and whether data from the stations are preferably a year before project implementation). available for a period before the mitigation is imple- mented. If yes, measurement can be used. If not, the There are large variations among the mitigation projects monitoring method cannot be used, unless the mitigation contained in the project portfolio under evaluation for this can be postponed until a pre-mitigation monitoring pro- report. The various assessment methodologies that can be gram can be carried out. used for different types of projects are described below. Furthermore, when the monitoring method is used to CASES WHERE MODELING IS NEEDED assess the concentration situation, at least two monitoring This is the most frequent situation, where the spatial stations should be located in the area: one station should variability of concentrations in the affected area is con- be located so as to measure the average air pollution level siderable. The use of models is often considered a more in the area, in an area that will not be affected directly complex undertaking than doing measurements. The by the mitigation. The purpose of this station is to meas- costs as well as the practicality of using modeling method ure the typical air pollution level in the area. One station in a given project are largely determined by the availabil- should be located in the area that is most affected by the ity of data: Are emissions and meteorology data available mitigation. For example, in a traffic pollution reduction for the area? If yes, modeling can be used. If not, model- project in an urban area (for example, a bus corridor), the ing gets difficult, as getting such data for modeling is a station should be located close to the trafficked streets/ considerable cost, and it takes time to collect them. In lanes that will experience the largest change (presum- the case that modeling is being used, the following steps ably a reduction) of the traffic and the resulting pollutant should be followed: concentrations. For a project involving reduction in emis- 1. An air pollution model should be set up for the area. sions from a stack, for example, from a power plant, or an 2. The necessary input data (emissions of the rele- industrial source, the station should be located in the area vant compounds, including all emissions affecting of maximum ground-level impact from the stack. Enhancing the World Bank's Approach to Air Quality Management 49 more, assessing the effectiveness of a pollution mitigation ASSESSMENT OF HEALTH project or intervention requires calculation of the reduc- IMPACT tion in adverse health impacts between the before- and Health impact, as introduced earlier, is assessed by apply- the after-mitigation. Two methodologies of health impact ing the concentration-response function to the population assessment are calculation and comparison to air quality exposed to a certain level of air pollution. The result is the standard, as described below. number of cases of mortality and/or morbidity. WHO describes health impact assessment as a multi-step process CALCULATION that “combines data on population exposure with infor- After concentration data are gathered either via monitor- mation on concentration-response functions derived from ing or modeling, the exposure would be assessed (with pop- epidemiological studies to estimate the extent of health ulation distribution data), then the health impact would be effects expected to result from the exposure to the hazard- calculated (with CRF). Reduction of health impact would ous pollutants in the population” (WHO/ECEH Techni- be determined by using the reduction of exposure, which cal Report 2001). is determined by using the reduction of concentrations. The concentration-response function is complicated by a multitude of factors: depending on the pollutant, the COMPARISON TO AIR QUALITY function may be linear or non-linear, where a non-linear STANDARDS One can assess the health impact via the simpler method function indicates that response is triggered by exceeding of comparing the exposure with air quality guidelines or a certain threshold. In addition, the kind of response— standards (for example, provided by WHO or national such as incidence of tumor, disease, or death—as well as authorities on public health), in which case the impact of the various local characteristics—such as climate, housing the project on public health becomes limited to calculat- characteristics, and so on—have an influence on the CRF ing the reduction in number of people who are exposed and its shape. Finally, characteristics of the affected popu- to air quality levels that exceed the air quality guideline lation such as age, time spent outdoors, habits of exercise value or standard. and diet, smoking status, socioeconomic status, access to health care, and baseline health status also affect con- centration-response function. This means that the dose- SOFTWARE TOOLS FOR response relationship found in one area with one subset MODELING IMPACTS of the population would not necessarily be applicable to Examples of existing state-of-the-art modeling software another population in another area. Selecting the correct tools for estimating various impacts are provided in this CRF can follow the steps: section. The various impacts include those on ecosystem, 1. Conduct a literature review to identify relevant agriculture, and human health. The assessment is made publications from input data and functions consisting of population, 2. Select the study most consistent with your needs air quality (that is, air pollution concentration), and CRFs. 3. Identify the CRFs and the associated health out- With such input, the software tools allow the user to sim- comes and exposure indicators ulate several scenarios of implementing different miti- 4. Perform sensitivity analysis by examining alter- gation measures such as percentage reduction, absolute native CRFs reduction, rollback to standard, and so on. As is the case in observing the change in air quality due While all three software tools introduced here perform to a mitigation project, assessing the health impact of the basic functions that include modeling of population an implemented project necessitates coverage of a time exposure, assessing the performance of pollution abate- period long enough to capture pollution peaks or long ment measures, and comparison of scenarios, certain enough to obtain a representative average concentration software tools may have more direct suitability for typical in case there is effect from long-term exposure. Further- World Bank projects that address pollution management 50 Clean Air and Healthy Lungs and environmental health theme. This is largely due to FIGURE B.2. SCHEMATIC the geographic scale at which the modeling software is REPRESENTATION OF AirQUIS applied. For example, the RAINS model was developed Air for analysis of transboundary pollution across countries and continents. BenMAP is similar but can be applied Monitoring Emission data collection also to smaller geographic-scale projects, as is the case for Abatement AirQUIS. The three tools are briefly described below. strategy Database Models (dispersion and exposure) RAINS, BY IIASA User interface Developed from the early 1970s to solve transboundary air pollution problems in Europe, the latest Regional Air Users (policy makers, researchers, etc.) Pollution Information and Simulation (RAINS) model Source: Adapted by authors based on Norwegian Institute for Air Research by the International Institute for Applied Systems Anal- presentation. ysis (IIASA) has evolved to estimate the effectiveness of emissions control measures in terms of its impacts on ecosystems, health, and economic costs. Its application has also been expanded to integrate greenhouse gas emis- project’s potential to reduce health impacts. Further- sions (Greenhouse Gas–Air Pollution Interactions and more, if the user can provide economic valuation func- Synergies, or GAINS, model). RAINS model simulates tions—which is the relationship describing monetary the transport path from the source (point of emission) savings from improved health outcomes—BenMAP to the receptor (point of deposition) and analyzes sce- can also estimate the economic benefits. BenMAP has narios to identify the optimally cost-effective emissions been used internationally to analyze the health benefits control mechanism. In terms of estimating the impact of city-level air quality management plans and alterna- on human health, RAINS model estimates the exposure tive air quality policies in cities in Bank client countries, of population and the change in mortality rates and life including Mexico City, Santiago, São Paulo, Mumbai, expectancy due to transboundary air pollution. and Pune. BenMAP BY USEPA AirQUIS BY NILU Environmental Benefits Mapping and Analysis Pro- The Air Quality Information System (AirQUIS) of the gram (BenMAP) of the USEPA is another tool that Norwegian Institute for Air Research (NILU) is an exam- estimates the change in the incidence of various health ple of an air quality management software platform that consequences due to changes in the air pollutant con- is suitable for the health impact assessment applications centration. The user can provide the population data, needed for typical Bank projects that address the theme air pollution concentration data—either from model- of pollution management and environmental health (see ing or monitoring—as well as apply the health function figure B.2). Similar to RAINS and BenMAP, AirQUIS (CRF) that has been selected for the situation. BenMAP assesses air pollution and exposure, as well as the effects of is preloaded with a CRF library that includes hundreds abatement options. The system incorporates state-of-the-art of CRFs from U.S. and Canadian studies. With these dispersion models and a versatile emissions inventory soft- inputs, BenMAP becomes the platform to conduct ware package that enables the establishment of a full emis- various analyses, such as comparing different mitiga- sions inventory of all sources in a project area. Together, tion measures and projecting future air quality using a these modules enable an efficient analysis of the effects of combination of modeling and monitoring data. The the project abatement to reduce the population exposure in results of air quality modeling are used to calculate a the area (Norwegian Institute for Air Research 2011). Enhancing the World Bank's Approach to Air Quality Management 51 APPENDIX C METHODOLOGY FOR REVIEW FOR THE WORLD BANK PROJECT PORTFOLIO THAT IS RELEVANT TO AIR POLLUTION The portfolio review was a desk-based iterative, expanded process, conducted in two stages, as detailed STAGE 2–REVIEW OF below. Project documents reviewed were primar- CLOSED AND ACTIVE PMEH ily Project Appraisal Documents, additional project PROJECTS APPROVED documentation, and, where applicable, Implementa- BETWEEN FY08–FY13 tion Completion Reports. The review covered projects Based on the findings of the first review, it was decided financed by IBRD/IDA resources (investments, TALs, to lower the threshold of the PMEH thematic content and DPLs) as well as projects supported by concessional to 5 percent, with a view to increasing the sample size finance (such as Global Environment Facility, Carbon and hence the possibility of capturing more projects Offset, and others). that would contain adequate information on improv- ing air quality outcomes and health outcomes, as well as projects that could have significant air pollution con- STAGE 1–REVIEW OF CLOSED trol activities but were allocated less than 25 percent PMEH PROJECTS APPROVED PMEH thematic content. Furthermore, it was decided BETWEEN FY02–12 to include both closed and active projects in the review and to capture more recent air-quality-related projects. The first stage consisted of a review of the World Bank’s It was also decided to incorporate all projects approved portfolio of closed projects approved between FY02 and in FY13 projects, which had become available by this FY12 with a PMEH thematic content of at least 25 per- stage. Based on this thinking, and in view of time con- cent, as allocated according to the World Bank’s system straints, the time period FY08–FY13 was considered of sector and theme codes. The total number of projects adequate to address all the foregoing criteria and also to in this initial sample was 114. The selection of projects allow observation of any trends. The number of projects was based on the expectation that closed projects would in this initial sample was 174. possess the most comprehensive collection of documen- tation for each project and that the 10-year period would allow a good amount of time to observe any long-term CAVEATS trends in the portfolio. Given the focus of the work on The following caveats relating to the methodological a specific environmental topic (air pollution), it was con- approach should be noted: sidered that a threshold of 25 percent thematic content would be reasonable for a project that would seriously Use of PMEH content for selecting portfolio address the topic of air pollution. This review revealed review sample: The Pollution Management and Envi- that in most cases the project documentation contained ronmental Health theme code (84), as specified in the very limited information that would allow an understand- World Bank’s Guidance on Sector and Theme Codes, ing of the extent to which the project had contributed to was used to select the projects covered under the portfolio improving air quality outcomes, and even less so to health review. Box C.1 lists activities that fall under the PMEH outcomes. thematic code. As can be seen, the listed activities cut Enhancing the World Bank's Approach to Air Quality Management 53 BOX C.1. ACTIVITIES UNDER POLLUTION CRITERIA FOR SELECTION OF MANAGEMENT AND AIR-POLLUTION-RELEVANT ENVIRONMENTAL HEALTH THEME CODE PORTFOLIO From the initial sample sets, a further selection of projects » mitigation of pollution and health effects from pesticide was made based on screening for their relevance to air pol- use lution reduction. Projects that included activities that have » reduction and elimination of the use of persistent the potential to reduce air pollution were identified as “air- organic pollutants and ozone-depleting substances » mitigation of non-point source pollution from agricul- pollution-relevant.” These projects spanned various World tural runoffs Bank sectors and themes, given the applicability of the » cleaner fuels PMEH code to various sector and thematic areas as noted » oil spill contingency planning and remediation above. Table 2.2 in the main body of the report provides a » rehabilitation of contaminated production sites and sur- summary of the projects that were reviewed. Of the total rounding areas number of projects reviewed, 83 were found to be air-pollu- » improved environmental management in mining and tion-relevant: 46 for FY02–12 and 37 for FY08–13. The total energy operations » cleaner production and eco-efficiency number of air-pollution-relevant projects does not include » industrial pollution control and prevention double counting of projects in the overlap period FY08–12. » hazardous waste treatment, management, storage, and disposal » reduction and elimination of the production of persis- DEVELOPMENT OF PROJECT tent organic pollutants and ozone-depleting substances TYPOLOGIES » pollution abatement from shipping activities » vehicle emissions monitoring and maintenance The development of the project typologies started based » water pollution abatement on the universe of activities described in the OPCS » sanitation and sewerage Guidelines on Sectors and Themes that were deemed to » wastewater management and treatment solid waste be relevant to air pollution. This typologies linked activi- management ties to Sector and Subsector Types. Subsequently, the » surface and groundwater quality management and portfolio of 83 air-pollution-relevant projects (covering monitoring FY02–12 and FY08–13 samples) (see tables C.2 and C.3) Source: World Bank OPCS Sector and Theme Code Guidelines, 2013. was mapped to the developed typologies based on project activities. It is important to note that the Sector and Sub- across various World Bank sectors and themes. The selec- sector Types included in table C.1 are intended to cap- tion of the theme or sector codes assigned to a project is ture the essence of the nature of project activities under typically decided by the Task Team Leader for the project. a single grouping rather than an approach based on the Therefore, projects for which a task team leader did not Bank’s current organizational nomenclature, and there- allocate any PMEH thematic content are not covered by fore they do not necessarily correspond to sector names as this project even though they may include air-pollution- identified in the OPCS Sector and Theme Guidance or relevant activities and components. the names of Managing Sector Boards. For example, an industry sector type was identified, although in the Bank Overlap between reviewed project portfolios: The there is no official sector or sector board named industry. overlap of projects in the period FY08–12, in both stages Given that one of the objectives of this exercise was to of the review, is acknowledged. Twenty-one projects were assess the potential health impacts of the projects, it was identified as common to both data sets. Each set of data considered important to group project activities based on is presented separately where appropriate. Importantly, air pollution sources. Therefore the typology of projects to avoid double counting, the projects that overlap are was chosen to more closely reflect air pollution source cat- counted only once so that the results of the analysis are egories, and so represents a slight modification from the not affected by double counting. World Bank-based sector and theme nomenclature. 