Abstract View
Assessment of Children's Personal and Land Use Regression Model-Estimated Exposure to NO2 in Springfield, Massachusetts
DONG GAO, Sarah Esenther, Laura Minet, Alexander De Jesus, Tina Savvaides, Marianne Hatzopoulou, Krystal Godri Pollitt, Yale University
Abstract Number: 684
Working Group: Instrumentation and Methods
Abstract
Ambient nitrogen dioxide (NO2) is a widely available measure of traffic-related air pollution and has been found associated with a variety of health outcomes. The land use regression (LUR) models, which are frequently used to estimate air pollution exposures, is able to describe small-scale spatial variation in air pollution levels based on geographical predictor variables. In this study, an NO2 monitoring network was established at 40 sites across the greater Springfield region in Massachusetts using cost-effective passive samplers and measured ambient NO2 levels during a five-day period in winter 2018. The NO2 measurements, along with the land use characteristics, were incorporated into a LUR model. A total of 25 children (age 12-13), meanwhile, were recruited from a local elementary school, and their personal NO2 exposures were measured by novel wearable samplers. The personal NO2 exposure was compared with LUR-derived NO2 exposure estimates in different microenvironments including homes, school and commute paths. The results show that the variability in personal NO2 exposure is greater than the typical individual outdoor exposures predicted by LUR, though the LUR model had a good prediction performance and could capture the NO2 emission hotspots in outdoor settings. The findings suggest that human NO2 exposure can be highly personalized based on different combustion sources and access to house ventilation, reaffirming the importance of measuring personal exposure.