Atmospheric Correction of Multispectral Satellite Images Based on the Solar Radiation Transfer Approximation Model

Resumo

A model of the solar radiation transfer in the Earth’s atmosphere is developed for explicit calculations of the transfer characteristics of the atmosphere (brightness coefficients, transmission, and spherical albedo) in the visible and near-IR atmospheric windows depending on the surface air pressure (or terrain elevation), aerosol optical parameters, and angles indicating the position of the Sun and the satellite system with respect to the Earth’s surface. The model is based on spectrally dependent approximations obtained using the atmospheric radiative transfer code DISORT. The effect of altitudinal stratification of the atmospheric optical parameters on the accuracy of underlying surface spectral albedo retrieval from the brightness coefficients at the top of the atmosphere is investigated. An algorithm is developed for radiometric correction of multispectral satellite images based on the model suggested. Approbation of the algorithm is performed using MERIS (Medium Resolution Imaging Spectrometer) data. It is shown that accounting for optical properties of the atmosphere and adjacency effects in processing satellite images substantially improves the visibility of the underlying surface and renders their spectral reflectance parameters.