The Research Of Validation Of Microwave Remote Sensing Soil Moisture Products

Soil moisture is a key variable in hydrological processes,bioecological processes,and biogeochemical processes.Long-term observations of soil moisture over large areas are critical to research on drought and flood monitoring,water resources management,and crop yield prediction.Microwave remote sensing has become the most effective means of continuous monitoring of soil moisture,and a large number of microwave remote sensing soil moisture products have been released by domestic and foreign institutions.At present,the authenticity test of quantitative remote sensing products generally has problems of global applicability and comprehensive verification,which seriously restricts the promotion and application of remote sensing products.With the continuous improvement of the accuracy requirements of quantitative remote sensing products in the process of business and engineering,the authenticity test of remote sensing products has become extremely important.In this study we examined the accuracy of four satellite soil moisture products by using dense in-situ measurements from the Little Washita?LW?and the Fort Cobb?FC?network over United States and the Yanco network over Australia.The four satellite soil moisture products include Chinese Fengyun 3 series soil moisture products of FY3B and FY3C,Japan Aerospace Exploration Agency Advanced Microwave Scanning Radiometer 2?JAXA AMSR2?and European Space Agency of Soil Moisture and Ocean Salinity?SMOS?.The paper focuses on scale effect in authenticity testing of satellite soil moisture products to conduct the research work.The main research contents and conclusions of this dissertation are as follows:?1?Validation of accuracy and error sources of four satellite soil moisture products based on traditional methodIn this study we examined the validation of four satellite soil moisture products via using three dense in-situ measurements from the LW,FC and Yanco networks by traditional methods.And we also discussed the possible error sources of these products.The results showed that FY3B and FY3C soil moisture products overestimated the in-situ soil moisture in the three observation networks,with the maximum wet bias of 0.091 m³ m^-3 and 0.093 m³ m^-3,respectively.JAXA AMSR2 soil moisture products underestimated the measured soil moisture most of the time in the three observation networks,with the maximum dry bias of-0.110 m³ m^-3.SMOS soil moisture products overestimated the in-situ soil moisture in the FC and the Yanco networks,with the maximum wet bias of 0.013 m³ m^-3.In general,SMOS soil moisture products outperformed FY3B,FY3C,and JAXA AMSR2 soil moisture products,with higher correlation coefficients value of 0.85 and the minimum root mean square of 0.05 m³ m^-3.The FY3B and FY3C products showed satisfactory performance with acceptable error metrics,the highest correlation between FY3B and FY3C soil moisture products and measured soil moisture is0.714,and the minimum unbiased root mean square value is 0.042 m³ m^-3.?2?Construction of a direct upscaling validation method based on ground-based soil moisture obvervationsUpscaling ground-based soil moisture obvervations by the kriging and inverse distance weighting?IDW?and simple average methods to validation the four satellite soil moisture products over heterogeneous ground stations.The results show that different scale-up methods affect the accuracy of satellite soil moisture product verification results.When the spatial heterogeneity of the stations are large,the results of satellite soil moisture products verified by the kriging upscaling method and IDW upscaling method are better than the simple averaging approach.When the spatial heterogeneity of the stations are relatively small,the validation results of satellites soil moisture products by the kriging method,inverse distance weighting method,and average method are not much different.?3?Construction of upscaling validation method based on satellite-aided dataSeven heterogeneous ground stations were used to upscale soil moisture over Yanco network by apparent thermal inertia upscaling approach,which has been developed with the aid of high-resolution satellite thermal data.The aid of high-resolution satellite thermal data include MODIS land surface temperature products and MODIS land surface reflectance products.The results show that the measured soil moisture based on apparent thermal inertia upscaling approach is highly consistent with the four satellites soil moisture products,and the validation results of JAXA AMSR2 and SMOS soil moisture products are superior to traditional methods.Ground-based soil moisture by apparent thermal inertia upscaling approach provides a new method for validating satellite soil moisture products over a large region,which has few ground stations.In conclusion,this thesis achieved satellite soil moisture product validation based on different upscaling approachs and analyzed the effectiveness of different upscaling methods by comparing the common satellite evaluation indicators.The research results can provide reference for the promotion and application of remote sensing products.