Research For Remote Sensing Retrieval Of Surface Soil Moisture By Coupling Soil Physical Properties In Guangxi

Soil moisture is a very important parameter affecting plant survival,climate change and global ecological environment,which is closely related to human life.Due to the special Karst landform and special soil types formed by karst weathering in Guangxi area(such as red soil,lateritic red soil and lateritic red soil),its natural environment is relatively fragile,and is prone to natural disasters such as drought,floods and so on.As an important indicator of disaster monitoring,it is necessary to accurately and dynamically monitor the change of soil moisture.Remote sensing technology has the advantages of high resolution,large range and real-time.Therefore,the research on soil moisture retrieval using remote sensing method has important practical significance.Taking Guangxi as the research area,this research comprehensively uses Guangxi soil physical property data,soil particle size data,monthly average temperature and precipitation data of Guangxi stations from 2015 to 2019,TRMM precipitation data,ERA-5 soil moisture data and soil relative humidity measured by meteorological stations in 2017 as the main data sources.Considering the advantages and disadvantages of optical remote sensing model in retrieving soil moisture,and combining the soil effective water content model(AWC＿m)of soil physical properties with the adaptive Palmer remote sensing index model(PDSI),the AWC＿m-sc＿PDSI model is constructed,and the AWC＿m-sc＿PDSI model is used to retrieve the soil moisture in Guangxi.The temporal and spatial changes of soil moisture in Guangxi from 2015 to 2019 are analyzed,and the practicability and accuracy of the AWC＿m-sc＿PDSI model are evaluated.The main research conclusions are as follows:(1)Based on the actual soil physical property data in Guangxi,the soil effective water content model AWC＿m in Guangxi is constructed by using the stepwise linear regression method,and the fitting determination coefficient of the model is 0.832.Using the soil moisture model of AWC＿Dai for verification,it is found that the correlation coefficient of the two models is 0.789,which proves that the AWC＿m model constructed in this research is suitable for Guangxi,and comes to the conclusion that the average AWC content of red soil,latosol and lateritic red soil in Guangxi is18.38%,19.75% and 19.3%.(2)Combining the previously constructed soil water content model AWC＿m with PDSI and sc＿PDSI models,the AWC＿m-PDSI and AWC＿m-sc＿PDSI models integrating soil physical properties are compared to retrieve the dry and wet changes in Guangxi.It is found that the inversion results of AWC＿m-sc＿PDSI model can better reflect the dry and wet conditions of soil in Guangxi in terms of distribution area and dry and wet reflection degree,and the inversion effect is better than AWC＿m-PDSI model.(3)The soil moisture retrieved by AWC＿m-sc＿PDSI model and SPI model is consistent with the actual situation,and the inversion results of the two models are verified by using ERA-5 reanalysis soil relative humidity data.It is found that the highest correlation between AWC＿m-sc＿PDSI model and soil relative humidity is 0.753,and the highest correlation between SPI model and soil relative humidity is only 0.507,The results show that the monthly and annual soil water retrieved by AWC＿msc＿PDSI model considering soil physical properties has a higher correlation with soil relative humidity products,and the average correlation coefficient is 0.325 higher than that of SPI model.(4)According to the soil moisture inversion,from 2015 to 2019,Guangxi gradually became humid,the areas with drought decreased,and the degree of humidity increased from north to south.