Accuracy assessment of Terra-MODIS aerosol optical depth retrievals

Moderate Resolution Imaging Spectroradiometer (MODIS) aerosol products have been widely used to address environment and climate change subjects with daily global coverage. Aerosol optical depth (AOD) is retrieved by different algorithms based on the pixel surface, determining between land and ocean. MODIS-Terra and Global Aerosol Robotic Network (AERONET) products can be obtained from the Multi-sensor Aerosol Products Sampling System (MAPSS) for coastal regions during 2000-2010. Using data collected from 83 coastal stations worldwide from AERONET from 2000-2010, accuracy assessments are made for coastal aerosol optical depth (AOD) retrieved from MODIS aboard the Terra satellite. AOD retrieved from MODIS at 0.55μm wavelength has been compared With the AERONET derived AOD, because it is reliable with the major wavelength used by many chemistry transport and climate models as well as previous MODIS validation studies. After removing retrievals with quality flags below1 for Ocean algorithm and below 3 for Land algorithm, The accuracy of AOD retrieved from MODIS Dark Target Ocean algorithms (correlation coefficient R 2 is 0.844 and a regression equation of τ M = 0.91·τ A + 0.02 (where subscripts M and A represent MODIS and AERONET respectively), is the greater than the MODIS Dark Target Land algorithms (correlation coefficient R 2 is 0.764 and τ M = 0.95·τ A + 0.03) and the Deep Blue algorithm (correlation coefficient R 2 is 0.652 and τ M = 0.81·τ A + 0.04). The reasons of the retrieval error in AOD are found to be the various underlying surface reflectance. Therefore, the aerosol models and underlying surface reflectance are the dominant factors which influence the accuracy of MODIS retrieval performance. Generally the MODIS Land algorithm implements better than the Ocean algorithm for coastal sites.