54 Clean Air and Healthy Lungs TABLE C.1. GROUPING OF PROJECT ACTIVITIES INTO TYPOLOGIES Sector Type Subsector Type Activity Type Energy Energy efficiency Improving efficiency in boilers and power plants, incl. conversion to CPH District heating and cooling systems Entities providing energy efficiency services Energy saving/conservation Reducing transmission and distribution losses Refitting of buildings: insulation and so on Fuel switch in existing installations LPG, CNG, biogas, and so on, to replace coal/oil/biomass Fuel cleaning and improvements Cleaning of coal Cleaning of petrol and diesel fuel Renewables to replace energy Energy production by renewables production by fossil fuel Rural electricity to replace local fuel (fossil or biofuel) Cook stove improvements Efficiency improvements, and fuel switch/renewables Improved management in mining Rehabilitation, closure of mines Industry Cleaning of emissions Cleaning of emissions (between the process and the outlet: end- of-pipe) Production process improvements Process improvements and their effects to reduce emissions Cleaner production and eco-efficiency Hazardous chemicals and waste Management, storage, disposal of hazardous chemicals Contaminated sites clean-up and remediation Transport Public transport Public transport systems Urban roads Repair, construction, maintenance, upgrade Urban transport fuels Improved fuels Traffic management systems Urban traffic management and planning Vehicles Cleaner transportation technologies Emission monitoring and maintenance Urban Urban planning/upgrading/ Urban planning to reduce transport demand construction activities Urban upgrading and construction to reduce PM suspension Wastewater management Wastewater and sewage collection and treatment Solid-waste management Establishment, management and control of municipal dumps and landfills Refuse incineration Reduction and elimination of small-scale open refuse burning Agriculture On-farm activities Livestock production and manure management Reduction of agricultural field burning Agricultural chemicals Pesticides control Land administration Desertification, land degradation Control of desertification to reduce PM suspension Public administration/ Policy and institutional Policy and institutional improvements Environmental improvements policies and institutions Source: Developed by authors. Enhancing the World Bank's Approach to Air Quality Management 55 56 LIST OF 83 AIR-POLLUTION-RELEVANT WORLD BANK PROJECTS REVIEWED (TABLES C.2 AND C.3) TABLE C.2. WORLD BANK AIR-POLLUTION-RELEVANT PROJECTS (APPROVED FY02–12: CLOSED PROJECTS) Air Quality Results or Air Outcome Serial Project Product Lending Quality Indicators Number Sector Type Activity Type ID Project Name Line Instrument PDO (Yes/No) 1 Industry Cleaning of emissions P070962 Zambia Copperbelt Environment Project IBRD/IDA SIL No No (end-of-pipe) (CEP) 2 Industry Cleaning of emissions P090666 Bosnia and Herzegovina Energy IBRD/IDA APL No No (end-of-pipe) Community of South East Europe 3 Industry Cleaning of emissions P093882 China Shandong Flue Gas Desulfurization IBRD/IDA SIL Yes Yes (end-of-pipe) 4 Industry Hazardous chemicals P049968 Russian Fed. Ozone-Depleting Substances GEF TAL No No and waste Production Phase-Out Technical management Assistance (GEF) 5 Industry Hazardous chemicals P059803 Kazakhstan Nura River Clean-Up Project IBRD/IDA SIL No No and waste management 6 Industry Hazardous chemicals P075776 Multi-focal area Africa Stockpiles GEF SIL No No and waste Programme—Project 1 (GEF) management 7 Industry Hazardous chemicals P082992 China demonstration of alternatives GEF SIL No No and waste to Chlordane and Mirex in Termite management Control Project (GEF) 8 Industry Hazardous chemicals P103189 Mali, Morocco, and Tanzania Africa GEF APL No No and waste Stockpiles Programme—Project 1 management 9 Industry Hazardous chemicals P105711 Ethiopia Africa Stockpiles Programme-P1 GEF APL No No and waste management 10 Industry Hazardous chemicals P112291 China Sichuan Earthquake Emgerency GEF ERL No No and waste Project (GEF) management Clean Air and Healthy Lungs Air Quality Results or Air Outcome Serial Project Product Lending Quality Indicators Number Sector Type Activity Type ID Project Name Line Instrument PDO (Yes/No) 11 Industry Production process P101615 Ukraine Alchevsk Steel Mill Revamping Carbon offset N/A Yes * - improvements and Modernization 12 Energy Energy efficiency P067337 China Energy Conservation Project, Phase II GEF FIL Yes * No 13 Energy Energy efficiency P068062 Romania Energy Efficiency (GEF) GEF SIL No * No 14 Energy Energy efficiency P075343 Serbia Energy Efficiency Project IBRD/IDA SIL Yes * Yes 15 Energy Energy saving/ P077575 Bulgaria District Heat Carbon Offset SIL Yes Yes * conservation 16 Energy Energy saving/ P106719 Belarus Social Infrastructure Retrofitting IBRD/IDA SIL No No conservation Project—Add’l Financing 17 Energy Fuel cleaning and P075730 China Hunan Urban Development IBRD/IDA SIL No Yes improvements/ Project Waste water Enhancing the World Bank's Approach to Air Quality Management and sewage management 18 Energy Renewables replacing P063644 Ecuador Power and Communications IBRD/IDA SIL No * No energy production Sectors Modernization and Rural by fossil fuels Services Project (PROMEC) 19 Energy Renewables replacing P071464 Croatia Renewable Energy Resources GEF SIL Yes * No energy production Project (GEF) by fossil fuels 20 Energy Renewables replacing P075042 Small Hydro Project (GEF) GEF SIL No * - energy production by fossil fuels 21 Energy Renewables replacing P084688 Moldova Renewable Energy from GEF N/A No * Yes energy production Agricultural Waste (Biomass) by fossil fuels 22 Public Institutional P074499 Iran Environmental Management Support IBRD/IDA SIL Yes Yes improvements 23 Public Institutional P081397 Colombia Programmatic Development IBRD/IDA DPL Yes Yes improvements Policy Loan for Sustainable Development (Continued ) 57 58 TABLE C.2. WORLD BANK AIR-POLLUTION-RELEVANT PROJECTS (APPROVED FY02–12: CLOSED PROJECTS) (Continued ) Air Quality Results or Air Outcome Serial Project Product Lending Quality Indicators Number Sector Type Activity Type ID Project Name Line Instrument PDO (Yes/No) 24 Public Institutional P095877 Colombia Second Programmatic IBRD/IDA DPL No Yes improvements Development Policy Loan for Sustainable Development 25 Agriculture Manure management P066065 Romania Agricultural Pollution Control GEF SIL No No (GEF) 26 Agriculture Manure management P075995 Moldova Agricultural Pollution Control GEF SIL No No (GEF) 27 Agriculture Manure management P084604 Serbia Danube River Enterprise Pollution GEF SIL No No Reduction 28 Agriculture Manure management P100639 Croatia Agricultural Pollution Control GEF SIL No No Project (GEF) 29 Agriculture On-farm activities P079610 East Asia Livestock Waste Management GEF SIL No No Project 30 Transport Public transport P059161 Mexico Introduction of Climate Friendly GEF SIL Yes * Yes measures in Transport 31 Transport Public transport P073985 Chile Sustainable Transport and Air GEF SIL Yes * Yes Quality for Santiago (GEF) 32 Transport Public transport P074021 Peru Lima Transport Project (GEF) GEF SIL Yes * Yes 33 Transport Public transport P086689 Chile Santiago Urban Transport IBRD/IDA TAL No No Technical Assistance Project 34 Transport Solid-waste P104937 Morocco Solid Waste Sector Development IBRD/IDA DPL No No management Policy Loan 35 Transport Urban roads P040599 China Second Tianjin Urban IBRD/IDA SIL No No Development and Environment Project 36 Urban Solid-waste P088934 Argentina Olavarria Landfill Gas Carbon Offset SIL No No development management Recovery Project 37 Urban Solid-waste P094495 Uruguay Montevideo Landfill Gas Carbon Offset SIL Yes * No development management Capture 38 Urban Wastewater P057933 China-Tai Basin Urban Environment IBRD/IDA SIL No No development and sewage Project management Clean Air and Healthy Lungs Air Quality Results or Air Outcome Serial Project Product Lending Quality Indicators Number Sector Type Activity Type ID Project Name Line Instrument PDO (Yes/No) 39 Urban Wastewater and P065416 Croatia Coastal Cities Pollution Control IBRD/IDA APL No No development sewage management Project APL I 40 Urban Wastewater P065920 Gaza II Emergency Water Project IBRD/IDA ERL No No development and sewage management 41 Urban Wastewater P066955 China Zhejiang Urban Environment IBRD/IDA SIL No No development and sewage Project management 42 Urban Wastewater P070191 China Shanghai Urban Environment IBRD/IDA APL No No development and sewage Project management 43 Urban Wastewater P075728 China Guangdong Pearl River Delta IBRD/IDA SIL No No Enhancing the World Bank's Approach to Air Quality Management development and sewage Urban Environment Project management 44 Urban Wastewater P079661 Philippines Manila Third Sewerage IBRD/IDA SIL No No development and sewage Project management 45 Urban Wastewater P081346 China Liuzhou Environment IBRD/IDA SIL No No development and sewage Management Project management 46 Urban Wastewater P086505 China Ningbo Water and Environment IBRD/IDA SIL No No development and sewage Project management Note: Asterisk indicates that PDO includes global air quality objectives. 59 60 TABLE C.3. WORLD BANK AIR-POLLUTION-RELEVANT PROJECTS (APPROVED FY08–13: ACTIVE AND CLOSED PROJECTS) Air Quality Results or Air Outcome Serial Project Product Lending Quality Indicators Number Sector Type Activity Type ID Project Name Line Instrument PDO (Yes/No) 1 Energy Improving efficiency P098654 China: GEF-Thermal Power Efficiency GEF SIL Yes* Yes* in boilers and (SIL) power plants, incl. conversion to combined heat and power plants (CHP) 2 Energy Improving efficiency P100531 India: Coal-Fired Generation GEF SIL No Yes* in boilers and Rehabilitation (SIL) power plants, incl. conversion to combined heat and power plants (CHP) 3 Energy Energy efficiency P117333 Poland: Green Investment Scheme— Carbon Offset Yes* Yes* Additional Financing (SIL) 4 Energy Efficiency P122320 Mongolia: Ulaanbaatar Clean Air (SIL) IBRD/IDA SIL Yes No improvements, and fuel switch/ renewables 5 Energy Rehabilitation, closure P100968 China: Shanxi Coal Bed Methane IBRD/IDA SIL Yes Yes* of mines Development (SIL) 6 Energy District heating and P120664 China: Urumqi District Heating (SIL) IBRD/IDA SIL No Yes cooling systems 7 Industry Management, P091031 India: Capacity Building for Industrial IBRD/IDA SIL No No storage, disposal Pollution Management (SIL) of hazardous chemicals 8 Industry Cleaner production P098151 Bangladesh: Clean Air & Sustainable IBRD/IDA SIL Yes Yes and eco-efficiency Environment (SIL) 9 Transport Public transport P083581 Vietnam: Hanoi Urban Transport (SIL) IBRD/IDA SIL No Yes* systems Clean Air and Healthy Lungs Air Quality Results or Air Outcome Serial Project Product Lending Quality Indicators Number Sector Type Activity Type ID Project Name Line Instrument PDO (Yes/No) 10 Transport Public transport P100589 India: Sustainable Urban Transport GEF SIL No Yes* systems Project (SIL) 11 Transport Public transport P106390 Brazil: SP Metro Line 4 (Phase 2) (SIL) IBRD/IDA SIL No No systems 12 Transport Public transport P114008 Argentina: Sustainable and Transport and GEF APL Yes* Yes* systems; Urban Air Quality Project (APL) traffic management and planning 13 Transport Urban traffic P127036 China: GEF Large-City Congestion GEF SIL Yes* Yes* management and Project (SIL) planning 14 Transport Cleaner transportation P119483 Egypt: Vehicle Scrapping and Recycling Carbon Offset Yes Yes Enhancing the World Bank's Approach to Air Quality Management technologies Prgrm (CF) 15 Transport Cleaner transportation P119654 China: GEF Green Freight Demonstration GEF SIL Yes* Yes* technologies (SIL) 16 Transport Repair, construction, P087630 Burkina Faso: GEF Ouagadougou GEF Yes* Yes maintenance, Transport Modal Shift (SIL) upgrade of urban roads 17 Transport Cleaning of petrol and P101301 Colombia: Third Programmatic IBRD/IDA DPL Yes Yes diesel fuel Development Policy Loan for Sustainable Development (DPL) 18 Transport Cleaning of petrol and P101471 Peru: Programmatic Environmental IBRD/IDA DPL Yes Yes diesel fuel; Cleaner Development Policy Loan (I) (DPL) transportation technologies; Emission inspection, monitoring and maintenance (I&M); Improved fuels (Continued ) 61 62 TABLE C.3. WORLD BANK AIR-POLLUTION-RELEVANT PROJECTS (APPROVED FY08–13: ACTIVE AND CLOSED PROJECTS) (Continued ) Air Quality Results or Air Outcome Serial Project Product Lending Quality Indicators Number Sector Type Activity Type ID Project Name Line Instrument PDO (Yes/No) 19 Transport Cleaning of petrol and P116152 Peru: Programmatic Environmental IBRD/IDA DPL Yes Yes diesel fuel; Cleaner Development Policy Loan (2) (DPL) transportation technologies; Emission inspection, monitoring and maintenance (I&M); Improved fuels 20 Transport Cleaning of petrol and P118713 Peru: Programmatic Environmental IBRD/IDA DPL Yes Yes diesel fuel; Cleaner Development Policy Loan (3) (DPL) transportation technologies; Emission inspection, monitoring and maintenance (I&M); Improved fuels 21 Transport Emission inspection, P092631 China: Xi’an Sustainable Urban IBRD/IDA SIL No No monitoring and Transport (SIL) maintenance (I&M) 22 Transport Emission inspection, P115639 Colombia: Colombia Sustainable IBRD/IDA SIL No No monitoring and Development Investment Project— maintenance (I&M) Additional Financing (SIL) 23 Urban Establishment, P104960 Jordan: Amman Solid Waste Management IBRD/IDA SIL No Yes* development management (SIL) and control of municipal dumps and landfills 24 Urban Establishment, P105404 Palestine Liberation Authority (West Bank SPF SIL No No development management and Gaza): Southern West Bank Solid and control of Waste Management (SIL) municipal dumps and landfills Clean Air and Healthy Lungs Air Quality Results or Air Outcome Serial Project Product Lending Quality Indicators Number Sector Type Activity Type ID Project Name Line Instrument PDO (Yes/No) 25 Urban Establishment, P106702 Brazil: Integrated Solid Waste & Carbon IBRD/IDA FIL No No development management finance (FIL) and control of municipal dumps and landfills 26 Urban Establishment, P107314 Kenya: Nairobi Metropolitan Services IBRD/IDA SIL No No development management (SIL) and control of Municipal dumps and landfills; Urban traffic management and planning; Enhancing the World Bank's Approach to Air Quality Management Wastewater and sewage collection and treatment 27 Urban Establishment, P111110 Belarus: Integrated Solid-Waste GEF SIL No No development management Management Project (SIL) and control of municipal dumps and landfills 28 Urban Establishment, P101253 Argentina: Salta Solid-Waste Management Carbon Offset Carbon Yes* Yes* development management Emission Reductions Project (CF) Finance and control of municipal dumps and landfills 29 Urban Establishment, P106857 Malaysia: Kota Kinabalu Composting N/A Carbon Yes* Yes* development management Project (CF) Finance and control of municipal dumps and landfills (Continued ) 63 64 TABLE C.3. WORLD BANK AIR-POLLUTION-RELEVANT PROJECTS (APPROVED FY08–13: ACTIVE AND CLOSED PROJECTS) (Continued ) Air Quality Results or Air Outcome Serial Project Product Lending Quality Indicators Number Sector Type Activity Type ID Project Name Line Instrument PDO (Yes/No) 30 Urban Establishment, P123323 China: Ningbo Municipal Solid-Waste IBRD/IDA SIL No No development management Minimization and Recycling Project and control of (SIL) municipal dumps and landfills 31 Urban Wastewater and P090376 China: Shanghai Agricultural and GEF SIL No No development sewage collection Non-point Pollution Reduction and treatment Project (SIL) 32 Urban Wastewater and P095925 Egypt: Alexandria Coastal Zone GEF SIM No No development sewage collection Management Project (SIM) and treatment 33 Urban Wastewater and P094311 Egypt: Integrated Sanitation & Sewerage IBRD/IDA SIL No No development sewage collection Infrastructure Project (SIL) and treatment 34 Urban Wastewater and P112626 China: Second Liuzhou Environment IBRD/IDA SIL No No development sewage collection Management Project (SIL) and treatment 35 Urban Wastewater and P112719 Uzbekistan: Bukhara and Samarkand IBRD/IDA SIL No No development sewage collection Sewerage Project (SIL) and treatment 36 Urban Reduction of small- P106885 Philippines: Integrated persistent organic GEF SIL No No development scale open refuse pollutants management project (SIL) burning 37 Urban Reduction and P118090 Egypt: Enhanced Water Resources GEF SIL No No development elimination of open Management Project (SIL) refuse burning Note: Asterisk indicates that PDO includes global air quality objectives. Clean Air and Healthy Lungs Seven Sector Types were identified, as shown in table C.1: » the emission conditions, for example, proximity to energy, industry, transport, urban, agriculture, land admin- where people are located, height of the emissions istration, and public administration. For each one, relevant » the size of the population affected by the emissions, Subsector Types are identified and corresponding activity such as in an urban area or rural area. types are grouped under similar Subsector Types. Each of the 83 air-pollution-relevant projects (46 from the METHODOLOGY FOR FY02–12 portfolio and 37 from the FY08–13 portfolio) was assigned to one of three classes: 3, 2, or 1 for the CLASSIFYING THE POTENTIAL potential adverse health impact reduction that could be PUBLIC HEALTH IMPACTS achieved by the project: OF THE AIR-POLLUTION- » 3 representing potentially large or significant health RELEVANT PORTFOLIO impact reduction » 2 representing a potentially medium potential for After identification and typing of air-pollution-related health impact reduction projects, they were reviewed to see if the potential public » 1 representing a low, but certain potential for health health impacts of the activities could be quantified. impact reduction. The review found that often missing was both a description It is difficult to indicate what “significant reduction” can and a quantification of the air pollution situation existing reasonably be expected from projects of, realistically, fairly before the intervention of the projects as well as the reduc- limited extent as typical Bank projects, many of which tion in air pollution and public health impact resulting include demonstration-level activities. Unless a baseline is from the project implementation. A few projects (for exam- clearly defined and a quantitative reduction target is set it, ple, Peru Lima Transport project, Mexico Introduction of is difficult to indicate what a significant reduction is. As an Climate Friendly Measures in Transport project) included illustration, achievement of a PM10 exposure reduction of measurements of air pollution concentrations before and 10–20 μg/m3 by project abatement would be considered after project interventions. However, the measurements significant. were not sufficient to make adequate assessments of either human exposure to air pollution or health impacts. Many of the reviewed portfolio projects deal with abate- ment of air pollution emissions from an activity or activi- The public health impacts of the projects could not be ties within an area where there are several air pollution fully assessed due to limited information on the data sources that contribute to the air pollution concentrations points required for such quantification in project docu- and the resulting air pollution exposure of the population. mentation. Thus, the extent of the reductions in air pollu- In most cases the health impact reduction can be assessed tion and the public health impact in the project areas that from the concentration reduction resulting from the proj- might result from pollution abatements in the projects can ect abatements. However, there is a need to also know be treated only in broad terms. For this reason, a qualita- the pre-project concentration level, especially in cases tive exercise was conducted to provide an indication of with very high particulate matter concentrations. Absent the size of reductions in air pollution concentrations and such information, one can mainly provide a subjective exposure, and the health impacts in the project area, that and qualitative indication of the potential of the project could be expected from abatement activities in projects. activity to reduce air pollution and health impacts in the Such a qualitative evaluation must be considered as indic- project area. ative, and was based upon the following criteria: » a qualitative evaluation of the amount of emissions reduced by the abatement17 interest to assess how large the emission and concentration reduction is com- 17 For some health end points (for example, cardiovascular disease), the dose- pared with the total emissions and concentration of the pollutant compound response curve flattens toward the high concentration levels. It is thus also of affecting the area and the population exposure there. Enhancing the World Bank's Approach to Air Quality Management 65 