Report No: AUS0002512 © 2021 The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved This work is a product of the staff of The World Bank. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank 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 The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because the World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution is given to this work. Attribution—Please cite the work as follows: “Xie, Jian, W. Jia, L. Croitoru, S. Guttikunda, and J. Grutter. 2021. Safe to Breathe? Analyses and Recommendations for Improving Ambient Air Quality Management in Ethiopia. Washington, DC: The World Bank.� All queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202- 522-2625; e-mail: pubrights@worldbank.org. Safe to Breathe? Analyses and Recommendations for Improving Ambient Air Quality Management in Ethiopia TABLE OF CONTENTS ABBREVIATIONS AND ACRONYMS ................................................................................. vi ACKNOWLEDGMENTS ...................................................................................................... viii EXECUTIVE SUMMARY ....................................................................................................... x 1. Introduction......................................................................................................................... 1 2. Overview of Urban Air Pollution Problems in Ethiopia ................................................ 4 2.1 Introduction to Air Pollution Problems ........................................................................... 4 2.2 Need for an Integrated AQM Approach .......................................................................... 5 2.3 Stakeholder Mapping and Institutional Arrangements for AQM .................................... 6 2.4 Air Quality in AA: A Review of Existing AQ Monitoring Efforts and Results .............. 9 2.5 Recommendations ......................................................................................................... 14 3. Health Impacts and Economic Costs ............................................................................. 16 of Air Pollution in Addis Ababa .......................................................................................... 16 3.1 The Health Impacts of Ambient Air Pollution .............................................................. 16 3.2 Economic Cost of Health Impacts due to Exposure to PM2.5 ........................................ 20 3.3 Recommendations ......................................................................................................... 21 4. Emission Inventory and Source Apportionment ......................................................... 22 4.1 Data resources ............................................................................................................... 22 4.2 The Airshed ................................................................................................................... 23 4.3 Air Pollution Emissions................................................................................................. 24 4.4 Source Apportionment................................................................................................... 27 4.5 Recommendations ......................................................................................................... 29 5. Zooming in Urban Transport: Vehicle Emission Control in Addis Ababa ............... 31 5.1 Urban Transport Development and Air Pollution ......................................................... 31 5.2 Identification and Assessment of Potential Mitigation Options .................................... 33 5.3 Conclusions and Recommendations .............................................................................. 38 6. Proposed Action Plan and Concluding Remarks ......................................................... 41 References.............................................................................................................................. 46 Annex 1. List of Background Paper to this Report ........................................................... 52 Annex 2. Supporting Information for the Estimation of the Health Impact of Ambient PM2.5 ....................................................................................................................................... 53 LIST OF TABLES Table 1. Location and PM2.5 Monitoring Results in Addis Ababa .......................................... 10 Table 2. Location and Results of AddisAir's PM2.5 Monitoring Network ............................... 12 Table 3. Monthly Average (and Standard Deviation) in the PM2.5 Concentrations Data in 2018-19 from All the Stations in Addis Ababa Airshed ......................................................... 13 Table 4. Annual Average PM2.5 Concentration in Addis Ababa ............................................. 17 Table 5. Impact of Air Pollution in Selected African Cities ................................................... 19 Table 6: Summary of Ethiopia and Addis Ababa geographical, economic, and environmental characteristics .......................................................................................................................... 23 Table 7. Estimated Total Emissions for AA's Airshed for Base Year 2018 ............................ 25 Table 8. List of Potential Transport Air Pollution Mitigation Options ................................... 33 Table 9. Summary of the Assessment of Mitigation Measures ............................................... 35 Table 10. Prioritization of Mitigation Measures ..................................................................... 38 Table 11. Action Plan for Promoting Integrated AQM in Ethiopia ........................................ 41 LIST OF FIGURES Figure 1. Relationship of the Components of the World Bank AQM Program ........................ 3 Figure 2. DPSIR Diagram for Air Pollution .............................................................................. 4 Figure 3. Air Pollution and Its Impacts ..................................................................................... 5 Figure 4. Integrated Air Quality Management .......................................................................... 6 Figure 5. Air Quality Stakeholder Mapping .............................................................................. 7 Figure 6. Location of AQ Monitors in Addis Ababa ............................................................... 10 Figure 7. A Summary of All PM2.5 Concentrations Data Showing Monthly Averages and the Variation in Hourly Data by Month ........................................................................................ 13 Figure 8. Estimated Number of Premature Deaths Due to PM2.5 Exposure, by Age Group.... 18 Figure 9. Addis Ababa City Airshed with Road Density ........................................................ 24 Figure 10. Estimated Gridded Annual PM2.5 Emissions for Addis Ababa's Airshed in 2018 . 25 Figure 11. Estimated Sector Shares to Addis Ababa Airshed's Total Emissions in 2018 for (a) PM2.5, (b) PM10, (c) CO, (d) SO2, (e) NOx, and (f) CO2 .......................................................... 26 Figure 12. Modeled Annual PM2.5 Concentrations for Addis Ababa's Airshed in 2018 ......... 27 Figure 13. Modeled Annual Average PM2.5 Concentrations in Addis Ababa's Airshed in 2018 ................................................................................................................................................. 28 Figure 14. Modeled Source Contributions to Annual Average PM2.5 Pollution in Addis Ababa ................................................................................................................................................. 29 Figure 15. Spatial Expansion of Addis Ababa ........................................................................ 31 Figure 16. Type and Age of Registered Vehicles in Addis Ababa in 2020............................. 32 ABBREVIATIONS AND ACRONYMS AA Addis Ababa ADB Asian Development Bank AQ Air Quality AQM Air Quality Management BAM Beta Attenuation Monitor BC Black Carbon CO Carbon Monoxide CO2 Carbon Dioxide CO2e Carbon dioxide equivalent EF Emission Factor EFCCC Federal Environment, Forest and Climate Change Commission, Ethiopia EPA Environmental Protection Agency EPGDC AA Environmental Protection and Green Development Commission ESA Ethiopian Standard Agency ESP Ethiopia State Petro Company FTA Federal Transport Authority, Ethiopia GDP Gross Domestic Product GEOHealth Global Environmental and Occupational Health GHG Greenhouse Gas GIS Geospatial Information Systems HDV Heavy Duty Vehicle ICS International Community School IM Inspection/Maintenance IMF International Monetary Fund LCS Low-Cost Sensors LCV Light Commercial Vehicle LDV Light Duty Vehicle LMIC Low- and Middle-Income Countries MOT Ministry of Transport NASA National Aeronautics and Space Administration NMA National Meteorological Agency NMT Non-Motorized Transport NO2 Nitrogen Dioxide NOx Nitrogen Oxides O3 Ozone OECD Organization for Economic Co-Operation and Development OSM Open Streets Maps PM10 Particulate matter under 10 micro-meter diameter PM2.5 Particulate matter under 2.5 micro-meter diameter ppm Parts per million PWC Population Weighted Concentration SCC Social Cost of Carbon vi SO2 Sulfur Dioxide SPARTAN Surface Particulate Matter Network UNEP United Nations Environment Program USDP United States Diplomatic Post VOC Volatile Organic Compounds WHO World Health Organization µg/m3 micro-grams per cubic meter vii ACKNOWLEDGMENTS This report summarizes the main findings and recommendations of the World Bank’s Advisory Services & Analytics program (ASA) entitled “Ethiopia: Air Quality Management and Urban Mobility.� Launched in September 2020, the ASA aimed to assist the Government of Ethiopia in deepening its understanding of ambient air quality management, with a focus on urban transport air pollution control. Specifically, it examined priority issues related to air pollution in Addis Ababa, developed policy recommendations and proposed an action plan for Ethiopia and Addis Ababa to step up AQM efforts. The ASA was developed under the general guidance of Ousmane Dione (Country Director for Ethiopia), Doina Petrescu (Operations Manager for Ethiopia), Iain Shuker (Practice Manager, Environment, Natural Resources and Blue Economy (ENB) Global Practice), and Maria Marcela Silva (Practice Manager, Transport Global Practice) at the World Bank. It was implemented by a team led by Jian Xie (Sr. Environmental Specialist, World Bank Task Team Leader) and Wenyu Jia (Sr. Urban Transport Specialist, Co-Task Team Leader) and comprised Tamene Tiruneh (Sr. Environmental Specialist), Bereket Belayhun Woldemeskel (Municipal Engineer), Fatima Barry (Health Specialist), Lelia Croitoru (Environmental Economist, Consultant), Sarath Guttikunda (AQM Specialist, Consultant), Jurg Grutter (Transport Specialist, Consultant), Zenebe Tilahun Abayneh (Transport Specialist, Consultant), and Worku Tefera (AQM Specialist, Consultant). In addition, Michelle Anne Winglee (Climate Change Specialist), Caroline Anitha Devadason (Health Specialist, Consultant), Kimberly Worsham (Environmental Specialist, Consultant), and Christopher Arthur Lewis (Environmental Economist, Consultant) participated either in early or late phases of the ASA program. This report was prepared by Jian Xie, Wenyu Jia, Lelia Croitoru, Sarath Guttikunda, and Jurg Grutter, with input from Tamene Tiruneh (institutional arrangements), Bereket Belayhun Woldemeskel (urban development), Fatima Barry (health), Zenebe Tilahun Abayneh (transport), Worku Tefera (air quality monitoring), Kimberly Worsham (regulatory and policy review), and Christopher Arthur Lewis (financial arrangements). This report benefits from the participation and support of government officials and experts from various Ethiopian government agencies and institutions, particularly the Federal Environment, Forest and Climate Change Commission (EFCCC), Ministry of Transport (MoT), Federal Transport Authority (FTA), Ministry of Health (MoH), National Meteorology Agency (NMA), Addis Ababa Environmental Protection and Green Development Commission (AAEPGDC), Addis Ababa Transport Bureau (AATB), Addis Ababa Vehicle and Driver Registration and Authorization Authority, Addis Ababa Health Bureau (AAHB), Ethiopia State Petro Company (ESP), and Addis Ababa Institute of Technology (AAIT). It also benefits from regular discussions and consultations with representatives and experts from a group of international organizations actively working on air quality management in Addis Ababa, such as C40 Cities, UNEP, U.S. Embassy and EPA, USAID’s Clean Air Catalyst Program (managed through the World Resource Institute and Environmental Defense Fund), Global Environmental and Occupational Health’s (GEOHealth’s) East Africa Hub, and Surface Particulate Matter Network (SPARTAN). viii The authors would like to thank Paul Jonathan Martin, Martin Heger, Fiona Collin, Rodger Gorham, James Robert Markland, Sameer Akbar, and Marius Vismantas for their valuable comments and suggestions. Special thanks also go to the participants of the consultation workshops held virtually in March, April, and June 2021 for their review and discussions on the findings and results of studies and technical reviews of the ASA. Ross Hughes helped coordinate within World Bank’s county management unit. Esther Bea and Sofia Said provided administrative and logistic assistance to the program. Demetra Aposporos provided editorial assistance and Guomeng Ni provided infographic design. ix EXECUTIVE SUMMARY 1. Air pollution presents a global problem that undermines health and economic productivity. Data from the World Health Organization (WHO) shows that 9 out of 10 people breathe air containing high levels of pollutants, with low- and middle-income countries (LMICs) bearing the brunt of poor air quality. Air pollution is the leading environmental risk factor for premature death. In 2019 alone, indoor and outdoor air pollution was estimated to have contributed to 6.67 million deaths worldwide, nearly 12% of the global total. The health impacts of air pollution are also reflected in morbidity levels, loss of income, decreased participation in the workforce, disability, and higher health care costs. 2. Poor air quality during the COVID-19 pandemic may further jeopardize hard-won gains in public health. Particulate matter (PM), especially fine particulate matter such as PM2.5 (PM with aerodynamic diameter less than 2.5�m), can affect respiratory, cardiovascular, cardiopulmonary, and reproductive systems, and can in some instances lead to cancer. Ongoing research suggests a correlation between poor air quality and the incidence of severe illness and death among people exposed to COVID-19 with pre-existing, non- communicable diseases (NCDs), such as cardiovascular and respiratory diseases. 3. Air pollution, exacerbated by urbanization and increasing mobility, is a growing concern across Africa. In Africa, indoor and outdoor air pollution has already become the most significant environmental contributor to premature death, outpacing malaria and HIV. Reports show that ambient air pollution causes 73 premature deaths per 100,000 urban residents in Cairo, 46 in Lagos, and 35 in Abidjan. Growing rural-urban migration, traffic congestion, and increasing industrialization are increasing air pollution and resulting in poorer air quality, aggravating health impacts. Ambient air pollution also affects the quality of soil and water resources, impeding flora development, reducing food production, and contaminating water surfaces. 4. Rapid economic growth and urbanization in Ethiopia, particularly in its capital Addis Ababa (AA), have resulted in environmental deterioration and health concerns. Ethiopia is one of the fastest- growing economies in the region, growing at 9.9% annually from 2008 to 2018. Ethiopia's urban population accounted for 21.2% of its total population of 112 million in 2018, and the urbanization rate is projected to increase to 39% in 2050. Growing economic activities, unmanaged urban sprawl, and increasing mobility and transportation in AA have increased pollution emissions, traffic congestion, land and environmental deterioration, and risks to public health. Analysis of visibility data in AA suggests that air quality has been declining since the 1970s, with average air quality now approximately 1.6 times worse than it was in the 1970s. There are increasing incidents of airborne diseases. The impact of air pollution on public health is becoming a growing concern for both governments and the public. 5. As is the case in many other countries, ambient air pollution in Ethiopia, particularly in AA, includes emissions from multiple pollutants and various sources. Common drivers of ambient air pollution in a city and its surrounding areas include activities such as transport, cooking and heating, industries, construction sites, bare earth open areas, agricultural activities, and solid waste management (including open waste burning). Unlike many polluted cities in the world, in which power is largely generated by coal-based thermal power plants, AA’s predominant power generation source is hydro and other sectors predominantly use biomass, diesel, and coal for fuel. Primary air pollutants that put pressure on airsheds include particulate matter (PM), carbon monoxide (CO), nitrogen oxides (NOx), sulfur oxides (SO2), volatile organic compounds (VOC), x and ozone. Among these pollutants, PM2.5 is considered the most relevant urban air quality indicator in AA and worldwide. PM2.5 is responsible for aggravating respiratory, circulatory system, and heart problems, as well as leading to premature mortality. Moreover, studies also associate air pollution with increases in communicable infections, such as influenza and COVID-19. 6. Air pollution and AQM involve numerous stakeholders, including households, vehicle owners, manufacturers, government agencies, and so on, across various sectors. Managing the different stakeholders involved in AQM is complex. Government agencies from different levels and across different sectors must develop and implement policies and regulations that consistently minimize air pollution across their jurisdictions while aligning with national and even international standards. This can be challenging, as agencies need to balance priorities while accommodating objectives from other agencies. Managing the public’s behavior in reducing air pollution is also complicated, as the public, such as vehicle divers and households, need awareness programs that encourage their participation and behavioral changes. 7. Integrated air quality management (IAQM) is necessary to comprehensively address air pollution problems and protect the public from health issues and economic losses from poor AQ. IAQM involves developing a knowledge base or analytical tools for air pollution emissions and AQ monitoring that can support pollution control planning. Analysis of the interplay between different sources of pollutants, their consequences on health, the environment, and the economy, supports the interpretation of data for different actors. Additionally, key to implementing an IAQM strategy is building institutional and human resource capacity for pollution monitoring and management; institutional coordination between different sectors; national standards, regulations, and policies for pollution; and compliance monitoring and regulatory enforcement. 8. This study aims to deepen the understanding of ambient air pollution and develop policy recommendations for improving AQM in Ethiopia. Conducted from September 2020 to June 2021, the study reviewed ambient AQ and institutional arrangements, assessed the impacts of ambient air pollution on health, analyzed emission inventory and source apportionment, and identified and prioritized urban transport air pollution mitigation measures. It focused on AA and the urban transport sector due to limited data available nationwide and limited time and resources. This report summarizes the main findings and recommendations of the study. 9. Although Ethiopia has environmental policies and regulations that touch on air quality, the country has no specific framework on AQM. Ethiopian national standards of ambient AQ were issued in 2003. Its standard for annual average ambient PM2.5 concentration is set at 15 µg/m3, higher than the WHO guideline of 10 µm/m3 and standards in many other countries. There are no rules or laws around fuel quality and vehicle emissions standards, which leads to the transportation sector and vehicle owners emitting unnecessarily high levels of air pollution with low-quality fuel. 10. The government agencies with mandates for AQM are facing weak institutional and financial capacity to enforce AQ standards and implement AQM programs. In the federal government, the Environment, Forest, and Climate Change Commission (EFCCC) leads air pollution control efforts. Other line agencies support AQM, including the ministries for health and transport and the Ethiopian Standard Agency (ESA). In AA, the AA Environmental Protection and Green Development Agency (AAEPGDA) coordinates the city's AQM planning and oversees air pollution controls. Other city agencies, such as the AA Transport Bureau and Health Bureau, participate in and support AQM. However, there is a lack of functional units in xi charge of AQM in AAEPGDA and other agencies, making it difficult to ensure AQM roles and responsibilities. Also, the mandates of some line ministries or agencies for AQM are not clearly defined. Ethiopia's enforcement of environmental regulations and standards has also been inconsistent and weak because of conflicting goals, insufficient data for government coordination, low capacity, and low persistence of enforcement. Other underdeveloped areas include financial arrangements for AQM. Environmental taxation and fees are rare, with little earmarked to fund air pollution control activities. Line ministries/agencies and municipal governments lack a specific budget for AQM. Private sector investments in AQM are few. There have been few investment activities in air pollution reduction, even in key polluting sectors (for example, urban transport and industry) though some investments have gone towards building AQ monitoring networks. 