Enhanced Deep Blue aerosol retrieval algorithm: The second generation

TLDR

The MODIS Deep Blue aerosol retrievals have provided useful aerosol information over bright land surfaces, but their reliance on a static surface reflectance database limits accuracy in mixed vegetated/non‑vegetated regions with strong seasonal changes. This study develops an enhanced Deep Blue algorithm that combines a precalculated surface reflectance database with NDVI‑based estimates, along with updated aerosol model selection and cloud screening, to produce MODIS collection 6 aerosol products. The hybrid surface reflectance estimation, coupled with revised aerosol modeling and cloud screening, is applied to MODIS and similarly adapted for SeaWiFS Deep Blue products. The enhanced algorithm extends aerosol coverage to all land areas, achieves an expected AOT error of better than 0.05 + 20 %, and places 79 % of high‑quality AOT values within this error bound.

Abstract

The aerosol products retrieved using the Moderate Resolution Imaging Spectroradiometer (MODIS) collection 5.1 Deep Blue algorithm have provided useful information about aerosol properties over bright‐reflecting land surfaces, such as desert, semiarid, and urban regions. However, many components of the C5.1 retrieval algorithm needed to be improved; for example, the use of a static surface database to estimate surface reflectances. This is particularly important over regions of mixed vegetated and nonvegetated surfaces, which may undergo strong seasonal changes in land cover. In order to address this issue, we develop a hybrid approach, which takes advantage of the combination of precalculated surface reflectance database and normalized difference vegetation index in determining the surface reflectance for aerosol retrievals. As a result, the spatial coverage of aerosol data generated by the enhanced Deep Blue algorithm has been extended from the arid and semiarid regions to the entire land areas. In this paper, the changes made in the enhanced Deep Blue algorithm regarding the surface reflectance estimation, aerosol model selection, and cloud screening schemes for producing the MODIS collection 6 aerosol products are discussed. A similar approach has also been applied to the algorithm that generates the Sea‐viewing Wide Field‐of‐view Sensor (SeaWiFS) Deep Blue products. Based upon our preliminary results of comparing the enhanced Deep Blue aerosol products with the Aerosol Robotic Network (AERONET) measurements, the expected error of the Deep Blue aerosol optical thickness (AOT) is estimated to be better than 0.05 + 20%. Using 10 AERONET sites with long‐term time series, 79% of the best quality Deep Blue AOT values are found to fall within this expected error.