Abstract View
Statistical Analysis and Geospatial Exposure Model of Air Pollution Derived from Brake and Tire Wear
JONATHAN LIU, Irish Del Rosario, Michael Jerrett, Jonah Lipsitt, Farzan Oroumiyeh, Suzanne E. Paulson, Beate Ritz, Yifang Zhu, University of California, Los Angeles
Abstract Number: 443
Working Group: Aerosol Exposure
Abstract
In the realm of air pollution, due to regulations and technological advancements reducing tailpipe emissions, there is an increased focus on non-exhaust automobile emissions, consisting of particulate matter which contain heavy metals capable of oxidative stress. The following reports statistical and geospatial analyses on a dataset generated from two sampling campaigns across 51 different sites in southern California.
Data analyzed include particulate matter and speciated metals data from gravimetric samplers, time-series data from low-cost Purple Air sensors, and filter-based passive NO2 samplers from OGAWA USA. Through a review of prior literature, we identified a priori 31 primary and secondary tracers of brake wear, tire wear, tailpipe emissions, and dust and soil particulates.
Primary statistical analysis reveals a high level of correlation between tracers of automobile activity (Pearson R2 ranging from 0.71 to 0.95 in the fine fraction, 0.70 to 0.96 in the coarse fraction) not found between automobile activity and tracers of dust and soil particulates in both fine and coarse fractions. Similarly, dust and soil tracers are extremely positively correlated with one another (Pearson R2 > 0.8). We also found that levels of correlation between NO2, a gas phase pollutant typically used as a marker for vehicle activity, and tracers of brake wear, tire wear, and tailpipe emissions exceeded that between NO2 and dust and soil particulates.
Incorporating data on intersection density, car traffic intensity, truck traffic intensity, slope semivariance, we have generated an exposure surface of brake and tire wear generated co-kriging within a Bayesian hierarchical framework. Moving forward, our exposure surface will be used within existing epidemiological cohort data and government birth records to better understand the relationship between heavy metals in air pollution and negative health outcomes.