tation of 640 MW power plant in India; financing of small POTENTIAL PUBLIC HEALTH energy efficiency projects in Romania, as well as support to IMPACTS OF THE AIR- development of an energy management industry in China. POLLUTION-RELEVANT The 3 energy savings/conservation projects include PORTFOLIO improved district heating in Sofia and Pernik, Bulgaria, and the retrofit of boilers and buildings (such as thermal insula- As mentioned previously, each of the 83 air pollution tion) in projects in cities in Belarus and Poland. There is 1 related projects were subjected to a qualitative classifica- fuel switch project, replacing coal with gas in 40 boilers in tion of possible impact on health outcomes. There were Belgrade, and also 1 fuel cleaning project—coal cleaning three project classifications: in the Changsha-Zhuzhou-Xiangtan urban area in Hunan, » 3 representing potentially large or significant health China. The 4 renewable energy projects to replace fossil impact reduction fuel include promoting the use of agricultural straw instead » 2 representing a potentially medium potential for of coal for heating in Moldova by removal of market bar- health impact reduction riers (Moldova Renewable Energy from Agricultural Waste » 1 representing a low, but certain potential for health project, P084688), development of market for renewables impact reduction (wind, biomass, co-generation) in Croatia (Croatia Renew- This qualitative classification put 11 projects in class 3, the able Energy resources Project, P071464), upgrade of intermediate class 2 had 24 projects, and 48 projects were hydropower in Raba River in Hungary (Hungary Small in class 1. Table C.4 summarizes how these projects are Hydro Project, P075042), and rural electrification in Ecua- distributed across the 19 Subsector Types that these proj- dor (Ecuador Power and Communications Sectors Mod- ects belong to. The following section discusses the poten- ernization and Rural Services Project, P063644). The one tial for reductions in adverse public health impacts from cookstove improvement project replaces coal stoves abatements in projects from each of the Sector Types. in 150,000 suburban Mongolian households or “Gers” in Ulaanbaatar Mongolia (Mongolia-Ulaanbaatar Clean Air Project, P122320). One improved management POTENTIAL PUBLIC HEALTH IMPACTS in mines project involves a mine gas (methane) recovery OF ENERGY SECTOR TYPE project in Shanxi Province in China, estimated to enable The 16 air-pollution-relevant projects in the energy Sec- replacement of 830,000 tons of coal. tor Type deal with the reduction of emissions from fos- sil fuel combustion—almost exclusively coal in various With respect to health impact reductions from the energy sources—through improvements in energy efficiency Sector Type, three projects have a potentially large health and energy saving, the introduction of renewable energy impact reduction, since they involve significant reduction sources, and cleaning of fuels. The projects exhibit a large of coal combustion in populated areas: Serbia Energy diversity, spanning from large power plants to small boiler Efficiency Project (P075343), which includes replacement improvements to market development support for renew- of coal with gas in 40 boilers in Belgrade; Bulgaria District able energy carriers. All the projects result in reduction Heat Project (P077575), which includes improvements in in emissions of particulate matter, in some cases concen- emission cleaning at a power plant and in district heating trated in tall stacks and in other cases distributed over systems in Sofia and Pernik; and Mongolia-Ulaanbaatar entire urban areas. This variation in turn results in very Clean Air Project (P122320). Eight projects have a small different potential health impact reductions, since the health impact reduction (Class 1), while five projects are in emission characteristics and distribution result in very dif- the intermediate class (Class 2). ferent PM concentration reductions. Based on Subsector Types, the 5 energy efficiency proj- POTENTIAL PUBLIC HEALTH IMPACTS ects include thermal efficiency improvements of large OF INDUSTRY SECTOR TYPE power plants (in Guangdong and Shandong provinces) and There are 13 air-pollution-relevant projects under the indus- district heating power plants in Urumqi in China; rehabili- try Sector Type. They deal with reduction of emissions as 66 Clean Air and Healthy Lungs TABLE C.4. QUALITATIVE CLASSIFICATION OF HEALTH IMPACT REDUCTION POTENTIAL OF ABATEMENTS IN AIR-POLLUTION-RELEVANT PROJECTS Number of Projects in each of the Classes for Potential Health Impact Reduction 3 2 1 (Potentially Large (Potentially (Potentially Small Sector Type and Health Impact Medium Health Health Impact Subsector Type Reduction) Impact Reduction) Reduction) Total FY02–12 FY08–13 FY02–12 FY08–13 FY02–12 FY08–13 FY02–13 Energy Energy efficiency 1 2 2 5 Energy saving/conservation 1 1 1 3 Fuel switch in existing 1 1 installations Fuel cleaning and 1 1 improvements Renewables to replace energy 1 3 4 production by fossil fuel Cook stove improvements 1 1 Improved mgt. in mining 1 1 Industry Cleaning of emissions 3 3 (end-of-pipe) Production process 1 1 2 improvements Hazardous chemicals and waste 2 5 1 8 Transport Public transport 4 1 2 7 Urban roads 1 1 2 Urban transport fuels Traffic management systems 2 2 Vehicles 4 4 Urban Urban planning/upgrading/ construction activities Wastewater/sewage 9 4 13 management Solid-waste management 3 10 13 Agriculture On-farm activities 5 1 6 Agricultural chemicals Land administration Desertification, land degradation Public administration Policy and Institutional 3 3 1 7 improvements Total 6 5 12 12 28 20 83 Enhancing the World Bank's Approach to Air Quality Management 67 a result of emissions cleaning (which involves installation in China (P082992), rehabilitation of hazardous waste of cleaning devices between the industrial process and the sites in the states of Andhra Pradesh and West Bengal emissions outlet, called end-of-pipe cleaning), industrial in India (P091031), and assessment of potential environ- process improvements, and the management of hazard- mental impacts and risks associated with releases of haz- ous industrial waste. ardous chemicals and waste after the Sichuan earthquake (P082992). While the health impact risk associated with The three end-of-pipe emission cleaning projects the ozone depletion compounds and African pesticides include cleaning of copper smelter emissions in Zambia dumps is considered to be small (Class 1), the risk associ- (Zambia Copperbelt Environment Project, P070962), ated with Nura river mercury and pesticides management desulfurization of power plant emissions in four large in China is considered higher (Class 2), due to the poten- plants in the Shandong province in China (China Shan- tial closeness to the exposed population. dong Flue Gas Desulfurization Project, P093882), and four power plants in Bosnia-Herzegovina (Bosnia and In summary, 5 of the 13 projects under the industry type Herzegovina Energy Community of South East Europe are considered to be of health impact reduction Class 2, Project, P090666). The main focus in the Zambia project and the remaining 8 are in health impact reduction Class 1. is regulation and cleanup of lead pollution in soils and water bodies from previous copper mining and smelt- POTENTIAL PUBLIC HEALTH IMPACTS ing operations, in addition to cleaning of emissions from OF TRANSPORT SECTOR TYPE smelters. These three projects are considered to have a The 15 air-pollution-relevant projects under the trans- medium potential for health impact reduction, being asso- port Sector Type deal with reduction of emissions from ciated with emissions from tall stacks but still in medium- the transport sector primarily in urban areas as a result dense populated areas. of improvements in public transport, urban traffic plan- ning and management, cleaner transport technologies, There are two production process improvement and vehicle inspection and maintenance and urban road projects, one regarding steel mill technology moderniza- projects. tion in Alchevsk, Ukraine (Ukraine Alchevsk Steel Mill Revamping and Modernization, P101615) and the other The seven projects under the public transport systems on modernization of brick production in Bangladesh activity include support to Bus Rapid Transit and trans- (Bangladesh Clean Air and Sustainable Environment port modal shifts in projects in Santiago, Chile (P073985, Project, P098151). Both projects are considered to have P086689), in Lima (P074021), in Mexico City (P059161), a medium potential (Class 2) for health impact reduction. and in Hanoi (P083581); public transport demonstration According to the project documents, the brick industry projects in four cities in India (P100589); and establishment project in Bangladesh could affect as much as 42 percent of a metro line 4 in São Paulo (P106390). These projects of the manufacturers in the Northern Dhaka cluster. often have reducing the greenhouse gas emissions from Uncontrolled brick production in the area results in rather urban transport systems as one of their main objectives. large, although intermittent, PM emissions distributed At the same time, the interventions would also reduce air within a relatively dense population area. pollution emissions and impact. Given that Bank projects are supporting development of and investments in the The eight hazardous chemicals and waste man- major public transport system in these cities, the potential agement projects include elimination of ozone depleting for improvements in air quality and thus reduced adverse compounds (plant shutdown) in Russia (Russian Federa- health impact for the urban population could be signifi- tion Ozone Depleting Substances Production Phaseout, cant for these projects. Five of them (P073985, P086689, P049968), mercury pollution cleanup on the banks of the P074021, P059161, and P083581) are classified as Class 3, Nura river in Kazakhstan (P059803), elimination of obso- while P106390 and P100589 are classified as Class 2, as lete pesticide storage and dumps in six countries in Africa their interventions are considered more limited in scale in (P075776; P103189; P105711), pesticides management terms of their impact on the cities as a whole. 68 Clean Air and Healthy Lungs The two projects under traffic management systems untreated sewage and wastewater. Under certain condi- include urban planning in Zuzhou, Chengdu, and Har- tions, ammonia combines in air with sulfur dioxide (SO2) bin cities in China (P127036) to reduce large city conges- and nitrogen oxides (NOX) to produce particles, called tion problems and a project in Argentina on integrated secondary inorganic aerosol particles (SIA) (Schaap et transport and land use planning to reduce travel demand, al. 2004). Emissions of ammonia, SO2, and NOX over increase efficiency of public transport, and promote non- extended areas can result in a large background concen- motorized transport. These projects will result in reduced tration of SIA that contributes to public health impacts emissions from traffic in city centers and thus have a of PM. The 13 projects were selected from several other health impact reduction potential (Class 2). similar projects because they are in geographic areas where climatic conditions and population and emission Under the vehicles Subsector Type, there are two proj- densities are such that the potential for SIA production ects on cleaner transportation technologies: the Green based on the ammonia releases from the projects is con- Freight Technology Demonstration project (P119654) sidered to be significant. The reductions in ammonia in Guangdong province in China and a vehicle (taxis emissions resulting from the project interventions could and micro-buses) scrapping and recycling program in lead to a reduction of the concentration of SIA particles Egypt (119483). And there are two projects on vehicle in the broader areas surrounding the project area, cover- inspection and maintenance systems: a motor vehicle ing up to hundreds of kilometers. However, such reduc- inspection program in Xi’an in China (P092631) and a tion in SIA concentration, given the limited extent of the vehicles inspection and maintenance and emissions test- projects, is considered to be very small. Thus the health ing laboratory in Colombia (P115639). These projects impact reduction potential is classified as small (Class 1). have a limited but still considerable potential for health impact reduction in cities; thus they are assigned as Class The 13 solid waste management projects under the 2 projects in terms of health impact reductions (although urban Sector Type all involve management and con- the green freight demo in Guangdong is not so much an trol of municipal dumps and establishment of landfills, urban project). and in some cases landfill gas recovery and flaring. They include projects in Argentina, Olivarria (P088934); Uru- There are two projects under the urban roads guay, Montevideo (P094495); Morocco (P104937); Jor- Subsector Type: one in Tianjin, China (P040599), involv- dan (P104960); Southern West Bank (P105404); Brazil ing infrastructure improvements (intersections), traffic (P106702); Kenya (P107314); Belarus (P111110); India management improvements, and bus priority and another (P091031); and Uzbekistan (P112719). In the case of the in Burkina Faso (road improvements, signals and signs, Morocco DPL, policy reforms to address governance, sus- intersections). These are small-scale and considered to tainability, and environmental and social aspects in solid have a marginal health impact reduction potential and waste services were envisaged to result in environmental are therefore assigned to Class 1. and climate benefits as a result of the capture and flar- ing of landfill gas. Carbon finance operations (such as POTENTIAL PUBLIC HEALTH IMPACTS Argentina, Olivarria, and Uruguay, Montevideo) have as OF URBAN SECTOR TYPE their primary focus the reduction of greenhouse gas emis- There are 26 projects under this Sector Type, including sions, notably methane. However, collection of landfill 13 wastewater and sewage management projects and 13 gas and its combustion will also have a reducing effect on solid waste management projects. diffuse emissions of landfill gas and any toxic contents, dependent on the waste mix. The combustion emissions Of the 13 wastewater and sewage management from the flaring itself are considered small and of little projects, 8 are in China and 1 each found in Manila, consequence. Since the air pollution and potential health Croatia (coastal cities) and Gaza. The air pollution con- impact is localized on and near the dumpsites, the poten- cerns relevant to sewage and wastewater management tial health impact reduction of solid-waste management is involve emissions (evaporation) of ammonia (NH3) from considered small-scale (Class 1). Enhancing the World Bank's Approach to Air Quality Management 69 POTENTIAL PUBLIC HEALTH IMPACTS DPLs, P101471, P116152, P118713) supporting institu- OF AGRICULTURE SECTOR TYPE tional and policy improvements with respect to air pollu- There are six projects under this Sector Type. Four proj- tion control. The Colombia projects include an emission ects deal with manure management, involving the control reduction action: a substantial reduction of sulfur in diesel of manure runoff to control nutrient enrichment in the fuel (from 1,200 ppm to a maximum of 500 ppm), which Danube River and subsequently the Black Sea. Similar should result in a substantial reduction in PM emissions, to the sewage and wastewater projects above, the air- primarily from buses and trucks. In addition, guidelines pollution-related concern is the contribution of emissions are developed for emissions measurements and inven- of ammonia from untreated manure. The interventions torying. The Peru program includes policies to improve of the projects are considered to contribute to the reduc- urban air quality through low-sulfur fuels and to promote tion of SIA particles in the project sites, relative to the compressed natural gas vehicles and the scrapping of old broader areas (up to hundreds of kilometers) surrounding vehicles, including a vehicle inspection and maintenance the projects. Two projects address livestock production, program. The project in Tehran and other Iranian cities both involving the management of organic waste from pig (P074499) involves assistance on developing an environ- and poultry production—one in China (P090376) and the mental support system and strategies and investments to other the East Asia Livestock Waste Management Proj- reduce air pollution impact. The technical intervention ect covering China, Thailand, and Vietnam (P079610). in the project is to set up air pollution monitoring in Teh- The air pollution concern also relates to the control of ran and other cities. While setting up monitoring stations emissions of ammonia from production activities. Where will not in itself improve the air pollution situation, it will the improved management practices are replicated on provide the opportunity to acquire the necessary data for a broader scale, such projects could result in significant assessing the air pollution situation. Together with emis- reduction in ammonia emissions. These six agriculture sions inventories, this can provide a basis for identifying Sector Type projects have a limited potential for health further actions to reduce air pollution and its adverse impact reduction, and all are assigned to Class 1 with health impacts. respect to health impacts. The seven projects in this sector involve policy develop- ments that when successfully implemented can have a POTENTIAL PUBLIC HEALTH IMPACTS OF significant potential for health impact reduction through PUBLIC ADMINISTRATION SECTOR TYPE emissions reductions from the transport sector in urban There are seven projects in the institutional improve- areas. Based on the significant declines in air pollution ment Subsector Type; six of them are DPL projects, concentrations achieved in highly polluted industrial and with 3 in Colombia (Sustainable Development Program- urban areas in Peru, those 3 DPLs (P101471, P116152, matic DPLs, P081397, P095877, P101301) and three in P118713) are classified as Class 3 and the remaining four Peru (Environmental Development Policy Programmatic projects are considered Class 2 projects. 