Aerosol optical depth remote sensing retrieval method

Abstract

The invention discloses an aerosol optical depth remote sensing retrieval method, comprising the following steps of: 1) building a lookup table according to a radiation transfer model; 2) calculating apparent reflectance; 3) extracting a pure pixel in a remote sensing image according to a pure pixel index, and generating a pure pixel mask; 4) determining surface reflectance of the pure pixel; and 5) retrieving radiance obtained by remote sensing into an aerosol optical depth. The aerosol optical depth remote sensing retrieval method disclosed by the invention can be effectively applied to urban aerosol monitoring and has many other advantages besides small influence of season, such as no blocking by high building in city, no influence of a mixed pixel and is easy to obtain bidirectional reflectance of the pure pixel.

Description

technical field [0001] The invention relates to the field of remote sensing, in particular to an aerosol optical depth remote sensing inversion method. Background technique [0002] The retrieval of aerosols from satellite data has been carried out for more than 30 years. Because the ocean surface is relatively uniform, the reflectivity is very small and approximately constant. Therefore, the early aerosol direct inversion research and application mainly focused on the ocean and large water body surface, and put it into the business application of NOAA / AVHRR products. Now it has developed There are two generations of algorithms, and the third generation algorithm is being developed (Husar et al., 1997). However, over land, due to the high heterogeneity of aerosol and surface reflectance in time and space, the land surface reflectance is relatively large. In addition, the radiation at the top of the atmosphere is highly sensitive to both aerosol and surface reflectance, so ...

AI Technical Summary

Problems solved by technology

In 2007, the MODIS algorithm (C5) corrected the relationship between the surface reflectance of visible light and near-infrared bands, and after using more diverse and more accurate aerosol models on a global scale (based on the global aerosol ground-based observation network AERONET ), the inversion accuracy has been greatly improved compared with the previous C4 version (see the figure below), but the inversion results of the upgraded algorithm in some urban areas are not satisfactory

Moreover, even though the dense vegetation algorithm is feasible in urban areas, since the CCD camera of the Environment-1 satellite does not have a mid-infrared band, it is impossible to accurately obtain the surface reflectance of dense vegetation, so this algorithm cannot be used for the Environment-1 satellite

Images