Yusef Reazaee; Mohammad Reza Mobasheri
Abstract
Due to the effect of aerosols present in the atmosphere on the satellite images, the study of the effect of local aerosols distribution on the satellite images is important. On the other hand, the study shows that the effect of aerosols on the greenhouse gases and consequently on climate is also undeniable ...
Read More
Due to the effect of aerosols present in the atmosphere on the satellite images, the study of the effect of local aerosols distribution on the satellite images is important. On the other hand, the study shows that the effect of aerosols on the greenhouse gases and consequently on climate is also undeniable and as a result, this puts more emphasize on the necessity of this study. Lack of information about modality of distribution of the aerosols in the atmosphere, brings uncertainties in the extraction of decent information from satellite images. Different methods of aerosol retrieval over the ocean have been deployed using AVHRR/NOAA image products. Over the continent, however due to the fact that the surface albedo is generally unknown and varies with time and wavelength, nothing much has so far been done and what is done is for desert dusts only. Some retrieval algorithms are developed based on the assumption of low reflectance of vegetation canopies in the blue and red portion of the electromagnetic spectra. Also few algorithms have been developed for multi-temporal studies based on constant surface reflectance. Application of all of these algorithms is tedious and very time consuming. On the other hand we need to extract decent information on a real time basis and this arises the need for an algorithm that can be performed as fast as possible and this was the objective of this study. The algorithm used for aerosol detection in this study is innovated using a combination of aerosol information model and optical thickness parameter calculated using visible and IR bands. In this study using Optical Thickness which is a measure of the amount of aerosols in the atmosphere, a fast method for relative correction of the satellite images is presented. In this method, surface reflectance in MIR is calculated from which, dark pixels are determined. Using surface reflectance of these dark pixels in red and blue, optical thicknesses are estimated. Finally using these calculated optical thicknesses, the effects of atmospheric aerosols has been removed from satellite imageries. The results show that in visible bands of MODIS where the effect of aerosols are more serious, the computed optical thickness can be used to estimate the effect of aerosols and help to eliminate its effects from images.