70 Clean Air and Healthy Lungs APPENDIX D REVIEW OF CASE EXAMPLES 3. Assessment based upon monitoring of air pollu- INTRODUCTION tion and assessing health impacts, including on Three cases have been selected for presentation in this poor groups: the Peru Environmental DPL report as examples of how operational and analytical activities have addressed air quality, using upfront air pol- lution assessment approaches (see table D.1). CASE EXAMPLE 1: IMPROVED PUBLIC TRANSPORT SYSTEM One of the cases in the reviewed PMEH portfolio, the IN SANTIAGO, CHILE Santiago urban transport project (in practice really three projects: one development policy loan, one technical assis- INTRODUCTION: MAIN AIR POLLUTION PROBLEM IN SANTIAGO AND tance loan, and one GEF grant), is being examined as a DEVELOPMENT OF TRANSANTIAGO case study for how air pollution assessment methodologies PROJECT18 can be used to assess the improvement in air quality that As identified in the 1990s, Santiago’s pollution problem has resulted from the project implementation. The project stemmed from emissions from the transport system, as itself did not include this assessment, but additional analy- well as industrial and residential sources, while its cli- sis and assessments have been carried out in Santiago that mate and topography was of such a nature that it aggra- can be used for this purpose. vated the pollution problem. A frequently occurring thermal inversion acted as a cap over the city during fall The Ulaanbaatar Air Monitoring and Health Impact Base- and winter months (April–August), inhibiting the disper- line (AMHIB) Study provides an example of how a com- sion of pollutants, which was further obstructed by the plete air quality and health impact assessment methodology mountains surrounding the city. For more than 10 years can be utilized to assist in the development of an investment the government declared pre-emergency or emergency program, in this case the Ulaanbaatar Clean Air project, days when ambient levels of PM reached unhealthy lev- and to create a baseline for future assessment of improve- els. While measures taken in the 1990s and early 2000s ments in air quality and improved health conditions. led to substantial improvements, reflecting a much lower occurrence of emergency days, air pollution levels were The Peru Environmental Development Policy Loan Pro- still far too high. Transport was the major contributor to gram promoted improvements in air quality and provides local emissions: in 2000, transport accounted for 91 per- an example of how a series of (DPL) operations, brought cent of CO emissions, 84 percent of NOX, 48 percent together under an overall environmental program, helped of PM10, 34 percent of SO2, and 38 percent of VOC to shape air pollution control measures informed by envi- emissions. ronmental priority-setting and estimation of health impacts. Due to this severe air pollution situation throughout the Hence, the cases reviewed provide examples of the 1990s, the government announced its 2000–10 Urban three different main methodologies for assessing air Transport Plan for Santiago (TRANSANTIAGO), where pollution and health impacts of abatement actions in the main goal was to improve the quality of life in the projects: metropolis of Santiago. This would be attained through 1. Assessment based upon monitoring of air pollu- better traffic flows, lower travel times, improved air quality tion: the Santiago project (DPL, TAL, and GEF) 2. Assessment based upon a combination of moni- toring and modeling: the Ulaanbaatar project 18 Introductory text taken mainly from PADs for the projects in Santiago. Enhancing the World Bank's Approach to Air Quality Management 71 TABLE D.1. CHARACTERISTICS OF THE THREE CASE STUDY PROJECTS Type of Set APC Analytical Health Impact Project Project Objective Sector Focus Foundation Study Project Cost 1. Santiago Urban 1 DPL, 1 Yes (in all 3) Transport Partly referred to Preliminary health DPL $30.16 Transport Projects TAL, in CAS, 2002 cost estimate million TAL 1 GEF $4.8 million GEF $6.8 million 2. Ulaanbaatar Air SIL (Based Yes Multi, but Full scale AQM Yes, upfront AMHIB: $1 Monitoring and on AAA) mainly plans million SIL: Health Baseline energy $15 million Project and Ulaanbaatar Clean Air Project 3. Peru Environmental 3 DPLs Yes Multi, but CEA Yes, in initial CEA $330 million, DPL program (based on mainly including CoED $70 million, CEA) transport study $75 million for three DPLs from reduced emissions of air pollutants, improved access including reduction of PM emissions, in the public bus to public transport, and improved mobility.19 transport system, increased use of bicycle as a transport mode, up-to-date emission testing, development of busi- In order to assist the national TRANSANTIAGO project, ness schemes for operation of the public transportation the Bank developed first a Santiago Urban Transport Pro- system, monitoring of the bus systems, and overall traffic grammatic DPL, followed by a Santiago Urban Transport planning based upon environmental assessments. Technical Assistance Project and a Sustainable Transport and Air Quality for Santiago GEF project. Beside focus- For the GEF project, the following specific targets were set: ing on making the overall public transportation system in » Maintain share of public transport (60 percent of Santiago more efficient, all the projects included air qual- total trips) ity improvement objectives, which included helping to » Promote rational management of transport de- improve Santiago’s air quality by reducing air pollutants mand, internalizing all costs car travel like SOx, CO, PM, and NOx. The GEF project also focused » Promote land use policies that take into account on reduction of greenhouse gases (GHG) from ground environmental and transport dimensions helping transportation in Santiago through a long-term modal shift reduce the average trip length to more efficient and less polluting forms of transport. » Promote better coordination between agencies dealing with transport-related policies and issues The interventions in the transport sector, which were » Reduce emissions of air pollutants from public mainly implemented in the 2003–09 period (the TAL transport (70 percent of PM10 and 45 percent of closed by the end of 2011), focused on improvements, NOX emissions from 1997 levels) 19 TRANSANTIAGO included the following programs: (1) Public Transport AIR POLLUTION IMPACT ASSESSMENT, Modernization; (2) Road Investments and Traffic Control; (3) Location of Edu- METHODOLOGY, AND RESULTS cational Institutions; (4) Promotion of New Commercial and Service Centers; The PAD of the 2004 TAL project states that “a prelimi- (5) Change in Residential Land-Use Trends; (6) Non-Motorized Transport; nary evaluation estimated the environmental health ben- (7) Immediate Action Program; (8) Urban Goods Transport; (9) Monitoring efits of the urban transport reform at $17.6 million per and Control; (10) Financing; (11) Communications and Citizen’s Participation; (12) Institutional Aspects. year in 1998 prices.” 72 Clean Air and Healthy Lungs FIGURE D.1. NUMBERS AND TYPES OF BUSES IN SANTIAGO, CHILE, 2006–13 Euro I Euro II Euro II Euro III+DPF Euro V No information 9000 7974 5975 6399 6572 6564 6165 6299 6299* 8000 7000 2% 5% 6% Número de buses 6000 2% 31% 34% 23% 5000 51% 65% 68% 43% 4000 32% 3000 59% 60% 2000 37% 30% 26% 48% 45% 1000 10% 3% 1% 10% 6% 2% 0 2006 2007 2008 2009 2010 2011 2012 2013* Source: M. Castillo and P. Oyola of Centro Mario Molina Chile. While a direct assessment of the actual health benefit out- system. After 2006, the PM2.5 reductions are associated come of the implementation of the project has not been with improved buses as shown in figure D.1.21 carried out, data from monitoring of air pollution, par- ticularly PM, can to some extent be used to assess the While the data and information that are available from the impact of the project implementation on the air quality TRANSANTIAGO and the World Bank project are not in Santiago. Information, data, and results on bus fleet sufficient to assess the actual reduction in health impacts in changes and air quality has been provided by M. Castillo Santiago resulting from the improved air quality, the PM and P. Oyola of Centro Mario Molina Chile (CMMCh). reductions in the city do point to reduced health impact. The TRANSANTIAGO project implementation started Although a direct link between the DPL, TAL, GEF, and to make an actual impact on the quality of the bus fleet TRANSANTIAGO projects and the drastically reduced after 2006. Figure D.1 shows that the number of buses has PM concentrations in Santiago cannot be directly assessed, been reduced from about 8,000 in 2006 to about 6,300 in it is notable that the various transport measures imple- 2013. While buses in 2007 and earlier were of Euro I, II, mented by the projects have usually been followed by lower and III classes, in 2013 more than 50 percent of the buses PM concentrations. From 1998/99 to 2011/12 the average were of Euro III+DPF20 and Euro V classes, with substan- PM2.5 concentrations at the referred monitoring station in tially lower PM emissions. Santiago were reduced from around 35 μg/m3 to around 15–17 μg/m3, which equals moving from an WHO Interim Figure D.2 shows annual average PM2.5 concentrations at Target 1 to 3 over 12–14 years, which is noteworthy. In a monitoring station close to a major bus street in cen- health impact terms, this means that from 1998/99, the tral Santiago (Parque O’Higgins station). It shows how long-term mortality risk relative to the WHO Air Quality the PM2.5 concentration has been steadily reducing in Guidelines (AQG) levels have reduced from about 15 per- steps since 1989, thus also before the TRANSANTIAGO cent above AQG to about 3 percent above AQG.22 effects started to come in particularly after 2006. Before 2006, improvements were associated with lower sulfur 21 M.Castillo and P. Oyola, CMMCh. 22 Several other monitoring stations in Santiago show a similar reduction as the contents of fuel and with initial improvements in the bus Parque O’Higgins Station between 2007/08 and 2010/11, including Stations F, D, and M (Chile Environmental Statistical Yearbook/Medio Ambiente. Informe Annual, 20 Diesel particle filter 2012, pp. 125–128). Enhancing the World Bank's Approach to Air Quality Management 73 FIGURE D.2. PM2.5 CONCENTRATIONS AT THE PARQUE O’HIGGINS STATION, SANTIAGO CHILE, 1989–2012 First public transport tender Withdrawal of 500 buses Reduction of fuel sulfar 5000 to 35% of the fleet with DPF 3000 ppm Low natural gas sales, (transantiago), LDV standard Sale of fuel increase gasoline and euro IV/V, Sale of gasoline with without Pb Natural gas, diesel use 15 pmm S reduction of fuel sulfar 3000 to 1500 Withdrawal of 3000 buses ppm Transantiago (buses euro 3) 60 31% of the fleet with DPF (transantiago) Reduction of Formal implementation of fuel sulfar 50 1500 to 1000 transantiago ppm Withdrawal of 2200 buses, 5% of the fleet 40 Sale of diesel with 50 ppm with DPF PM2.5 (g/m3) sulfar (transantiago) 43% of the fleet with DPF, 6% euro 30 5 (transantiago) 20 10 0 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: Prepared by M.Castillo and P. Oyola of Centro Mario Molina Chile (CMMCh) based on CMMCh (2008, 2013) and Jhun et al. 2010. There are about 180,000 (150,000 when the project CASE EXAMPLE 2: OPTIONS began) Ger tents, and in each of them people burn about FOR IMPROVING THE AIR four to five tons of coal and several m3 of wood per year QUALITY IN ULAANBAATAR, (mainly during the winter). This coal and wood burn- MONGOLIA, TO REDUCE ing, with the emissions practically at breathing height (two to three meters above ground), constitutes the main HEALTH IMPACTS FROM source of air pollution for the population. Other sources AIR POLLUTION23 are wind-blown and suspended dust, as well as coal-fired power plants located in the central urban area. These INTRODUCTION have rather tall stacks (100–200 meters), resulting in The main air pollution problem in Ulaanbaatar is very only a small contribution to the ground-level concentra- high concentrations of particulate matter (PM). Ulaan- tions in Ulaanbaatar. The emissions from the coal and baatar has a population of about 1.2 million people. The wood use in the Ger tents result in Ulaanbaatar being city has a traditional urban commercial and residential one of the most PM-polluted cities in the world, particu- center where about 40 percent of the population lives, and larly in the Ger areas. Around 2007–08, the World Bank an extended surrounding area populated with traditional started to plan a program to replace the stoves in the Ger tents (called Gers), where the rest of the population lives. areas. However, this plan was met with resistance from Ulaanbaatar government officials who were not certain 23 Sources for all figures and tables in this case example were taken from World that they wanted to prioritize stove removal, particularly Bank 2011, with the exception of figure D.9 and table D.6, which are from communication with Professor Lodoysamba Sereeter (engaged on the World if it meant taking out loans to finance the project. The Bank-financed Mongolia Ulaanbaatar Clean Air Project). World Bank and Ulaanbaatar government therefore 74 Clean Air and Healthy Lungs decided to undertake a full-scale air quality management FIGURE D.3. MONTHLY AVERAGE PM2.5 (AQM) study in order to get a complete understanding CONCENTRATIONS IN of sources, concentration levels, and health impacts and ULAANBAATAR, JUNE 2008– an outline of most cost-effective abatement options in the MAY 2009 short, medium, and long term. The study was named the 1400 1350 NRC (2) Ulaanbaatar Air Monitoring and Health Impact Baseline 1300 100 ail (3) 1250 study. 1200 1150 6 buudal (4) 1100 1050 Bayan hoshuu(7) 1000 950 Airport(8) 900 HEALTH IMPACT ASSESSMENT 850 800 METHODOLOGY 750 700 650 The AMHIB study included a health impact assessment 600 550 500 that covered both air quality monitoring and air pollution 450 400 350 modeling. The applied methodology had the following 300 250 steps: assessment of the air pollution concentrations and 200 150 100 their variations in time and space through measurements 50 0 and modeling, giving air pollution concentrations in each 08 8 08 08 8 8 8 08 08 9 9 9 l/0 /0 /0 /0 /0 r/0 /0 n/ g/ p/ n/ b/ ct ov ec ar ay Ju Ap Ju Au Se Ja Fe O M N D M of 1 km2 grid cells in the Ulaanbaatar area; assessment Source: World Bank 2011. of exposure by combining concentrations and population numbers in each of the grid cells; assessment of health impact through the use of dose-response relationships calibrated against standard PM measurement equipment. (partly established for local conditions through a local State-of-the-art monitors had also been established at health impact study); analysis of costs and effects of vari- four different monitoring points, financed and operated ous abatement scenarios; comparison of costs of abate- by another donor organization. In order to cover possible ment with the monetary value of the health benefits; and variations in air pollution concentrations throughout the selection of the most cost-effective scenario. year, the AMHIB station network was run over 12 months from June 2008 to May 2009. DESCRIPTION OF THE AIR POLLUTION CONCENTRATIONS AND EXPOSURE Together, these monitoring efforts provided an assessment ASSESSMENT PROCESS IN of the present air pollution (PM10 and PM2.5) situation ULAANBAATAR AIR and established that the PM concentration level in Ulaan- Pollution Monitoring Data: Before the AMHIB study, baatar was very high. Figure D.3 shows, as an example, air pollution in Ulaanbaatar had been measured to some monthly average PM2.5 concentrations at five of the eight extent, but only in the traditional city center and not in monitoring stations. The figure shows the very large dif- surrounding Ger areas. The monitoring system that had ference between winter and summer concentration lev- been in operation for a number of years had concentrated els. It also shows that the concentrations are very much on SO2 and NOX. Only some sporadic measurements had lower in the central areas of Ulaanbaatar (represented by been carried out for PM. Those measurements showed the NRC station) than in or near the Ger areas (the other high concentrations of PM, but they had only been con- stations). The measured concentrations were among the ducted in the city center. highest ever measured in any city in the world (showing monthly concentrations of PM2.5 above 1400 μg/m3). Through the AMHIB baseline study, a PM monitoring network of eight stations was established. The stations Air Pollution Emissions Inventory, Modeling, were equipped with inexpensive PM collectors that were and Exposure Assessment: The spatial distribution of spread across the city by redistributing already available the population in Ulaanbaatar was established using avail- collectors, covering both the Ger areas and the urban able data on family residences and household size in the central area. The measurements were, to some extent, central and Ger areas. This was the basis for distributing Enhancing the World Bank's Approach to Air Quality Management 75 FIGURE D.4. POPULATION DISTRIBUTION IN ULAANBAATAR, 2008 Source: World Bank 2011. Note: The range in the rightmost pane goes from close to zero (yellow) to more than 20,000 per km2 (darkest brown). TABLE D.2. EMISSIONS FROM MAIN AIR POLLUTION SOURCES IN ULAANBAATAR, 2008 (TONS/YEAR) Height of Source PM10 PM2.5 SO2 Emissions, Meters Spatial Distribution Ger households 19,731 15,785 8,784 3–5 Throughout Ger areas Heat-only boilers (HOB) 6,480 3,888 4,360 20–30 Dispersed over Ulaanbaatar surroundings Power plants 18,589 7,436 33,600 100–250 3 point sources to the west of Ulaanbaatar canter Vehicle exhaust 1,161 1,161 1,354 <1 Dispersed along main road system mainly throughout the central city areas Dry dust from roads Paved roads 5,142 771 <1 Mainly throughout the central city areas Unpaved roads 4,812 722 <1 Mainly throughout the Ger areas Source: World Bank 2011. the population in 1 km2 grid cells across the Ulaanbaatar Input to the air pollution model was mainly data on AMHIB study area. Results can be seen in figure D.4. emissions as well as meteorology and topography. The amount of the emissions from all major sources in An air pollution model was established for the study area. Ulaanbaatar was established using a typical emis- Because of the hilly topography of the area, as well as sions inventory procedure, based on amounts of fuel the considerable time variation of the emission sources used in various sources, amount of traffic on road (due to the temperature variations from season to season segments, and emission factors (for example, amount as well as the daily variation in Ger tent activities [heat- of emitted compound, here PM, per fuel unit or km ing, cooking]), it was necessary to use a model that could traveled). Special considerations were used to esti- follow the spatial and time variations in concentrations. mate the wind-blown and suspended dust emissions. A Eulerian transport model was selected that gives time The results of emissions inventory are shown in table variable (hourly) concentrations, calculated as hourly D.2. Meteorology data were available from meteorologi- averages in each of the 1 km2 grid cells of the study area cal stations operated by the Mongolian meteorology that was established (a 30x30 km grid). department. 