Retrieval of Aerosol Optical Depth From INSAT‐3D Imager Over Asian Landmass and Adjoining Ocean: Retrieval Uncertainty and Validation

Abstract

AbstractAn algorithm for aerosol optical depth (τ) retrieval from imager data onboard geostationary Indian National Satellite (INSAT‐3D) is described. The basic principles of the algorithm are adopted from Geostationary Observational Environmental Satellite Aerosol and Smoke Algorithm, where darkest observations over a time period and constant background aerosol optical depth (τb) value are used to derive surface reflectance. However, in INSAT‐3D algorithm spatially and temporally dynamic τb derived from Moderate Resolution Imaging Spectroradiometer (MODIS)‐aqua is used. The theoretical simulations suggest retrieval uncertainty of around ±30–45% depending on the certainty of aerosol optical properties and other parameters used in radiative transfer calculations. Retrieval uncertainty is less over dark regions and lower scattering angles. INSAT‐3D τ at 0.55 μm is derived operationally for the first time over Asia and adjoining ocean at 30‐min temporal resolution. The retrieval algorithm is validated against in situ Sun‐sky radiometer measurements at 26 Aerosol Robotic Network sites and Sun‐photometer measurements during Sagar‐Sampada cruise in the northern Arabian Sea. The validation study encompassing 3,803 and 844 data points over land and ocean stations show correlations ranging from 0.70 to 0.91 and 0.79 to 0.91, respectively. The uncertainty of retrievals is within 45% over land and within 30% over ocean. The diurnal variability of INSAT‐3D and Aerosol Robotic Network τ in India and southeast Asia is also presented. The comparison of INSAT‐3D and MODIS‐aqua monthly average τ shows correlation ranging from 0.74 to 0.89 and 0.65 to 0.97 over Indian landmass and adjoining ocean, respectively. The validation suggests that INSAT‐3D can be used for aerosol monitoring and can be merged with MODIS‐aqua/terra aerosol product to generate combined aerosol product at high temporal resolution.