Integration of Community Sensor Network into High Resolution PM2.5 Exposure Assessment: A Case Study in Central California

Peizhi Hao, Wenfu Tang, Joost A. de Gouw, XUAN ZHANG, University of California, Merced

     Abstract Number: 39
     Working Group: Aerosol Exposure

Abstract
The Central Valley, home to the largest population of disadvantaged communities in California, has long experienced some of the worst air quality in the United States. However, the sparse distribution of regulatory monitoring stations in the region limits our ability to fully characterize the temporal dynamics and spatial patterns of air pollution. Community-driven initiatives deploying low-cost air quality sensors play a critical role in addressing this gap by identifying pollution sources and hotspots. In this study, we evaluated how integrating a community sensor network improves the accuracy of PM2.5 exposure mapping in Fresno County, located in the heart of the Central Valley. We combined data from 214 low-cost sensors with WRF-Chem model simulations using the Enhanced Voronoi Neighbor Averaging (eVNA) method. This data fusion approach produced a higher-resolution spatial distribution of PM2.5 concentrations, enabling more precise characterization of pollutant variability and improved exposure assessments. Additionally, we assessed the effect of sensor density on mapping accuracy through repeated random sampling experiments. Our results indicate that increased sensor density significantly enhances the accuracy of PM2.5 spatial distribution maps. These findings offer valuable insights for optimizing sensor deployment strategies and support data-driven interventions aimed at protecting public health.