Publications

Weber, SA, Engel-Cox, JA, Hoff, RM, Prados, AI, Zhang, H (2010). An Improved Method for Estimating Surface Fine Particle Concentrations Using Seasonally Adjusted Satellite Aerosol Optical Depth. JOURNAL OF THE AIR & WASTE MANAGEMENT ASSOCIATION, 60(5), 574-585.

Abstract
Using satellite observations of aerosol optical depth (AOD) to estimate surface concentrations of fine particulate matter (PM2.5) is a well-established technique in the air quality community. In this study, the relationships between PM2.5 concentrations measured at five monitor locations in the Baltimore, MD/Washington, DC region and AOD from Moderate Resolution Imaging Spectroradiometer (MODIS), Multi-Angle Imaging Spectroradiometer (MISR), and Geostationary Operational Environmental Satellite (GOES) were calculated for the summer of 2004 and all of 2005. Linear regression methods were used to determine the direct quantitative relationships between the satellite AOD values and PM2.5 concentration measurements. Results show that correlations between AOD and surface PM2.5 concentrations range from 0.46 to 0.84 for the analyzed time period. Correlations with AOD from MODIS and MISR were higher than those from GOES, likely because of variations in the algorithms used by the different instruments. To determine the relative usefulness of platform- and season-specific AOD PM2.5 regression analysis, the results from this study were used to estimate surface PM2.5 concentrations for two representative case studies. This analysis of case studies demonstrates that it is necessary to include season and satellite platform information for more accurate estimates of surface PM2.5 concentrations from satellite AOD data. Consequently, tools that currently use a constant relationship to estimate surface PM2.5 concentrations from satellite AOD data, such as the Infusing satellite Data into Environmental Applications (IDEA) website, may need to be revised to include parameters that allow the relationships to vary with season and satellite platform to provide more accurate results.

DOI:
10.3155/1047-3289.60.5.574

ISSN:
1047-3289