Using Remote Sensing To Understand Urban Air Quality Exposures And Inequities

Author : Matthew Bechle
Publisher :
ISBN 13 :
Total Pages : 110 pages
Book Rating : 4.:/5 (129 download)

DOWNLOAD NOW!

Download or read book Using Remote Sensing to Understand Urban Air Quality Exposures and Inequities written by Matthew Bechle and published by . This book was released on 2021 with total page 110 pages. Available in PDF, EPUB and Kindle. Book excerpt: Outdoor air pollution is one of the leading causes of morbidity and mortality in the United States and around the world, but these impacts are not distributed equally. Countries, communities, and households that are socially and economically deprived often experience higher levels of air pollution. Yet too often these locations remain unmonitored or insufficiently monitored by traditional ground-based measurements. In this dissertation I employ satellite-based remote sensing of nitrogen dioxide (NO2), a major contributor to urban air pollution and a proxy for a toxic mix of pollutants associated with traffic and combustion emissions, to explore air pollution levels globally and within the US. Within the last two decades, satellite air pollution measurements have considerably expanded the capability to measure air pollution in previously unmonitored locations and across administrative boundaries. Cities serve as focal points, concentrating social and economic opportunities, but may also concentrate hazards, including air pollution. Strategic, compact urban design may be a way to improve a cities air quality, yet global empirical evidence has historically been limited by data availability and consistency. Here I use satellite-based measurements of NO2 and built-up land area to explore the relationship between city-wide NO2 levels and urban form characteristics (i.e., contiguity, circularity, percent impervious surfaces, percent vegetation coverage) for a global sample of 1,274 cities. Three of the urban form metrics (contiguity, circularity, and vegetation) have a small, but statistically significant relationship with city NO2 levels; however, the combined effect of these three attributes could be sizeable. For example, a city at the 75th percentile for all three metrics could accommodate, on average, twice the population as a city at the 25th percentile, while maintaining similar air quality. This work also shows that country level factors such as economic conditions and environmental policies may impact the urban form - air pollution relationships. Moreover, the impact of urban form on air quality may be larger for small cities, an important finding given the large portion of current and projected future population that lives in small cities. Satellite air pollution measurements are limited by their spatial resolution. For example, they are well suited for exploring NO2 levels between cities, as described above; however, alone they typically cannot capture the fine-scale spatial variability needed to characterize population exposure to air pollution. Satellite-based empirical models combine the regional concentrations from satellite measurements with ground-based measurements and local land use and land cover information to predict air pollution concentrations with high spatial resolution (typically 1 km or less). These models have become ubiquitous, yet few studies have investigated how satellite and other regional air pollution covariates impact these models. In this dissertation, I address this gap by exploring the effect of several regional NO2 covariates in an empirical model for annual average NO2 over the contiguous US and find that inclusion of a regional covariate improves model predictive power, yet choice of covariate has limited impact. Additionally, empirical models can be data and computationally intensive, and are often limited to long-term averages and a small number of years. Here, I address these issues by developing a straightforward and easy to implement spatiotemporal scaling technique to extend the temporal coverage of a year-2006 annual NO2 model to over a decade (2000-2010) of monthly NO2 estimates. The resulting estimates are data publicly available online. The spatiotemporal scaling technique and these data have since been used in several publications exploring health effects and residential exposure disparities associated with outdoor NO2 levels. Residential air pollution disparities in the contiguous US have become a topic of recent interest. Children are a particularly vulnerable population and disparities in their air pollution exposure could have lasting impacts. Despite this, little has been done to track outdoor air pollution levels at schools throughout the US. In this dissertation, I add to this body of work by exploring a criteria pollutant, NO2, and by considering home and school locations to better understand the role of public schools in students' total exposure. I find that, on average, racial and ethnic minority students live in and attend schools in areas with higher NO2 levels than their non-Hispanic, white peers, and that impoverished students (defined here as those eligible for school lunch programs) attend, on average, schools with higher NO2 levels than their non-impoverished peers. Minority students are much more likely than their white peers to live in areas above the World Health Organization's annual outdoor NO2 guideline, and this likelihood is larger at schools than at home locations, particularly when comparing predominately minority schools to predominately white schools. This finding -- that public schools may exacerbate disparities -- has important implications for addressing childhood inequities. Notably, strategies that do not address school exposure inequities may fail to address overall exposure inequities. Moreover, strategies to reduce school segregation or to identify and mitigate NO2 levels at the most at-risk schools could have a significant impact on children's overall NO2 inequities. This work also shows that race and income are intertwined; independently, more impoverished schools and schools with more minority students tend to be in areas with higher NO2 levels than more well-off schools and schools with fewer minority students. Schools in large urban areas exhibit disparities by race/ethnicity alone, even when controlling for school-level income. This work highlights NO2 disparities at public schools throughout the contiguous US. Those national disparities are driven largely by disparities in the 50 largest urban areas, which provides motivation for additional exploration and tracking of air pollution levels at these locations. In summary, in this dissertation I have demonstrated how satellite measurements and empirical models that incorporate satellite measurements vastly improve the capability of uncovering and monitoring air pollution exposure disparities for a global and US-wide analysis. Recently launched and soon to be launched satellite-borne sensors promise higher spatial and temporal resolution air pollution measurements. Those measurements will allow for better understanding of concentrations and emission sources, as well as improve satellite-based empirical models, facilitating further tracking and characterization of exposures and exposure disparities from global to local scales.