Abstract View
Evaluation of MODIS MAIAC AOD Retrievals against AERONET AOD over Different Land Cover Types
SOMAYA FALAH, Alaa Mhawish, Meytar Sorek-Hamer, David Borday, Technion
Abstract Number: 15
Working Group: Remote and Regional Atmospheric Aerosol
Abstract
The Multi-Angle Implementation of Atmospheric Correction (MAIAC) algorithm enables retrieving aerosol and surface bidirectional reflectance products over the land from the Moderate Resolution Imaging Spectroradiometer (MODIS) measurements. We performed critical assessment of MAIAC AOD550nm at 1km spatial resolution using ground-truth AOD550nm data from 21 Aerosol Robotic Network (AERONET) sites in Northern Africa (NA), California (CA) and Germany (GR) from the years 2007-2017. For this, we centered the MAIAC AOD, spatially averaged over 1, 9, 25 and 81 km2 (corresponding to 1x1, 3x3, 5x5 and 9x9 pixels of 1x1 km), around the AERONET stations. AERONET Version 3, Level 2 (V3, L2) Direct Sun AOD retrievals were averaged over 5, 15 and 30 min around the satellite overpass time. This enabled us to investigate the effect of different spatial and temporal averaging windows on the MAIAC performance.
The MAIAC accuracy was found to depend on the surface properties, with ~70% of the retrievals falling within the expected error, EE = ±(0.05 + 0.05 × AOD), and the correlation coefficient, R2, exceeding 0.897 over highly vegetated (i.e. dark) regions (NDVI > 0.6). However, the accuracy of AOD retrievals over bright surfaces was poorer, with EE = 58% and R2 = 0.772. In addition, the MAIAC accuracy is also affected by the aerosol loading in the atmospheric column, estimated in terms of high/low AOD, and by the aerosol type. The retrieval bias found higher at high aerosol loading (high AOD) and coarse aerosol types (low AE).
In general, Aqua MAIAC AOD retrievals showed good agreement with the AERONET measurements, with correlation coefficients of 0.768, 0.634 and 0.747, and expected errors of 50.3%, 73.54% and 60.4%, for North Africa, California and Germany, respectively.
The results contribute to a comprehensive evaluation of MAIAC AOD retrievals, with the latter being used for estimation particulate matter over different land cover types.