76 Clean Air and Healthy Lungs FIGURE D.5. MODELED SPATIAL DISTRIBUTION OF PM2.5 IN ULAANBAATAR PM2.5 concentrations 2008.06.01 00:00–2009.06.01 01:00 (value) Less than 10 10–20 20–50 50–100 100–150 150–200 200–300 300–400 400–450 450–500 500–550 Greater than 550 Source: World Bank 2011. Note: Annual average Concentrations (μg/m3). As an example of results from the air pollution mode- TABLE D.3. POPULATION WEIGHTED ling, figure D.5 shows the modeled annual average PM2.5 EXPOSURE TO PM IN concentrations in the grid cells. The modeled concentra- ULAANBAATAR AS CALCULATED tions were compared with the concentrations provided by BY THE AIR POLLUTION MODEL the measurements. With a view to the uncertainties con- Source PM10 PM2.5 nected with the measurements, due to the simple measure- ment equipment used, the model represented the real PM Gers stoves 195.6 156.5 concentrations in Ulaanbaatar with reasonable accuracy. HOBs 9.0 5.4 CHPs 0.3 0.1 The exposure of the population to PM pollution was calculated by combining the PM concentrations calculated for each EP 9.2 9.2 grid cell and the population in that grid cell. These data Road dust 29.9 9.0 provided the basis for assessing the health impact from PM Windblown dust 134.4 40.3 on the population. In the Ulaanbaatar case, the exposure in Waste burning 48.9 39.1 each grid cell was combined, giving one number to charac- terize the population exposure for the whole area: the pop- Total 427.3 259.6 ulation-weighted exposure (PWE). The calculated PWE for Source: World Bank 2011. Note: Contributions from the main sources (μg/m3). PM2.5 is presented in table D.3, which also shows the con- tributions to the PWE from the main sources. Clearly, coal and wood from the Gers and the wind-blown dust are the and to establish local (Ulaanbaatar) dose-response relations dominant PM2.5 sources in Ulaanbaatar. between PM pollution concentrations and various health end points. An example of results from the dose-response Health Impact Assessment: A health impact study was relationships is presented in table D.4, showing both car- carried out to establish the air-pollution-related health diovascular and all-cause mortality based upon full-year/ impacts for the existing situation in Ulaanbaatar (baseline) warm season/cold season periods, the change in the health Enhancing the World Bank's Approach to Air Quality Management 77 TABLE D.4. STATISTICALLY SIGNIFICANT OR NEAR-SIGNIFICANT MORTALITY EFFECT ESTIMATES FOR PM AND NO2, JUNE 2008–MAY 2009 Percent Change per Each 10 μg/m3 95% Confidence Season Outcome Variable Lag (days) in PM10 or 1 μg/m3 of NO2 Interval Full year Cardiovascular CP 1 0.25 0, 0.51 Full year Cardiovascular CP 3 0.25 –0.05, 0.55 Full year All-cause NO2 1 0.84 –0.10, 1.78 Full year Cardiovascular NO2 1 1.23 –0.25, 2.71 Warm All-cause PM2.5 2 1.38 0.42, 2.44 Warm All-cause PM10 2 0.53 –0.04, 1.15 Cold Cardiovascular CP 1 0.27 0.01, 0.54 Cold Cardiovascular CP 3 0.28 –0.04, 0.50 Source: World Bank 2011. end point occurrence corresponding to a certain change FIGURE D.6. ANNUAL HEALTH in the PM (PM2.5, PM10, or course fraction PM and NO2). BENEFITS FROM 5 OUT OF Connected to the health impact part of the study, a study 8 ABATEMENT SCENARIOS of the population’s willingness to pay (WTP) to avoid cer- IN ULAANBAATAR FROM tain health impacts was carried out, as part of the basis for 2010–23 Scenario 1 startup emissions Scenario 2 certified performing cost-benefit analysis of the various pollution (lighting technique) stoves reduction options. Scenario 3 SCC + stoves Scenario 4 electric 300 heating Scenario 5 relocation into apartments 250 Annual health benefit (mill USD) AIR POLLUTION ABATEMENT OPTIONS AND COST-BENEFIT CALCULATIONS 200 Under the AMHIB study it was established that the small stoves (in about 150,000 Gers at the time of the study) were 150 the main sources of PM in Ulaanbaatar and that the most cost-effective abatement options were various ways to reduce 100 the emissions from stoves in Gers (see figure D.6). Most of the emissions from the stoves arose during the start-up 50 procedure (i.e., when starting the fire). This finding resulted 0 in several options to deal with improving the start-up pro- 10 11 20 2 20 3 20 4 20 5 20 6 20 7 20 8 20 9 20 0 20 1 20 2 23 cedure. The options investigated were, compared with a 1 1 1 1 1 1 1 1 2 2 2 20 20 20 business-as-usual scenario: 1a: reduce start-up emissions Source: World Bank 2011. through backlighting the fire; 1b: reduce start-up emissions As a final result from the analysis of health impacts and through stove modifications; 2: replace existing stoves with their costs for each of the abatement scenarios, for a grad- cleaner stoves, no fuel change; 3: replace existing stoves and ual implementation over time, table D.5 gives the net ben- fuel with cleaner stoves and semi-coked coal; 4: install elec- efit (monetary value of reduced health impact minus the tric heating in existing Ger homes; 5: relocate Ger house- cost of the abatement) for each scenario. The immediate holds into apartment buildings; 6: improve the heat-only term option with the stove start-up modifications gave the boilers; 7: reduce road dust suspension: street cleaning; and largest net benefit, while improved stoves and fuels and 8: green urban areas to prevent dust suspension. the medium-term option with electric heating in the Gers 78 Clean Air and Healthy Lungs TABLE D.5. COMPARISON OF PRESENT VALUE (PV) OF HEALTH BENEFITS (BASE CASE) WITH NET PRESENT VALUE (NPV) OF IMPLEMENTING COSTS, AND NET BENEFIT (PV MINUS NPV) FOR THE EIGHT ABATEMENT SCENARIOS, 2010 ($ MILLIONS) Present Value Net Present Value (NPV) (PV) of Health of Costs of Implementing Net Abatement Scenario Benefit the Option Benefit Ger stoves: Scenario 1a Reduce startup emissions (propellant) 865.8 –53 918.8 Scenario 1b Reduce startup emissions (lighting technique) 1,599.0 –35.8 1,634.8 Scenario 2 Certified stoves 1,605.0 –0.3 1,605.3 Scenario 3 SCC + stoves 1,028.5 36.7 991.8 Scenario 4 Electric heating of Gers 1,802.9 1,410.0 392.9 Scenario 5 Relocation into apartments 597.1 4,094.0 –3,496.9 Heat only boilers (HOB) emission reduction: Scenario 6 HOB improvements 19.7 5.9 13.8 Road dust reduction: Scenario 7 Street cleaning 66.9 66.7 0.2 Windblown dust reduction: Scenario 8 Greening 58.1 2.5 55.6 Source: World Bank 2011. gave significant net benefits. The long-term option of EVALUATION OF EXPERIENCES WITH moving Ger households into apartments was very costly, THE USE OF THE METHODOLOGY IN whereas road cleaning and city greening had only limited THE ULAANBAATAR HEALTH IMPACTS effects on health impacts. REDUCTION PROJECT The Ulaanbaatar project was an urban area air pollution In order to determine the cost of delaying, or not apply- abatement project. There are a multitude of air pollu- ing, short-term measures compared with only applying tion source sectors that each have a significant contribu- a long-term measure (that is, the cost of inaction), the tion to the urban air pollution, although it was clear that health benefits associated with immediate interventions the coal burning in the Ger stoves was a major source were compared with long-term options (see figure D.7). (see figure D.8 illustrating the reduction potential in the One illustration shows, for example, that by only apply- Ger stoves sector, for heat-only boilers (HOBs) and soil/ ing the relocation into apartment options and not apply- dust both with regard to PM10 and PM2.5 reduction). Such ing the certified stoves option, the accumulated lost health a project requires a rather comprehensive data basis, to benefit for the 2010–13 period would be about $1 billion enable abatement option studies to be carried out. for Ulaanbaatar city due to lives and years of life that would be lost during the long interval before the popula- The project was carried out over three years. At the start tion could be shifted into apartments. Following the illus- of the project, there was already an air quality monitoring trations of the lost health benefits (cost of not applying system in place (albeit in the traditional city center only), immediate abatement options), the Ulaanbaatar govern- an emissions inventory had been established, and an air ment decided to go ahead with the replacement program pollution model had been established that had provided a for certified stoves. first view of the air pollution situation in Ulaanbaatar and Enhancing the World Bank's Approach to Air Quality Management 79 FIGURE D.7. THE COST OF DELAYING SHORT-TERM MEASURES— COMPARING HEALTH BENEFITS BETWEEN THE CERTIFIED STOVES AND THE RELOCATION INTO APARTMENTS SCENARIOS IN ULAANBAATAR Scenario 2 certified stoves 300 Scenario 5 relocation into apartments 250 Health benefit (mill USD) 200 Accumulated health benefit for 150 2010-2023 period Senario 2 $ 1605 million 100 Senario 5 $ 597.1 million 50 $ 1007.9 million 0 10 12 10 16 18 20 12 20 20 20 20 20 20 20 Source: World Bank 2011. its spatial and temporal variations (Guttikunda 2007). This such as fuel consumption, population distribution, and provided a starting base for the study. However, it was evident emission factors. from the previous results that substantive improvements had to be made to the monitoring system, the emissions inven- Air pollution modeling: An improved state-of-the-art tory, and the air pollution model in order to present a more urban air pollution model was established for the city area. accurate picture of the air pollution situation. Also, it was seen as desirable to attempt to establish, through a special Air pollution and health impact study: An epi- health impacts study, locally valid dose-response relation- demiology study was carried out using one year of ships between the exposure of the population to PM and mortality and hospitalization data for eight different dis- various health outcomes and a locally based WTP study. tricts (including a medical record of more than 52,000 patients) in Ulaanbaatar together with the PM measure- The following improvements were made in the data basis ment data. for evaluating the cost-effectiveness of various investment options. As a result of the improvements, it was possible to estab- lish a situation where the results of the air pollution mode- Air pollution monitoring: Several additional PM ling agreed with the measurement data within acceptable measuring sites were established (through relocation from accuracy. Based upon this, one could establish a quanti- old to new sites). The measurement equipment used was tative analytical process where the comparisons between simple and inexpensive, and thus of limited accuracy. abatement costs and health benefits could be established However, calibration activities were carried out with state- with acceptable confidence, as well as differences in cost- of-the-art equipment, to bring the measurement results up effectiveness of the various abatement options. to an acceptable standard. Since the pre-project data basis for the city was consid- Air pollution emissions: An updated inventory of ered to be lacking in completeness and quality, improve- emissions was carried out based upon updated basic data ments needed to be made, and this led to the need for a 80 Clean Air and Healthy Lungs FIGURE D.8. NEEDED REDUCTION IN PM10 AND PM2.5 CONCENTRATION LEVELS TO REACH AIR QUALITY STANDARDS Population weighted average concentration for given emission reductions, PM10 500 –30% –50% –80% –13% –12% –26% 400 –37% –31% –69% Gers 300 46% 200 Soils 38% 100 HOBs WHO IT-1 Mongolian standard Others WHO guideline value 0 baseline Ger stoves HOBs Soils all 3 sectors Population weighted average concentration for given emission reductions, PM2.5 300 –30% –50% –80% 250 –18% –15% –24% –65% 200 Gers –48% 60% 150 100 Soils 19% 50 HOBs WHO IT-1 Others Mongolian standard WHO guideline value 0 Baseline Ger stoves HOBs Soils all 3 sectors Source: World Bank 2011. prolonged project duration (three years)24 to establish the Results of the Investment Program: Ger Stove needed data base and carry out the analytical procedures Replacement and Air Pollution Reduction. Since and calculations that were necessary. The project can 2010, a program of Ger stove replacement has been be considered a success, in that a clear basis for guiding carried out. It started with funding by the Millennium investment strategies was established. Challenge Corporation (MCC) and gradually expanded with funding through the Ulaanbaatar Clean Air Project, which also included other abatement measures. The joint 24 Activities conducted during the three-year period included: identification of MCC and World Bank–funded stove replacement pro- counterparts; identification of monitoring equipment; location of monitors and gram resulted in the placement of 69,422 new stoves in establishment of a monitoring network; one full year of monitoring at all loca- tions to allow for capture of seasonal variations; analysis of monitoring data; 2011, followed by 33,533 new stoves in 2012 and 38,357 scientific research on sources and impacts; and writing a final report. new stoves in 2013. Enhancing the World Bank's Approach to Air Quality Management 81 FIGURE D.9. DEVELOPMENT IN PM10 TABLE D.6. AVERAGE PM10 CONCENTRATIONS CONCENTRATIONS AT AT ZUUL AIL STATIONS 2008–13 ZUUL AIL STATION IN Average PM10 ULAANBAATAR 2010–13 Year/Period Concentration 2010 2011 2012 2013 1800 2008/09 (AMHIB pilot period) 558 1600 2010 (June–Dec. only) 1400 2011 453 1200 1000 2012 495 800 2013 357 600 Source: Prepared by authors based on information received from Lodoysamba 400 Sereeter, Professor at National University of Mongolia. 200 0 the program is being fully implemented in order to ensure n b ch ril ay ne ly g pt ct ov ec Ja Fe Au the full effect on PM concentration reduction and thereby Ju O Ap Se M N ar D Ju M Source: Prepared by authors based on information received from Lodoysamba optimal health improvements. Sereeter, Professor at National University of Mongolia. This means that by the end of 2013, most of the Ger households had received new stoves (that is, about 141,312 CASE EXAMPLE 3: Gers out of 180,000 at the time). In addition, 21,000 Ger PERU ENVIRONMENTAL households had received better insulation. Also, about DEVELOPMENT POLICY LOAN 200 of the heat-only boilers had been abolished (mainly PROGRAM (2009–13) through funding from the Japanese International Coop- eration Agency), and the HOB-served buildings had been INTRODUCTION connected to the district heating system supplied by the The Government of Peru has made broad strides in power stations in the city. improving air quality, a significant cause of disease and death in Peru. This case study describes the World Bank’s Continued monitoring of PM10 and PM2.5 has so far shown support to Peru in addressing air pollution, as part of a some reduction in the PM concentrations in Ulaanbaatar broader process characterized by continual dialogue compared with the AMHIB baseline from 2008–09 com- between the Bank and the government of Peru, a solid pared with years 2010–13. It is well known that differ- analytical foundation, and engagement of relevant stake- ing climatic and meteorology conditions in Ulaanbaatar holders across the Peruvian society. vary considerably between years and that a fairly long period of monitoring over several years may be needed Between 2009 and 2013, the World Bank provided to establish a trend in the concentrations. Yet indications financial assistance to the government of Peru through are that the average yearly PM10 concentrations have been a program of three development policy loans (total- reduced. This is, for example, exemplified at the Zuul Ail ing $475 million), aimed more broadly at supporting monitoring station in a Ger area that was established dur- the government’s efforts to strengthen environmental ing the AMHIB period (see figure D.9 and table D.6). governance and institutions in Peru and incorporat- ing environmental sustainability into the development It should also be noted that the stove exchange process agenda of key economic sectors (including mining, has met with challenges. Indications are, for example, that urban transport, energy, and fisheries). By the end of new stoves provided to some families are sold in areas out- the DPL program, air quality in Peru had undergone a side of Ulaanbaatar, while the families keep using their marked improvement, particularly in the country’s most old stoves. Further control has to be applied to ensure that polluted cities. 82 Clean Air and Healthy Lungs FIGURE D.10. COST OF ENVIRONMENTAL TABLE D.7. ESTIMATED ANNUAL HEALTH DAMAGE IN PERU HEALTH IMPACT OF URBAN 2.5 AIR POLLUTION FROM 2 PARTICULATE MATTER Cost (bn. soles) Total Total 1.5 Health End-Points Cases/Year DALYs/Year 1 Premature mortality 3,900 29,253 Chronic bronchitis 3,812 8,386 0.5 Hospital admissions 12,834 205 0 Emergency room visits/ 251,765 1,133 Water supply, Outdoor air Lead exposure Indoor air outpatient hospital visits sanitation, and pollution pollution hygene Restricted activity days 43,347,360 13,004 Source: World Bank 2007b. Lower respiratory illness in 533,457 3,467 children Respiratory symptoms 137,957,686 10,347 HEALTH AND ECONOMIC DAMAGE Total 65,796 FROM AIR POLLUTION IN PERU Source: Larsen and Strukova, 2005. Air pollution is one of the most widespread and serious envi- ronmental problems in Peru, particularly in urban centers, where almost 50 percent of the population lives. Compared FIGURE D.11. HEALTH IMPACTS OF with other cities in Latin America, the levels of air pollution AMBIENT AIR POLLUTION were higher than in Mexico City and Santiago. In the Lima- PER UNIT OF INCOME IN Callao metropolitan region, average particulate matter con- LIMA-CALLAO1 centrations were more than 80 μg/m3. The two major air 300 Poor Non-poor pollutants of concern to health in Peru are particulate mat- ter and lead, both of which come principally from transport 250 and industrial sources. A Country Environmental Analysis 200 conducted in 2007 by the Bank in collaboration with the government of Peru estimated that health damage from 150 environmental problems accounted for about 2.8 percent of GDP and over 70 percent of the total cost of environmental 100 degradation in Peru. Poor air quality—from particulate mat- ter and ambient lead pollution—accounted for the largest 50 share of the health damage, estimated at approximately 2.81 0 billion soles or 1.38 percent of GDP in 2003 (see figure D.10). “High-case” Base-case “Mid-case” scenario scenario scenario The problem of air pollution is most serious in industrial Source: Larsen and Strukova, 2005. 1 The health impacts are indexed to 100 for the non-poor in the base-case corridors such as Lima-Callao, which bears more than scenario. The health impact per person is divided by income per person, 75 percent of the cost of health impacts of poor ambient normalized to 100. air quality in Peru. Ambient air pollution caused the pre- mature death of about 3,900 Peruvians and accounted for by environmental health problems in Peru and are least the loss of approximately 66,000 disability-adjusted life able to afford measures to address its health impacts: the years per year, attributable to mortality (44 percent) and CEA found that in the Lima-Callao area, health impacts chronic bronchitis (13 percent), restricted activity days of air pollution could be more than three times higher for (20 percent), and respiratory symptoms (16 percent) (see the poor than for the non-poor, relative to income (Larsen table D.7). Poor people are disproportionately affected and Strukova, 2005) (see figure D.11). Enhancing the World Bank's Approach to Air Quality Management 83 FIGURE D.12. MARGINAL COSTS AND transport system in Lima). More generally, most poli- BENEFITS OF ACTIONS TO cies had implications for the welfare of transport users and other affected sectors (for example, an increase in REDUCE PARTICULATE the price of cars or a limitation on their use). Rank- MATTER EMISSIONS IN PERU ing options along a cost-effectiveness curve to show how 18 CNG much reduction each option can make and at what cost 16 per ton provides policy makers with a clear description 14 of the complex analysis required to evaluate different 12 alternatives. Of the 12 options considered, the follow- Million USD 10 ing 5 were amenable to such analysis (see figure D.12). 