Search

In the present study, we report the first comparison of the aerosol properties measured with sun photometer at Camagüey, Cuba, with the MODerate resolution Imaging Spectroradiometer (MODIS) instruments on Terra and Aqua satellites. We compared the aerosol optical depth at 550 nm (AOD) and the Ångström Exponent (AE) from the sun photometer for the period 2008 to 2014 with the same variables measured by both MODIS instruments, that are spatially and temporally coincident. The comparison includes AOD derived with both Deep Blue (DB) and Dark Target (DT) algorithms from MODIS Collection 6. The AOD derived with DT algorithm for Terra and Aqua agrees better with AOD from the sun photometer than the AOD derived with DB. Additionally there is little difference between AOD from both satellite instruments, when they are compared with sun photometer AOD, allowing to combine AOD from Terra and Aqua for more comprehensive climatological statistics. The comparison of the AE showed similar results with reports in the literature about the little skills of the current DT and DB algorithms for its retrieval. In addition, we report the comparison of the broadband AOD (BAOD) from pyrheliometer measurements located at Camagüey site and other three meteorological stations along Cuba, with AOD measurements from the sun photometer and from MODIS onboard Terra and Aqua. The comparison of the BAOD from the four sites as a whole with coincident AOD from MODIS onboard Terra and Aqua showed similar results than the ones of the comparison between the sun photometer AOD and the AOD from the two satellite instruments. In the comparison between the BAOD and the AOD at each one of the eight individual sun photometer wavelengths, the results improve in the spectral range 400 to 675 nm, with the best result at 500 nm. The BAOD typical uncertainty ranges from 0.04 to 0.06 at this band. The results from the BAOD comparisons demonstrate its reliability for characterizing AOD at sites with no sun photometer and for extending backward in time AOD estimates. [ABSTRACT FROM AUTHOR]