Improved MODIS aerosol retrieval over urban scenes

Determination of aerosol optical depth from satellite remote sensing measurements is extremely complex due to the large variability of aerosol optical properties. Significant simplification occurs when measurements are taken over water since the ocean reflection signal can be taken as negligible in the NIR. Unfortunately, over land, most of the signal can be attributed to ground reflectance. While conventional approaches look for "dark" pixels in an image to isolate aerosols, these pixels are subjected to increased noise. In this paper, we focus on the feasibility of the MODIS sensor to determine aerosol optical depth. In particular, an intercomparison between MODIS and CIMEL shows a significant trend for MODIS to overestimate optical depth. We show that this may be explained through an inaccurate assumption on the correlation between the VIS and NIR surface albedos. In particular, we show through an analysis of hyperspectral high resolution Hyperion data that the correlation coefficient assumption underestimates ground albedo resulting in an overestimate of the VIS optical depth. Using a series of coincident measurements between aeronet and MODIS, we estimate a more accurate angle dependant albedo and use it to determine aerosol optical depth. The results show significantly less overbias. Our efforts to reduce space resolution to 1.5km are also discussed.