11. AQ monitoring is relatively new in Ethiopia and AQ data is scarce . AQ monitoring began in 2014 when three AQ monitoring stations were established by the National Meteorological Administration (NMA): in AA, Adama, and Hawassa. The AQ monitoring network in AA has since gradually expanded and the past two years have seen a boom of installing low-cost sensors. As of March 2021, about 20 AQ monitors have been installed inside AA. Most are low-cost sensors that encounter operational challenges due to power and internet connectivity-related problems. So far, only a few stations release continuous monitoring data to the public. 12. AQ monitoring data available in recent years shows that AA’s annual average PM2.5 is 2-3 times higher than WHO guidelines and, as a result, the city’s air is unhealthy. Air pollution concentrations vary by location and time, and some AQ monitoring stations indicate that PM2.5 can reach several times above the WHO standards. A short AQ study done by the Eastern Africa Global Environmental and Occupational Health (GEOHealth) Hub in AA's city center from November 2015 to November 2016 shows that the mean (±SD) daily PM2.5 concentration was 53.8 (±25.0) µg/m3, with 90% of sampled days exceeding the WHO's guidelines. The World Bank program’s review of AQ monitoring data reveals that the annual average PM 2.5 concentration in AA varies between 30 µg/m3 and 36 µg/m3 across 2016-2020. 13. Ambient PM2.5 causes health damages estimated at the equivalent of 1.3% of the city’s GDP each year. In Ethiopia, air pollution is the second leading risk factor for premature death, after malnutrition. Ambient PM2.5 pollution is a growing public health problem in the country’s urban areas, particularly AA. This study estimated the health impacts caused by the ambient PM2.5 for a typical non-COVID year. Using ground monitored PM2.5 data from a few locations and the population exposed in each, the study estimates the annual average PM2.5 concentration at about 34 µg/m3 – slightly higher than that of other African capitals, such as Abidjan (Côte d’Ivoire) and Cotonou (Benin). Then, based on the Global Burden of Disease methodology, it showed that long-term exposure to this pollution level causes about 1,600 premature deaths a year in AA. Stroke, ischemic heart disease, and lower respiratory infections are the leading causes of PM2.5-related mortality; the elderly are the most affected. In addition to deaths, exposure to ambient PM2.5 causes morbidity (disability), related to a plurality of health outcomes, e.g., chronic bronchitis, hospital admissions, etc. The study estimated that these effects correspond to about 4,100 Years Lived with Disability (YLDs) annually. Overall, the total health damage due to air pollution in AA is estimated at $78 million, which is equivalent to about 1.3% of the city's GDP. Although the estimate is lower than that in very polluted megacities - 2.1% of Lagos’ GDP, 4.5% of Greater Cairo’s GDP, 5.5% of New Delhi’s GDP – it points to the need to investigate the main pollution sources and hotspots in AA. 14. An air pollution source apportionment study further shows that source contributions to PM2.5 pollution in AA are dominated by vehicle exhaust, residential use, and industrial sources. The transport xii sector–including urban transport and aviation–emits 29% of PM2.5, 97% of NOx, 71% of SO2, and 96% of CO2. The transport sector’s contribution to ambient PM2.5 concentration further reaches 35% because PM from vehicle emissions are fine and stay in the air. The remainder of the ambient PM2.5 pollution originates from residential fuel combustion in rural and urban areas using a mix of biomass, coal, and LPG with a share of 28%; industrial sources with a majority using biomass as fuel and quarries using diesel; open waste burning; and some long-range sources outside the city’s airshed. A closer look at the spatial distribution of industrial sources found that heavy industrial polluters are scattered along the edge of the city, requiring the attention of a concerted effort in managing PM2.5 by the city and surrounding regions. 15. The AQ issues in the transport sector in AA are characterized by rapid motorization, an aging vehicle fleet, high-sulfur fuels, lack of emission standards, ineffective and unenforceable emission inspections, and increasing emissions. In the past five years, registered vehicles in the city increased by 40 percent, reaching 630,000 vehicles in 2020 and accounting for nearly 50% of Ethiopia’s registered vehicles. Most registered vehicles are over 10 years old. Ethiopia currently imports 500ppm sulfur diesel and 10ppm sulfur gasoline; this high sulfur content in fuels pollutes directly, but also prevents the use of newer vehicles with better pollution control technologies. Emission inspection is ineffective and unenforceable for vehicle emission control. In the next decade, should the current vehicle growth trend continue at 8% a year, AA would expect 1.3 million vehicles. Even a conservative estimate, which assumes vehicle growth at the GDP growth rate, projects 900,000 vehicles in AA by 2030. In these scenarios, the level of vehicle emissions will increase in the range of 50% to over 100%. 16. Continuing business-as-usual (BAU) will further degrade air quality in AA, threatening the city's livability and ability to grow. Without changes to transport standards, policies, and technologies, transport air pollution in AA will have profound local and national impacts. Meanwhile, the current approach of accommodating vehicle growth has not effectively improved accessibility, but has exacerbated road fatalities, worsened traffic congestion, increased greenhouse gas emissions and deteriorated air quality locally. It is imperative for transport, a contributing sector, to identify and act on air quality management measures. 17. The Government has undertaken studies to understand the need for air pollution control measures for the transport sector in AA and in Ethiopia. The Government of Ethiopia and the City Government of AA have prepared studies and plans to identify and implement measures for reducing vehicle emissions. This study reviewed these documents and looked into ongoing urban transport investments and policy works in AA relating to short-term mitigation of emission and air pollution. Based on the review of government studies and ongoing sector activities, assessment of local context, consultation with relevant government agencies, and international experiences, a list of potential mitigation measures in the transport sector in AA was identified for further assessment. These mitigation options fall into three groups: fuel and emission standards, vehicle measures, and public transport. 18. Overall, a set of policy recommendations are suggested to improve AQM in Ethiopia. They cover the following areas. • Regulation and Policy Reforms: Ethiopia needs to strengthen its institutional arrangements for AQM, particularly its current regulatory and policy framework, including AQ standards and governmental organization structure. xiii • AQM strategy and plan: Although AA, with external support, has recently developed its AQM plan, no national strategy or plan exists. It is necessary for the country and other cities to develop their strategy and action plan to guide their AQM activities in the future. • Budgeting and capacity building: Low financial and institutional capacity will make sustainable AQM impossible. The governments should attach a high priority to strengthening financial and implementation capacity of line agencies responsible for AQM. • AQ monitoring: Ethiopia must establish a robust AQ monitoring program and network nationally and strengthen existing AQ monitoring. • Data collection and analytics: Ethiopia needs to deepen its analytical work, including AQ data collection and analysis, air pollution dispersion modeling, emission inventory and source apportions, transport modeling and spatio-temporal analysis of traffic and vehicle air emissions, impact assessment, and economic valuation and analysis of AQM programs to better inform AQM planning and decision making. • Awareness raising and behavioral change: The country should conduct communications and public awareness activities to increase understanding and change behaviors that worsen air pollution. • Vehicle emissions control: It is important to prioritize and implement air pollution control measures from the transport sector because of its significant and growing contributions to ambient air pollution. • Point and area sources control: Although this study focused on transport air pollution reduction, it is necessary to point out that point and area sources from other sectors are also important to ambient AQ. Therefore, it is important to develop and implement air pollution control measures for point- and area- sources including (a) controlling air pollutant emission from industries, (b) improving SW collection and reducing open burning of trash, (c) raising public awareness of the health impacts of burning plastic and hazardous wastes, and (d) implementing best practices for control of construction and road dusts. Table ES-1 highlights the highest-priority and most actionable recommendations in each area. Table ES-1. Summary of Recommendations for AQM Category Recommendation Clearly define or clarify the roles and mandates of key government agencies at the federal and municipal level for AQM through functional review of relevant agencies Introduce taxation and pricing policies or other fiscally neutral instruments or incentives to encourage newer and cleaner vehicles such as hybrids and electric Regulation vehicles and phase out polluting vehicles and Policy Reforms Upgrade AQ standards (including ambient PM2.5 standards) Introduce low sulfur fuel standards (start with 50 ppm; progress to 10 ppm) with the Euro 4/IV vehicle emission standards nationwide, along with other transport standards AQM Strategy and Develop a national urban AQM strategy and action plan Plan Strengthen institutional capacity of responsible agencies for AQM, particularly for regulating and enforcing AQ regulations and standards xiv Develop mechanisms for promoting inter-agency and cross-sectoral coordination Budgeting and collaboration as well as experience sharing and learning on AQM and Capacity Building Increase government revenue and budgets for AQM by reviewing and revising environmental taxation and fee systems Develop standardized and unified systems for monitoring and reporting ambient AQ and emission levels at critically polluted source regions in AA and other AQ major cities Monitoring Make verified and standardized AQ data open to public access via the internet and mobile apps Systematically collect and analyze AQ and air emission data Adopt the top-down chemical analysis-based approach in AA to further update and verify the results of the bottom-up emissions inventory and source Data apportionment study Collection Improve and extend the economic valuation of air pollution impacts beyond health and impacts and develop the guidelines of the cost-benefit analysis or cost- Analytics effectiveness analysis of AQM programs, investment projects, and interventions Develop an abatement cost curve and a plan to implement AQM interventions to cost-effectively reduce air pollution emissions and incorporate the results of health impact assessment and economic analysis Introduce low-sulfur (max 50ppm) diesel fuel and Euro 4/IV vehicle emission standards or their equivalent Ban importing secondhand diesel passenger cars or light commercial vehicles and Vehicle restrict imports to new and secondhand imported diesel vehicles weighing more Emissions than 3.5t gross (i.e., continue to allow imported heavy diesel vehicles, such as Control trucks or buses) Promote and strengthen public transport and non-motorized transport (NMT) measures, with possible transport demand measures and/or transit-oriented development 20. There are areas to further improve the analytical results and recommendations. This rapid assessment has faced some limitations. The first is the lack of reliable and comprehensive AQ data. The study behind the report had to rely mostly on open-source data, literature, and limited access to government data in its analytics including emission inventory and source apportionment and health impact assessment. Secondly, the study was carried out within a short period from October 2020 to June 2021 during the Covid-19 pandemic and, as a result, it was impossible to conduct field trips and in-person discussions with stakeholders for the firsthand and in-depth assessment of the local situation. Thirdly, due to data availability and time and resource constraints, most of the analytical work was done in AA only and air pollution emission control measures focus on the transport sector. Therefore, to enhance its policy analysis and decision-making capacity, Ethiopia should continue conducting in-depth analyses for various aspects of AQ, such as ambient air pollution dispersion modeling, spatial distribution of populations vulnerable to AQ, wider assessment of the health and non-health xv impacts of air pollution, and cost-effectiveness or cost-benefit analysis of AQM interventions. While this report used an established bottom-up approach to estimate source contributions, a complimentary program that includes a top-down chemical analysis-based approach at a representative number of locations across the airshed of AA would help strengthen and verify the results and increase confidence in policy applications. From the additional analyses, Ethiopia can develop a more evidence based IAQM action plan that includes recommendations for nationwide action and corresponding investment activities. xvi 1. Introduction Air pollution is a global problem that undermines health and economic productivity. Air emissions are a risk in tandem with greenhouse gas (GHG) emissions. Data from the World Health Organization (WHO) shows that 9 out of 10 people breathe air containing high levels of pollutants, with low- and middle-income countries (LMICs) bearing the brunt of poor air quality levels.1 Air pollution is the leading environmental risk factor for premature death. In 2019 alone, air pollution globally caused 6.7 million premature deaths, corresponding to about 19 percent of the total premature deaths (IHME, 2020). The health impacts of air pollution are also reflected in morbidity levels, loss of income, decreased participation in the workforce, disability, and higher health care costs. Air pollution has also been known to impede cognitive development in children, with long- term implications on human capital development. Poor air quality during the COVID-19 pandemic may further jeopardize hard-won gains in public health. PM2.5 can affect respiratory, cardiovascular, cardiopulmonary, and reproductive systems, and in some instances can lead to cancer. Ongoing research suggests a correlation between poor air quality and the incidence of severe cases and death among people exposed to COVID-19 who have pre-existing, non-communicable diseases (NCDs), such as cardiovascular and respiratory diseases (Conticini, E., et al. 2020; Fattorini, D. and Regoli, F. 2020). Air pollution, exacerbated by urbanization and increasing mobility, is a growing concern across Africa. Growing rural-urban migration, traffic congestion, and increasing industrialization will increase air pollution emissions and may result in poorer air quality, aggravating the health effects. In Africa, indoor and outdoor air pollution have already become the most significant environmental contributors to premature death, outpacing malaria and HIV.2 Reports indicate the numbers of premature deaths in a year per 100,000 population (urban) due to ambient air pollution was 73 in Cairo, 46 in Lagos, and 35 in Abidjan.3 Ambient air pollution also impacts the quality of soil and water resources, impeding flora development, reducing food production, and contaminating water surfaces.4 Over recent decades, Ethiopia has achieved strong economic growth, accompanied by urbanization and mobility. The country is one of the region’s fastest-growing economies, growing 9.9% annually from 2008 to 2018, compared to a regional annual average of 5.4%. Extreme poverty fell markedly, from 55% in 2000 to 26.7% in 2016. Ethiopia's urban population accounted for 21.2% of its total 112 million population in 2019. A UN study on world urbanization prospects estimates Ethiopia’s annual average rate of urbanization at 2.0% 1 World Health Organization. 2018. “9 out of 10 people worldwide breathe polluted air, but more countries are taking action.� Available at: https://www.who.int/news-room/detail/02-05-2018-9-out-of-10-people-worldwide-breathe-polluted-air-but-more-countries-are- taking-action#:~:text=New%20data%20from%20WHO%20shows,containing%20high%20levels%20of%20pollutants.&text =%E2%80%9CAir%20pollution%20threatens%20us%20all,%2C%20Director%2DGeneral%20of%20WHO. 2 World Health Organization. March 2018a. Ambient (outdoor) air pollution. Available at: https://www.who.int/en/news-room/fact- sheets/detail/ambient-(outdoor)-air-quality-and-health. World Health Organization. March 2018b. Household air pollution and health. Available at: https://www.who.int/en/news-room/fact- sheets/detail/household-air-pollution-and-health. 3 Larsen (2019) for Cairo; Croitoru et al. (2020) for Lagos; Croitoru et al. (2019) for Abidjan. 4 “Air pollution: effects on soil and water.� Government of Canada. Available at: https://www.canada.ca/en/environment-climate- change/services/air-pollution/quality-environment-economy/ecosystem/effects-soil-water.html. 1 from 2018 to 2050 and the urbanization rate is projected to increase to 39% in 2050.5 Urban areas—housing only 15% of the national workforce—contribute over 38% of national GDP.6 Addis Ababa (AA), Ethiopia’s capital, is the country’s most economically and politically significant city. The city was estimated to have a population of about 4.8 million in 2020,7 or 4.1% of the national population. Its annual average population growth rate in 2015-2020 is 4.4%. Its GDP has grown on average by over 15% in recent years8 and is about 8% of the national GDP. As the demographic and economic center of the country and a major regional hub, AA’s growth will continue. Rapid economic growth and urbanization in AA have resulted in environmental deterioration. Urban air pollution is an increasing problem, among others, and threatens public health and local ecosystems. Growing economic activities (e.g., construction and industrial development), unregulated urban sprawl, and increasing mobility and transportation in AA have increased pollution emissions, traffic congestion, land and environmental deterioration, and risks to public health. Transportation-related air pollution due to increasing demands and traffic congestion is emerging as the most important source. Ethiopia’s deteriorating air quality (AQ) undermines its citizens’ quality of life, but the country has a limited capacity for AQ monitoring and management. An analysis of AA’s visibility data suggests that air quality has been declining since the 1970s, with the average air quality now approximately 1.6 times worse than in the 1970s.9 Although the capital has the largest air quality monitoring network in the country, it lacks the long-term data required to determine air quality variations temporally and spatially; AQM hardly exists in other cities. The Ethiopian Constitution protects the right to a clean and healthy environment (Article 44/1). The country’s National Environmental Law Development and Enforcement Programme (NELDEP) identifies air pollution as a key area for improvement from 2020-2030 under its vision to integrate environmental laws into Ethiopia’s development strategies.10 The city of AA is also taking steps to develop its first Air Quality Management Plan (AQMP), which has identified a lack of basic information and analytical work on ambient AQ monitoring and source apportionment and a weak institutional capacity. Ethiopia has few AQM studies, even in AA, especially on ambient/outdoor air pollution. Limited AQ monitoring data are available, although the last two years has seen an installation boom in low-cost sensors in AA. Information on source apportionment and chemical compositions of particulate matter (PM) and other air pollutants is scarce. There are a few reports that cover source apportionment of PM2.5 and PM10 sources in AA (Etyemezian et al., 2005; Gebre et al., 2010; Tefera et al., 2020), though there are no published sources identifying the PM and other pollutant sources nor CO2 via receptor modeling or principal component analysis. 5 UN, 2019. World Urbanization Prospects: The 2018 Revision. Available at https://worldpopulationreview.com/world-cities/addis- ababa-population�. 6 Ethiopia Urbanization Review: Urban Institutions for a Middle-Income Ethiopia. World Bank. 2015. Available at: https://openknowledge.worldbank.org/handle/10986/22979. 