HR: 1340h
AN: A33C-1424    [Abstracts]
TI: Improved Surface albedo correlation models at high spatial resolution for urban areas using combined MODIS and sky radiometer measurements
AU: * Oo, M
EM: gross@ccny.cuny.edu
AF: City College of New york, 140th St and Convent Ave, New York, NY 10031, United States
AB: 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. In previous work, it was shown that the Collection 4 algorithms severely underestimated urban ground albedo leading to a significant overestimate of aerosol optical depth. This effect was partially compensated in collect 5 who used a dynamic surface albedo model to distinguish surface types and enhance albedo. In this paper, we reassess the accuracy of collection 5 aerosol optical depth in comparison to CIMEL AOD at different spatial resolutions. To obtain these resolutions, the cloud mask must be slightly modified. We find that at 10km resolution, the Collection 5 retrieval has greatly improved retrieval accuracy over collection 4 although a significant overestimate bias is seen. However, when processed at 4km resolution, the vegetation pixels are dramatically reduced in proportion to the urban pixels and the aerosol optical depth from MODIS dramatically overestimates aerosol optical depth. To address this issue, simultaneous matchups from MODIS reflectance at 4km and 1.5km resolution are matched to CIMEL AOD retrievals allowing us to retune the surface albedo model at these higher resolution. Based on large statistical data sets, we find that the angular variation of the albedo is negligible relative to the mean albedo justifying the use of a lambertian model which is then used to develop surface correlation maps for the NYC Metropolitan area. To eliminate the contamination due to local rivers, an additional 2130 mask is used after our cloud clearing has been developed These maps clearly identify the changes of correlation coefficient due to different land surface types. Finally, we demonstrate that using this retuned algorithm results in a significantly improved AOD product at 4km. Finally, simultaneous measurements between neighboring sky radiometers at 2km separation are used to show that the horizontal correlation scale for urban environments are on the order of 4km showing that further improvement of spatial resolution of satellite products is not crucial.
DE: 0305 Aerosols and particles (0345, 4801, 4906)
SC: Atmospheric Sciences [A]
MN: 2007 Fall Meeting