Retrieval And Verification Of Land Surface Temperature And Vegetation Optical Depth In China Based On FY-3D Microwave Radiation Imager

Land surface temperature is a key parameter of energy exchange between the surface and atmosphere,and it plays an important role in global climate change,ecological monitoring and other research fields.vegetation optical depth is a seasonal cycle parameter that describes the growth and aging of vegetation canopy.Both of them play an important role in global climate change,ecological monitoring and other research fields.Therefore,how to quickly obtain high-precision and wide-range information of land surface temperature and vegetation optical depth to meet the needs of practical application has become a key concern.Compared with optical-infrared remote sensing,microwave remote sensing has the advantages of penetrating clouds and fog,all-day observation,etc.It is one of the most effective means to monitor surface parameters on a large scale in the world.At present,FY-3D microwave land surface temperature product data has the problems of wide range missing and low accuracy.Therefore,this paper introduces UMT(University of Montana Land Surface Retrieval Algorithm)algorithm to retrieve the surface temperature from the brightness temperature data of FY-3D microwave imager.The accuracy is verified and evaluated by using related products and measured values on the land surface.At the same time,the vegetation optical depth is retrieved and verified.The main research contents of this paper are summarized as follows:(1)Retrieval and verification of land surface temperature based on FY-3D microwave imagerThe UMT retrieval results and FY-3D microwave imager land surface temperature products are verified and analyzed with the measured values and MODIS land surface temperature products respectively.The results show that,compared with the FY-3D microwave imager land surface temperature products,the UMT retrieval results can not only ensure the retrieval accuracy,but also increase the amount of retrieval data.In the verification of land-water mixed pixel land surface temperature retrieval results,the verification accuracy of UMT retrieval results increases with the increase of the proportion of open water in the pixel.When the proportion of open water in [0.05,0.2),the root mean square error and unbiased root mean square error of UMT retrieval results reach the minimum values of4.415°C and 4.377°C,respectively.When the proportion of open water in pixels is large,the UMT retrieval results can still reflect the time series changes of the measured data.The UMT retrieval results can reflect the changing trend and distribution pattern of global land surface temperature,and at the same time make up for the data missing of FY-3D microwave imager land surface temperature products when there is a high proportion of open water in pixels.(2)Retrieval and verification of global vegetation optical depth based on FY-3D microwave imagerSelecting the brightness temperature data of FY-3D microwave imager from 2018 to2020,according to the information of land surface temperature,atmospheric water content,open water fraction parameters and so on obtained from the retrieval of surface temperature model,introducing the land-water slope parameter ?,using UMT algorithm to retrieve the global vegetation optical depth,and using MODIS vegetation index(NDVI?EVI)products?GLOBMAP leaf area index products and LPDR products to analyze the correlation of the retrieval results.The results show that there is a good time series correlation between the vegetation optical depth retrieved by UMT algorithm and vegetation index and leaf area index,which can reflect the seasonal and periodic changes of global vegetation phenology.The correlation of low vegetation coverage area is better than that of high vegetation coverage area,and the correlation between vegetation optical depth and vegetation index(NDVI?EVI)under grassland type is the highest,with correlation coefficients of 0.955 and 0.952 respectively.The correlation coefficient between vegetation optical depth and leaf area index under woody savannas is the highest,which is 0.912.The correlation coefficient between vegetation optical depth and LPDR products under evergreen needleleaf forest is the highest,which is 0.898.Vegetation optical depth can better reflect the global spatial distribution of vegetation.In areas with high vegetation coverage,such as equatorial rainforest,Northeast Asia and America,the value of vegetation optical depth is higher.In areas with low vegetation coverage,such as West Asia,North Africa and western Australia,the value of vegetation optical depth is lower.