An Empirical Study of Particulate Matter Exposure for Transit Users at Bus Stop Shelters

摘要

Congested traffic corridors in dense urban areas are key contributors to the degradation of urban air quality. While waiting at bus stops, transit patrons may be exposed to greater amounts of vehicle-based pollution, including particulate matter, due to their proximity to the roadway. Current guidelines for the location and design of bus stops do not take into account air quality or exposure considerations. This thesis provides a unique contribution to roadside air quality studies and presents an innovative method for the consideration of bus shelter placement. Exposure to roadside pollutants is estimated for transit riders waiting at three-sided bus stop shelters that either: 1) face roadway traffic, or 2) face away from roadway traffic. Shelters were instrumented with particulate matter monitoring equipment, sonic anemometers for wind speed and direction, and vehicle counters capable of categorizing vehicles by length. Temperature and relative humidity were gathered from a nearby monitoring station. Data were collected for two different days at three shelters during both the morning and afternoon peak periods for a total of eleven data periods. Bus shelter orientation is found to significantly affect concentration of four sizes of particulate matter: ultrafine particles, PM1, PM2.5, and PM10. Shelters with an opening oriented towards the roadway were observed to have significantly higher concentrations inside the shelter than outside the shelter. In contrast, shelters oriented away from the roadway were observed to have significantly lower concentrations inside the shelter than outside the shelter. The differences in average particulate matter concentrations are statistically significant across all four sizes of particulate matter studied. Additional correlation and linear regression investigation reveals interactions between particulate concentrations and built environment characteristics, vehicle flow, and weather conditions. Temperature and relative humidity played a large role in the diurnal variation of average concentration levels. In all instances, particulate concentrations were greater during the morning period, often substantially so. Particulate concentrations are shown to vary based on both wind speed and direction. Vehicle flow is correlated with particulate levels, though significance is not consistent. Lagged vehicle flow is demonstrated to be more consistently significant. Regression analysis suggests weather factors such as wind, temperature, and relative humidity explain roughly 70% of particulate variation, while vehicle flow explains less than 6%.