Estimating Soil Moisture in Semi-Arid Region by Remote Sensing Based on TM Data

The drought is the most common natural disaster in the world, and the soil moisture content is one of the main indicators of drought monitoring. This study took the semi-arid region in northern Henan province as the study area and used Landsat-TM of 2008 remote image to set up the empirical model for getting the reflectance of the earth surface. The influence of vegetation on the pixel reflectance was eliminated by optical vegetation coverage model. Quantitative analysis model between remote data and ground measured data was established using spectroscopy which is based on the visible and near-infrared. And then we inversed the soil moisture content of surface ground and produced distribution maps of soil moisture. The accuracy of the retrieved results was validated by in situ data. The results showed that the theoretical and practical precision were 82.07% and 91.69%, respectively, it was suggested that the remote sensing quantitative retrieval of soil moisture content was scientifically available in semi-arid regions. All these researches can provide a basis support for the spatial distribution regulation of soil moisture, ecological protection and restoration of water resources.