Using MODIS Satellite Imagery to Estimate Particulate Matter (PM2.5)

Abstract

With air pollution increasingly becoming one of the most deadly environmental issues worldwide, this study looks to observe a specific type of air pollution called particulate matter (PM2.5). PM2.5 consists of inhalable particles, less than 2.5 microns or 0.0025 millimeters in size and is a leading environmental cause for cardiovascular diseases and mortality. Since many places in the world, especially in developing countries that are rapidly industrializing and urbanizing, do not have adequate access to PM2.5 ground monitors, or even normal weather stations to directly observe and measure high concentration rates, this study strives to determine how accurately particulate matter can be estimated using only readily available Moderate Resolution Imaging Spectroradiometer (MODIS) satellite imagery that has global coverage since 1999 at 1-2 day intervals. To validate the accuracy, the MODIS data was compared with ground monitor stations across the contiguous United States. Correlation trends were observed temporally (period of time), seasonally, and latitudinally. While results did not show significant evidence of any temporal trends, there were significant seasonal patterns with highest correlations in the summer, followed closely by the fall. The model also showed that higher latitudes tend to produce higher correlations, but generate less satellite data, especially during the winter months. Although satellite imagery passover cannot solely estimate particulate matter, using the trends observed in this study along with meteorological variables and land use could prove to be useful in countries without ground monitors.