8 In evaluating the options, the damage costs associated 6 with a ton of emissions of particulate matter and other Low sulfur diesel particles is compared with the cost of a specific option Retrofit 4 I & M Program Industry for abating that ton. 2 » Introduction of low-sulfur diesel 0 0 5 10 15 20 25 » Inspection and maintenance (I & M) programs Health cost redution (%) » Retrofit particle-control technology Source: ECON 2005. » A shift in due course from low-sulfur diesel to com- pressed natural gas (CNG) » Reduction of emissions from industry sources ANALYSIS OF INTERVENTIONS TO COMBAT AMBIENT AIR POLLUTION ACTIONS SUPPORTED BY THE PERU Having identified ambient air pollution as a significant cause PROGRAMMATIC ENVIRONMENTAL of premature death and economic damage in Peru, a study DPLS TO ADDRESS AIR POLLUTION IN was conducted that examined 12 options for reducing ambi- PERU ent air pollution (ECON 2005). The following 12 options As part of the program of Environment DPLs, the World were considered: introduce low-sulfur diesel; encourage use Bank supported specific policy and institutional reforms of gasoline cars at the expense of diesel through various to address air pollution under the cover of two objectives: tax incentives; convert some gasoline/diesel cars to natural strengthening the framework for environmental quality gas; convert some vehicles to ethanol or biofuel; develop a standards, emission levels, and environmental monitoring new public transport system in Lima; provide tax incentives and incorporating environmental sustainability principles in to scrap older high-use cars (for example, taxis); strengthen the urban transport sector and in industry, the main drivers inspection and maintenance programs; retrofit catalytic of air pollution in Peru (see table D.8). Some actions sup- converters on cars and particle control technology on diesel ported specific interventions to strengthen the capacity of vehicles; ban imports of used cars for taxi use; ban use of relevant Peruvian institutions to address air pollution; others diesel cars and/or two stroke engines as taxis; implement were intended to more broadly enhance policies aimed at various city planning interventions such as “green traf- improving governance relating to air quality management. fic light waves” and bike lanes; and introduce measures to reduce emissions from industry sources. The first DPL, a $330 million loan approved in early 2009, supported government actions—notably, issuance At the time of examining the above options, there were of a decree to establish air quality standards and maxi- some important aspects considered. Some of the options mum permissible levels for air emissions; reduction of the were either in the process of being implemented or con- sulfur levels in diesel fuel and establishment of a scheme sidered by the Peruvian authorities. Some of the options for converting vehicles from diesel to natural gas; enact- would not generally be considered for environmental ment of a law requiring reduction of sulfur content in reasons (for example, the option to develop a new public diesel; issuance of a decree establishing requirements for 84 Clean Air and Healthy Lungs TABLE D.8. PERU DPL PROGRAM DEVELOPMENT OBJECTIVES, KEY INDICATORS, BASELINE AND PROGRAM OUTCOMES (Continued) Program Actions Supported by Under Each DPL Development Program Outcome Objective Indicators Baseline DPL I DPL II DPL III Program Outcomes Strengthen the Air quality data for Data were not Issuance of decree that Publication and Regulations for a framework for the Lima-Callao disseminated in set environmental dissemination on harmonized air quality environmental Metropolitan Region real time quality standards SENHAMI’s and monitoring network were quality standards, is widely published (ECAs) and DIGESA’s issues in 2010. Data is emission and disseminated (in maximum websites of daily air disseminated in real time levels, and real time) through an permissible levels quality monitoring for Lima-Callao http:// environmental integrated monitoring (LMPs) for air data for the cities of www.digesa.sl d.pe/ monitoring network emissions Lima and La Oroya PVSCA/ Air quality contingency Contingency plans Contingency plans are plans are developed are in place in place for La Oroya, and implemented only for La (Dióxido de azufre y (when pollution levels Oroya material particulado) and Enhancing the World Bank's Approach to Air Quality Management largely exceed quality Clean Air Action Plans standards) in the 5 most are under development polluted cities in Peru: for Iquitos, Chimbote, Lima, Arequipa, Trujillo, and Ilo Chimbote, Ilo, and La Oroya Mainstreaming of At least 80,000 vehicles 35000 Establishment of a Continuance of the More than 86,000 vehicles environmental converted to natural gas vehicles temporary national promotion and converted to natural gas sustainability converted to scheme promoting implementation of principles in the natural gas vehicle conversion vehicle conversions urban transport from diesel to to natural gas, as sector natural gas evidenced by an allocation in the proposed budget for calendar year 2010 90 service stations are Zero stations in More than 90 service installed and operating Lima- Callao stations supply natural supplying natural gas in supply natural gas in Lima Lima gas (Continued ) 85 86 TABLE D.8. PERU DPL PROGRAM DEVELOPMENT OBJECTIVES, KEY INDICATORS, BASELINE AND PROGRAM OUTCOMES (Continued) Program Actions Supported by Under Each DPL Development Program Outcome Objective Indicators Baseline DPL I DPL II DPL III Program Outcomes At least 30% of gas Zero stations in Enactment of law Issuance of decree Adoption of an By 2010, all stations in stations (approx. 750) Lima-Callao establishing that prohibited the investment Lima-Callao and by in main cities supplying supply clean reduction of sulfur supply of diesel with plan for the 2012 all stations in clean diesel (with less diesel in diesel to 50ppm more than 50 ppm modernization of Arequipa, Puno, Cuzco than 50 ppm of sulfur by 2010 of sulfur content in PETROPERU’s and Madre de Dios content) by 2010 the cities of Lima refinery in Talara supplied clean diesel with and Callao that reduces the less than 50ppm sulfur sulfur content of content the diesel fuel. Vehicle inspection and About 60,000 Issuance of decree Creation of Implementation By 2010, 585,000 vehicles maintenance system vehicles which established regulation for the of an effective were inspected in Lima operating in Lima and inspected in requirements for implementation of inspection and in the three largest cities Lima diesel vehicles to a vehicle inspection maintenance access to economic and maintenance system in Lima incentives, and system in the Lima and three standards for the metropolitan region additional cities scrapping process through the introduction of an information and communication system Source: World Bank 2013d. Clean Air and Healthy Lungs FIGURE D.13. ANNUAL AMBIENT PM2.5 CONCENTRATIONS IN LIMA-CALLAO 2003–12 (μg/m3; 3-YEAR MOVING AVERAGES) North East South Callao Center 90 80 70 60 50 40 30 20 10 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: Macizo and Sanchez (forthcoming). diesel vehicles to access economic incentives; and stand- end of the last DPL; they intended to have 30 percent of ards for the scrapping process (World Bank 2009a). the fuel stations in major cities supplying clean low-sulfur diesel by 2010, and by 2010 all stations in Lima-Callao The second DPL, a $70 million loan approved in the sum- supplied clean diesel, with 100 percent coverage achieved mer of 2009, supported actions to publish and disseminate in four additional major cities by 2012; and the number air quality monitoring data in the highly polluted cities of of service stations supplying natural gas in Lima rose from Lima and La Oroya; issuance of a decree to prohibit the zero to over 90. supply of high-sulfur diesel in the metropolitan areas of Lima and Callao; institutional measures to ensure con- AIR QUALITY IMPROVEMENTS AND tinued funding of the implementation of vehicle conver- ENVIRONMENTAL HEALTH sion; and regulations for the implementation of a vehicle OUTCOMES ACHIEVED inspection and maintenance system that was designed to By the time the DPL series closed, Peru had experienced remove highly polluting vehicles from the streets of Lima- significant improvements in air quality. As shown in Callao (World Bank 2009b). figure D.13, ambient levels of PM2.5 air quality in all zones of Lima-Callao displayed a downward trend. In the Cen- The third DPL, approved in the summer of 2010, was a tral zone, PM2.5 levels decreased from about 75 μg/m3 in $75 million loan that supported the implementation of 2009 to less than 40 μg/m3 in 2012. It is estimated that the the vehicle inspection and maintenance system, the crea- improvements in air quality between 2001 and 2003 and tion of an information and communication system and between 2010 and 2012 resulted in a decline in popula- the adoption of an investment plan for the modernization tion exposure of 15 percent, notwithstanding population of PETROPERU’s refinery that reduces the sulfur con- growth. Given the association between PM2.5 and health, tent in diesel (World Bank 2010). the air quality improvements achieved may be linked to improvements in health outcomes. Table D.8 summarizes the actions that the DPL program supported in helping the government address air pollution. In an effort to quantify the overall impact of the Envi- As shown in the table, targets were achieved and in many ronment DPL program on environmental health out- cases exceeded: from a baseline of 60,000 inspected vehi- comes, a second cost of environmental degradation cles in Lima, almost 10 times as many vehicles (585,000) study for Peru was conducted in 2013. Recent evidence were inspected; they intended to have 80,000 CNG vehi- of the impact of environmental conditions on human cles on the streets and ended up converting 83,000 by the health, however, makes comparisons between the 2006 Enhancing the World Bank's Approach to Air Quality Management 87 and 2012 data very difficult. Additional contributing Peru, not only in cities with a population greater than factors to this difficulty include introduction of meth- 100,000 inhabitants (World Bank 2013d). The 2013 odological improvements in the 2013 study; the use study showed that if the methodology used on the 2012 of additional parameters that were absent in the 2006 data were applied to the 2006 data, the resulting cost analysis; increased urbanization; and inclusion of health of environmental degradation would have been much effects among the entire urban and rural population of higher than estimated. APPENDIX E IDEA NOTE FOR METHODOLOGY FOR INTEGRATION OF SLCP COMPOUNDS IN AIR QUALITY MANAGEMENT MODELING Some SLCPs also have impacts on human health. This INTRODUCTION is true for BC, OC, and sulfate as well as ozone. A World Short-lived climate pollutants (SLCPs) include particulate Bank report (2013e) quoted the following from the UN matter (PM) pollutants such as black carbon (BC), organic Environment Programme (UNEP 2011): carbon (OC), sulfate and nitrate; gaseous pollutants such as methane (CH4), nitrous oxide (N2O), and ozone (O3); The United Nations Environment Programme (UNEP) and some hydrofluorocarbons (HFCs). Precursor gases estimates that fast action to reduce SLCP emissions to some of these pollutants that are not included in the could avoid an estimated 2.4 million premature deaths SLCP definition include NOx, VOC, and CO (for ozone from outdoor air pollution annually by 2030 and avoid and organic carbon) and SO2, NOX and NH3 (for sulfates over 30 million tonnes of crop loss per year. and nitrates). Air quality management aims to reduce the health SLCPs, as their group name implies, have an impact on impacts of air pollution in a cost-effective way. It is poten- global climate. They have fairly short lifetimes in the tially useful, when analyzing the cost-effectiveness of atmosphere compared with long-lived climate pollutants various air pollution abatement options, to include the cli- (mainly CO2). The reason for the recent focus on SLCPs mate impacts of SLCPs and try to balance them with the and their control is precisely their short life-time: the con- health impacts of SLCPs and of other health-affecting trol of their emissions and concentrations in the atmos- air pollutants—that is, to balance the health and climate phere can bring about important reductions in the global effects of air pollution control options. total warming potential on a much shorter time scale than control of CO2 would. For example, as summarized by Bond et al. (2013): THE HEALTH EFFECTS A large fraction of the atmospheric black carbon OF SLCPs concentration is due to anthropogenic activities. Con- The health effects of SLCPs are mainly due to PM com- centrations respond quickly to reductions in emissions pounds (BC, OC, sulfate) and to ozone. because black carbon is rapidly removed from the atmosphere by deposition. Thus, black carbon emis- BLACK CARBON AND PM sion reductions represent a potential mitigation Two fairly recent works have examined the health effects strategy that could reduce global climate forcing from of black carbon and other PM pollutants. A study by Smith anthropogenic activities in the short term and slow et al. (2010) looked at both ozone and PM pollutants: the associated rate of climate change. In this report we review the health effects of three Some SLCPs are warming agents, namely BC, ozone, and short-lived greenhouse pollutants—black carbon, HFCs, while others are cooling agents, namely OC and ozone, and sulfates. We undertook new meta-analyses sulfate. A PM mixture containing BC, OC, and sulfate, of existing time-series studies and an analysis of a which is often the case in urban and regional air pollution, cohort of 352,000 people in 66 U.S. cities during could then be warming or cooling, depending upon the 18 years of follow-up. This cohort study provides relative amounts of the agents in the mixture. This would estimates of mortality effects from long-term exposure be an important aspect of assessing the climate effect of to elemental carbon, an indicator of black carbon PM pollution. mass, and evidence that ozone exerts an independent Enhancing the World Bank's Approach to Air Quality Management 89 risk of mortality. Associations among these pollut- available toxicological studies suggested that BC may ants make drawing conclusions about their individual not be a major directly toxic component of fine PM, health effects difficult at present, but sulfate seems to but it may operate as a universal carrier of a wide have the most robust effects in multiple-pollutant mod- variety of chemicals of varying toxicity to the lungs, els. Generally, the toxicology of the pure compounds the body’s major defense cells and possibly the sys- and their epidemiology diverge because atmospheric temic blood circulation. A reduction in exposure to black carbon, ozone, and sulfate are associated and PM2.5 containing BC and other combustion-related could interact with related toxic species. Although sul- PM material for which BC is an indirect indicator fate is a cooling agent, black carbon and ozone could should lead to a reduction in the health effects associ- together exert nearly half as much global warming as ated with PM. carbon dioxide. The complexity of these health and climate effects needs to be recognized in mitigation Thus, the studies find some evidence that BC is a better policies. indicator of harmful effects of PM than undifferentiated PM mass (WHO 2012) or that BC shows stronger effects More specific key messages in their report regarding mor- for elemental carbon than for undifferentiated fine par- tality effects of sulfate and black carbon (measured as ticles (PM2.5) (Smith et al. 2010). But the evidence is not “black smoke”) were: clear and conclusive. » “Meta-analyses of time-series studies of short- term exposure suggest larger mortality effects per The WHO 2012 Task Force study thus concluded as fol- unit mass of sulfate than of black smoke.” lows in the Executive Summary: » “Our analysis of a 66-city, 18-year nationwide The Task Force on Health agreed that a reduction in U.S. cohort provides estimates of the mortal- exposure to PM2.5 containing BC and other combus- ity effects of long-term exposure to elemental tion-related PM material for which BC is an indirect carbon, the best available measure of black indicator should lead to a reduction in the health carbon. This analysis shows stronger effects for effects associated with PM. The Task Force recom- elemental carbon than for undifferentiated fine mended that PM2.5 should continue to be used as the particles (PM2·5), but the model estimates are primary metric in quantifying human exposure to PM unstable with respect to inclusion of other pol- and the health effects of such exposure, and for pre- lutants.” dicting the benefits of exposure reduction measures. The use of BC as an additional indicator may be use- A WHO Task Force (WHO 2012) concluded: ful in evaluating local action aimed at reducing the This report presents the results of a systematic population’s exposure to combustion PM (for exam- review of evidence of the health effects of black ple, from motorized traffic). carbon. Short-term epidemiological studies pro- vide sufficient evidence of an association of daily The recommendation that PM2.5 should continue to be variations in cardiopulmonary hospital admissions. used as a primary metric for PM exposure and health Cohort studies provide sufficient evidence of all effects quantification reflects the fact that there is at present causes and cardiopulmonary mortality with long- not a good enough basis for providing exposure-response term average BC exposure. Studies of short-term relationships on health for BC, and also that PM2.5 would health effects suggest that BC is a better indicator capture the main health impact associated with PM. of harmful particulate substances from combustion sources (especially traffic) than undifferentiated par- OZONE ticulate matter (PM) mass, but the evidence for the The health effects of ozone are well established and have relative strength of association from long-term stud- been the basis for developing air quality guidelines and ies is inconclusive. The review of the results of all standards for ozone. 