7 World Urbanization Prospects - United Nations population estimates and projections of major Urban Agglomerations. Available at https://worldpopulationreview.com/world-cities/addis-ababa-population. 8The State of Addis Ababa. UN Habitat. 2017. Available at: https://www.urbanafrica.net/wp-content/uploads/2017/07/State-of-Addis- Ababa-2017-Report-web-1.pdf. 9 ASAP East Africa, 2019.-Air Quality Briefing Note: Addis Ababa (Ethiopia). https://assets.publishing.service.gov.uk/media/5eb16f4b86650c4353446282/ASAP_-_East_Africa_-_Air_Quality_Briefing_Note_- _Nairobi.pdf. 10 Ethiopian National Environmental Law Development and Enforcement Programme 2020 – 2030. Ethiopia Environment, Forest and Climate Change Commission. 2020. 2 Studies on air pollution outside of AA are also hard to find (Kume et al., 2011; Embiale et al., 2019; Amhayesus, 2019; Bulto, 2020). Transport air pollution mitigation measures have not been systematically assessed, despite some efforts to study the causes of transport air pollution and prepare a strategy for transport environmental control in AA. AQM has been one of the areas that the World Bank has been supporting. The World Bank launched an Advisory Services & Analytics (ASA) program entitled “Ethiopia: Air Quality Management and Urban Mobility� in September 2020. The ASA program aims to assist the Government of Ethiopia in deepening its understanding of ambient AQM through analytical studies and develop policy recommendations and an action plan for institutional strengthening and physical investments. Given the complexity and magnitude of AQM issues and time limitations, the program’s scope of work, as shown in Figure 1 below, focus on a few main issues in AA: reviewing ambient AQ and institutional arrangements, assessing health, valuing its economic cost, analyzing emission inventory and source apportionment, and identifying and prioritizing urban transport air pollution mitigation measures. This report summarizes the main findings and recommendation of the studies carried out under the program. It is not feasible for the program to conduct the studies nationwide in a short period and lacking data; the AQM program largely focuses on AA. The Program builds on the ongoing work of AA’s AQM planning and AQ monitoring carried out by the City Government of AA, the US Embassy/USEPA, UNEP, and other partners. Due to limited time and resources, the modeling work on air pollution dispersion, spatio-temporal analysis of travel time and vehicular air emissions, and in-depth economic and financial analyses of AQM programs and projects will have to be done in the future. Figure 1. Relationship of the Components of the World Bank AQM Program Comp 2: Pollution Source AQM Plan in AA Apportionment (using satellite (drafted by AA with the support of US Government) images/open source data) and Emission Inventory Comp 1: Overview of air Comp 3: Recommendations Comp 2: Transport air pollution pollution issues (AQ, institution, and roadmap for Ethiopia mitigation measures health, economic costs) AQ Monitoring (ongoing AQM Modeling: air In-depth Economic programs by AAEPGDC and dispersion, transport and Financial Analysis other partners) congestion (future activities) (future activities) This report is the synthesis of the key findings and recommendations of the studies carried out under the World Bank’s Ethiopia AQM ASA program. The rest of the report is organized as follows. Chapter 2 is an overview of air pollution issues, including air pollutants and impacts, the concept of integrated AQM, and the air quality situation in AA. Chapter 3 assesses the health impact of air pollution in AA and valuates its economic costs. Chapter 4 summarizes the study results of emission inventory and source apportionment, while Chapter 5 proposes and prioritizes the mitigation measures for transport air pollution control. The final Chapter provides recommendations and a road map that Ethiopia might take to address emerging air pollution problems and improve AQM nationwide and in AA over the next ten years. 3 2. Overview of Urban Air Pollution Problems in Ethiopia This chapter provides a brief overview of air pollution problems and impacts, introduces the IAQM approach, reviews institutional arrangements and gaps for AQM in Ethiopia, and examines AA’s AQ monitoring programs and results. 2.1 Introduction to Air Pollution Problems Ambient air pollution is a complex issue related to emissions from multiple pollutants from various sources involving numerous stakeholders across multiple sectors. These different pollutants create a chain reaction of health and environmental impacts that require addressing. Figure 2 below uses the generic DPSIR diagram to show the drivers (D), pressure (P), state (S), impact (I), and response (R) and their interaction in AQM. Figure 2. DPSIR Diagram for Air Pollution Common drivers of ambient air pollution in a city and its surrounding areas include activities such as cooking and heating; industries, including thermal power plants; construction sites; unpaved roads; and bare earth open areas; agricultural activities (especially the burning of agricultural residuals), and solid waste management, including the open burning of trash. Some studies in other African cities demonstrate that the transport sector is dominant in air pollution emissions. For example, a study in Nairobi, Kenya, shows traffic is responsible for 39% of ambient PM2.5 concentrations (Gaita et al., 2014). Air pollution from transport includes emissions from inefficient, aging vehicles, incomplete combustion from diesel vehicles, traffic congestion, and unpaved roads. Limited air pollution emission data exists in Ethiopia, and there are no systematic studies on air pollution source 4 apportionment; this is also true for AA. A lack of information and data of air emissions inventory and source apportionment hinders pollution control planning. Therefore, the World Bank AQM program conducted pollution emission inventory and source apportionment studies, with the results summarized in Chapter 4. Primary air pollutants that put pressure on airsheds include PM, carbon monoxide (CO), nitrogen oxides (NOx), sulfur oxides (SO2), and volatile organic compounds (VOC). PM is a complex pollutant and further divided by its aerodynamic diameter in micrometers (µm); most commonly, PM2.5 with diameters less than or equal to a nominal 2.5µm and PM10 less than or equal to 10µm. PM2.5 is considered the most relevant indicator for urban air quality (Cohen et al., 2005) and an important risk factor for premature death worldwide. It can pass the barriers of the lung, enter the bloodstream, and destroy the integrity of the blood-brain barrier, thus causing premature deaths, as well as respiratory, cardiovascular, and neurological diseases (Brook et al., 2010; Bowe et al., 2019; Shou et al., 2019; Peeples, 2020). CO, NOx, and VOC are mainly from vehicular or industrial activity emissions, and SO2 is usually the byproduct of coal and burning fuel. These pollutants also cause respiratory, circulatory system, and heart problems, and can be fatal. Moreover, studies recently associated air pollution with increased infections, such as influenza and COVID-19 (Petroni et al., 2020; Zivin et al., 2021). Air pollution negatively impacts human health, the economy, and ecosystems. Figure 3 below shows the main air pollutants and their impacts from the corrosion of man-made structures and cultural heritage, impaired health, and cognitive development–particularly in children–to their damage to ecosystems, for example, excess nitrogen in waterways harms aquatic life. Figure 3. Air Pollution and Its Impacts 2.2 Need for an Integrated AQM Approach Air pollution issues involve a complex network of sectors and stakeholders, which creates challenges for aligning policies, regulations, and government structures that address air pollution. An integrated approach to AQM is necessary to address complexities around AQM – including challenges in working across sectors, engaging a diverse set of public and private stakeholders, and other technical and financial constraints. 5 Figure 4 depicts an Integrated air quality management (IAQM) approach, which involves developing a knowledge base or analytical tools for air pollution emissions and AQ monitoring that can support pollution control planning. An effective AQM plan will require actions from multiple stakeholders, targeting different pathways and pollutants. Analysis of the interplay between different sources of pollutants, their consequences on health, the environment, and the economy, supports the interpretation of data for different actors. Additionally, it is key to implement an AQM strategy by building institutional and human resource capacity for pollution monitoring and management; institutional coordination between different sectors; national standards, regulations, and policies for pollution; and compliance monitoring and regulatory enforcement. Improved public awareness on sources and impacts of pollution can further support successful AQM. Figure 4. Integrated Air Quality Management IAQM has been a key approach for countries and cities to comprehensively address air pollution problems and protect the public from health issues and economic losses resulting from poor AQ. International experiences in most developed countries, and progress in some developing countries (e.g., Mexico, China, and Thailand), demonstrate that IAQM has successfully reduced air pollution and GHG emissions and dramatically improved air quality. 2.3 Stakeholder Mapping and Institutional Arrangements for AQM The Government of Ethiopia recognizes that long-term economic growth must incorporate sustainable, climate- aligned development to combat challenges related to urbanization and pollution problems. The City Government of AA has also started to pay attention to its growing air pollution problems and launched its AQM plan in May 6 2021. To provide an accurate assessment and relevant recommendations for AQM in Ethiopia, this study reviewed the stakeholders and institutional arrangements currently available for air pollution. Stakeholder Mapping Air pollution and AQM involve numerous stakeholders, including households, vehicle owners, manufacturers, and so on, across various sectors. Figure 5 maps out key stakeholders with whom AQM needs to interact. They include those that emit air pollution, are affected by air pollution, and are in charge of AQM, such as government agencies, public institutions, households, and business sectors. Figure 5. Air Quality Stakeholder Mapping Households cook and openly burn their trash, emitting and inhaling pollution from different sources. The transportation sector has the largest air pollution emissions, and commercial activities, developers, and construction projects emit pollution. Agricultural stakeholders are impacted by air quality because pollution can negatively affect their crop yields. Many government agencies are responsible for parts of AQM. Academic and research institutions and NGOs study air pollution impacts, deepening the understanding of air pollution issues, advocating for improved AQM, and raising public awareness. The public and private sectors as well as bilateral and multilateral organizations, play a role in financing air quality-related initiatives at the national and local levels. Managing the different stakeholders involved in air quality management is complex. Different layers of participation must happen–from government agencies to institutions and even to households. Government agencies from different levels and across different sectors must develop and implement policies and regulations 7 that consistently minimize air pollution across their jurisdictions while aligning with international standards. This can be challenging, as agencies need to balance priorities while accommodating objectives from other agencies. The public can also be complicated to manage, as they need awareness programs that encourage participation and behavioral changes in reducing air pollution. Strong political will, leadership, and senior officials' commitments are essential to ensure that agencies plan, implement, and enforce air quality measures. Highlights of Regulatory and Policy Review Ethiopia has many different environmental policies and regulations at the federal and city levels. These focus on environmental protection and health or address environmental concerns for specific industries, such as the energy and transportation sectors, with the ultimate purpose of protecting citizens' health by providing healthy living environments and balancing economic development with preventing environmental pollution (UNEP & ECI, 2018). Although these policies and regulations touch on air quality (including vehicle emissions testing) in different ways (UNEP & ECI, 2018; AATB, 2020), the country has no AQM-specific policy as well as no national AQM strategy or plan–aside from its reference in the CRGE Strategy. Ethiopian national standards of ambient AQ were issued in 2003 as guideline values which cover SO2, NO2, CO, ozone (O3), lead (Pb), PM10, and PM2.5. The guideline value for annual average PM2.5 is set at 15 µg/m3, higher than the WHO guideline of 10 µm/m3 as well as standards in many other countries (UNEP & ECI, 2018). There are few rules or laws around fuel quality and vehicle emissions standards. There are only two parameters in Ethiopia around vehicle emissions – smoke and carbon monoxide (CO) and they were however introduced within the Standards for Industrial Pollution Control. AA currently has no vehicle emission standards. The lack of fuel quality and vehicle emission standards leads to the transportation sector and vehicle owners emitting unnecessarily high levels of air pollution through low-quality fuel (EPSE, 2021). To create aggressive economic development, the objectives in different policies conflict with promoting environmental protection, including air quality, and often are not evidence- based (UNEP & ECI, 2018). Furthermore, the country’s regulatory and policy framework lacks explicit AQM measures that intersect different sectors. In sum, the current regulatory and policy framework provides an insufficient incentive to manage and penalize air pollution activities from any particular stakeholder group (AACPPO, 2017). Ethiopia needs to enhance current ambient air quality standards, introduce air pollution emissions and fuel quality standards, and develop a more comprehensive and integrated policy package specifically addressing and guiding AQM, including providing clear financial incentives and penalties on air pollution activities from the urban transport, industrial and business sectors. Governmental Organizational Setup for AQM The Government of Ethiopia has designated a few agencies responsible for different aspects of AQM at the federal and municipal levels. In the federal government, the Environment, Forest, and Climate Change Commission (EFCCC) is the national environmental protection agency (EPA) that leads air quality efforts by coordinating national policies and laws and overseeing AQM (Kumie et al., 2014; UNEP & ECI, 2018). The Ethiopian Standard Agency (ESA) prepares air quality standards in collaboration with the EFCCC. Other agencies support AQM, including the ministries for health and transport. As the city-level EPA, the AA Environmental Protection and Green Development Agency (AAEPGDA) coordinates the city's AQM strategies and oversees air pollution controls (Tariku et al., 2019). Other city agencies support air pollution, such as the bureaus for transport and health, matching the federal level (AATB, 2020). However, there is a lack of functional units in charge of AQM in leading agencies such as EFCCC and EPGDA, making it difficult to ensure AQM 8 roles and responsibilities. Also, the mandates of some line ministries or agencies (e.g., Transport Bureau or Health Bureau in AA) for AQM are not clearly defined, despite the organizational structure of the governments for AQM. Ethiopia’s enforcement of environmental regulations and standards, including those related to AQ, has been inconsistent and weak because of conflicting goals, insufficient data, low capacity for government coordination and implementation, and low persistence of enforcement (UNEP & ECI, 2018). In particular, agencies with AQM roles normally have limited personnel and capacity to develop, plan, and enforce AQM-specific policies and regulations and prepare and implement AQM activities. For example, Federal EFCCC and AA EPGDA have little means to monitor AQ and enforce ambient AQ standards and policies. The AA Transport Bureau supporting vehicle emission control in the city only has two staff members who are without a sufficient budget, equipment, or guidance on their work to manage vehicle emission control (AATB, 2020). As a result, there is limited data for benchmarking standards and existing ambient air quality standards are barely enforced in AA (UNEP & ECI, 2018). The situation is undoubtedly even worse in the country’s regional and secondary cities. Government agencies or units also face limited capacity to coordinate and implement policies, regulations, and plans among themselves and across sectors. Weak coordination, collaboration, and communication among various government agencies and stakeholders are common horizontally and vertically. Government bodies are not used to collaborating on air quality efforts, and existing governmental structure and policy frameworks do not offer guidance. Some ministries that would be effective in the organizational setup–such as the ministries for agriculture, energy, trade, and industry–are not yet involved in AQM, hampering AQM efforts. Another underdeveloped area is finance and financing arrangements for AQM. Environmental taxation and fees are rare, with little earmarked to fund air pollution control activities. Line ministries/agencies and municipal governments commonly have no specific budget for AQM. Private sector investments in AQM are minimal, except for in recent years some partnerships of private entities and public or international organizations to finance AQ monitoring programs. Few investment activities have been made in air pollution reduction even in key polluting sectors, for example, urban transport and industry. 2.4 Air Quality in AA: A Review of Existing AQ Monitoring Efforts and Results AQ monitoring in AA AQ monitoring in Ethiopia is fairly new. The NMA established the first batch of three AQ monitoring stations in 2014–one at a meteorological station in AA and one each in Adama and Hawassa. NMA's stations only measure the gaseous pollutants CO, NO2, and O3, with the AQ monitoring network in AA gradually expanding since then. In 2016, the US Government (through the USDOS and EPA) set up two AQ monitoring stations in AA–one inside the US Embassy and the other at International Community School (ICS). The Eastern Africa GEOHealth Hub also set up a station in 2017. These three reference-grade monitoring stations use Beta Attenuation Monitors (BAM) and monitor PM2.5. In 2019, there was a dramatic increase in AQ monitoring stations when AddisAir and AAEPGDC/UNEP installed low-cost sensors. As of March 2021, there have been about 20 AQ monitors installed in AA (Figure 6). C40 Cities, SPARTAN Network, and GIZ are also planning to support AQ monitoring in AA; more AQ monitoring devices–stationary or mobile–should be available in the coming years. 9 Figure 6. Location of AQ Monitors in Addis Ababa During the period from 2016 to 2019, only three PM2.5 monitoring sites operated in AA, all of which were financed by development partners, i.e., the two installed by the US Diplomatic Post (USDP) at US Embassy and International School, respectively, and one by East Africa GEOHealth Hub at Black Lion Hospital (BLH) on the campus of AA University. Only the two US-operated monitoring stations, which are part of the US DOS's global ambient air quality monitoring network, have since disseminated hourly PM2.5 data to the public in near real-time via AirNow.gov since 2016 (Dhammapala, 2019). However, the ICS monitor has had operational challenges for months due to power and internet connectivity-related problems. Starting in late 2019, additional AQM sensors were installed by AAEPGDA with the support of UNEP and by AddisAir, a network of citizens and organizations. These new sensors are low-cost sensors, some of which malfunction for various reasons, and provide limited uncalibrated data. The GEOHealth Hub's monitoring station at BLH started operating in April 2017. Its monitoring data is not yet available to the public due to the project’s research nature; this report only obtained an aggregated average of PM2.5. Table 1 presents the coordinates, aggregated PM2.5 concentration, and operational period of the three reference grade monitoring stations at the ISC, US Embassy (Central), and BLH. It shows that BLH's PM2.5 monitoring results are higher than those from ISC and US Embassy location monitors, likely due to its location in the urban center. Table 1. Location and PM2.5 Monitoring Results in Addis Ababa Location ICS US Embassy TASH/Black Lion 10 Hospital / AAU Longitude 38.727623 38.763649 38.7425 Latitude 8.996519 9.058076 9.0154 Average Conc. 30.7 24.3 43 PM2.5 (ug/m3) Year reported 2020 2020 2020 4/1/17- Period -- 01/101/20-12/31/20 3/30/20 Total number of -- 348 -- days monitored Largely residential Urban background Traffic/Institutions/ Land use zoning area area residential AAEPGDC worked on another AQ monitoring effort in AA with the technical support of UNEP as part of the US-funded project. It has installed 6 Kunak-cloud low-cost sensors (LCS) (out of 10 planned) at different locations in AA. The monitors have measured AQ since September 2019, and output data utilizing the AQCSV standard to enable interoperability. However, sensor availability for continuous measurement has been challenging for various reasons including stability of the electrical power supply. Limited monitoring data is accessible through several applications situated on the UNEP WESR platform (https://wesr.unep.org/topic/index/1). The system will be integrated into a heterogeneous data management system under development in collaboration with the US EPA to take advantage of reference instruments within the wider network for quality assurance. AddisAir launched its AQ monitoring network program in late 2019 to understand AA's AQ and increase citizens' awareness of AQ and its health effects. As of March 2021, It had ten sensors installed at different locations in AA. Their sensors are real-time and low-cost, using a light-scattering method to measure PM2.5 and send monitoring data regularly to AddisAir's website (https://airquality.addisabeba.info/). Though AddisAir’s sensors have provided some data since late 2019, most of them are not functioning as of April 2021. Table 2 details the features of its ten sensors and shows that the PM2.5 average concentrations across the city, from 22.4 µg/m3 in Arat-Kilo to 55.5 µg/m3 in the EU Delegation to AU location. However, a lack of calibration of the monitoring results raises some concern over the data’s reliability. 