Using Big Data Techniques to Better Understand High-Resolution Cumulative Exposure Assessment of Traffic-Related Air Pollution
Minmeng Tang, Deb Niemeier
Abstract
In this paper, we calculate exposure concentrations of traffic-related air pollutants for different travel modes in the urban environment. Using recent high-resolution mobile sensor measured air pollution concentration data, we simulate bicycle, transit, and vehicle trips within Oakland, California and calculate exposure concentrations for nitric oxide (NO), nitrogen dioxide (NO2), and black carbon (BC). We draw on highly resolved sensor data (on the order of seconds) collected by Aclima and Google, which was then aggregated to the annual median for every 30-m road segment in the study area of Apte et al. For each bicycle, transit, and vehicle trip that we simulate, we calculate the average concentration and the cumulative exposure for all three pollutants. The cumulative exposure is calculated as the total mass of pollutants inhaled during a trip. Our results show that cumulative exposure, rather than the more typical average ambient concentration, may be a better metric for assessing travel pollutant exposure. For all three travel modes, the average concentrations of each trip are not significantly different. When we account for trip duration and route variations for different travel modes and inhalation rates, the cumulative exposure varies dramatically. Cumulative exposure for vehicle passengers tends to be the lowest, as well as having the lowest average per meter and per minute exposure. Bicyclists and public transit users tend to experience higher cumulative exposure due to increased inhalation rates for bicyclists and longer trip duration for public transit users. Last but not least, our study shows that total trip duration is more influential than total trip distance when estimating air pollution exposure. Our use of big data and modern simulation techniques point toward better metrics for assessing air pollutant exposure.