Abstarct: Most measureable important factors in water and soil conservation and hydrology are soil volumetric moisture and plants water available. Thus measuring these parameters is very essential. Regard to the difficulties of direct measurement of soil volumetric moisture, using indirect methods for estimation of this parameter, has received attention in recent years. In this study use soil parameters such as bulk density, porosity, organic matter and day in three soil texture (clay, silt and sand) to estimate soil volumetric moisture with artificial neural network and its result compared with nonlinear regression model and TDR method. moreover time variation of volumetric moisture in soil textures had survived. In order to evaluate the performance of these models, the statistical parameters such as root mean square error (RMSE), mean absolute erroe (MAE) and correlation coefficient (R2) were used. The result show artificial neural network with R2 (0.97, 0.99, 0.97), RMSE (0.860, 0.401, 0785) and MAE (0.685, 0.312, 0.551) for test stage of sand, silt and clay texture in estimating soil moisture volumetric was better than nonlinear regression model and TDR method, respectively.