11 Table 2. Location and Results of AddisAir's PM2.5 Monitoring Network Location EU EU ICS/ Arat- Churchill Delegation ILRI/ Lycée Peacock Park- Delegation Bisrate Kirkos Lambert Kilo* Avenue** to CCAFS Francais Bole Atlas to AU Gebriel Ethiopia Longitud 38.72762 38.7579 38.7520 38.7411 38.7570 38.7579 38.7520 38.7411 38.7570 38.8143 e 3 Latitude 9.0351 9.0221 8.999 8.9819 9.0351 9.0221 8.999 8.9819 8.996519 9.0163 Average Conc. 22.4 45.5 55.5 24 24.4 23.9 29.2 26 45.3 23.6 PM2.5 (µg/m3) Year 2020 2020 2020 2020 2020 2020 2020 2020 2020 2020 reported 12/3/20 12/4/19- 12/3/2019- 19- 12/4/19- Period 11/6/202 12/4/19-8/1/2020 6/30/2020 9/3/202 12/3/2020 0 0 Number 207 200 124 356 239 of days Green Green Traffic/ AU HQ parks Commer Residenti Land use Residenti Traffic/ area/ Commerc Parks/ area/ backgroun Residentia cial/ al area at the al area Commercial/ Instituti ial/ River Kebena/ Residential/ d l area Resident (Hilly location School area ons/ School Urban agriculture Traffic area/Traffi ial side) Offices area c 12 Evaluation of AQ in AA According to the WHO and US EPA standards, PM2.5 concentrations should not exceed 10-12 µg/m3. The Government of Ethiopia's national AQ standard for annual average PM2.5 has been 15 µg/m3 since 2003. The available monitoring results indicate the annual average of PM2.5 concentrations shows that AA's AQ exceeds the standards. Some AQ monitoring stations indicate that PM2.5 can be 4-5 times above the WHO standards. The results from the US-monitored sensors indicate that AA's rainy season AQ is routinely Unhealthy for Sensitive Groups (USG) and unhealthy even during periods of stronger wind. Winds of all speeds, mostly from the southwest quadrant, degrade air to USG levels, while other directions are mostly associated with Good or Moderate Air. Table 3 and Figure 7 below further present monthly average concentrations and their variations. The table considers all the hourly averages within a month available from all the stations to construct the variations. Table 3. Monthly Average (and Standard Deviation) in the PM2.5 Concentrations Data in 2018-19 from All the Stations in Addis Ababa Airshed Unit: µg/m3 January 29.2 ± 14.2 July 42.5 ± 11.9 February 24.0 ± 11.8 August 42.4 ± 12.6 March 25.4 ± 9.6 September 46.9 ± 13.7 April 26.3 ± 7.4 October 31.3 ± 11.3 May 30.5 ± 8.7 November 27.8 ± 15.5 June 50.2 ± 12.1 December 28.9 ± 14.6 Figure 7. A Summary of All PM2.5 Concentrations Data Showing Monthly Averages and the Variation in Hourly Data by Month Source: Authors’ interpretation Typically, experts do not expect high concentrations of PM2.5 during the rainy months (June through August), but the trends in AA appear to deviate from expectations. There was also a short AQ monitoring and research from November 2015 to November 2016, where the research collected 24- hour PM2.5 samples in AA's city center every 6 days. Its results show that the mean (±SD) daily PM 2.5 concentration was 53.8 (±25.0) µg/m3, with 90% of sampled days exceeding the WHO's guidelines (Tefera et al., 2020). Typically, it is not expected to see highs in PM2.5 during the rainy months (Jun- Jul-Aug), but the trend in AA is an aberration which can be linked to lower surface temperatures, lower mixing layer heights and wind speeds under 2 m/s causing limited dispersion of emissions. These observations are consistent with the prevalent amount of anthropogenic air pollution sources characterized by low buoyancy. For example, enhanced biomass burning during the rainy season is one possible explanation for high concentrations of PM2.5 (Dhammapala, 2019). A summary of daily trends and a comparison of monthly average data from reference-grade and low- cost sensors present similar qualitative trends. However, there are differences in absolute values. Low- cost sensor data indicate PM2.5 levels average 30% higher than the reference-grade sensor data. However, it must be noted that the locations for these sensors have different characteristics and land use at their location and lack information to calibrate the sensors. Since the overall quantity and quality of sensor data was not good but this collective data is useful in qualitative model verifications. The overall comparison and validation exercise can be replicated in the coming years when more data is available from the expanded AQ monitoring network. In summary, the air quality of AA is unhealthy and its annual average of PM2.5 in recent years exceeds WHO guidelines by 3-4 times. Air pollution concentrations vary by location and time. However, a limited number of functioning and calibrated AQ monitoring stations in AA hinders the thorough assessment of ambient AQ in the city. AA needs to continue to expand its AQ monitoring network and develop a unified and standardized data management system for AQ monitoring. Epidemiology studies have found consistent association with both mortality and morbidity by exposure to fine particulate matter (PM2.5) air pollution. The next chapter will further introduce the population- weighted PM2.5 concentrations and assess the health impact of AA's air pollution and economic costs. 2.5 Recommendations Ethiopia must strengthen its institutional arrangements to support AQM. To start, authorities should review, strengthen, and harmonize existing regulatory and policy frameworks at the national and local levels to address AQM requirements well. Along with reviewing frameworks, the government should conduct a functional review of government agencies to further clarify line ministries and their roles, mandates, and functions for AQM. This could include improving details on which government agencies should have AQM units or staff specifically allocated to manage their mandates. Once government agencies have clear AQM mandates and functions communicated, building institutional capacity in the line ministries about AQM. Building capacity will be particularly important to ensure agencies can properly regulate and enforce AQ standards, monitor air pollution control activities, and manage the daily tasks required to ensure AQM is successful. Along with this clarification of roles and capacity-strengthening efforts, authorities should introduce a mechanism for inter-agency and cross-sectoral coordination and collaboration on AQM activities that also fosters learning and experience sharing. These mechanisms can also build a coalition to mobilize different stakeholders and incentivize them to change behaviors that emit significant air pollution. These stakeholder groups include the general public (including households and vehicle owners), the private sector, civil society, and other groups that can significantly contribute to the financing and management of air pollution prevention. These institutional reforms will only succeed if key line ministries have adequate budget and support to accomplish these reforms. Budgets can be boosted through examining environmental taxation and fee systems and develop the tax base and sources for AQM’s public financing. For institutional reform, Ethiopia also needs to introduce new standards for ambient air quality and air pollution emissions. This work would need to include increasing national guideline standards of ambient AQ, particularly updating the PM2.5 standard to follow WHO guidelines. It would also need to 14 include new fuel quality and emission standards specifically for the transportation sector that is clear and enforceable so that vehicle owners and fuel users can comply with them. Part of this process should also include supporting the government's petroleum authority, the Ethiopian State Petro Company (ESP)’s efforts to import fuel with lower sulfur contents, started with 50 ppm. Apparently, the regulations and standards on ambient air pollution control in other sectors such as industries, solid waste management, and agriculture are under development and more studies are necessary on these areas. Secondly, Ethiopia must establish a most robust AQ monitoring program and network nationally and strengthen existing AQ monitoring networks. The AQ monitoring network in AA would also require the key line ministries to strengthen information management of AQ data. Authorities can establish better data management by developing standardized and unified systems to report AQ and emission levels. This reporting system should make it easy for the public to access the internet and mobile apps. In AA, the capital city’s ambient AQ monitoring network should have reference-grade systems at multiple representative locations for the analysis of long-term pollution trends and support them with calibrated sensors to increase the overall data density and produce a pollution map. This network can support regulations and validate the baseline emissions and pollution estimates similar to those presented in this report. 15 3. Health Impacts and Economic Costs of Air Pollution in Addis Ababa As introduced earlier, air pollution is a threat to public health and environmental quality. Air pollution affects lung function and can trigger asthma, among other health conditions, and can lower productivity. Air pollution also has a distinct impact on vulnerable groups, including women and children, as studies have shown that high exposure to air pollution can affect the ovaries and fertility.11 In Ethiopia, PM pollution is the second leading risk factor for death after malnutrition (IHME, 2020). Local residents in urban areas, especially in a large city like AA, are at considerable risk of heart and lung diseases and premature death. While household air pollution is the most prominent contributor to mortality,12 ambient PM2.5 pollution is a growing public health problem in urban areas (Berhane et al., 2016), primarily due to vehicle exhaust, resuspended dust on roads, and industrial activities (see Chapter 4 for more details). Several studies have assessed the consequences of ambient air pollution on people’s health in Ethiopia (see Tefera et al., 2016 for a review; and IEC, 2019); however, no economic valuation has been done of the health impacts of air pollution. This chapter addresses this gap by providing a physical and economic assessment of the health impacts caused by ambient PM2.5 pollution in AA. It uses the most updated methodology developed by the Institute of Health Metrics and Evaluation (IHME-2020) regarding the Global Burden of Disease (GBD) and the WHO for the health impact assessment, and the World Bank/IHME (2016) guidelines for the economic valuation of the damages to health. The valuation focuses on a typical year,13 then briefly highlights potential linkages among the current pandemic, air pollution, and health. As the analysis was conducted during a short period (December 2020-March 2021), it is based on secondary information, complemented by some primary PM2.5 data reported in the report’s overview chapter. The following sections present the main results of the analysis, while Annex 2 provides further details. 3.1 The Health Impacts of Ambient Air Pollution The health effects of long-term exposure to ambient PM2.5 include ischemic heart disease, lung cancer, chronic obstructive pulmonary disease (COPD), lower respiratory infections (such as pneumonia), stroke, type 2 diabetes, and adverse birth outcomes (GBD 2019 Risk Factors Collaborators, 2020). This chapter assesses these impacts based on the four steps presented below. Estimate the PM2.5 exposure Exposure to PM2.5 is commonly estimated using the population-weighted PM2.5 concentrations (PWC) for a given year, which are based on the number of people living within a certain area and the PM2.5 concentrations they are exposed to. PWC provides better population exposure estimates than other types of averages (e.g., arithmetic mean, median, mode). It gives proportionally greater weight to the air pollution experienced where most people live (HEI/IHME, 2020). Data derived from ground monitors are often seen as the “gold standard� globally. However, their spatial coverage is usually limited (Zeger et al., 2000); many efforts have focused on measuring PM2.5 concentration using other methods (e.g., satellite-based imagery and atmospheric chemical models) to 11 European Society of Human Reproduction and Embryology. "Air pollution found to affect marker of female fertility in real- life study: Decline in ovarian reserve related to particulate matter and nitrogen dioxide in atmosphere." June 2019. 12 In 2019, there were 67,800 deaths due to household air pollution vs. 8,960 deaths due to ambient air pollution in Ethiopia (http://ghdx.healthdata.org/gbd-results-tool, accessed March 2021). 13 In the context of this chapter, a typical (regular) year is intended as a year without dramatic changes, such as those brought by COVID-19. Assessing the impact of an atypical year would require consideration of additional factors for which current data are not available, e.g., changes in population exposed to ambient PM2.5. 16 overcome this problem. That said, these methods complement surface ground-monitored data rather than fully replacing them (Duncan et al., 2014). Ideally, integrating data from ground-based monitors, satellite imagery, and other models should leverage the benefits of each data source –thus providing PM2.5 concentration estimates over a wide geographic scale with better accuracy (Diao et al., 2019). This study did not conduct this type of data fusion14–in its absence, the average population-weighted PM2.5 concentration was estimated based on available ground-level monitored data and the population exposed to air pollution in the proximity of each station. Table 4 highlights the results, which indicate that the annual average PM2.5 concentration varies between 30 µg/m3 and 36 µg/m3 across 2016-2020. These estimates are in the same broad range; the differences among them result from several factors, such as differences in the number of stations from which data were collected and possible changes in seasonal parameters across years (e.g., rainfalls, wind speed, etc.) This analysis aims to estimate PM2.5’s impact on health during a typical year by using exposure averages across 2016-2019 as the basis, which corresponds to 34 µg/m3. Table 4. Annual Average PM2.5 Concentration in Addis Ababa Population-weighted PM2.5 Year Observations on monitoring stations concentration (µg/m3) 2016 2 stations: USDP Central, USDP School 36 2017 3 stations: USDP Central, USDP School and Black Lion 34 2018 3 stations: USDP Central, USDP School and Black Lion 35 2019 2 stations: USDP Central and Black Lion 30 14 stations: USDP central, USDP school, Black Lion, 2020 33 NMA Geo Health, and 10 Addis Air stations. Quantify the impacts of ambient PM2.5 on premature mortality Several epidemiological studies revealed strong correlations between long-term exposure to PM2.5 and premature mortality (Apte et al., 2015; Cohen et al., 2017; Wu et al., 2020). Recent research associated PM2.5 exposure with mortality related to several other health outcomes: lower respiratory infections; tracheal, bronchus, and lung cancer; ischemic heart disease; stroke; chronic obstructive pulmonary disease; type 2 diabetes mellitus; and adverse birth outcomes (IHME GBD 2019 Risk Factor Collaborators, 2020). The number of premature deaths attributable to ambient PM2.5 pollution in Addis Ababa is estimated for each of the above outcomes using data on (i) mortality by disease and age group at the national level, based on IHME (2020); (ii) adjustment for AA, considering the differences in population distribution by age group between the city and the national level (projections from CSA, 2013); (iii) proportion of deaths attributable to PM2.5, calculated based on specific relative risk factors for the PM2.5 concentration of AA (34 µg/m3), which differ by the outcome. It is important to note that the most recent relative risk factors developed by the IHME GBD 2019 Risk Factors Collaborators (2020) differ considerably from those used in the previous GBD methodology.15 The results show that exposure to ambient PM2.5 causes about 1,600 premature deaths per year, on average. While the estimate provides a good indication of the magnitude of the air pollution problem, it should not be interpreted as a precise quantification of the impact of air pollution in the city. To reflect this uncertainty, a range of 1,200 to 2,100 premature deaths has been estimated, based on the lower and 14 IEC (2019) used air quality data from ground monitors and satellite sources for ground-truthing the PM2.5 information for 2016. However, as the present analysis refers to a more recent period (2016-2019), it did not use the IEC information. 15 This is because the most updated risk curves for ischemic heart disease, stroke, chronic obstructive pulmonary disease, lower respiratory infections, and type II diabetes do not include studies of active smoking data. For more details, see supplementary appendix 1 to GBD Risk Factors Collaborators (2020). 17 upper bounds of mortality provided by the GBD 2019 in Ethiopia. Stroke, ischemic heart disease, and lower respiratory infections are the leading causes of PM2.5-related mortality. The elderly (people between 60 and 84 years of age) are most affected by PM2.5, accounting for about 58% of premature deaths (Figure 8). The same group accounts for a relatively large share of the total premature deaths due to ambient PM2.5 at the national level, according to the IHME GBD 2019 study (Annex 2). Figure 8. Estimated Number of Premature Deaths Due to PM2.5 Exposure, by Age Group 250 200 150 Premature deaths 100 50 0 Age groups LRI and adverse birth outcomes Lung cancer COPD IHD Stroke Diabetes mellitus type 2 Source: Authors, based on data from IHME (2020) and GBD 2019 Risk factors collaborators (2020). Notes: IHD = ischemic heart disease; LRI = lower respiratory infections; COPD = chronic obstructive pulmonary diseases; the adverse birth outcomes include neonatal disorders due to low birth weight and short gestational age. Quantify the impacts of ambient PM2.5 on morbidity In addition to mortality, exposure to PM2.5 causes disability related to a plurality of health outcomes (e.g., chronic bronchitis, respiratory and cardiovascular hospital admissions, lost workdays), suffered by various agents (e.g., patient, family, co-workers) (World Bank/IHME, 2016). Based on recent guidelines on quantifying non-fatal outcomes (Robinson and Hammitt, 2018), this study estimates the morbidity due to exposure to ambient PM2.5 based on the number of Years Lived with Disability (YLDs). The YLD is defined as the equivalent of one full year of healthy life lost due to disability or ill health (WHO, 2021). It measures the burden of living with a disease or disability. Similar to the valuation of premature deaths, the number of YLDs is valued using data on (i) the total number of YLDs by disease and age group, based on the IHME (2020); (ii) adjustment for AA, considering the differences in population distribution by age group between the city and the national levels (projections by CSA, 2013); (iii) proportion attributable to PM 2.5, calculated based on specific relative risk factors for the PM2.5 concentration of AA. The results indicate that exposure to ambient 18 PM2.5 is responsible for about 4,100 YLDs annually,16 on average (with a range between 2,800 YLDs and 5,700 YLDs). Middle-aged people (40–69 years of age) are most affected by illnesses due to PM2.5, accounting for about 58% of the total morbidity related to PM2.5. This result is consistent with the 2019 IHME GBD study at the national level, which found that the same age group accounts for a similar proportion in the total ambient PM2.5-related morbidity cases (Annex 2). To put the health assessment results into a regional context, Table 5 illustrates the ambient PM2.5 concentrations and their impact in selected African cities. It shows that the PM2.5 concentration and its effects on health in AA are slightly higher than those in other African capitals, namely Cotonou (Benin), Lomé (Togo), and Abidjan (Côte d’Ivoire). At the same time, it is lower than that in very polluted megacities, such as Lagos (Nigeria) and Cairo (Egypt). Table 5. Impact of Air Pollution in Selected African Cities Ambient PM2.5 Deaths due to air Cities Deaths/100,000 people concentration (µg/m3) pollution Dakar 21 270 25 Cotonou 32 200 32 Lomé 32 490 31 Abidjan 32 1,500 35 Addis Ababa 34 1,600 35 Lagos 68 11,200 46 Cairo 76 12,570 73 Sources: Authors for Addis Ababa; Croitoru et al. (2020) for Lagos; Croitoru et al. (2019) for Abidjan, Cotonou, Dakar, Lomé; Larsen (2019) for Cairo. Note: A portion of these estimates represents deaths due to the joint effect of exposure to ambient and household air pollution. Adjusting the estimates to capture only the impact of ambient air pollution would require the quantification of the number of people exposed to both household and ambient air pollution; in-depth knowledge about the causes of household air pollution; and data on household PM2.5 concentration in the affected areas. As this chapter did not focus on household air pollution, it does not incorporate this adjustment. The average estimate obtained–about 1,600 premature deaths–should be considered conservative because it is much lower than the estimate for AA AQM planning from US EPA and C40 Cities (2020) of 2,700 premature deaths. However, the US EPA estimate is related to different years, methodologies (Burnett et al., 2018), and tools (BenMap), and thus is difficult to directly compare with the result of this study. Finally, it is important to highlight that this section did not address the links between COVID-19, air pollution and health. This relationship is quite complex. First, COVID-19 affected air pollution in different ways: in many cities, the lockdown triggered lower vehicular traffic and industrial emissions (He et al., 2020); after the transition to partial relaxation, these concentrations often returned to pre- pandemic levels (Rybarczyk and Zalakeviciute, 2021). The situation is different in AA, where available research suggests that PM2.5 concentration was higher during the pandemic (March-July 2020) compared to prior months (October 2019–March 2020), primarily due to the lack of lockdown measures (Bulto et al., 2020). Second, it is important to note that long-term exposure to air pollution affected COVID-19 related mortality: for example, a study from Harvard Chan School of Public Health found that a 1 µg/m3 increase in air pollution was associated with an 11 percent increase in mortality from COVID-19 infection in the United States (Wu et al., 2020). Moreover, other studies found significant relationships between short-term exposure to air pollution and COVID-19 morbidity, for example in China (Zhu et al., 2020). 