Abstract: Qualitative and quantitative parameters of underground water resources could be studied by taking discrete samples at different locations in study area. In this way, for change procedure study of mentioned parameters and to gain a deeper understanding of the phenomena, a continues surface need to be generated. The main purpose of Geo-Statistical analysis is creating or interpolating a continuous surface from discrete sample points; in the other hand, Geo-statistical Analyst, derives a surface using values from the measured sampling points to predict values for each location in the landscape. The ultimate goal is to produce a surface of predicted target values. In Geo-Statistics, there are two main categories for interpolation: deterministic and geostatistical. The first category, based on the similarity and distance from measured points (Inverse Distance Weighting, Global Polynomial Interpolation, and Local Polynomial Interpolation) or the degree of smoothing (Radial Basis Functions); and second category employs some statistical properties of observed samples (Kriging and Co-Kriging). This paper utilizes Geo-statistics for the estimation of underground water table at location of interest where measured values are not available. In this regard, by employing Kriging, IDW and RBF procedures, underground water table contours have been created. For best estimator selection, results have been validated by some statistical indices such as Mean Absolute Error (MAE), Mean Absolute Relative Error (MARE), Root Mean Square Error (RMSE), Mean Biased Error (MBE) and Coefficient of Correlation. Absolute relative error variation curve for the number of sample points used for examination of efficiency and accuracy of each Geo-Statistical estimators. Eventually, Kriging method has been selected as the best method to estimate the underground water table of Razan-Ghahavand plain. In order to Kriging estimations, direction of underground flow, which is the major relevant issue to Hydro-Geologic studies, has been gained.