Understanding and addressing exposure inequality in ambient air pollution

Abstract

Air pollution is the largest environmental risk factor in the world, causing ~6.5 million deaths per year. However, people don’t breathe the same air. Air pollution exposures and associated health impact are unevenly distributed across countries, regions, communities, and individuals. Between-country inequalities in ambient air pollution indicate that people living in low- and middle-income countries disproportionately experience the higher pollution exposures and larger burdens of ambient air pollution. Within-country inequalities, including how ambient pollution levels correlate with socio-demographic attributes, are poorly studied other than in the US and a few other high-income countries. Most literature in the US indicates higher air pollution exposures for people of color and low-income populations. During the past decades, the air has gotten cleaner in the US, however, the disparities persist. While ample studies have documented patterns of air pollution exposure inequality in the US, there is almost no scientific literature that explore possible solutions and policies to eliminate the systemic and long-standing inequalities. In addition, inequality patterns in the US may or may not be applicable to other countries, owing to the different historical, social, economic, political, urbanization characteristics. This dissertation consists of four original studies of ambient air pollution concentration and exposure inequality (Chapters 2-5), plus an introduction to these topics (Chapter 1) and a summary of findings with potential implications for future research and for policy (Chapter 6). There are three main objectives: (1) quantify spatial sources of ambient air pollution (Chapter 2); (2) explore exposure inequality patterns in countries other than US (specifically, in China; Chapter 3); and (3) investigate policies/approaches to address the persistent exposure inequalities in the US (Chapters 4 and 5). I focus on two important air pollutants: fine particulate matter (PM2.5), and nitrogen dioxide (NO2). The approaches to estimate air pollution concentration include both empirical and mechanistic air quality models. Chapter 1 provides relevant background on air pollution distribution and exposure inequality, highlights the motivation, objectives, approaches, and structure of this dissertation. Chapter 2 quantifies the spatial sources of ambient NO2 and PM2.5 concentrations. I develop a readily scalable algorithm based on “spatial-increment” to decompose air pollution concentrations into four spatial components: long-range, mid-range, neighborhood, and near-source. I apply the algorithm to the 2010-2015 annual-averaged concentrations from empirical predictions for all census blocks in the contiguous US. I find that NO2 is of urban origin and varies by urbanicity; ~90% of the concentration differences are driven by “neighborhood” and “mid-range” components; climate or geographic regions have less effects on the NO2 concentrations. In contrast, PM2.5 is a regional pollutant with a strong secondary component; the concentrations are dominated by “long-range” components (>50% in most geographic regions) and vary at state and regional level; urbanicity has modest effects on PM2.5 concentrations and minor effects on concentration differences. Chapter 3 quantifies the relationship between ambient air pollution exposure and socioeconomic status (SES) in China. I combine estimated year 2015 annual-average ambient levels of NO2 and PM2.5 from empirical models with national demographic information, which is derived from both China Health and Retirement Longitudinal Study (CHARLS) cohort and gridded gross domestic product (GDP). I find that in contrast to the typical patterns for the US, ambient air pollution concentrations in China are higher for higher-SES populations and communities than for lower-SES populations, and higher for long-standing urban residents than for rural-to-urban migrant populations. The positive relationship holds for different SES measurements (individual SES score, community-averaged SES score, gridded GDP per capita), in rural and urban locations, across geographic regions, across a wide range of spatial resolution from 1-100 km, and for modeled vs. measured pollution concentrations. Exposure inequalities are higher for NO2 than PM2.5. My findings are consistent with the idea that in China’s current industrialization and urbanization stage, economic development is positively correlated with both SES and air pollution levels. Chapters 4 and 5 investigates approaches and policy scenarios to address US national racial-ethnic inequalities of air pollution exposure. I use InMAP (Intervention Model for Air Pollution) Source-Receptor Matrix (ISRM) to predict how changes in emissions impact annual-average PM2.5 concentrations and exposure inequalities. Chapter 4 investigates three emission-reduction approaches, and compare their optimal ability to address both average exposure for the overall population and racial-ethnic exposure inequalities. I find that US national inequalities in exposure can be eliminated with minor emission-reductions (optimal: ~1% of total emissions) if targeting specific locations. In contrast, achieving that outcome using existing regulatory strategies would require eliminating essentially all emissions (if targeting specific economic sectors) or is not possible (if requiring urban regions to meet concentration standards). In addition, there is no tradeoff between reducing overall average and reducing national inequalities; rather, the approach that does the best for reducing national inequalities (i.e., location-specific strategies) also does as well as or better than the other two approach (i.e., sectors-specific; meeting concentration standards) for reducing overall averages. Chapter 5 expands the investigation in Chapter 4 to a specific government environmental justice (EJ) policy -- the Biden Administration’s Justice40 Initiative (“J40”), which is a general policy to address environmental injustice. Climate and Economic Justice Screening Tool (CEJST) is the signature element and ongoing approach of J40. I investigate whether emission-reductions brought about by CEJST/J40 investments will eliminate disparities in PM2.5 exposure by race-ethnicity and other attributes in 20 years, through comparing a Business As Usual (“BAU”) scenario against two scenarios wherein CEJST-identified locations (“J40 communities”) experience accelerated emission-reductions. BAU simply continues historical rates of emissions and emission-changes into the future; in the two CEJST scenarios, I double or quadruple emission reduction in J40 communities, relative to BAU. I find that under BAU scenario, disparities remain in place. The two accelerated scenarios will only eliminate inequalities for J40 communities and for low-income populations in 20 years; yet they do not reduce relative disparities by race-ethnicity. The results indicate that additional and more targeted actions, beyond CEJST/J40, will be needed to end racial-ethnic exposure disparities in the next decades. Chapter 6 summarizes the findings of the studies presented in Chapters 2-5 and discusses implications for future research and policy. Overall, these studies provide new knowledge and insights for air pollution exposure inequalities by: (1) extending the EJ knowledge to China -- indicating a consistent positive relationship between SES and ambient air pollution exposures; and (2) advancing the EJ knowledge in the US -- implying a possible general solution to address the persistent racial-ethnic inequalities in the US, and providing a useful example of regulatory impact analyses for assessing the effectiveness of the EJ policies.