16Considering a population of 4.6 million in 2019 of Addis Ababa (https://worldpopulationreview.com/world-cities/addis- ababa-population), this is equivalent to about 0.3 days lived with disability per person per year, on average. 19 3.2 Economic Cost of Health Impacts due to Exposure to PM2.5 The economic valuation of the health impacts estimated in the previous section refers to both mortality and morbidity, as follows. Cost of mortality. This is estimated based on the number of premature deaths caused by the exposure to ambient PM2.5 and the Value of Statistical Life (VSL), based on the World Bank/IHME (2016) guidelines. The VSL reflects the society’s willingness to pay to reduce the risk of death, or rather, the local trade-off rate between fatality risk and money (Viscusi and Masterman, 2017; Kniesner and Viscusi, 2019). The valuation of the VSL for Ethiopia is based on the benefits transfer method, following the World Bank/IHME (2016) guidelines. Accordingly, the VSL for Ethiopia is estimated at $43,600 for 2019. It should be noted that the VSL result is a conservative estimate, which does not capture the full value of life.17 Using the number of premature deaths estimated in the previous section (1,600 deaths), the economic cost of premature mortality reaches $70 million. Cost of morbidity. This is estimated based on the number of YLDs due to exposure to ambient PM2.5 and the Value of Statistical Life Years (VSLY). The VSLY is derived by dividing the VSL by the discounted expected life years remaining for an individual at the mean age of the population studied (Robinson and Hammitt, 2018). Accordingly, the VSLY for Ethiopia was estimated at $1,800 for 2019. Using the number of YLDs estimated in section 3.1 (4,100), the cost of morbidity is valued at about $8 million. Economic value of health impacts. Adding up the values obtained above, the total health cost is estimated at $78 million. This is equivalent to 0.1% of the country’s GDP in 2019, or about 1.3% of AA’s GDP.18 The estimate is lower than that obtained in very polluted megacities, such as Lagos (2.1% of the state’s GDP), Greater Cairo (4.5% of its GDP), and New Delhi (5.5% of its GDP).19 The estimate for AA points to the need to further investigate the main pollution sources and particularly, the pollution hotspots in the city. For example, Kumar et al. (2021) examined the health risks caused by in-car open- window exposure to PM2.5 pollution in several hotspots of ten world cities. The authors found that in AA, the annual deaths caused by in-car exposure to PM2.5 pollution represents more than 10 percent of the total ambient PM2.5-related deaths among car commuting population; this is a very high share compared to that found for other polluted cities – such as Dhaka (Bangladesh), Chennai (India), and São Paolo (Brazil) – which is less than 3 percent. Considerable effort was made to calculate meaningful estimates of the impact of air pollution based on reliable information. Nonetheless, the analysis remains subject to several limitations. These relate to the use of (i) population-weighted PM2.5 concentrations based on data collected from monitoring stations with different–sometimes low–coverage across the period 2016-2019; (ii) relative risk factors that, while reflecting the newest development in the field (IHME, 2020), are not specifically adjusted to Ethiopian conditions; (iii) VSL and VSLY concepts, which are based on benefits transfer from other countries, due to lack of primary studies in Ethiopia. 17 The concept of the VSL is very different from and should not be compared with the notion of “salary� or “wage .� 18 Ethiopia’s GDP was $95.9 billion in 2019 (https://data.worldbank.org/, accessed March 2021). Addis Ababa’s GDP was estimated at 6.4% of Ethiopia’s GDP by the World Bank (2018). 19 For Lagos, the estimate refers to 2018, based on Croitoru et al. (2020); for Greater Cairo, it refers to 2017, based on the health cost of 1.35% of Egypt’s GDP, estimated by Larsen (2019); and the contribution of Greater Cairo to the country’s GDP of 30% (UN-Habitat et al., 2019); for New Delhi, it is based on https://energyandcleanair.org/revealing-the-cost-of-air- pollution-in-real-time/, and excludes the cost related to NO2. The estimate for New Delhi refers to the first half of 2020. 20 3.3 Recommendations Overall, this analysis estimates the health damage due to exposure to ambient PM2.5 at about $78 million, or 1.3% of AA’s GDP in 2019. Air pollution causes about 1,600 premature deaths a year, on average. Stroke, ischemic heart disease, and lower respiratory infections are the leading causes of PM 2.5- related mortality. People between 60-84 years of age are most affected by PM2.5, accounting for about 58% of the total premature deaths. The results point to the need to: • improve the assessment of health and other environmental impacts (e.g., visibility, changes in property prices) of ambient air pollution. To do so, it is important to strengthen the collection of data related to air pollution, e.g., PM2.5 concentrations at other representative locations; carry out epidemiological studies in Ethiopia using local hospital information, to adapt current concentration- response functions to the country conditions; conduct primary studies that assess the Value of Statistical Life and the willingness to pay to avoid morbidity, based on local information. • conduct new studies that apply modelling to estimate the relationship between urban mobility and air quality. This would involve data on traffic measurements (volumes and vehicle mix) along principal corridors as input to vehicle emissions models (like MOVES or COPERT), followed by a linear dispersion model to estimate street-level concentrations, matched to daytime population estimations. • identify the AQM interventions that reduce ambient air pollution, focusing on the most contributing sectors; and conduct economic analysis (e.g., cost-effectiveness or cost-benefit analysis) of these interventions - with the aim of reducing ambient air pollution to specific targets, such as those identified by the annual ambient air quality guidelines of Ethiopia (15 µg/m3) and of WHO (10 µg/m3) (UN, 2018; WHO, 2005). 21 4. Emission Inventory and Source Apportionment A prerequisite to creating an AQM plan is to have an idea of the main sources of pollution and their potential for controlling emissions. This would help in prioritizing both local and regional efforts for effective management. Therefore, the World Bank AQM program conducted studies on air emission inventory and source apportionment. This Chapter presents the emissions inventory results and quantifies the source contributions to PM2.5 pollution in AA. 4.1 Data resources In addition to the published reports on air quality in AA and Ethiopia, to support the background information on energy consumption and sectoral activities, the study also accessed a number of open- source databases listed below. • OpenAQ website to access ambient PM2.5 concentrations data measured at 2 reference grade monitoring stations (https://openaq.org) • AddisAir website to access ambient PM2.5 concentrations data (from November 2019 to November 2020) at 10 low-cost sensor stations (https://airquality.addisabeba.info) • EthioInfo dashboard of the Statistical Office of Ethiopia, for information on population, gross domestic product, fuel usage, and vehicle registrations (http://www.dataforalldemo.org/dashboard/v1/ethioinfo/ethioinfo#/) • FTA Ethiopia via staff communications for information on vehicle registrations by vehicle type and age • STATISTA is a commercial data service site, which provides limited information on vehicle sales, registration by vehicle type and year, population, and GDP (https://www.statista.com) • Open Street Maps (OSM) database for information on road network covering highways, arterial, and feeder roads; and commercial activity points such as hotels, hospitals, apartment complexes, industries, parking lots, fuel stations, malls, markets, office and commercial complexes, banks, cafes, restaurants, and convenience stores (http://overpass-turbo.eu) • European Space Agency (ESA)’s global human settlements (GHS) program for information on the built-up urban area in the airshed for the years of 1975, 1990, 2000, and 2014 (https://ghsl.jrc.ec.europa.eu/datasets.php) • LANDSCAN program for information on gridded population at 30 sec resolution for the entire city airshed (https://landscan.ornl.gov/landscan-datasets). This database uses official estimates from the respective governments at district and ward level, which is further segregated to finer grids using information on commercial, land use, and night light data fields. We also received the ward-level population totals from the Ethiopian office • FlightStats is a commercial data service, which provides information on domestic and international flight schedules for airports in the airshed (https://www.flightstats.com) • VIIRS satellite retrievals for information on open biomass-burning fires spotted daily. The satellite resolution is 375m and capable of detecting fires during daytime and nighttime (http://viirsfire.geog.umd.edu) • Google Earth, for information on interesting features for which GIS fields are not readily available • All meteorological data was processed through the Weather Research Forecasting (WRF) modeling system (https://www.mmm.ucar.edu/weather-research-and-forecasting-model). All the data necessary for emissions and the pollution modeling system is available at a spatial resolution of 0.01° and one-hour temporal resolution 22 • The Washington University in St. Louis program for long-term PM2.5 concentration data based on a global chemical transport model coupled with satellite retrievals (https://sites.wustl.edu/acag/datasets/surface-pm2-5) • For baseline emissions inventory development, an emission factors database was extracted from the GAINS modeling system (https://gains.iiasa.ac.at/models). For reference, a summary of material extracted from these reports is presented in Table 6. Source information is detailed in the following sections. All the extracted data files are available upon request, including the meteorological fields. Table 6: Summary of Ethiopia and Addis Ababa geographical, economic, and environmental characteristics Category Remarks Ethiopia National GDP per capita $974 (nominal, 2020 est.) Ethiopia national population (2018) 10.9 million Total registered vehicle fleet (2020) 1.2 million Estimated premature mortality from PM2.5 1,060-10,700 in 1990 and 4,180-16,200 in 2019 Predominant fuel source All biomass Predominant electricity source Hydro Addis Ababa District population (2020) 2.8 million District land area 527 km2 Airshed size 60 x 50 grids (~1 km2 each at 0.01° resolution) Airshed area 3,646 km2 No. of sub-cities 10 Urban share of the airshed land mass 18% Airshed population (2018) 4.9 million Urban share of airshed population (2018) 87% No. of grids with pop > 30,000 56 % increase in built-up area (1990-2014) 190% Total registered vehicle fleet (2020) 0.7 million Total mapped highways length 1,350 km Total mapped roads length 17,600 km Total mapped landfill area 0.5 km2 Total mapped quarries area 9.9 km2 Total mapped commercial activity points 10,700 Predominant wind direction Westerly for Jun-Jul-Aug and easterly for the rest Predominant rainy months Jun-Jul-Aug No. of official AQ monitoring stations 2 Max. PM2.5 monthly average (in 2018-19) 50.2 ± 12.1 mg/m3 in June Min. PM2.5 monthly average (in 2018-19) 24.1 ± 11.8 mg/m3 in February Annual average (in 2018-19) 24.9 ± 21.2 mg/m3 Long-term PM2.5 trend (1998 and 2018) 15.4 and 21.8 mg/m3 No. of airport landings & take-offs per day 450 (2018 average ) 4.2 The Airshed For the analysis of AA’s AQ, an airshed spanning 60 x 50 grids between 38.5°E to 39.1°E in longitudes and 8.7°N and 9.2°N in latitudes was selected (Figure 9). This covers the main city area and the neighboring satellite cities, industrial estates, landfills, and quarries. The spatial grid resolution is 0.01°; each grid is equivalent to 1 km2. All the collated information and analyzed results from the study are maintained in standard GIS-ready formats at this grid resolution. 23 Figure 9. Addis Ababa City Airshed with Road Density This report used 2018 as the base year for all the meteorology, emissions, and pollution analyses and then projected forward using business-as-usual conditions. While some basic information is available for 2020, the COVID-19 lockdown caused limited activity for multiple months for most sectors, so it was not used as a representative baseline year. For example, while total vehicle registration numbers were available for the beginning of 2020, they were affected by lockdown conditions. With little or no information available on the change in usage, a 2020 baseline cannot be normalized to support long- term planning. In AA’s airshed, the prevalent wind direction is westerly for the rainy months of June through August and easterly for the rest of the year. Wind speed variation is consistent over all months, with a brief reduction in averages for June through August. The nighttime surface temperatures are under 10°C during winter months and consistently under 15°C in all others, indicating some need for space heating. In AA, space heating is an integral part of cooking and baking, often by burning wood, crop residue, coal, and cow dung. Air mixing heights are lowest between June and August, which restricts the movement of emissions. In general, the nighttime mixing layer height is half that of daytime. Figure 7 in Section 2.4 presents monthly average PM2.5 pollution and its variation within a month, which will be used in this chapter to verify the emissions results. 4.3 Air Pollution Emissions This study established a first-of-its-kind emissions inventory at a spatial resolution of 1 km 2 by using the data collected from Ethiopia’s environmental and statistical offices and open resources (Figure 10). These data sources included (i) satellite retrieval feeds on urban and rural built-up areas, land use and land cover classification, road networks, commercial activities, and population; (ii) other resources such as high-resolution gridded population and energy demand measures, and consumption statistics; and (iii) published literature by academic and non-governmental organizations on air pollution in AA. 24 As described above, all the calculations were conducted with 2018 as the base year and projected forward using business-as-usual conditions. Figure 10. Estimated Gridded Annual PM2.5 Emissions for Addis Ababa's Airshed in 2018 The emissions inventory includes PM in two bins (PM10 and PM2.5), SO2, NOx, CO, non-methane VOCs, and CO2. Major sources of emissions are road transport; residential activities like cooking and lighting; resuspended dust from roads, construction, and erosion; open waste burning; and industrial activities. Particularly, transportation contributes 29% of PM2.5, 16% of PM10, 97% of NOx, 71% of SO2 and 96% of CO2. A summary of estimated total emissions is presented in Table 7, and percent shares are presented in Figure 11 below. Table 7. Estimated Total Emissions for AA's Airshed for Base Year 2018 tons/year PM2.5 PM10 NOx CO VOC SO2 CO2 All transport 7,850 8,050 120,100 94,200 10,750 4,550 8,683,250 Residential 7,300 7,450 100 101,500 15,100 900 157,800 All industries 6,650 7,700 2,950 47,000 13,700 850 92,250 All dust 3,950 26,100 - - - - - Open waste 850 900 - 4,050 800 - 5,400 burning Diesel generator 200 200 1,250 3,950 1,750 50 120,000 sets Total 26,800 50,400 124,400 250,700 42,100 6,350 9,058,700 25 Figure 11. Estimated Sector Shares to Addis Ababa Airshed's Total Emissions in 2018 for (a) PM2.5, (b) PM10, (c) CO, (d) SO2, (e) NOx, and (f) CO2 For PM2.5 and PM10, the major sources of emissions are biomass burning in domestic and industrial sectors, vehicle exhaust, and resuspended dust from roads and construction activities. Freight vehicles and old passenger buses contribute to 80% of the total PM2.5 emissions by clubbed vehicle types. On a fuel basis, more than 95% of the emissions are associated with diesel-based vehicles. A comprehensive 26 summary of emission factors by fuel and by vehicle type is included in the annex material. The emissions inventory also includes contributions from open waste burning and diesel generator sets. These emission shares, while small at the airshed level, are mostly concentrated in the core urban parts of the city. Due to the coarse nature of crustal elements in dust, dust contributes primarily to PM10 emissions. Since the country’s overall energy needs are dominated by biomass burning in the domestic and industrial sectors, this is also reflected in the total CO emissions, along with the transport sector, which remains a key contributor to all the pollutants. VOC emission rates follow the same trend as CO emissions. SO2, NOx and CO2 emissions are mostly from the transport sector. Using satellite observations, open biomass fires were detected over Western Ethiopia, Northern Uganda, and Eastern South Sudan. While these fires are an important emission source in Ethiopia, fire instances were limited in AA’s airshed. All the calculations and databases are maintained in chemical transport model-ready formats at the grid level. The spatial distribution of the emissions combines total emission calculations, and multiple layers of GIS feeds for various sectors. There is no published official or unofficial gridded emissions inventory to represent AA’s AQ at this spatial resolution. Therefore, no direct or indirect comparisons can be made with any studies discussing emissions and pollution levels in the city. 4.4 Source Apportionment For this study, the open-source Comprehensive AQ Model with extensions (CAMx) (http://www.camx.com), coupled with the meteorological data processed through the Weather Research Forecasting (WRF) model, was utilized to transform emissions into concentrations for further analysis (Figure 12). The maximum annual average concentration appears in the most populated area of the city, with most emissions originating from transport and residential activities. In the month-by-month maps, the winter months (Dec-Jan) and rainy months (Jun-Jul-Aug) stand out, which also represent the months with the highest fraction of hours with temperatures under 15°C, slow moving winds, and lower mixing layer heights limiting the overall dispersion of emissions. Figure 12. Modeled Annual PM2.5 Concentrations for Addis Ababa's Airshed in 2018 27 The pollution modeling exercise replicated the spatial and temporal variations in the available ambient monitoring data, thus providing baseline confidence in the estimated emissions inventory for AA’s airshed (Figure 13). The blue band represents the 10th to 90th percentile range over all urban grids within each month (approximately 600 grid points). The solid black line represents modeled monthly average concentration, and the dashed orange line represents the measured monthly average concentration and its variation within the month. Going forward, this emissions inventory can be used for the analysis of possible emission control mechanisms and likely benefits of a reduction in ambient concentrations, health impacts, and environmental degradation. Figure 13. Modeled Annual Average PM2.5 Concentrations in Addis Ababa's Airshed in 2018 In addition to annual and monthly average concentrations, the modeling system also allows for source apportionment. The term “source apportionment� refers to the estimation of source contributions to ambient pollution levels and is important to differentiate from an emissions inventory, which also provides information at the source level. Pollution source apportionment data is critical for policy discussions. The modeled source contributions for AA (Figure 14) highlight various key findings: • On an annual basis, source contributions to PM2.5 pollution are dominated by vehicle exhaust, residential activities, resuspended dust, and industries. Without exception, vehicle exhaust remains the main source, and a constant one, of pollution in all months. • On an annual basis, dust is a major contributor to PM10. Resuspended dust on roads due to the constant movement of vehicles is substantial in both PM2.5 and PM10 fractions. It will increase with an increasing number of vehicles (irrespective of the engine/fuel technology) if dust control programs like paving, watering, and greening the roads are not introduced. • The use of biofuels dominates all non-transport sources. Biomass in the form of wood, crop residue, cow dung, and charcoal is the dominant fuel source for all residential activities like cooking and baking, and some space heating in the outlying parts of the city. • Open waste burning and diesel generator sets are not significant contributors but are not negligible. • The term “boundary� represents the contribution of sources outside the modeling airshed and is an indicator of regionally transported pollution. Overall boundary contribution is under 10% for all months, indicating a limited influence of sources outside the city airshed. This contribution is higher during the months of easterly winds. 