90 Clean Air and Healthy Lungs the various agents differs from those looking at the past CLIMATE EFFECTS OF SLCPS period 1750–2011. The cooling or warming potentials of various SLCPs is discussed in the recent reports of the Intergovernmental Another way of comparing GHGs against each other is to Panel on Climate Change in terms of the radiative forcing use their global warming potential (GWP).28 GWP is a rel- (RF)25 of the various GHG compounds (gases and partic- ative measure of how much heat a greenhouse gas traps in ulates). All SLCPs are significant warming agents, except the atmosphere. It compares the amount of heat trapped the sulfate, nitrate and organic carbon (OC) contents of by a certain mass of the gas in question to the amount of PM, which are cooling agents. Based on their total emis- heat trapped by a similar mass of carbon dioxide. A GWP sions between 1750 and 2011, the RFs of methane (CH4), is calculated over a specific time interval, commonly 20 or Black carbon (BC), nitrous oxide (N2O) and tropospheric 100 years. ozone (O3T) have been calculated to be about 38, 38, 11, and 15 percent, respectively of the RF of CO2, the domi- Table E.1 shows that, for a 20-year time horizon, methane nant warming agent in the atmosphere (IPCC 2013).26 has a GWP 84 times higher than CO2, while black carbon has a GWP 3,200 (uncertainty 270–6,200) times higher Tropospheric ozone is produced by chemical reactions, than CO2—in other words, per mass unit, methane and partly associated in the troposphere with methane and black carbon are much stronger heating agents than CO2. CO, NOX, and NMVOC and in the stratosphere with halocarbons and N2O (where it is slightly cooling). The These examples of RFs and GWPs are provided here tropospheric ozone represents a significant warming in order to visualize the warming (and cooling) poten- agent. tials of the different pollutants and to show their relative Regarding PM pollutants, black carbon represents a importance in the various scientific assessment method- significant warming agent—third after CO2 and meth- ologies. If the SLCPs are to be included in AQM method- ane, while sulfate, nitrate and organic carbon represent ologies, the different SLCP emission strength in different significant cooling agents. The resulting RF of PM source types must be weighted against each other as well emissions from various source types, such as indus- as against long-lived GHGs such as CO2 and methane, trial boilers, diesel vehicles, cookstoves, and so on, will where regional effects should also be considered. When depend upon the mix of all these compounds in the integrating SLCPs in an AQM methodology, the weight- emissions. ing method between SLCPs and other GHGs must be developed on a scientific basis, to arrive at representative Similarly, the integrated radiative forcing of the global global warming “factors,” so that different sources can be emissions of various compounds can be compared in compared. For example, Bond et al. (2013) provide emis- terms of the effects of their emissions on the radiative sion mix estimates from various sources that will be useful balance into the future. The integrated RF of the global in arriving at such factors. emissions of each compound in 2000, has been calculated over various future periods, for example, 20 years and 100 years (IPCC 2007).27 The relative integrated future RF of 28 Greenhouse gases are often compared with one another using their warming potency. For example, methane is a more powerful warming agent than CO2. 25 Radiative forcing measures the change in Earth’s energy budget (that is, the In most GHG accounting, one ton of methane is equal to 25 tons of CO2. warming effect) caused by the total atmospheric concentration of a GHG or par- This, however, assumes a 100-year period. Because methane only lasts in the ticulate. For instance, the radiative forcing of CO2 (over pre-industrial conditions) is atmosphere for 12 years, the impact ratio changes as a variable of time: over a 1.66 watts per square meter (W/m2)*. Methane’s radiative forcing is 0.48 W/m2. In period of 20 years, for example, one ton of methane has the warming effect of other words, the current atmospheric concentration of methane causes a warming 72 tons of CO2. The warming impact of a climate pollutant over a designated effect equal to 29 percent of the effect caused by the current concentration of CO2. timeframe, as a ratio of an equal mass of CO2, is known as global warming 26 IPCC (2013) Figure 8.17, p. 698. potential. GWPs of 20 years or less are better indicators of the short-term cli- 27 IPCC (2007) Figure 2.22, p. 206. mate impact of emissions. Enhancing the World Bank's Approach to Air Quality Management 91 TABLE E.1. LIFETIMES, RADIATIVE EFFICIENCIES, AND GWPs RELATIVE TO CO21 Global Warming Potential for Given Time Horizon Chemical Radiative Efficiency Compound Formula (W/m–2 ppb–1) 20-year 100-year Carbon monoxide CO2 1.4×10 –5 1 1 Methane CH4 3.6×10–4 84 28 Nitrous oxide N2O 3.00×10–3 264 265 Black carbon 3,200 (270–6,200) 900 (100–1,700) CFC-13 CClF3 0.25 10,800 14,400 Black carbon (non-standard 4,470 1,055–2,240 method) Soot (non-standard method) 2,530 840–1,280 Source: IPCC (2013). Numbers in brackets are uncertainty intervals. 1 IPCC, 2013, Tables 8.A.1-8.6, pp. 731–40. A further step that is presently often included in AQM INTEGRATING SLCPs IN AQM analysis is to take account of greenhouse gas emissions METHODOLOGIES in parallel with the air pollution concentration exposure Present methodologies for air quality management in and impact assessment. Abatement for air pollution often urban areas are geared toward assessing the exposure results also in reduced GHG emissions, referred to as a of an urban population to certain air pollutants and cal- “co-benefit.” It is possible to compare the reduced health culating the reduced exposure and subsequent reduced impact from a certain abatement action (in terms of (health) impact of various abatement measures operated reduced exposure, or reductions in the number of induced on the main emission source categories or individual deaths per year) with the parallel reduction in GHG emis- sources affecting the urban area. Present scientific under- sions (for example, in tons per year). Methodologies for standing is that in most cases it is the concentration of comparing/weighting these two entities can be devised, particulate matter, PM, in the air that dominates the to provide a quantitative basis for decisions, for selecting health impact. the “best” abatement strategy, taking both local air pollu- tion and global climate into account. Figure E.1 illustrates the concept of analytical AQM. Pol- lution drivers (societal activities resulting in emissions to Incorporating consideration of SLCPs in AQM method- air) create air pollution in the form of concentrations in ology would involve introducing the health and climate air of various air pollution compounds. This often creates impacts of SLCPs. Based on the discussion above, the fol- an impact in the form of various types of damage (impact lowing can be stated: on people’s health, on the environment, on materials). 1. Regarding the health impacts of SLCPs: SLCPs with Abatement actions leading to reduced emissions result in health effects are the PM species (black carbon, reduced concentrations and damage. organic carbon, and sulfate), as well as ozone. The health impacts of the PM species can be Analytical AQM represents a standard methodology for represented by undifferentiated PM, such as PM2.5 or analyzing effects of abatement, comparing them with the PM10 in the form of mass concentration (μg/m3), costs, and on this basis selecting the most effective strate- which is already the most used air pollution met- gies to reduce the pollution. Software tools exist that ena- ric in AQM work. The health impact of the SLCP ble efficient AQM analytical work. ozone is also a well-used metric in this work. 92 Clean Air and Healthy Lungs FIGURE E.1. PRINCIPAL CONCEPT OF ANALYTICAL AIR QUALITY MANAGEMENT Pollution drivers/ emissions Monitoring Dispersion of air modeling pollution Solutions to Air pollution reduce Emissions concentrations Air pollution pollution Abatement options Exposure Optimized abatement costs strategies Cost and benefit Damage–Effects value comparison Impact/ health damage Source: World Bank 2011. 2. Regarding the climate impact of SLCPs: The climate A concept of the further procedure of the AQM method- impact of chemical compounds in air can be repre- ology would be: sented by their radiative force or their global warm- » The population exposure to air pollutants, mainly to ing potential. The emissions from air-polluting PM2.5 and/or PM10, would be calculated according sources consist of a mixture of compounds, differ- to state-of-the-art AQM procedures (such as applied ent for each type of source. The relevant climate in the Ulaanbaatar air quality management study impact metric to be included in AQM methods, (World Bank 2011). The emissions from the source(s) to represent the climate impact of the emissions, in question are assessed, and their dispersion is mod- would be a weighted average of the source’s RF eled to produce PM concentration fields and popu- or GWP, weighted according to the relative abun- lation exposure in the area as input to health impact dance of the various GHG/SLCP agents in the assessments. The reduction in emissions as a result emissions. The weighting procedure must be de- of candidate abatement options are calculated, and veloped according to climate impact science and it produces a reduction in adverse health impact. understanding as well as to knowledge of the » The climate impact of the emissions from the mixture of GHG agents in each type of source. source(s) is calculated by the procedure described Examples of a basis for this weighing are includ- above as a concept. Various abatement options ed in table E.1. In this weighting procedure, both produce different climate impact reductions. SLCPs and long-lived GHGs, mainly CO2, must » The “best” abatement strategy is selected by all be taken into account. combining and weighting the health impact and Enhancing the World Bank's Approach to Air Quality Management 93 FIGURE E.2. POPULATION WEIGHTED AVERAGE CONCENTRATION FOR GIVEN EMISSION REDUCTIONS, PM2.5 300 –30% –50% –80% 250 –18% –15% –24% –65% 200 Gers –48% 60% 150 100 Soils 19% 50 HOBs WHO IT-1 Others Mongolian standard WHO guideline value 0 Baseline Ger stoves HOBs Soils all 3 sectors Source: World Bank 2011. climate impact reductions. The weighting proce- for different PM sources (coal combustion in Ger tents, dure has to be developed. In a simple situation, the heat-only boilers, and soil suspension). The columns option that produces the largest reduction in both show the reduction in PWE compared with the baseline health impact and climate impact will be chosen. situation. More often, however, the situation would not be so straightforward. It would also be possible to include columns showing the reductions in climate impact next to the PM2.5 PWE An illustration of air pollution exposure reductions for columns, as an illustration of the combined situation. various abatement options is shown in figure E.2. It shows The selection of the “best” abatement option might be the population-weighted exposure (PWE) to PM2.5 in straightforward, or it might need a further developed Ulaanbaatar for various emissions reduction options and selection method. 94 Clean Air and Healthy Lungs APPENDIX F PROJECT TYPOLOGIES, AIR POLLUTION ASPECTS, AND SOME RELEVANT EMISSIONS DATA REQUIREMENTS Sector Subsector Activity Type Type Type Air Pollution Aspects Energy Energy efficiency Improving efficiency Reduction in emissions and pollutant concentrations resulting in boilers and power from the efficiency improvements. plants, incl. conversion Main pollutants: PM, BC, SO2, NOX to combined heat and Data related to emissions: Emissions before and after the power plants (CHP) improvements. District heating and Typically replacing smaller air polluting units (for example, cooling systems boilers) with a larger/more energy efficient power plant. Main pollutants: PM, BC, SO2, NOX. Data related to emissions: The amount and location of replaced emissions and of new/increased emissions from the power plant. Entities providing energy Enterprises providing expert services for industries and efficiency services companies to improve energy efficiency and reduce energy consumption. Typically related to reduced combustion of fossil fuel. Would ideally result in reduced emissions from their clients’ operations. Main pollutants: PM, BC, SO2, NOX Data related to emissions: In this case, the data of interest would be to assess the total reduction in fossil fuel consumption that results from the services of the entity. Energy saving/ Reducing transmission Results in reduced fuel consumption in the power plant, giving conservation and distribution losses reduced emissions. Main pollutants: PM, BC, SO2, (NOX). Data related to emissions: Emissions before and after the changes. Refitting of buildings: Same as above. insulation and so on Improved Rehabilitation, closure of Would result in reduced emissions in the mine’s area. These management mines would be diffuse emissions (distributed over an area). in mining Main pollutants: PM, BC Data related to emissions: Estimation of emissions before and after the changes. Fuel switch LPG, CNG, biogas, and Results in reduced emissions. in existing so on replacing coal/ Main pollutants: PM, BC, SO2, possibly not NOX. installations oil Data related to emissions: Emissions before and after the changes. Enhancing the World Bank's Approach to Air Quality Management 95 Sector Subsector Activity Type Type Type Air Pollution Aspects Fuel cleaning Cleaning of coal Results in reduced contents of ash (PM, BC) and SO2 in and the coal, and thus in reduced emissions where the coal is improvements combusted. Main pollutants: PM, BC, SO2, (NOX). Data related to emissions: − the reduction in ash and SO2 contents, and total for the annual coal production and consumption, giving the reductions in emissions. − emissions before and after the changes. − amounts and locations where the cleaned coal is used. Cleaning of petrol and At present, the interest is in reducing the contents of sulfur in ) (Continued diesel fuel the fuel, to: − reduce the SO2 emissions per se; − allow for improved efficiency of exhaust emissions cleaning systems on the vehicles. (It is assumed that lead has been removed from petrol, and that the diesel fuel is of otherwise good quality, recetane number, and so on). This second point is the more important of the two, since improved cleaning systems result in significantly reduced emissions of PM, NOX and VOC. The reduction of SO2 emissions is of little consequence, since the SO2 contribution from automotive fuels is normally small. Main pollutants (re. the exhaust emissions): Petrol cleaning: PM, NOX, VOC Diesel cleaning: PM, BC, NOX, VOC. Data related to emissions: − the sulfur contents of the cleaned fuel. − fuel amounts and locations were the fuels are used. − the distribution across vehicle/cleaning system classes in the city, before and after the project. Renewables to Energy production by Renewable energy can always be said to replace fossil energy. replace energy renewables Renewable energy can be clean. that is, no emissions (solar, production by wind, hydro) or associated with combustion emissions fossil fuel (biomass, biogas) in new installations. Main pollutants: PM, BC, SO2, (NOX). Data related to emissions: − the emissions (amount and locations) associated with the saved fossil energy production. − the emissions (amount and location) associated with the renewable energy production (when applicable). Rural electricity to Electricity replacing local use of combustion energy (fossil or replace local fuel biomass). Main pollutants: PM, BC, SO2. Data related to emissions: − amount of fuel, and fuel type, replaced. − emissions replaced. 96 Clean Air and Healthy Lungs Sector Subsector Activity Type Type Type Air Pollution Aspects Cook stove Efficiency improvements, Cook stove improvements lead to reduced emissions. A variety improvements and fuel switch/ of abatement options and combination of options, such as: renewables − improved or new stove types with less emissions − improved fuel, for example, cleaned coal − fuel switch to gas or renewables (combined with changed stove technology). Main pollutants: PM, BC, NOX Data related to emissions: − number of stoves of each type (incl. old, new) used in the area − emission factors for each stove/fuel combination (kg/ton of fuel) − types and amounts of fuel used in the area. Industry Cleaning of Cleaning of emissions: Installation of cleaning equipment in the stack or flue in front of the emissions ‘end-of-pipe’ actual point of emissions to air (for example, top of stack). Power plants, industrial process plants, incineration plants, and so on Examples: flue gas desulfurization (FGD) to reduce SO2 emissions; electrostatic precipitator (ESP) to reduce PM emissions; catalytic reduction (SCR, NSCR) to reduce NOX. Pollutants: all health relevant pollutants, gaseous and particulates, dependent upon the process and cleaning equipment. Data related to emissions: − emissions before and after the changes, for each pollutant. − changes in emission conditions (outlet temperature, gas velocity). Production Process improvements to Process improvements that lead to reduced emissions. Process process reduce air pollution emissions can be emitted to air through stacks (as above) or improvements emissions more diffuse/distributed. Improvements may include the collection of diffuse emissions to stacks. Pollutants: all health relevant pollutants, gaseous and particulates, dependent upon the process and cleaning equipment. Data related to emissions: − emissions before and after the changes, for each pollutant. − changes in emission conditions (for example, collection to stacks, outlet temperature, gas velocity). Cleaner production and Company specific environmental protection actions to eco-efficiency minimize waste, energy use and emissions, while maximizing product output. Pollutants: all health relevant pollutants, gaseous and particulates, dependent upon the industrial process and reduction strategies. Data related to emissions: − emissions before and after the changes, for each pollutant. − changes in emission conditions (for example, collection of diffuse emissions to stacks, outlet temperature, gas velocity). (Continued ) Enhancing the World Bank's Approach to Air Quality Management 97 Sector Subsector Activity Type Type Type Air Pollution Aspects Hazardous Management, storage, Many different chemicals can be involved here, such as chemicals and disposal of hazardous pesticides, persistent organic compounds (POPs), stratospheric waste chemicals ozone depleting chemicals, radioactive materials. The question is whether there is leakage/emissions of the material to air (from for example, storage piles, contaminated soil, store houses). Such emissions will in most cases be diffuse (not from stacks). In each case, the emissions from the storage, if any, must be assessed. Presumably, the emissions after the disposal will be zero. Pollutants: depending upon the actual case, that is, specific chemicals that are involved. Data related to emissions: − reduction of emissions to air, for each chemical. Contaminated sites Removal of pollutants/contaminants from a site. In this(Continued ) clean-up and context the interest is in the possible release from the site remediation. of pollutants to air, through release of vapors (to outdoor or indoor areas) or suspension of particulates. The clean- up action itself may result in release of otherwise stable deposits. A range of clean-up technologies exist, dependent upon the actual site and type of contamination (for example, excavation, desorption and destruction, pumping and treating, solidification/stabilization, soil vapor extraction). Main pollutants (air): PM, VOC, POP Data related to emissions: − emissions before, during and after clean-up. Transport Public transport Public transport systems. Public transport systems (PTS) to replace individual transport by cars. PTS development, improvements, extensions will reduce and/ or redistribute (spatially) transport related emissions within the area (city), from car emissions to emissions from the PTS itself. Buses will have their own emissions, while metro and tram systems will not have emissions (excluding the emissions involved in the production of the electricity, which will in many cases fall outside the city area). Development of public transport corridors will involve infrastructure development to improve PTS traffic flow. Main pollutants: PM, BC, NOX, VOC. Data related to emissions: − emissions before and after the changes, and their redistribution within the city, from the entire transport sources. − emission amount from each transport mode (individual, public, each vehicle type). − emissions and its location, associated with the production of electricity to metro/tram systems. 