Evaluation of OLS regression models and GWR regression for modeling Spatial Soil Moisture, Case Study: Fars Province

spatial modeling is one of the method for understanding and predicting environmental variables .soil surface moisture is a key variable for drought description , water and energy exchanges between the earth and the earth .soil moisture can affect many environmental phenomena such as runoff , soil erosion and crop production , but because of the lack of spatial and temporal conditions of soil parameters , soil moisture is highly changeable .the purpose of this paper is to evaluate the overall regression model and geographically weighted regression in spatial modeling of soil moisture in Fars province . soil moisture distribution as dependent variable and precipitation, snow equivalent water, vegetation index and topographic wetness index were selected as independent variables and then, using the general regression model and geographically weighted regression is used to model the spatial modeling .based on the evaluation criteria, the results showed the GWR model has better explanatory power with the R2=0/71 and a better estimate than the overall regression model with the R2=0/66. The spatial factors of precipitation and topographic wetness have the most positive effect and evapotranspiration had negative effect on soil moisture in the study area.The spatial factors of precipitation and topographic wetness have the most positive effect and evapotranspiration had negative effect on soil moisture in the study area.