28 • With limited variation in the monthly weather patterns, the variation in the sectoral contributions is also limited. Figure 14. Modeled Source Contributions to Annual Average PM2.5 Pollution in Addis Ababa Similar concentration trends were observed in a chemical analysis-based source apportionment study with samples collected at one station (Tefera et al., 2020). Since that paper discusses results from only one location, it is not possible to generalize the results for the city airshed, but the presence of higher organic and black carbon components supports the conclusion of higher contributions from biomass and diesel combustion, followed by dust from the presence of crustal elements in the collected samples.20 The fraction of secondary components such as sulfates, nitrates, and ammonium are small, indicating limited contributions from regional sources. 4.5 Recommendations As emission inventories and source apportionments are fundamental to planning and prioritizing emission reduction activities, AA and other cities need to develop and/or update them. This exercise would build confidence in the modeling platform. For example, data in the road transport emissions analysis relies on various reports and the basic registrations database which can be further improved. This exercise could include conducting surveys at fuel stations to understand in-use vehicle characteristics like age mix, vehicle distance traveled, and fuel efficiency. Similarly, a survey to understand fuel mixes and usage in the domestic sector will help in bettering the emission estimates. A source apportionment exercise is a technically and financially taxing project. While this report used an established bottom-up approach to estimate source contributions, a complimentary program that includes a top-down chemical analysis-based approach at a representative number of locations across the AA airshed would strengthen and validate the results and increase confidence in adopting policy applications. Besides using this analysis for long-term policy “what-if� scenarios, the modeling system can be extended to conduct short-term air quality forecasting. The platform simulates weather and 20 The findings of this chemical analysis-based study and the bottom-up analysis presented in this study are comparable to studies conducted in cities with a similar mix of urban-rural classification and similar mix of emission sources (like transport, industries, residential, etc.). For example, a study conducted in six cities by India’s Central Pollution Control Board (https://cpcb.nic.in/source-apportionment-studies, 2011), concluded that the transport sector is responsible for 20-40% of the pollution in the cities, with the remaining contributions spread between industrial, residential, open waste burning, and dust. 29 pollution patterns every day for the following 3 to 5 consecutive days and uses the results to alert the public on air quality. 30 5. Zooming in Urban Transport: Vehicle Emission Control in Addis Ababa Prior to this study, there was limited transport air pollution data and few systematic studies on vehicle emission inventory in AA, even though urban transport is widely recognized as a key contributor to ambient air pollution. The emission inventory analysis of this study estimated the transport sector contributed 29% of PM2.5 emissions and more than 90% of NOx and CO2. The health impact analysis provided data-driven evidence that air pollution in AA has become a public health crisis, with a societal cost equivalent to $78 million or 1.3% of AA’s GDP in 2019. This chapter focuses on transport air pollution control in AA. Building upon the overview of the urban transport sector context, documents from the government and development partners, and ongoing interventions, the chapter identifies mitigation options for transport emissions in consideration of global practices in the local context, assesses their effectiveness and applicability and recommends priority measures for short- and mid-term action, and presents key findings and recommendations on transport air quality mitigation.21 5.1 Urban Transport Development and Air Pollution AA is at the forefront of urbanization and motorization. The city expanded spatially three-fold in the past two decades, from 99 km2 to 284.9 km2. This increase decreased urban density in AA from 17,000 (people/km2 built-up area) in 1987 to 13,000 in 2017 (see Figure 15). As described above, the urbanization and motorization trend in AA has also come at a price of environmental degradation, with significant loss of green space and farmland and increasing vulnerability to climate change. Figure 15. Spatial Expansion of Addis Ababa Source: WB multi-sector ASA This rapid expansion has drastically increased the demand for motorization and increased needs to move both people and freight. The number of registered vehicles in the city has increased by 40% in the past 21 The details of the study on the transport mitigation can be found in a separate report titled “Steering Towards Cleaner Air: Measures to Mitigate Transport Air Pollution in Addis Ababa� (Grutter, J., W. Jia and J. Xie, 2021). 31 five years. The city’s registered vehicle stock reached 630,000 in 2020, accounting for nearly 50% of all registered vehicles in Ethiopia. Motorization is posing huge challenges for AA in all dimensions. The previous approach to addressing motorization through expanding main roads and accommodating vehicle growth has not effectively improved accessibility but has nonetheless generated negative externalities. In AA, 54% of the population walks and 31% rides on public transit, yet the city has low accessibility; only 17% jobs can be reached via public transport within one hour, and only 15% on foot. Walking can be dangerous, too, as studies by the World Health Organization (WHO) and the government show the city has disproportionately high pedestrian fatalities for the country and for Africa. In addition to the exacerbation of road fatalities, rapid motorization also worsens traffic congestion, increases greenhouse gas emissions, deteriorates air quality locally, and results in health impacts and economic costs. This calls for integrated, comprehensive management of motorization and air quality. Air pollution problems from the transport sector in AA are characterized by high-sulfur fuels, lack of emission standards, ineffective and unenforceable emission inspections, and an aging vehicle fleet. Ethiopia currently imports 500ppm sulfur diesel and 10ppm sulfur gasoline, allowing for operation of Euro 3/III vehicles. Emission inspection, while mandatory, only detects smoke and carbon monoxide and does not set any threshold for acceptance. Therefore, it does not influence the pass/fail decision for an inspection and is not enforceable. Most registered vehicles are over 10 years old, including buses, passenger cars and light duty vehicles, many of which are not subject to emission control under the current practice. Mini- and midi-buses are predominantly old and lack vehicle emission controls (Figure 16). Figure 16. Type and Age of Registered Vehicles in Addis Ababa in 2020 Note: MUV – Multipurpose vehicle; Bus1 – mini/midi bus; Bus2 – standard bus, HDV/LDV – heavy/light duty vehicle Continuing business-as-usual (BAU) management of motorization and air quality will further degrade air quality in AA, threatening the city's livability and ability to grow. As discussed in previous chapters, in AA, the transport sector – including urban transport, aviation, and roads – currently contributes 29% of PM2.5 emissions, 71% SO2, 97% NOx, and 96% of CO2. In the next decade, should the current vehicle growth trend continue at 8% a year, AA would expect 1.3 million vehicles. Even a conservative estimate, which assumes vehicle growth at the GDP growth rate, projects 900,000 vehicles in AA by 2030. In these scenarios, the level of vehicle emission would exponentially increase in proportion to vehicle growth, increasing at least 50% and possibly over 100%. Without a change of course, transport air pollution in AA will create profound local and national impacts. Furthermore, there exist co-benefits between local pollution control and climate change at a global level, as increased particle emissions damage air quality and increase Black Carbon (BC) emissions. Scientific assessments of BC emissions and impacts have found that they are the second most important emission that drives climate change, second to CO2. 32 Managing motorization and controlling vehicle emissions will help AA to reduce air pollution, improve public health and boost quality of life while providing climate co-benefits through the reduction of GHG emissions. They are a pathway to the city’s livability, competitiveness, and growth. 5.2 Identification and Assessment of Potential Mitigation Options The Government of Ethiopia and the City Government of AA have conducted studies on air pollution problems to identify measures for reducing vehicle emissions. The documents, reviewed during this study, include but are not limited to: the Air Quality Management Plan (AQMP) prepared by the city government in 2021 (through its AAEPGDC with the support from the US Government); the 2016 Greenhouse Gas Emissions Inventory Report by C40 Cities in 2020; Motorization Management in Ethiopia, prepared by the World Bank in 2017; an Air Quality Policy and Regulatory Situational Analysis (UNEP, 2018); an Electric Bus Study by World Resource Institute and Lucy Partners; and multiple urban transport studies by AA City. Ongoing urban transport investment and policy work in AA will come to fruition in the short-term and contribute to the mitigation of emissions and air pollution. The city of AA is working with the World Bank to modernize its public transport and traffic management systems, and with the French Development Agency to develop its first Bus Rapid Transit Line, all of which will be operationalized in 2-3 years. The city also built its first dedicated bike lane in 2020. Meanwhile several key studies covering comprehensive transport development, paratransit transformation, and other strategic areas will inform AA’s policies and priorities for the management of motorization and air pollution. Based on the review of government studies and sector interventions, assessment of local context, consultation with relevant government agencies, and learning from international experience, nine potential mitigation measures were identified for AA’s transport sector (Table 8). The identification also aligns with the urban transport policy framework “Enable, Avoid, Shift and Improve� (EASI),22 recommended by the Sub-Saharan Africa Transport Policy Program (SSATP). The potential measures are further categorized into three clusters or groups: (i) standards, (ii) vehicle measures, (iii) public transport. Table 8. List of Potential Transport Air Pollution Mitigation Options Category Option Proposed Options with Abbreviated Name Group No. 1 Low-sulfur fuels Vehicle emission standards for newly registered Fuel and vehicle 2 vehicles emission standards 3 Maximum emission levels for in-use vehicles (I/M) 4 Fuel economy or CO2 emission standards 5 Vehicle retrofits with emission control equipment Maximum vehicle age for imports and in-use 6 vehicles Vehicle measures Ban on import and new registration of light diesel 7 vehicles Promote low-carbon vehicles (electric and hybrid 8 vehicles) 22 https://www.ssatp.org/topics/urban-mobility 33 Public transport Improve public transport/non-motorized transport 9 measures (NMT)/transport demand management (TDM) Under each option group, there are varied, inter-connected measures. Among them, most were identified in the studies done by the government and development partners. Only two options – vehicle retrofits with emission control equipment and restricting sales of diesel vehicles – are proposed as new measures in this study, as they have been implemented in other countries with success and have potential applicability for Ethiopia. Various reports also included biofuels, though they alone will not significantly impact air quality because they do not affect vehicle emission standards. The impact on GHG emissions depends on the type of biofuel used and requires a life-cycle assessment, including direct and indirect impact on land-use change, which this study does not provide. Potential mitigation options were assessed based on the following criteria: (i) efficacy and environmental impact on local air pollutants and GHG emissions in AA; (ii) efficiency expressed in terms of economic cost-benefit and social impact; (iii) technical feasibility for AA in consideration of implementation capacity; and (iv) effort and time required to implement the measure in a sustainable manner in the local context. For the study’s economic analysis, the economic benefits of emission reduction from each mitigation measure are estimated based on the cost of air pollutants in Ethiopia presented in an International Monetary Fund (IMF) publication (IMF, 2014). The IMF study calculated the cost of a few main pollutants, namely, PM2.5, SO2, and NOx, based on local pollution at the ground level and cost and health impact of each pollution type in Ethiopia. By taking the cost values for Ethiopia from the IMF publication, the study further updated unit costs to their US$ equivalent as of 2019 according to the GDP per capita of Ethiopia. The following costs per ton of air pollutant are used in the study: US$4,014 per ton of PM2.5 emitted, US$132 per ton of SO2 emitted, and US$28 per ton of NOx emitted.23 The economic analysis also included global warming externality costs through the social costs of carbon (SCC), which estimate the economic damages associated with increases in CO2 emissions. Valuating the economic damage of CO2 emissions is complex and depends on discount rates. The Asian Development Bank (ADB) reported an SCC unit value of US$36 per ton of CO2e in 2016 prices for emissions, which increases by 2% annually in real terms to reflect the potential of increasing marginal damage of global warming over time. Updated to 2019 real US$ and including the annual increase results in around US$40 per ton CO2e for 2019. Table 9 highlights key aspects of the assessment outcomes. 23 Please note that the cost of $4,014 per ton of PM2.5 used in the assessment of transport emission mitigation measures is different from what one may derive from calculations of economic cost of the health impact of PM2.5 presented in the previous chapters of this report. The economic cost presented in Chapter 3 aims to demonstrate the scale of economic cost of mortality and morbidity due to ambient PM2.5 and represents a fraction of total costs. Therefore, the assessment in this chapter adopts the more complete cost data from an IMF study (IMF, 2014). 34 Table 9. Summary of the Assessment of Mitigation Measures Implementation Category Option Groups Air pollution impact GHG impact Cost-Benefit Overall assessment and recommendation complexity 1. Low-sulfur • Stand-alone: • Technically • Current level: 500ppm • Direct: strong SO2 fuels: significantly simple as fuels reduction; small PM2.5 • A pre-condition for most other measures Introduction of higher costs than are imported reduction; no NOx • Euro 4 fuel (50ppm) recommended; Euro 5 fuel 50ppm sulfur benefits • Fuel sur-cost is 1- impact No impact (10ppm) not recommended yet due to costs (stage 1) and • Combined with 2 US$ cents per • Indirect combined with • Low-sulfur fuels without subsequent issuance of 10ppm sulfur vehicle emission liter which is not (stage 2) diesel. vehicle emission tighter vehicle emission standards will only have a regulations: see considered a regulations: see below limited impact below huge barrier • Current level: none but primarily Euro 2 vehicles • Technically • Euro 4 emission standard for new or used vehicles • Euro 4 together simple as is recommended once 50ppm sulfur fuel is 2. Emission with 50ppm vehicles are available standards for sulfur fuels cost- imported and • Euro 6 level not recommended yet due to higher • Direct: large reduction newly registered neutral certificates of costs than benefits of all pollutants vehicles (used or • Euro 6 together conformity can • Euro 3/5 standards not recommended due to Fuel and • Measure requires low- No impact new units): Euro with 10ppm be used marginal benefits compared to Euro 2/4 sulfur fuels as pre- vehicle 4 (stage 1) and sulfur fuels has • No investment in • This measure can only be implemented together emission condition Euro 6 (stage 2) significantly national vehicle with low-sulfur fuels standards higher costs than testing center is • Vehicle emission standards cannot be replaced benefits required with in-use vehicle emission controls as the latter are only to identify faulty vehicles with excessive emissions 3. Maximum • Support the government in the establishment of emission levels Low impact due to In theory more maximum levels for in-use vehicle emission for in-use limited vehicle benefits than costs controls including measurement procedures and vehicles (I/M); degradation in emission but in practice equipment standards emissions testing (i.e., increasing many systems • Strengthen the process of integration of emission is already Highly complex to emissions over time due No impact only incur costs testing in road-worthiness tests with appropriate performed enforce and control to aging of vehicles) and due to lack of quality control and enforcement measures together with the roadworthiness complexity of effective adequate • In-use vehicle inspection cannot be used to implementation & implementation determine the vehicle emission standard or the test and the enforcement and enforcement compliance of a vehicle with a given vehicle government is in the process of emission standard, but only to identify gross determining polluters significantly in excess of prescribed maximum pass limits. levels for in-use vehicle emission controls • Fuel efficiency labels are • Small impact if no fuel complex to switch design 4. Fuel • Highly negative if this No general • Implementation is efficiency or results in dieselization Moderate Fiscal measures to promote low-emission vehicles statement complex for CO2 emission • Positive impact if impact possible should be designed to be fiscally neutral. second-hand standards switch towards electric vehicles vehicles (EVs) • Simple if focused on hybrids and EVs Small increase in • Requires Euro 4 fuel fuels 5. Vehicle consumption Significantly • Technically retrofits with Significant reduction of Not recommended due to highly negative cost-benefit but reduction higher costs than complex emission control PM2.5 relation and high implementation complexity equipment of Black benefits • On-road Carbon: total enforcement and small controls required reduction Vehicle age is not an • Pure age • GHG Vehicle adequate proxy for limitations are emissions of measures vehicle emissions and as simple to regulate 6. Maximum HDVs are a stand-alone measure but for in-use vehicle age for not related • Age is not an adequate proxy and is not correlated could increase Vehicle scrapping vehicles seldomly imports and in- to age well with emissions emissions, as vehicle programs have enforced (e.g., use vehicles • Small age • Emission standards are the appropriate parameter to deterioration rates are significantly multiple countries eventually deterioration limit vehicle emissions lower than changes in higher costs than have maximum combined with emission standards (a • Car GHG benefits ages for buses • Scrapping programs are not recommended due to scrapping emissions very high costs with limited benefits 10- or even 20-year-old which are never programs more related Euro 4 vehicle has lower enforced) to CO2 emissions than a brand- • Scrapping new Euro 2 unit) standards programs are 36 than to complex to vehicle age design Neutral for 7. Ban on import country but and new slightly higher • Highly recommended due to simplicity