Spatial scales of pollution from variable resolution satellite imaging

The Moderate Resolution Imaging Spectroradiometer (MODIS) provides daily global coverage, but the 10 km resolution of its aerosol optical depth (AOD) product is not adequate for studying spatial variability of aerosols in urban areas. Recently, a new Multi-Angle Implementation of Atmospheric Correction (MAIAC) algorithm was developed for MODIS which provides AOD at 1 km resolution. Using MAIAC data, the relationship between MAIAC AOD and PM2.5 as measured by the EPA ground monitoring stations was investigated at varying spatial scales. Our analysis suggested that the correlation between PM2.5 and AOD decreased significantly as AOD resolution was degraded. This is so despite the intrinsic mismatch between PM2.5 ground level measurements and AOD vertically integrated measurements. Furthermore, the fine resolution results indicated spatial variability in particle concentration at a sub-10 km scale. Finally, this spatial variability of AOD within the urban domain was shown to depend on PM2.5 levels and wind speed.

► The correlation between PM2.5 and AOD decreases as AOD resolution is degraded.
► High resolution MAIAC AOD 1 km retrieval can be used to investigate within-city PM2.5 variability.
► Low pollution days exhibit higher spatial variability of AOD and PM2.5 then moderate pollution days.
►  AOD spatial variability within urban area is higher during the lower wind speed conditions.