Simultaneous Retrieval Of Land Surface Temperature And Soil Moisture Using Passive Microwave Observations

Land surface temperature(LST)and soil moisture(SM)are two important parameters in land surface ecosystem at regional and global scale.The accurate acquisition of LST and SM can benefit various fields,including agriculture and climate which are closely related to human life.The development of remote sensing technique enables the measurement of LST expanding from point to regional measurements.Currently,the acquisition of LST and SM using remote sensing technique majorly relies on thermal infrared(TIR)and microwave measurements.Compared to TIR signals,microwave signals are penetrative to clouds,mist,snow,rain,vegetation and some other arid surface objects.The capability of acquiring land surface information in all-weather conditions makes microwave technique a primary method to obtain land surface features in cloudy and rainy conditions.Current LST and SM retrieval methods are mutually exclusive and highly depending on auxiliary data,leading to a serious problem of low accuracy.To solve this problem,this dissertation fully utilized multiple channel and dual polarization microwave data,built a simultaneous retrieval model of LST and SM using passive microwave data based on microwave radiative transfer model.The goal of this model is to barely rely on auxiliary data and improve the retrieval accuracy of LST and SM.Main contents and conclusions of this dissertation includes:(1)This research generated a simultaneous retrieval model of LST and SM based on passive microwave radiative transfer model by solving the underdetermination problem of retrieval equations by means of parameterization of emissivity and combination of unknowns(vegetation transmittance and land surface roughness).To further improve the accuracy of the model,this research made correction to the items of equations.Compared to LST and SM in simulated database,the accuracy of retrieved LST is approximately 1.63 K and the accuracy of retrieved SM is approximately 0.063 m3/m3.(2)This research estimated global LST and SM using AMSR-E passive microwave brightness temperature data and validated their accuracy using products and in-situ measurements.In cloud-free conditions,the comparison of MODIS LST product and retrieved LST shows an overall accuracy of 5.68 K in terms of RMSE,which is positively related to vegetation coverage.In cloudy conditions,the comparison of ISD air temperature and retrieved LST indicates a vegetation cover irrelevant accuracy of 4.29 K.Compared to ESA CCI SM product and SM observation network,the accuracy of retrieved SM varies from 0.0157 m3/m3 to 0.1152 m3/m3,which is negatively related to vegetation coverage.Therefore,the simultaneous retrieval model obtains a favorable accuracy under moderate vegetation cover.(3)This research estimated global daily average LST during 2003-2010 using the global estimated LST at 1:30 am and 1:30 pm,and further estimated annual average LST,annual amplitude and annual phase using annual temperature cycle model.The research revealed the spatial distribution of LST by analyzing the correlation between three annual cycle parameters and elevation,land cover type,climate type and other factors.This research also uncovered the annual variation tendency of LST using Theil-Sen slop and Mann-Kendall significance test method,highlighted the regions in which the land surface parameters varies significantly and analyzed the primary cause.