spatial resolution. support vector classification

Soil moisture is an important variable in climatic, hydrological, and ecological systems that link atmospheric processes to the earth's surface. The rainfall systems of the water year 2017-2018 abnormally caused more than normal rainfall in Fars province and caused major changes in the surface moisture of the soil. The purpose of this research is to analyze the spatial changes of soil moisture before and after heavy spring rains in Fars province using Downscale RADAR images. Using the backscatter bands of VV and VH polarizations as well as the incident angle band (𝜃) extracted from Sentinel 1 radar images and land use extracted from the MODIS sensor, a training layer was created. Furthermore, using the support vector machine algorithm, the downscaling soil moisture map was obtained. The results showed that the volumetric soil moisture with high resolution is between 0.18 and 0.38 in the rainy year and between 0.12 and 0.24 in the long term. An anomaly map showed that between 0.14 and 0.18 m3 increased soil moisture. The positive anomaly is more in the east and south of the province, and less humid areas have experienced a greater share of positive anomaly. Moran's index statistic with a value of 0.99 has also confirmed the spatial autocorrelation of soil moisture anomaly and clustering of moisture increase. In general, it can be concluded that by using the results of this method, it is possible to monitor areas with low or high soil moisture anomalies after different rainfalls and to improve the decision-making process.