Study on spatiotemporal characteristics and influencing factors of pedestrian-level PM2.5 concentrations in outdoor open spaces of Harbin in winter, using a generalized additive model (GAM)

Abstract

Urban outdoor PM2.5 pollution is becoming a growing global problem due to the accelerated urbanization, especially in heating seasons of cold regions. In the winter of northeast China, the poor air quality due to the high PM2.5 concentrations at pedestrian-level, greatly threatens residents' daily activity and health. However, little has been done to investigate the spatial-temporal characteristics and the influential factors of PM2.5 distribution at the pedestrian-level. In this paper, the spatiotemporal fluctuation patterns of pedestrian-level PM2.5 concentrations in the open spaces of Harbin were investigated employing weather-station monitoring, and the influential factors are evaluated using the generalized additive model (GAM). The results revealed that daily PM2.5 concentrations at the four open spaces followed a U-shaped pattern, with lows peaking between 15:00 and 16:00. PM2.5 concentrations were at their maximum on Monday, while they were quite low from Tuesday to Thursday before increasing somewhat on weekends. Among the four locations, PM2.5 concentrations are highest in an urban business district and lowest in a garden area. In addition, the interactions of meteorological parameters exhibited strong impacts with PM2.5 concentrations. GAM results showed that air temperature (AT) was negatively correlated with PM2.5 concentration. PM2.5 concentrations rose at first, then declined as relative humidity (RH) increased. On static or high-speed winds, PM2.5 concentrations were similarly high.