and registration of Marginal High reduction of PM2.5 cost for private immediate impact diesel passenger increase to Simple measure and NOx diesel car users • Petrol cars are a cost-effective alternative to diesel cars and light neutral due to diesel units commercial being subsidized vehicles and petrol not Higher upfront • Recommended for urban buses but requires more High reduction per High impact Simple measure 8. Promote low- costs and in in-depth evaluation first vehicle but low impact due to low but requires new carbon vehicles general still • Develop an e-mobility strategy across multiple due to low levels of carbon grid business models to (hybrids and higher total sectors, which could include designing and/or vehicle renovation and factor of be commercially EVs) lifetime financial implementing a demonstration project low vehicle numbers Ethiopia viable costs • Often requires Dependent on high initial • Recommended measures to ensure a sustainable concrete measure investment All measures low-emission transportation system but in general • TOD and TDM potentially • Priority on implementing ongoing interventions and highly positive in measures are All measures result in result in high key policy recommendations: (i) Sidewalks and Public 9. Improve economic terms highly complex to mode shift and have a GHG bike lanes, (ii) Anbessa and BRT projects under DP transport public transport, due to congestion design and high potential for emission financing, measures NMT and TDM reduction, time implement reducing local pollutants reductions (iii) Public transport network optimization due to modal savings, reduced • Benefits often including paratransit minibus reform, vehicle operating only accrue in the shift (iv) Parking management pilot project, and costs and reduced long-term due to (v) Completing BRT and TOD studies. health costs requiring behavioral change 37 5.3 Conclusions and Recommendations Based on the results of the assessment shown in Table 9, the potential mitigation options are screened and prioritized in Table 10. Table 10. Prioritization of Mitigation Measures Priority Measure Rationale Low-sulfur diesel (50ppm; at a This measure is a pre-condition for many other later stage 10ppm) (option no. 1) measures including stricter vehicle emission standards or import of low-emission vehicles. This measure should be linked with vehicle emission standards as otherwise its impact will be limited. Vehicle emission standards: once This measure directly results in lower emissions 50ppm sulfur fuels are available and is also a good criterion for selecting which introduce Euro 4/IV; long-term vehicles to import. This measure is dependent on Euro 6/VI (option no. 2) low-sulfur fuels. Improve public transport and This is important for achieving sustainable low- NMT, if possible combined with emission transportation systems and the economic High-priority TDM measures (option no. 9) benefits in general outweigh the costs by far. measures Establish maximum levels for in- Maximum in-use vehicle emission levels are under use vehicle emissions, development by Addis Ababa; adequate measurement procedures, measurement procedures, standards and a high- equipment standards, and quality supervision and enforcement system strengthen its integration into the including on-road spot checks can reduce the roadworthiness test (part of potential to circumvent the system. option no. 3) Ban on import and new The measure has a significant impact on reducing registration of diesel passenger pollutants with a positive cost-benefit ratio and is cars, and light commercial very simple to implement vehicles less than 3.5t (option no. 7) Foster electric and hybrid In the long-run carbon neutrality in transportation vehicles (option no. 8) will only be achieved with EVs. The measures could start with electric buses and then be expanded to other commercial vehicles used in urban settings Age limitation for in-use urban This option allows for periodic renovation of the fossil public transport buses (part bus fleet and thereby also improves the Medium- of option no. 6) attractiveness of public transport and gives an priority incentive to usage of electric units. However, the measures cost-benefit of the measure is negative and it will only have significant positive impact if the vehicles meet higher emission standards This measure will depend on the establishment of Integrate emission testing minimum emission standards, and effectiveness including data access and sharing and soundness of emission inspections. in roadworthiness test control Introducing this measure faces weak capacity, high centers (part of option no. 3) complexity, and lower cost-benefit. Low priority, Fuel efficiency standards (option High complexity with a low impact for Ethiopia not no. 4) plus the potential to result in a further dieselization recommended with resultant worsening of air quality 38 Vehicle scrapping programs (part High cost, negative cost-benefit, high complexity of option no. 6) and limited impact Vehicle retrofit programs with High technical complexity and highly negative diesel particle filters (option no. cost-benefit 5) Age limitations for import of new This is not an adequate proxy for vehicle or in-use vehicles (part of option performance no. 6) The recommended priorities from this study are further screened by their magnitude and rapidity of impacts and categorized into short-term, medium-term and long-term measures. Sometimes, high priority does not necessarily translate into quick implementation. Short-term refers to implementation with impact within 1- 2 years, medium-term for impact within 3-5 years, and long-term beyond 5 years. Proposed short- and mid- term measures are: Short-term measures include the following: (i) Introducing 50 ppm-sulfur diesel fuel, combined with Euro 4/IV vehicle emission standards or equivalent, which is a high priority for the government. (ii) Fostering public transport and NMT measures, if possible, combined with transport demand measures and transit-oriented development. This is also consistent with government priorities, as AA city administration recently announced an initiative to add 3,000 buses. Public transport, at today’s 31% mode share, carries high passenger volume and results in significantly lower GHG emissions per passenger-kilometer than private means of transport. However, due to usage of diesel buses public transport is also a major source of local pollutants. Moreover, without increasing the efficiency of public transport, fostering of public transport alone will not result in air quality improvements. Measures such as operational improvements and restructuring to allow for replacement of minibuses with larger units on heavy demand routes are the most relevant in terms of air quality improvement. (iii) Establishing maximum in-use vehicle emission levels and measurement procedures and strengthening the integration of emission testing in road-worthiness tests with quality control and enforcement measures, which are being developed by the Federal Transport Authority. (iv) Introducing a ban on importing all diesel vehicles with less than 3.5t gross vehicle weight, including both new and second-hand vehicles. At present, 36% of registered vehicles in Ethiopia are diesel vehicles. Even with the best available diesel technologies, the real-world performance of diesel engines results in high PM and NO2 emissions. Among these measures, (i), (iii) and (iv) are identified as high priority by the government. Medium–term measures: (i) Promote hybrids and EVs with fiscally neutral instruments and other policies (short term can start with developing multi-sector e-mobility strategy and understanding power infrastructure investment to support e-mobility). (ii) Integrate emission inspection including data access and sharing for vehicle roadworthiness test centers. (iii) Limit the age of in-use fossil buses for urban public transport to speed up public transport renovations and incentivize switching to electric units. 39 Other measures are considered low priority and not recommended at this stage. Some non-recommended options are parts or sub-components of other options: (i) fuel efficiency standards, due to high complexity and potential dieselization of the vehicle fleet and a subsequent increase in air pollution; (ii) vehicle scrapping programs, due to their highly negative cost-benefit; (iii) vehicle retrofit programs with diesel particle filters, due to their considerable technical complexity and highly negative cost-benefit ratio; and (iv) age limitations for the import of new or second-hand vehicles, due to them not being an adequate proxy for vehicle performance. The correlation between vehicle age and vehicle emissions only exists in countries with advanced vehicle emission regulations. In Ethiopia, used as well as new vehicles arrive from different parts of the world. Second-hand vehicles made in Europe since 2006 comply with Euro 4/IV and with Euro 6/VI for vehicles since 2015, while new heavy-duty vehicles assembled locally in Ethiopia may only meet emission standards of Euro 0. Therefore, introducing an age limitation for importing vehicles may not be effective in Ethiopia at this time. 40 6. Proposed Action Plan and Concluding Remarks Ethiopia needs a clear set of action steps to move forward towards IAQM. This final chapter summarizes the policy recommendations already discussed in previous chapters in an action plan. It also points out the limitations of the study and areas for improvement. To help the governments set priorities and timeframe to take actions, Table 10 below groups policy recommendations in eight categories and suggests the responsible agencies and implementation period for each recommended action. Table 11. Action Plan for Promoting Integrated AQM in Ethiopia Short- Long- Medium- Responsible term term Table 1 term (3-5 agency (0-2 (6-10 years) years) years) Regulation and Policy Reforms • Review, strengthen, and harmonize existing regulatory and policy frameworks for AQM EFCCC X EFCCC and • Clearly define or clarify the roles and mandates Municipal of key government agencies at the federal and EPB in municipal level for AQM through functional consultation X review of relevant agencies with other line agencies • Introduce taxation and pricing policies or other fiscally-neutral instruments or incentives to encourage newer and cleaner vehicles such as MoF, X hybrids and electric vehicles and phase out EFCCC polluting vehicles • Upgrade AQ standards (including ambient PM2.5 ESA, standards) X EFCCC • Introduce low sulfur fuel standards (start with 50 ESA with ppm; progress to 10 ppm) with the Euro 4/IV the support X vehicle emission standards nationwide of ESP • Introduce other fuel quality and emission ESA, standards for the transport sector X X EFCCC AQM Strategy and Plan Municipal • Develop urban ambient air quality monitoring EPA with strategy and implementation plan X the support of EFCCC 41 Municipal EPA with • Develop urban ambient AQM strategy and plan the support X X of line agency and EFCCC Budgeting and Capacity Building • Strengthen institutional capacity of responsible agencies for AQM, particularly for regulating Relevant X X and enforcing AQ regulations and standards. agencies EFCCC and • Develop mechanisms for promoting inter-agency relevant and cross-sectoral coordination and collaboration agencies at as well as experience sharing and learnings on X X federal and AQM. municipal levels • Increase government revenue and budgets for MoF and AQM by reviewing and revising environmental Municipal X X taxation and fee systems. governments AQ Monitoring • Develop standardized and unified systems for ambient AQ monitoring and reporting in AA and EFCCC and X X other polluted cities in Ethiopia local EPA • Develop a spatially and temporally representative emissions inventory and reporting AAEPGDC X system in AA • Strengthen the capacity of key line agencies in Relevant AQ information management X X agencies • Make verified and standardized AQ data open to EFCCC and public access via the internet and mobile apps X X local EPA Data collection, Research & Modeling • Systematically collect and analyze AQ and air EFCCC and emission data for better informing AQM local EPA planning and decision-making in AA and in X X and relevant Ethiopia agencies • Collect data to update emissions inventory EFCCC and X X local EPA • Conduct surveys to understand in-use vehicle characteristics and domestic fuel mixes and MoT and X X usage local TB • Adopt the top-down chemical analysis-based Local EPA X X approach in AA to further update and verify the 42 results of the bottom-up emissions inventory and source apportionment study • Strengthen the collection and analysis of air MoH, pollution related health data and establish the EFCCC and empirical relationship between airborne diseases X X local HB and air pollution in Ethiopia and EPA • Improve and extend the economic valuation of air pollution impacts beyond health impacts develop EFCCC and the guidelines of the cost-benefit analysis or cost- relevant X effectiveness analysis of AQM programs, agencies investment projects, and interventions EFCCC • Develop an abatement cost curve and a plan to with the implement AQM interventions to cost- participation X effectively reduce air pollution emissions of other relevant agencies Awareness Raising and Behavioral Change • Develop and implement AQM education and outreaching activities, including communications EFCCC and and dissemination of AQM policies, plans and X X X local EPA programs • Engage and mobilize private sector, general EFCCC and public and other stakeholders to change relevant X X X behaviors that emit air pollution agencies Vehicle Emissions Control • Introduce low sulfur (max 50ppm) diesel fuel and Euro 4/IV vehicle emission standards or their ESA X equivalent • Support efforts to import cleaner fuels with lower sulfur contents ESP X • Ban importing secondhand diesel passenger cars or light commercial vehicles and restrict imports to new and secondhand imported diesel vehicles MoT and weighing more than 3.5t gross (i.e., continue to local EPA X X allow imported diesel vehicles, such as trucks or and TB buses) • Establish maximum emission levels for in-use vehicles, measurement procedures and standards, EFCCC, quality control & enforcement system of MoT, local X X emission inspection performance TB and EPA 43 • Limit the age of in-use fossil buses for urban public transport to speed up public transport renovations and incentivize switching to hybrids Local TB X X and electric vehicles with fiscally-neutral and EPA instruments and other policies • Promote and strengthen public transport and NMT measures, with possible transport demand Local TB X X X measures and/or transit-oriented development Point and Area Sources Control • Enhance the capacity of cities to point-source monitor facilities, enforcement of city-level regulations for industrial sources, including Local EPA X X boilers and oil generator sets • Develop and implement area source control measures, including (a) controlling air pollutant EFCCC and emission from industries, (b) improving SW Local EPA collection and reducing open burning of trash, (c) with support raising public awareness of health impact of X X of other burning plastic and hazardous wastes, and (d) relevant implementing best practice for control of agencies construction and road dusts As already discussed in the report, the study faced a few limitations. The first is the lack of AQ data, a common phenomenon in many developing countries including Ethiopia. The study behind the report had to rely mostly on open-source data, literature, and limited access to government data in its analytics including emission inventory and source apportionment and health impact assessment. The health impact assessment adopted the methodology developed for GBD and estimate the health impact of annual average PM2.5 concentrations available from a few AQ monitoring stations. The economic costs of health impacts were valued by the benefit transfer approach and the Value of Statistical Life (VSL) available from literature. The cost-benefit assessment of transport mitigation options only intends to give an indication or first approximation and its results are insufficient to derive an abatement cost curve that would benefit AQM planning and decision making. Secondly, the study was entirely carried out within a short period from October 2020 to June 2021 during the Covid-19 pandemic and, as a result, it was impossible to conduct field trips and in-person discussions with stakeholders for the firsthand and in-depth assessment of the local situation. Thirdly, due to data availability and time and resource constraints, most of the analytical work was done on AA only and air pollution emission control measures focus on the transport sector. Therefore, to enhance its policy analysis and decision-making capacity, Ethiopia should continue conducting in-depth analyses for various aspects of AQ, such as ambient air pollution dispersion modeling, spatial distribution of populations vulnerable to AQ, wider assessment of health and non-health impact of air pollution, and cost-effectiveness or cost-benefit analysis of AQM intervention. While this report used an established bottom-up approach to estimate source contributions, a complimentary program that includes a top-down chemical analysis-based approach at a representative number of locations across the airshed of 44 AA would help strengthen and verify the results and gain confidence in policy applications. For the urban transport sector, additional analyses could include the spatial-temporal distribution of traffic congestion and travel time to localize pollution hotspots. Lastly, the future study could also include data from the impacts of changes during the COVID-19 pandemic to provide a better benchmark for scenarios when countries and cities reduce emissions under certain circumstances. From the additional analyses, Ethiopia can develop a more evidence-based IAQM plan that includes recommendations for nationwide action and corresponding investment activities. This report highlights that rapid urbanization and mobility and inadequate environmental management increase air pollution in AA and other major urban areas in Ethiopia. As discussed, Ethiopia faces several challenges in its urban AQM, which threatens its public health, economies, and ecosystems. This report endeavored to assess these challenges and recommend initial actions for Ethiopia and the City of AA to improve AQM. Fortunately, the Government of Ethiopia and the general public are increasingly aware of air pollution problems and are willing to change. With further research and strategic planning, the country can choose a path to balance economic growth and environmental protection and enhance AQM. 45 References Addis Ababa City Administation Transport Bureau. 2018. Addis Ababa Non-Motorised Transport Strategy 2019-2028. Addis Ababa, Ethiopia: Addis Ababa City Administation Transport Bureau.Addis Ababa Environmental Protection and Green Development Commission. 2021. Addis Ababa Air Quality Management Plan. Addis Ababa Institute of Technology. 2012. Final report on Pilot Global Fuel Economy Initiative Study in Ethiopia. Addis Ababa, Ethiopia: Addis Ababa Institute of Technology. Amhayesus, A. D. 2019. The Dynamics Relationship Of Carbon Dioxide, Nitrous Dioxide, Sulfur Dioxide Emissions And Economic Growth Of Ethiopia Addis Ababa, Ethiopia. St.Mary’s University. Available at: http://repository.smuc.edu.et/handle/123456789/5147. Apte, J. S., J. D. Marshall, A. J. Cohen, and M., Brauer. 2015. Addressing Global Mortality from Ambient PM2.5. Environmental Science and Technology 49 (13): 8057–66. Berhane, K., A. Kumie, and J. Samet. 2016. Health Effects of Environmental Exposures, Occupational Hazards and Climate Change in Ethiopia: Synthesis of Situational Analysis, Needs Assessment and the Way Forward. Ethiop. J Health Dev. 30 : 50-56. Blumberg K., M. Walsh, and C. Pera. 2003. Low-Sulfure Gasoline and Diesel: The Key to Lower Vehicle Emissions. https://theicct.org/sites/default/files/publications/Low-Sulfur_ICCT_2003.pdf.Bond, T.C., et al. 2013. Bounding the Role of Black Carbon in the Climate System: A Scientific Assessment. Journal of Geophysical Research. doi:10.1002/jgrd.50171 Borken-Kleefeld, J. et. al. 2015. New emission deterioration rates for gasoline cars – results from long- term measurements. Atmospheric Environment. Vol. 101, 01/2015, 58-64 Bowe, B., Y. Xie, Y. Yan, and Z. Al-Aly, 2019. Burden of Cause-Specific Mortality Associated with PM2.5 Air Pollution in the United States. JAMA Network Open, 2, e1915834. Brook, R.D., S. Rajagopalan, C.A. Pope, J.R. Brook, A. Bhatnagar, and A.V. Diez-Roux,. 2010. Particulate Matter Air Pollution and Cardiovascular Disease. Circulation, 121, 2331-2378. Bulto, T. W. 2020. Impact of Open Burning Refuse On Air Quality: In The Case Of “Hidar Sitaten� At Addis Ababa, Ethiopia, Environmental Health Insights, 14, P. 117863022094320. Doi: 10.1177/1178630220943204. Bulto, T. W., A. Chebo, A. Ede and B. Werku. 2020. Implications of COVID-19 on the fine particulate matter (PM2.5) in Ethiopia. Research Square. https://doi.org/10.21203/rs.3.rs-66750/v1. Burnett, Richard, et al. 2018. Global Estimates of Mortality Associated with Long-term Exposure to Outdoor Fine Particulate Matter. https://www.pnas.org/content/115/38/9592. C40 Cities. 2020. Addis Ababa City: 2016 Greenhouse Gas Emissions Inventory Report. New York City, USA: C40 Cities. C40 Cities. 2020. Benefits of Urban Climate Action. C40 Cities Technical Assistance Report. New York City, USA: C40 Cities. 46 California Air Resources Board. 