Retrieval And Application Of Soil Moisture In The Source Region Of The Yellow River Based On RS&GIS

Ecological degradation,such as land desertification,lakes and wetlands shrinking or "black-soil patch",has occurred in the source region of the Yellow River from 1980s.As an essential index of monitoring vegetation degradation and the main parameter in the fields such as climate,hydrological,global changing et al,soil moisture plays an important role in the material and energy exchange between the land surface and atmosphere.Remote sensing,a convenient method which can continuously observe the earth,is of great importance for soil moisture monitor in large scale.The thesis aims at constructing quantitative remote sensing inversion models which are suitable for the soil moisture monitoring in the source region of the Yellow River,by using the temperature vegetation drought index and"water-cloud" model.Based on these inversion results,analysis of soil moisture spatial heterogeneity of the study area is carried out with geostatistics,and a brief discussion is made about the relationship between soil moisture content and various environmental factors.The main conclusions are as following:(1)The quality testing of raw remote sensing image is conducted by building time-series data sets of MOD11A2 and MOD13A2 during 2009 to 2013.Data reconstruction with Savitzky-Golay filter effectively reduces the impact of clouds,water vapor and other noise on the data sets quality and fills in the missing values of raw image.And then,it's effective to further improve data quality that conversion from 8-day composite data to 16-day composite data for MOD11A2 with maximum value composites method.(2)Before dry&wet edges fitting process in TS-NDVI feature space,it's carried out that comprehensive analysis that the relationship between the vegetation index(MOD 13 A2),surface temperature(MOD 11A2)and soil moisture,the influences of low vegetation coverage rate upon linear fitting results.Then,we elaborate the reasons that fitted wet and dry edges show similar contours at different times,and preliminary analyze variation of the slope and intercept of dry&wet edges throughout vegetation growing season.(3)By using TVDI mean images collected from May to September in 2013,this research studies soil moisture heterogeneity of the study area with geostatistical method and analyzes the influences of sampling interval upon spatial autocorrelation and trends of soil moisture content during vegetation growing season.(4)Under Geographic Information System(GIS)technology and related analysis method,by using the topographic factors(attitude,slope and aspect),soil type,vegetation cover type and other data,this study reveals the relationship between soil moisture and these environmental factors.It also analyzes the influence of various soil&vegetation types,which are classified by different grades of soil moisture,in the soil moisture storage and transportation.(5)With the help of "water-cloud" model,which eliminates the effect of the vegetation in the whole radar backscattering by using ENVISAT-ASAR data,the relationship between backscattering coefficient and soil water content has been improved in this research.