Retrieval And Verification Of Aerosol Optical Depth Using Remote Sensing Data Of China High Resolution Satellites

In recent years,with the rapid development of industrialization and urbanization,smog pollution has appeared frequently throughout the country and has become the focus of attention from all walks of life.Atmospheric aerosol optical depth(AOD)is an important measure of air quality.It can describe the spatial distribution of haze.Retrieval of AOD based on remote sensing satellite imagery is an effective means to dynamically monitor large-scale air pollution.Since 2013,the successful launch of domestic high-resolution series satellites has met the high-precision and timely monitoring requirements for Earth observation.In order to better expand the industrial application of domestic high-resolution Earth observation satellite data,this paper takes Beijing,Taihu,Changsha,Zhuzhou and Xiangtan.For the study area,based on the band characteristics of the high-resolution 1(GF-1)multi-spectral remote sensing data,combined with the dark target algorithm and the dark blue algorithm,the method of constructing the reflection point stability point is explored around the red-blue band relationship of the GF-1 WFV image.A higher resolution surface reflectivity database establishes a fast and accurate AOD remote sensing retrieval method for GF-1 data land area operations.The main research results are as follows:(1)The GF-1 WFV images in different seasons were selected for rapid atmospheric correction,and the linear relationship between the GF-1 data red and blue band surface reflectance in the dense vegetation area in summer and winter was obtained by NDVI and vegetation coverage.Apply this relationship to dark target retrieval.The difference in the blue-band width of different sensors leads to the difference in the band response of the same feature.The atmospheric corrected GF-1 WFV data and the minimum monthly MODIS surface reflectance synthesis data are used to calculate the same surface vegetation coverage under GF-1 data blue.The relationship between the band and the MODIS blue band is applied to the retrieval of the dark blue algorithm.(2)In this paper,through the data analysis of MOD09A1 in 2014-2017,the relationship between red and blue bands of stable points in different months is obtained,and a set of 250 m MODIS surface reflectivity database in eastern China is established,which is used as the real surface input and retrieval of the retrieval part of the deep blue algorithm.The results show that after the spatial resolution is improved,the absolute error of different underlying surfaces except grassland can be kept within 14%,which is equal to the retrieval error of 500 m resolution surface reflectance data.(3)Combine advantages of the two algorithms,the dark target algorithm and the deep blue algorithm,achieve 250 m aerosol optical thickness retrieval in three regions.78.3% of AOD retrieval results are within the error line.Different aerosol model retrieval results have good correlation with AERONET measured values(R>0.75),and AOD results have the highest winter accuracy,and MODIS aerosol products.With a uniform spatial distribution(R>0.85),the use of a custom aerosol mode has a limited effect on improving the accuracy of the retrieval results.When the surface reflectance changes by more than ±0.01,the accuracy of the AOD retrieval results is reduced by 7%.