2015. EMFAC2014 Volume III – Technical Documentation. https://ww3.arb.ca.gov/msei/downloads/emfac2014/emfac2014-vol3-technical-documentation- 052015.pdf. Clean Air Asia. 2016. Vehicle Inspection and Maintenance in Asia: Status and Challenges. http://baqconference.org/2012/assets/Uploads/BAQ2012Parthaa-Bosu-and-Bert-Fabian.pdf. Central Statistical Agency. 2013. Population projections for Ethiopia (2007 – 2037). CSA. July 2013. Center for Science and Environment. 2016. “Urban Air Quality Management in Ethiopia: A Guidance Framework�.New Delhi, India: Center for Science and Environment. Clougherty, J. 2010. A Growing Role for Gender Analysis in Air Pollution Epidemiology, Environmental Health Perspectives, 2010 Feb., 118(2), 167-176 Cohen, A. et al. 2017. Estimates and 25-Year Trends of the Global Burden of Disease Attributable to Ambient Air Pollution: An Analysis of Data from the Global Burden of Diseases Study 2015. The Lancet 389 (10082): 1907–18. Cohen, A.J., Anderson, H.R., Ostro, B., Pandey, K.D., Krzyzanowski, M., Künzli, N., et al. 2005. The Global Burden of Disease Due to Outdoor Air Pollution. Journal of Toxicology and Environmental Health, Part A. 68, 1301-1307. Conticini, E., Frediani, B. and Caro, D. 2020. Can atmospheric pollution be considered a co-factor in extremely high level of SARS-CoV-2 lethality in Northern Italy? Environmental Pollution, 261. 114465. doi: 10.1016/j.envpol.2020.114465. Croitoru, L., Chang, J.C. and Kelly, A. 2020. The Cost of Air Pollution in Lagos. The World Bank, Washington DC. https://doi.org/10.1596/33038. Croitoru, L., J.J. Miranda, and M. Sarraf. 2019. The Cost of Coastal Zone Degradation in West Africa. Washington DC: The World Bank. https://doi.org/10.1596/31428. Diao, M., Holloway, T., Choi, S., O’Neill, S.M., Al-Hamdan, M.Z., Van Donkelaar, A., et al. 2019. Methods, Availability, and Applications of PM2.5 Exposure Estimates Derived from Ground Measurements, Satellite, and Atmospheric Models. Journal of the Air & Waste Management Association. 69, 1391-1414. https://doi.org/10.1080/10962247.2019.1668498. Duncan, B.N., Prados, A.I., Lamsal, L.N., Liu, Y., Streets, D.G., Gupta, P., et al. 2014. Satellite Data of Atmospheric Pollution for U.S. Air Quality Applications: Examples of Applications, Summary of Data End-User Resources, Answers to FAQs, and Common Mistakes to Avoid. Atmospheric Environment, 94, 647-662. https://doi.org/10.1016/j.atmosenv.2014.05.061. EC. 2010. On The Monitoring And Reporting of Data on The Registration of New Passenger Cars. EC. 2014. Regulation (EU) No. 443/2009 to define the modalities for reaching the 2020 target to reduce CO2 emissions from new passenger cars. EC. 2016. Revision of Regulation (EU) No 443/2009 and Regulation (EU) No. 510/2011 regulating CO2 emissions from light-duty vehicles. Embiale, A. Et Al. 2019. Commuter Exposure to Particulate Matters and Total Volatile Organic Compounds At Roadsides in Addis Ababa, Ethiopia. 16, Pp. 4761–4774. 47 Embiale, Asamene Et Al. 2019. Short-Term Exposure Assessment To Particulate Matter And Total Volatile Organic Compounds In Indoor Air During Cooking Ethiopian Sauces (Wot) Using Electricity, Kerosene And Charcoal Fuels. Indoor And Built Environment. Doi: 10.1177/1420326X19836453. Environmental Protection Agency, US. 2015. America’s Children and the Environment. https://www.epa.gov/americaschildrenenvironment.Etyemezian, V. Et Al. 2005. Results from A Pilot-Scale Air Quality Study in Addis Ababa, Ethiopia, Atmospheric Environment, 39(40), Pp. 7849–7860. Doi: 10.1016/J.Atmosenv.2005.08.033. European Environment Agency. 2020. Air Pollutant Emission Inventory Guidebook 2019. https://www.eea.europa.eu/publications/emep-eea-guidebook-2019. Fattorini, D. and Regoli, F. 2020. Role of the chronic air pollution levels in the Covid-19 outbreak risk in Italy. Environmental Pollution, 264, 114732 doi: 10.1016/j.envpol.2020.114732. Federal Office for the Environment, Switzerland. 2019. Study on the durability of European passenger car emission control system utilizing remote sensing data. https://www.ivl.se/download/18.34244ba71728fcb3f3fabb/1591706073882/C387.pdf. Gaita, S. M., J. Boman, M. J. Gatari, J. B. C. Pettersson, and S. Janhäll. 2014. Source apportionment and seasonal variation of PM2.5 in a Sub-Sahara African city: Nairobi, Kenya. Atmos. Chem. Phys. Discuss., 14: 9565–9601. GBD. 2019 Risk Factors Collaborators. 2020. Global burden of 87 risk factors in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet 396: 1135-59. Gebre, G., Feleke, Z. And Sahle-Demissie, E. 2010. Mass Concentrations and Elemental Composition of Urban Atmospheric Aerosols In Addis Ababa, Ethiopia’, Bulletin Of The Chemical Society Of Ethiopia, 24(3), Pp. 361–373. Doi: 10.4314/Bcse.V24i3.60787. Goel R. et al. 2015. Benchmarking vehicle and passenger travel characteristics in Delhi for on-road emissions analysis, Travel Behaviour and Society 2 (2015), 88-101 Grutter, Jurg, Wenyu Jia, and Jian Xie. 2021. Steering Towards Cleaner Air: Measures to Mitigate Transport Air Pollution in Addis Ababa. Washington, DC: The World Bank. He, G., Y. Pan and T. Tanaka. 2020. The short-term impacts of COVID-lockdown on urban air pollution in China. Nature Sustainability (3), 1005-1011. Health Effects Institute. 2020. State of Global Air 2020: Special Report. Boston, MA: Health Effects Institute. Hunt, A., J. Ferguson, F. Hurley, and A.Searl. 2016. Social Costs of Morbidity Impacts of Air Pollution. OECD Environment Working Papers, Paris, France: OECD Publishing. IEC. 2019. Introduction to Health Benefits Analysis and Air Quality Session, Review and Planned Updates to Baseline Assessment. Presentation Addis Ababa workshop. IHME. 2020. http://www.healthdata.org/. 48 International Council of Clean Transportation. 2016a. Costs of Emission Reduction Technologies for Heavy-Duty Diesel Vehicles. http://admin.indiaenvironmentportal.org.in/files/file/ICCT_costs- emission-reduction-tech-HDV.pdf. International Council of Clean Transportation. 2016b. NOx Emissions from Heavy-duty and Light-duty Diesel Vehicles in the EU: Comparison of Real-world Performance and Current Type-approval Requirements. https://theicct.org/sites/default/files/publications/Euro-VI-versus- 6_ICCT_briefing_06012017_revised.pdf. International Council of Clean Transportation. 2016c. A Technical Summary of Euro 6/VI Vehicle Emission Standards. https://theicct.org/sites/default/files/publications/ICCT_Euro6- VI_briefing_june2016.pdf. International Council of Clean Transportation. 2015a. Cost and Benefits of Motor Vehicle Emission Control Programs in China. https://theicct.org/sites/default/files/publications/ICCT_China_MVEC_benefits- costs_20150629.pdf. International Council of Clean Transportation. 2015b. Survey of Best Practices in reducing Emissions Through Vehicle Replacement Programs. https://theicct.org/sites/default/files/publications/ICCT_HDVreplacement_bestprac_20150302.pd f. International Council of Clean Transportation. 2013. Overview of India’s Vehicle Emissions and Control Program. https://theicct.org/sites/default/files/publications/ICCT_IndiaRetrospective_2013.pdf. International Energy Agency. 2016. Key World Energy Statistics 2016. www.oecd- ilibrary.org/energy/key-world-energy-statistics-2016_key_energ_stat-2016-en. International Monetary Fund. 2014. Getting Energy Prices Right. IMF: Washington, DC, USA. www.elibrary.imf.org/view/books/071/21171-9781484388570-en/21171-9781484388570-en- book.xml. Intergovernmental Panel on Climate Change. 2006. Guidelines for National Greenhouse Gas Inventories. www.osti.gov/etdeweb/biblio/20880391. International Transport Forum. 2016. Decarbonising Transport Initiative. https://www.itf- oecd.org/decarbonising-transport. Kniesner, T. and W. K. Viscusi. 2019. The Value of Statistical Life. Working Paper 19–15. Legal Studies Research Paper Series. Vanderbilt University Law School. Kumar, P. et al. 2021. Potential health risks due to in-car aerosol exposure across ten global cities. Environment International. 155 (2021) 106688. Kume, A. Et Al. 2011. Magnitude and Variation of Traffic Air Pollution as Measured by CO in the City of Addis Ababa, Ethiopia, Ethiopian Journal Of Health Development, 24(3). Doi: 10.4314/Ejhd.V24i3.68379. Larsen, B. 2019. Arab Republic of Egypt—Cost of Environmental Degradation. World Bank, Washington DC. https://doi.org/10.1596/32513. 49 Mitchell G. and Dorling D. 2003. An environmental justice analysis of British air quality, Environment and Planning A, vol. 35, 909-929 Peeples, L. 2020. News Feature: How Air Pollution Threatens Brain Health. Proceedings of the National Academy of Sciences 117, 13856-13860. Petroni, M. D. Hill, L. Younes, L. Barkman, S. Howard, I. Howell, J. Mirowsky, and M. Collins. 2020. Hazardous air pollution exposure as a contributing factor to COVID-19 mortality in the United States. Environmental Research Letters. 15. 0940a9. Quilter-Pinner, H. et al. 2016. Lethal and Illegal: London's Air Pollution Crisis. www.ippr.org. Robinson, L. and J. Hammitt. 2018. Valuing Nonfatal Health Risk Reductions in Global Benefit-Cost Analysis. Guidelines for Benefit-Cost Analysis Project. Working paper No. 2. https://sites.sph.harvard.edu/bcaguidelines/. Roland Berger. 2016. Integrated Fuels and Vehicles Roadmap to 2030+. file:///C:/Users/wb154559/Downloads/roland_berger_integrated_fuels_and_vehicles_roadmap_to _2030_v2_20160428%20(1).pdf. Rybarczyk, Y. and R. Zalakeviciute. 2021. Assessing the COVID-19 Impact on Air Quality: A Machine Learning Approach. Geophysical Research Letters. 48. Shou, Y., Huang, Y., Zhu, X., Liu, C., Hu, Y. and Wang, H. 2019. A Review of the Possible Associations between Ambient PM2.5 Exposures and the Development of Alzheimer’s Disease. Ecotoxicology and Environmental Safety, 174, 344-352. Tarekegn, M. And Gulilat, T. 2018. Trends ff Ambient Air Pollution and The Corresponding Respiratory Diseases in Addis Ababa, Medical Insight, 2. Tefera, W. et al. 2020. Chemical Characterization and Seasonality of Ambient Particles (PM2.5) In the City Centre of Addis Ababa, International Journal of Environmental Research and Public Health, 17(19), P. 6998. Doi: 10.3390/Ijerph17196998. Tefera, W. et al. 2016. Indoor and Outdoor Air Pollution- Related Health Problem in Ethiopia: Review of Related Literature, Ethiopian Journal of Health Development, 30 (Special Issue 1), Pp. 5–16. The Health Effects Institute/The Institute for Health Metrics and Evaluation (HEI/IHME). 2020. State of Global Air. See https://www.stateofglobalair.org/data/estimate-exposure. UNEP. 2018. Addis Ababa City Air Quality Policy and Regulatory Situational Analysis. Nairobi, Kenya: UNEP. United Nations Habitat, MetroHUB, and RMPU. 2019. Global State of Metropolis 2020. Experts Group Meeting. UN Habitat/MetroHUB/RMPU. US DOT 2011. Summary of Travel Trends: 2009 National Household Travel Survey Viscusi, W. K., and C. J. Masterman. 2017. Income Elasticities and Global Values of a Statistical Life. Journal of Benefit-Cost Analysis. 8 (2): 226–50. World Bank 2014, Reducing Black Carbon Emissions from Diesel Vehicles: Impacts, Control Strategies, and Cost-Benefit Analysis. Discussion Paper. Washington D.C.: The World Bank. 50 World Bank. 2017. Motorization Management in Ethiopia. Discussion Paper. Washington D.C.: The World Bank. World Bank. 2018. Enhancing Economic Development and Job Creation in Addis Ababa: the Role of the City Administration. W Washington D.C.: The World Bank. World Bank/IHME. 2016. The Cost of Air Pollution. Strengthening the Economic Case for Action. Washington D.C.: The World Bank. World Health Organization. 2005. Air Quality Guidelines for Particulate Matter, Ozone, Nitrogen Dioxide and Sulfur Dioxide: Global Update 2005. WHO. World Health Organization. 2006. WHO Air Quality Guidelines for Particulate Matter, Ozone, Nitrogen Dioxide and Sulphur Dioxide. Geneva, Switzerland: World Health Organization. World Health Organization. 2018. “9 out of 10 people worldwide breathe polluted air, but more countries are taking action.� Available at: https://www.who.int/news-room/detail/02-05-2018-9-out-of-10- people-worldwide-breathe-polluted-air-but-more-countries-are-taking- action#:~:text=New%20data%20from%20WHO%20shows,containing%20high%20levels%20of %20pollutants.&text=%E2%80%9CAir%20pollution%20threatens%20us%20all,%2C%20Direct or%2DGeneral%20of%20WHO. World Health Organization. 2021. https://www.who.int/data/gho/indicator-metadata-registry/imr- details/160. Wu, X., D. Braun, J. Schwartz, M. Kioumourtzoglou, and F. Dominici. 2020. Evaluating the impact of long-term exposure to fine particulate matter on mortality among the elderly. Science Advances. Vol. 6, No. 29, eaba5692 Xue, T., T. Guan, Y. Zheng, G. Geng, Q. Zhang, Y. Yao and T. Zhu. 2021. Long-term PM2.5 exposure and depressive symptoms in China: a quasi-experimental study. The Lancet. Vol. 6, 100079. Zachariadis, T. 2013. Gasoline, diesel and climate policy implications—Insights from the recent evolution of new car sales in Germany. Energy Policy, 54, 23–32 Zeger, S. L., D. Thomas, F. Dominici, J. M. Samet, J. Schwartz, D. Dockery, and A. Cohen. 2000. Exposure measurement error in time-series studies of air pollution: Concepts and consequences. Environ. Health Perspect. 108: 419–26. doi:10.1289/ehp.00108419. Zhu, Y., J. Xie, F. Huang and L. Cao. 2020. Association between short-term exposure to air pollution and COVID-19 infection: Evidence from China. Science of The Total Environment. Vol. 727. 138704. Zivin, J. G., Neidell, M., Sanders., N. and G. Singer. 2021. When Externalities Collide: Influenza and Pollution. Presentation for Mostly Harmless Econometricians Series. World Bank, March 16, 2021. 51 Annex 1. List of Background Paper to this Report No.1: An Overview of Ambient Air Quality Monitoring and Air Quality in Addis Ababa No.2: An Overview of Institutional and Financial Arrangements for Air Quality Management in Ethiopia No.3: Addis Ababa: Air Quality & Source Apportionment No.4: Steering Towards Cleaner Air: Measures to Mitigate Transport Air Pollution in Addis Ababa 52 Annex 2. Supporting Information for the Estimation of the Health Impact of Ambient PM2.5 The methodology and estimation of the impact of ambient PM2.5 on health This annex provides a description of the main methodological approaches and sources of data used to assess the health impact of ambient PM2.5 in Addis Ababa. (1) Estimate the PM2.5 exposure. This is calculated based on the following formula: ∑𝑛 1 �𝑜�𝑢𝑙𝑎𝑡𝑖𝑜𝑛𝑖 ∗ PM2.5 �𝑜𝑛�𝑒𝑛𝑡𝑟𝑎𝑡𝑖𝑜𝑛𝑖 Population-weighted PM2.5 = ∑𝑛1 �𝑜�𝑢𝑙𝑎𝑡𝑖𝑜𝑛𝑖 where i = number of monitoring stations; population i = population located in the proximity of the monitoring station at location i; PM2.5 concentration i = annual average PM2.5 concentration measured in location i; and n = number of monitoring stations. (2) Quantify the impacts of ambient PM2.5 on premature mortality. This quantification uses the following information: • Baseline mortality data at the national level, by cause and age group: Table A2-1. Baseline mortality data at the national level, by cause and age group Age groups Deaths Deaths Deaths Deaths Deaths Deaths LRI lung COPD IHD stroke diabetes type 0-4 18,591 cancer 0 3 0 5 II 3 5-9 622 0 0 0 1 0 10-14 351 0 0 0 2 0 15-19 347 0 2 1 10 25 20-24 356 0 4 28 84 36 25-29 377 3 25 165 204 30 30-34 355 7 37 280 276 70 35-39 393 14 50 436 391 123 40-44 492 31 78 672 723 193 45-49 563 51 127 960 900 416 50-54 702 80 205 1,289 1,326 704 55-59 889 115 382 1,721 1,649 921 60-64 1,296 173 656 2,351 2,428 1,353 65-69 1,635 230 925 2,970 3,088 1,426 70-74 2,202 268 1,343 3,660 3,541 1,844 75-79 2,773 250 1,427 3,696 4,085 1,893 80-84 2,564 131 1,253 3,351 3,125 1,562 85-89 1,569 45 593 1,998 1,548 822 90-94 681 11 266 773 706 357 95+ 249 2 53 219 195 79 All ages 37,006 1,411 7,429 24,570 24,289 11,857 Source: GBD 2019 (http://www.healthdata.org, accessed February/March 2021) • Adjustment of the above data for Addis Ababa, based on the following: (i) the national crude death rate reported by the World Bank for Ethiopia (6.55 per 1000 people, https://data.worldbank.org/, using national statistics and other sources) compared to that used in the GBD (2019) (5.2 per 1000 people); (ii) differences in population distribution by age group in Addis Ababa compared to the 53 population distribution at the national level, based on projections provided by the CSA (2013);24 (iii) share of Addis Ababa’s population (4.6 million in 2019, https://worldpopulationreview.com/world-cities/addis-ababa-population) in the total country’s population (112.1 million in 2019, data.worldbank.org). According to various studies, large cities often have higher mortality rates compared to those at the national level, partly due to the high demand for the higher quality of health services available in these areas (e.g., Baseera/NPC/UNFPA, 2016). Thus, many rural people travel to these cities for treatment when they have serious health conditions. When deaths occur, they are registered in these cities, resulting in higher recorded crude mortality rates. Thus, we did not adjust the baseline mortality data to reflect the crude death rate in the capital. • Estimate the proportion of deaths attributable to PM2.5, calculated based on specific relative risk factors which depend on the outcome, age and PM2.5 concentration. The relative risk factors used are related to the PM2.5 concentration of 34 �g/m3 and can be found at http://ghdx.healthdata.org/record/ihme-data/gbd-2019-relative-risks. In addition, deaths related to low birth rate and short gestation (LBWSG) attributable to the PM2.5 were estimated based on baseline mortality data related to LBWSG (0-27 days) and the attributable fraction specific for Addis Ababa, provided by the IHME (communication with IHME, February 2021). (3) Quantify the impacts of ambient PM2.5 on morbidity. This quantification uses similar steps and input sources as reported above, with reference to YLDs. (4) Estimate the economic value of health impacts due to exposure to PM2.5. Estimating the economic value of premature deaths is based on the VSL. To estimate the VSL for Ethiopia, we use the formula and input data suggested by the World Bank/IHME (2016). 𝑌�,𝑛 𝑒 VSLc,n = VSLOECD * ( ) 𝑌𝑂𝐸𝐶𝐷 Where: VSLc,n is the estimated VSL for country c in year n, VSLOECD is the average base VSL in the sample of OECD countries with VSL studies (US$3.83 million), Y c,n is GDP per capita in country c in year n (adjusted for price inflation and converted to 2011 US dollars at PPP rates); Y OECD is the average GDP per capita for the sample of OECD countries (US$37,000), and e is the income elasticity of the VSL (with a central value of 1.2 for low- and middle-income countries). This approach provides a VSL for Ethiopia in the amount of US$43,600 (2019, current prices) based on a GDP per capita of US$856 (2019, current prices). Estimating the economic value of premature morbidity is based on the VSLY. It is derived by dividing that VSL by the expected (discounted) life years remaining for an individual at the mean age of the population studied (Robinson and Hammitt, 2018). For Ethiopia, this was estimated at US$1,800, based on an average life expectancy of 66 years (World Bank, data.worldbank.org), an average age of 24 years (estimated based on CSA data), and a discount rate of 3 percent. 24 For Ethiopia, the distribution of population by group age was estimated as: 14% (0-5 years), 13% (5-9 years), 12% (10-14), 11% (15-19), 10% (20-24), 9% (25-29), 7% (30-34), 6% (35-39), 5% (40-44), 4% (45-49), 3% (50-54), 2% (55-59), 2% (60-64), 1% (65-69), 1% (70-74), 1% (75-79) and 0% (80+). For Addis Ababa, the distribution of population by group age was: 10% (0-5 years), 10% (5-9 years), 7% (10-14), 6% (15-19), 8% (20-24), 10% (25-29), 12% (30-34), 11% (35-39), 8% (40-44), 6% (45-49), 4% (50-54), 3% (55-59), 2% (60-64), 2% (65-69), 1% (70-74), 1% (75-79) and 0% (80+). (CSA, 2013 – based on projections for 2017 and 2022). 54 Results of the Global Burden of Disease Study 2019 for Ethiopia The GBD 2019 study estimated that ambient PM2.5 was responsible for about 8,960 premature deaths in in Ethiopia in 2019. Overall, the study showed that the main causes of premature deaths related to ambient PM2.5 were lower respiratory infections (30 percent of total deaths), stroke (20 percent), neonatal disorders (19 percent), ischemic heart disease (16 percent)– followed by COPD (8 percent), diabetes mellitus (5 percent), lung cancer (1 percent) and other. Two groups of age were particularly affected: elder people of 60-84 years old (44 percent), due to stroke, lower respiratory infections, and ischemic heart disease; and children under five years old (30 percent of total premature deaths), due to lower respiratory infections and neonatal disorders (Figure 2). Figure A2-1. Premature deaths due to ambient PM2.5 pollution in Ethiopia (2019) 3,000 2,500 2,000 Premature deaths 1,500 1,000 500 0 <5 5 -9 10 - 14 15 - 19 20 - 24 25 - 29 30 - 34 35 - 39 40 - 44 45 - 49 50 - 54 55 - 59 60 - 64 65 - 69 70 - 74 75 - 79 80 - 84 85 - 89 90 - 95 95 plus LRI Neonatal disorders Lung cancer COPD Stroke IHD Diabetes mellitus 2 Other Source: IHME (2020), http://ghdx.healthdata.org/gbd-results-tool In addition, ambient PM2.5 was responsible for about 16,036 YLDs in the same year. The group of 40 – 69 years was most affected by disability, concentrating about 55 percent of the total YLDs due to ambient PM2.5 (Figure 3). Figure A2-2. Years Lived with Disability due to ambient PM2.5 pollution in Ethiopia 1800 1600 1400 1200 1000 YLDs 800 600 400 200 0 <5 5 - 9 10 - 14 15 - 19 20 - 24 25 - 29 30 - 34 35 - 39 40 - 44 45 - 49 50 - 54 55 - 59 60 - 64 65 - 69 70 - 74 75 - 79 80 - 84 85 - 89 90 - 95 95 plus LRI Lung cancer COPD Stroke IHD Diabetes mellitus 2 Other Source: IHME (2020), http://ghdx.healthdata.org/gbd-results-tool 55 56