| Soil moisture is an important part of land surface water cycle of the ecosystem,can influence the growth of soil properties and vegetation,which have an impact on soil erosion and soil erosion.Remote sensing quantitative retrieval of soil moisture,soil moisture distribution information to facilitate access to a large area in a short period of time,to analyze the temporal and spatial distribution characteristics of soil moisture,soil and water conservation work,the construction of ecological environment and the sustainable utilization of water resources to provide a theoretical basis and guidance.Polarimetric synthetic aperture radar(Polarimetric SAR PolSAR)has all day long and all-weather,the ground has certain penetrability,geometric characteristics and surface soil moisture sensitive and other advantages,can record the polarimetric scattering information of targets completely,so as to improve the classification accuracy.However,the vegetation volume scattering and surface roughness effects on the scattering coefficient of radar,reduces the surface soil moisture inversion accuracy,application limits of soil moisture inversion model.In this paper,Dingxing County of Hebei city in Baoding Province as the study area,the area in 2013 four months 3~6 Radarsat-2 polarimetric SAR image in 2013 and 4~6 three month Landsat-8 quasi synchronous optical remote sensing image,and the vegetation parameters,soil moisture,soil moisture,surface roughness parameters of synchronous field measurements.With the help of ENVI5.3,ArcGIS10.1,MATLAB 2004 a and SPSS software platform,the related processing of experimental data acquisition.Then select 70% sample data as training set,the forward parameter acquisition model fitted values,with the remaining 30% of the sample data as a validation set,to verify the accuracy of the inversion model.The last generation of the temporal and spatial distribution map of study area of soil moisture,and its spatial and temporal characteristics of soil moisture change analysis.The main research contents and conclusions are as follows:(1)Firstly,the water cloud model is used to correct the effect of backscattering from the vegetation,and then based on Chen model,using the polarization ratio(HH / VV)to eliminate the roughness of the surface with the rear surface of the scattering coefficient can be avoided input parameters of surface roughness,thereby reducing the workload of field measurements,but also to better fit the inversion of the surface covered with soil moisture winter wheat,and the measured value at about 75%,indicating that the model is able to meet the area soils moisture retrieval accuracy requirements;(2)Herein,respectively NDWI,NDVI and RVI vegetation water content VWC parameters are calculated to eliminate the influence of backscatter coefficients of vegetation,humidity inversion model obtained bare surface of the soil,and then to evaluate the accuracy of the retrieval,analysis of three vegetation layer by the effect of eliminating the influence of radar backscatter coefficients,the experimental results show that,compared with the single use soil moisture inversion SAR data,multi-source remote sensing data can be combined to achieve better results in inversion,so as to increase the accuracy of soil moisture inversion of vegetation cover surface;(3)Based on the vegetation index,Moran I index and the global distribution of temporal and spatial analysis of soil moisture,soil moisture has found in the study area and spatial aggregation characteristics significantly positive correlation,and with the passage of time,the winter wheat coverage will increase,coupled with the increase of rainfall,the soil moisture content will be increased. |
PDF Full Text Request