Remote Sensing of Surface Visibility from Space

The last decade has seen a rapid growth in the use of satellite-derived Aerosol Optical Depth (AOD) measurements for estimating surface particulate matter (PM) concentrations for air quality and health studies. Although the two properties may be correlated, physically connecting the two quantities requires prior knowledge of aerosol type and relative humidity. AOD is an ambient, column-integrated optical property, and PM is a dry mass quantity measured at the surface. Surface visibility, however, is an ambient optical property, and may have closer physical ties to AOD. Since ambient visibility is so important for the safety of both aviation and ground transportation, there are near-continuous measurements at airports and other sites of commercial interest. Yet, like ground PM monitors, measurements of visibility have limited spatial coverage. In this study, we evaluate whether remote sensing techniques can help to determine surface visibility. AOD measurements from the MODerate Resolution Imaging Spectroradiometer (MODIS) and the Aerosol Robotic Network (AERONET) are compared with one-minute extinction coefficient data (visibility = 3.0/σ_ext) from the Automated Surface Observing System (ASOS). However, since ASOS data lack quality control regulations, we first develop methods for quality control, including steps to limit unrealistic variability, poor calibration, and inconsistent formatting. Then we test different protocols for temporal averaging of the ASOS data (±10, 30, and 60 minutes) and spatial averaging of the MODIS data (1x1, 3x3, and 5x5 pixels) to collocate and compare the two data sets. Preliminary results for the U.S. mid-Atlantic show overall moderate correlation between MODIS AOD and ASOS extinction coefficient with higher correlations during summer months and lower correlations during winter months.