Research On The Retrieval Of Surface Soil Moisture Based On The SAR Data

Soil moisture is closely related to human life. Soil moisture is not only the basic conditions of crop growth and development, but also affect the formation, conversion and consumption of water resources. It plays an important role in the process of surface runoff. Study on soil moisture has a very important significance for the entire watershed runoff, drought mitigation, flood control, monitoring and early warning on wet waterlogging. Since soil moisture changes obviously along with time and space in the natural environment, it’s difficult to measure and monitor on it in large-scale and continuously with a high precision. Microwave Remote Sensing compared with other bands have more advantages, making it an effective way to monitor the area in the temporal and spatial variation of soil moisture. Due to the special nature of the vegetation cover surface, the backscatter coefficients of the rear canopy can’t link directly with the soil moisture. It is necessary to remove the effect of vegetation. Therefore, if we can consider the type of crop coefficient substituted into the model, removing the influence of vegetation on soil moisture, it will provide an important reference for many methods that study on soil moisture of vegetation cover surface.This paper chooses Jiangsu Province as the study area. Firstly, using the ASAR GM data and MODIS data, calculate the relevant experience coefficient of water-cloud model related with the type of crop based on soaker situation. Secondly, establish a suitable water-cloud model for study area in Jiangsu Province to calculate the soil backscattering coefficient. Thirdly, by the relationship between the backscattering coefficient and soil moisture, we can obtain the relative humidity of the soil surface. Then, use the measured data of automatic soil moisture observation stations and the product of humidity data provided by AMSR-E separately to do the correlation analysis and accuracy assessment on the relative humidity of the soil surface. Finally, respectively from time scale and spatial scale, perform a spatial distribution and time changes analysis on the soil relative humidity distribution of the study area in Jiangsu Province.As research indicated, based on the formula of water-cloud model and in soaking rain situation, the results of relevant experience coefficient with type of crop have a certain reference value. The backscatter coefficient model also has a high precision established on this basis. After using the measured data of automatic soil moisture observation stations and the product of humidity data provided by AMSR-E to perform an accuracy assessment on the model, we found that the measured data of automatic soil moisture observation stations has a higher precision and it is more suitable for the accuracy assessment of study area in this paper. Apart from this, due to the low resolution of AMSR-E, it’s more suitable for precision verification on large-scale region. For smaller study areas, the work of using AMSR-E to verify still needs further study.