Retrieving The Soil Moisture Based On TM Data In Yan River Basin

Soil moisture is an important part of the soil, and, is an important factor to impact plant and crop growth, and, runoff. Monitoring soil moisture by remote sensing can reflect the distribution in space and time changes. It is practical and scientific to research the methods to monitor soil moisture.In this paper, after analyze various studies about soil moisture retrieval and compare different quantitative retrieval models, we choose the Temperature Vegetation Dryness Index to monitor soil moisture of Yan River Basin. This method combines the vegetation index (NDVI) and land surface temperature (LST) to describe the Earth's surface features, which can clearly express vegetation cover and soil moisture content change information, and contribute to accurately recognizing dynamic evolution of soil moisture.Firstly, based on the spectral features of vegetation, we estimate the normalized difference vegetation index model and calculate NDVI distribution. The author selects the single-window algorithm for retrieving surface temperature basis to four laws of thermal radiation, and gets brightness temperature, surface emissivity, transmissivity and the average atmospheric temperature of the study area. Then, using the method of air temperature lapse and solar radiation to get temperature model of the actual terrain, the results show that the air temperature distribution not only reflects the decreasing trend of temperature with the increase of altitude, but also makes the temperature distribution more precise by radiative correction. The author use space modeling tools to estimate land surface temperature retrieval model and obtain surface temperature distribution, which the results show that the low LST distribute in water and high vegetation cover area and the high LST in rocks and bare soil area.This study chooses TVDI to monitor the soil moisture basis to Ts-NDVI feature space theory. By fitting the dry edge and the wet edge of feature space, we obtain TVDI distribution and do the drought classification. The results show that the TVDI distribution have the consistent trend with terrain. From the macro aspect, the TVDI is in accordance with the underlying surface. TVDI expresses higher value in the low vegetation coverage and bare rock areas, and lower value in the high vegetation coverage area. To analysis the scatter map between TVDI and Ts, NDVI, we see Ts is more indicative than NDVI for monitoring soil moisture.