98 Clean Air and Healthy Lungs Sector Subsector Activity Type Type Type Air Pollution Aspects Urban roads Repair, construction, Improvement of urban roads for the purpose of improved maintenance, upgrade environment will often lead to improved traffic flow, less of urban roads congestion. It may also lead to increased traffic amount, so it is not certain the emissions will be reduced. Main pollutants: PM, BC, NOX, VOC. Data related to emissions: − emissions before and after the changes, and their redistribution within the city as a result of the road project. − emission amount from each transport mode (individual, public, each vehicle type). Urban transport Improved fuels See above under Cleaning of petrol and diesel fuel. fuels Traffic Urban traffic Air pollution aspects of traffic management and planning management management and include: systems planning − improved traffic flow within an area (city); − reduced traffic within an area; − modification of the spatial distribution of traffic; − modification of the distribution of traffic between transport modes (individual, public, bicycle, walking). Main pollutants: PM, BC, NOX, VOC. Data related to emissions: − emissions before and after the changes and their redistribution within the city, from the entire transport sources. − emission amount from each transport mode (individual, public, each vehicle type). Vehicles Cleaner transportation Introduction of vehicles with reduced emissions compared technologies to the existing vehicle fleet. Usually resulting from the introduction of stricter vehicle emissions regulations as well as introduction of new clean vehicle technologies. Main pollutants: PM, BC, NOX, VOC. Data related to emissions: − emissions before and after the changes and their redistribution within the city. − emission amount from each transport mode (individual, public, each vehicle type). Emission inspection, Would result in better technical condition of the vehicles in the monitoring and area of implementation, leading to reduced emissions. maintenance (I&M) Main pollutants: PM, BC, NOX, VOC Data related to emissions: Emissions from the vehicle fleet before and after the introduction of the I&M scheme (Continued ) Enhancing the World Bank's Approach to Air Quality Management 99 Sector Subsector Activity Type Type Type Air Pollution Aspects Urban Urban planning/ Urban planning to Involves land and area use and traffic changes, presumably to upgrading/ reduce transport reduce transport demand and thus emissions. construction demand Changes the spatial distribution of emissions in the city/ activities area. Main pollutants: PM, BC, NOX, VOC Data related to emissions: The emissions and their spatial distribution before and after the changes. Urban upgrading and Examples of interventions: construction to reduce Road upgrading/surfacing; construction site dust PM suspension containment. Results in reduction of PM emissions. Main pollutant: PM Data related to emissions: The spatial emissions distribution and its change. Waste water Waste water and Reduces emissions of ammonia (NH3) to air, resulting in(Continued a ) management sewage collection and certain reduction in secondary PM formation over a larger treatment downwind area. Main pollutants: NH3 (together with SO2 and NOX emissions in the area). Solid waste Management and control Typical abatement: The reduction in emissions from flaring, management of Municipal dumps and the reduced public health impact as a result. Main pollutants: PM, SO2, VOC Data related to emissions: The emissions resulting from the abatement process: amounts, location. Refuse incineration to Burning of municipal refuse in incineration plants. Often replace municipal combined with heat production and utilization, for example, dumps for district heating, which will reduce the need for fossil (or other fuel) combustion in the city. Modern incineration plants have strict emission standards, and emission cleaning. Main pollutants: PM, BC, NOX, HM, POP. Data related to emissions: − emission factors, and amount incinerated − emissions in the city replaced by the incineration plant (from reduced combustion for energy production in the city), if applicable. Reduction of small scale Open refuse burning, distributed/small scale, produces open refuse burning significant amounts of emissions of PM. Banning open refuse burning, or replacing it with refuse collection and incineration, reduces the emissions. Main pollutants: PM, BC. Data related to emissions: − amount burned annually, and its spatial and temporal distribution − emission factors. − amount of emissions from incineration in plants (if applicable). 100 Clean Air and Healthy Lungs Sector Subsector Activity Type Type Type Air Pollution Aspects Agriculture On-farm Livestock production and Similar aspects to waste water and sewage. See above. activities Manure management Reduction of agricultural Burning of straw on fields produces large amounts of PM field burning emissions to air. Banning reduces emissions, but may lead to increased need for fertilizer. Main pollutants: PM, BC Data related to emissions: − area burned before/after ban − time of burning − emission factors. Agricultural Pesticides control Reduces emissions of pesticides to air, reducing local (and to chemicals some extent) regional pesticides exposure. Main pollutants: the pesticide involved Data related to emissions: emissions of pesticides to air Land Desertification, Control of desertification Suspension of dry particles into the air from dry areas can administration land to reduce PM cause large air pollution problems through very high degradation suspension PM concentrations in air (such as in China, where PM suspension from dry areas west and northwest of Beijing causes sandstorms with very high PM concentration episodes in Beijing). Tree planting is an example of abatement. Main pollutants: PM Data related to emissions: Estimation of PM suspension amounts before and after the abatement. Public Institutional Common for such project activities is that they will to some administration/ improvements extent, as a secondary effect, result in reduced environmental Environmental impact of activities in an area where policies and institutions policies and are improved, such as a city, contingent to the successful institutions implementation of the policies introduced. Source: Developed by authors. Enhancing the World Bank's Approach to Air Quality Management 101 REFERENCES Bell, M. L., D. L. Davis, N. Gouveia, V. H. Borja-Aburto, and L. A. Cifuentes. 2006. “The Avoidable Health Effects of Air Pollution in Three Latin American Cities: Santiago, Sao Paulo, and Mexico City.” Environmental Research, 100(3), 431–440. Bond, T. C., S. J. Doherty, D. W. Fahey, P. M. Forster, T. Berntsen, B. J. DeAngelo, M. G. Flanner et al. 2013. “Bounding the Role of Black Carbon in the Cli- mate System: A Scientific Assessment.” Journal of Geophysical Research: Atmospheres, 118(11), 5380–5552. Brauer, M., M. Amann, R. T. Burnett, A. Cohen, F. Dentener, M. Ezzati, S. B. Hen- derson et al. 2012. “Exposure Assessment for Estimation of the Global Burden of Disease Attributable to Outdoor Air Pollution.” Environmental Science and Technology 46, 652–660. CAI-Asia (Clean Air Initiative for Asian Cities). 2010. Air Quality in Asia: Status and Trends—2010 Edition. Pasig City, Philippines. Clean Air Institute. 2012. “Summary.” Air Quality in Latin America: An Overview. Avail- able at: http://www.cleanairinstitute.org/calidaddelaireamericalatina/summary- airquality-la.pdf CMMCh (Centro Mario Molina Chile). 2008. “Calidad del Aire en la Región Metro- politana. Análisis Retrospectivo 1997–2007.” Centro Mario Molina Chile. ———. 2013. “Análisis y Evaluación del Impacto del Sistema de Transporte Público en la Calidad del Aire de Santiago, Año 2013.” Centro Mario Molina Chile. EAI (Energy Alternative India). 2013. “Replacing Diesel with Solar: A Comprehen- sive Guide for Indian Businesses.” Tamil Nadu, India. ECON (2005). “Urban Air Pollution Control in Peru,” Draft Report submitted to the World Bank. EEB (European Environmental Bureau), ClientEarth, AirClim, 2013. “Ammonia Control Options for Better Quality. Recommendations to the EU for its 2013 ‘Year of Air.’” March 12. Fuller, T. 2007. “Bangkok’s Template for an Air Quality Turnaround.” New York Times. February 23. GOI (Government of India). 2013. All India Study on Sectoral Demand of Diesel & Petrol. Petroleum Planning & Analysis Cell (PPAC), New Delhi, India. Guttikunda, S. 2007. “Urban Air Pollution Analysis for Ulaanbaatar.” Washington, DC: World Bank. HEI (Health Effects Institute). 2010. Outdoor Air Pollution and Health in the Developing Countries of Asia: A Comprehensive Review. Special Report 18. Boston. Hoornweg, D., and P. Bhada-Tata. 2012. What a Waste: A Global Review of Solid Waste Management. Urban development series; Knowledge Papers no. 15. Washington, DC: World Bank. IARC (International Agency for Research on Cancer). 2013. “Outdoor Air Pollu- tion a Leading Environmental Cause of Cancer Deaths.” Press release 221. 17 Enhancing the World Bank's Approach to Air Quality Management 103 October 2013. Available at http://www.iarc.fr/en/media- centre/pr/2013/pdfs/ pr221_E.pdf. IPCC (Intergovernmental Panel on Climate Change). 2007. IPCC Climate Change 2007 reports (the AR4 WG1 report: The Physical Science Basis). IPCC (Intergovernmental Panel on Climate Change). 2013. IPCC Climate Change 2013 reports (the AR5 WG1 report: The Physical Science Basis). JERP (Joint Economic Research Program)/World Bank. 2013. “Towards Cleaner Industry and Improved Air Quality Monitoring in Kazakhstan. Washington.” World Bank and Ministry of Environment and Water Resources of the Republic of Khazakhstan. Washington, DC. Jhun, I., P. Oyola, F. Moreno, M.A. Castillo and P. Koutrakis. 2010. “PM2.5 Mass and Species Trends in Santiago, Chile, 1998–2010: The Impact of Fuel-Related Inter- ventions and Fuel Sales.” Journal of the Air & Waste Management Association, 63:2, 161–169. KEPA (Kosovo Environmental Protection Agency). 2013. “Monitoring Air Quality in KEK Area.” Kosovo Hydro-Meteorology Institute: Prishtina. Kojima, M., and M. Lovei. 2001. “Urban Air Quality Management: Coordinating Transport, Environment and Energy Policies in Developing Countries.” World Bank Technical Paper No. 508. Pollution Management. Washington, DC: World Bank. Larsen, B., and E. Strukova. 2005. “Peru: Cost of Environmental Damage: An Analy- sis of Environmental Health and Natural Resources. Final Report.” Background Report for the Peru Country Environmental Analysis. Washington, DC: World Bank. Le Courtois, A. 2012. “Municipal Solid Waste: Turning a Problem into Resource.” Private Sector & Development. Issue 15. Legros, G., I. Havet, N. Bruce, and S. Bonjour. 2009. “The Energy Access Situation in Developing Countries: A Review Focusing on the Least Developed Countries and Sub-Saharan Africa.” UNDP and World Health Organization. Macizo, G., and S. Sanchez. Forthcoming. “Lima and Callao Case Study.” In Air Qual- ity Management in Latin American Cities. E. Sánchez-Triana and S. Sanchez (eds.). Washington, DC: World Bank. Norwegian Institute for Air Research, 2011. AirQUIS. Accessed on 17 March 2014. Available at http://www.airquis.com/ O’Connor, Anne-Marie. 2010. “Mexico City Drastically Reduced Air Pollutants since 1990s.” Washington Post. April 1. OPEC (Organization of the Petroleum Exporting Countries). 2013. World Oil Outlook 2013. Vienna. Parrish, David D., Hanwant B. Singh, Luisa Molina, and Sasha Madronich. 2011. “Air Quality Progress in North American Megacities: A Review.” Atmospheric Envi- ronment 45(39), 7015–7025. Patdu, K. 2012. “Compliance and Enforcement of Air Quality Standards in Asia Implications to Climate Change Mitigation.” Presentation. 2012 AECEN Forum. Bangkok, Thailand. March. PCFV and CAI-Asia (Partnership for Clean Fuels and Vehicles and Clean Air Ini- tiative–Asia). 2011. “Managing 2 & 3 Wheelers in Asia.” Partnership for Clean 104 Clean Air and Healthy Lungs Fuels and Vehicles, United Nations Environment Programme and CAI-Asia Cen- ter. Available at: http://www.unep.org/transport/pcfv/pdf/23-Wheelers-in-Asia- Book.pdf PEP (Poverty-Environment Partnership), 2008. Joint Agency Paper. Poverty, Health and Environment. “Placing Environmental Health on Countries’ Development Agendas.” Washington, DC. PPAC (Petroleum Planning and Analysis Cell). 2014. All India Study on Sectoral Demand of Diesel and Petrol. Ministry of Petroleum and Natural Gas, Government of India. Sánchez, S., and R. Lacy. Forthcoming. “Valley of Mexico Metropolitan Area Case Study.” In Air Quality Management in Latin American Cities. E. Sánchez-Triana and S. Sánchez (eds.). Washington, DC. Sánchez-Triana, E., and Y. Awe. 2007. “Environmental Degradation and Environ- mental Health Policy in Peru.” In An Opportunity for a Different Peru: Prosperous, Equi- table, and Governable. Washington, DC: World Bank. Sánchez-Triana, E., and S. Sánchez (eds.). Forthcoming. Air Quality Management in Latin American Cities. Washington, DC: World Bank. SEI (Stockholm Environment Institute). 2003. “Air Pollution and Health in Rapidly Developing Countries.” G. McGranahan and F. Murray (eds.). Earthscan Publica- tions Limited, USA. Simachaya, W. 2012. “Successful Air Quality Control Programs in Thailand.” Pre- sentation delivered at 2012 Regional Forum of Asian Environmental Compliance and Enforcement Network (AECEN). Available at: http://www.aecen.org/sites/ default/files/forums/2012/GHG%20-%20Air%20Quality%20Control%20Pro- gram-%20Thailand%20(Wijarn).pdf. Smith, K., M. Jerrett, H. Anderson, R. Burnett, V. Stone, R. Derwent, R. Atkinson et al. 2010. “Public Health Benefits of Strategies to Reduce Greenhouse-Gas Emis- sions: Health Implications of Short-lived Greenhouse Pollutants.” The Lancet 374, no. 9707: 2091–2103. Smith, K. R., N. Bruce, K. Balakrishnan, H. Adair-Rohani, J. Balmes, Z. Chafe, M. Dherani et al. 2014. “Millions Dead: How Do We Know and What Does It Mean? Methods Used in the Comparative Risk Assessment of Household Air Pollution.” Annual Review of Public Health 35: 185–206. Smith, K. R., M. G. Apte, M. Yuqing, W. Wongsekiarttirat, and A. Kulkarni. 1994. “Air Pollution and the Energy Ladder in Asian Cities.” Energy 19: 587–600. Sun, K. 2012. “Ammonia: An Overlooked Byproduct from Energy Generation.” Princeton University China Energy Group. Available at: http://www.princeton. edu/~puceg/perspective/ammonia.html UNEP (United Nations Environment Programme). 2011. Near-term Climate Protection and Clean Air Benefits: Actions for Controlling Short-Lived Climate Forcers. Nairobi, Kenya: United Nations Environment Programme (UNEP). ———. 2014. “Leaded Petrol Phase-out: Global Status April 2014.” Available at: http://www.unep.org/Transport/PCFV/pdf/Maps_Matrices/world/lead/ MapWorldLead_April20 14.pdf.Integrated Approach. Geneva. UN-Habitat. 2009. “Global Report on Human Settlements 2009. Planning Sustain- able Cities.” Nairobi. ———. 2010. “Urban Trends: Urban Sprawl Now a Global Problem.” Nairobi. Enhancing the World Bank's Approach to Air Quality Management 105 Vahlsing, C., and K. R. Smith. 2012. “Global Review of National Ambient Air Qual- ity Standards for PM10 and SO2 (24 h).” Air Quality, Atmosphere & Health. 5(4): 393–399. Wangwongwatana, S. 2013. “Air Quality Management: Thailand’s Experiences.” EANET. Also available at: http://www.iges.or.jp/isap/2013/PDF/L1/ISAP_ L1_4_Supat.pdf WHO (World Health Organization). 2006. “Air Quality Guidelines, Particulate Mat- ter, Ozone, Nitrogen Dioxide and Sulfur Dioxide Global Update 2005.” WHO Regional Office. Europe. ———. 2010. “Why Urban Health Matters.” Geneva. ———. 2012. Health Effects of Black Carbon. Prepared by Janssen, N., M. E. Gerloffs- Nijland, T. Lanki, R. Salonen, F. Cassee, G. Hoek, P. Fischer, B. Brunekreef, and M. Krzyzanowski. World Health Organization Regional Office for Europe. ———. 2014a. “News Release dated 25 March 2014.” Available at: http://www.who .int/mediacentre/news/releases/2014/air-pollution/en/. ———. 2014b. “Ambient (Outdoor) Air Pollution in Cities Database.” http://www .who.int/phe/health_topics/outdoorair/databases/cities/en/. WHO/ECEH Technical Report, 2001. Health Impact Assessment of Air pollution in the WHO European Region. World Bank, 1997. Urban Air Quality Management Strategy in Asia: Guidebook. J. Shah, T. Nagpal, and C. Brandon (eds.). Washington, DC. ———. 2002. “Thailand Environment Monitor 2002.” Washington, DC. ———. 2007a. “Review of Urban Air Quality in Sub-Saharan Africa.” Prepared by D. Schwela. Washington, DC. ———. 2007b. “Environmental Sustainability: A Key to Poverty Reduction in Peru. Country Environmental Analysis.” Washington, DC. ———. 2008. Environmental Health and Child Survival: Epidemiology, Economics, Experiences. Washington, DC. ———. 2009a. Program Document for Peru First Environmental Development Policy Loan. Washington, DC. ———. 2009b. Program Document for Peru Second Environmental Development Policy Loan. Washington, DC. ———. 2010. Program Document for Peru Third Environmental Development Policy Loan. Washington, DC. ———. 2011. Air Quality Analysis of Ulaanbaatar: Improving Air Quality to Reduce Health Impacts. Discussion Papers Sustainable Development—East Asia and Pacific Region. Washington, DC. ———. 2012a. “Integrated Air Pollution Management in Chinese Cities: Proposed Particulate Matter Control in Taiyuan, Lishi, Xiaoyi, and Urumqi (Seminar Edi- tion).” Washington, DC. ———. 2012b. Integrated Air Pollution Management in China: Developing Particulate Matter Control. Washington, DC. ———. 2012c. Poverty Environment Nexus–Sustainable Approaches to Poverty Reduction in Cam- bodia, Lao PDR, and Vietnam. Washington, DC. ———. 2013a. “United Mexican States. Reducing Fuel Subsidies: Public Policy Options.” Washington, DC. 106 Clean Air and Healthy Lungs ———. 2013b. Egypt—For Better or for Worse: Air Pollution in Greater Cairo. Washington, DC. ———. 2013c. “Sector Air Pollution Paper: Macedonia.” Macedonia Green Growth and Climate Change Analytic and Advisory Support Program. April. Washing- ton, DC. ———. 2013d. Implementation and Completion Report—Peru: Environmental Development Pol- icy Loan Program. Washington, DC. ———. 2013e. Integration of Short-lived Climate Pollutants in World Bank Activities. Wash- ington, DC. ———. 2014a. “Road Sector Diesel Fuel per Capita (kg of oil equivalent).” World Development Indicators. Data available at http://data.worldbank.org/indicator/ IS.ROD.DESL.PC (accessed on 16 June 2014). World Bank and the Development Research Center of the State Council, P.R. China. 2014. “Urban China: Toward Efficient, Inclusive, and Sustainable Urbanization.” Washington, DC. Yip, M., and P. Madl. 2001. “Air Pollution in Mexico City.” Project-Study paper. Uni- versity of Salzburg. Department of Biophysics. May 23. Available at http://bio- physics.sbg.ac.at/mexico/air.htm. Enhancing the World Bank's Approach to Air Quality Management 107 E N V I R O N M E N T A N D N AT U R A L R E S O U R C E S G LO B A L P R A C T I C E D I S C U S S I O N PA P E R 03 W O R L D B A N K G R O U P R E P O R T N U M B E R ACS9035 1818 H Street, NW Washington, D.C. 20433 USA Telephone: 202-473-1000 Internet: www.worldbank.org/environment