deterministic interpolator

In this study, some interpolation methods were evaluated for estimating groundwater depth in the Iranshahr- Bampour Plain during 2003, 2007 and 2012 in the months of May and October. Data used belonged to 42- 48 wells scattered across the study area. The methods used contained geostatistical approaches of Ordinary Kriging (OK) and Universal Kriging (UK), and deterministic approaches of Inverse Distance Weighting (IDW), Radial Basis Function (RBF) and Local Polynomial Interpolation (LPI). The performance of the prediction methods was evaluated through cross-validation with comparison criteria of determination coefficient (R2), root mean square error (RMSE) and mean bias error (MBE). The statistical analysis showed a high variance and coefficient of variation of groundwater depth and an increase in the average depth to groundwater during bygone years especially during 2003-2007 period. Directional semivariograms were calculated to find out the drift direction. UK method, with the first and second-order polynomials as drift, and different semivariogram models was examined. According to cross-validation results, the best geostatistical method for estimating groundwater depth was OK (with spherical semivariogram) for 1382 and 1391, and UK with J-bessel semivariogram model and second and first drift orders, respectively, for May and October 2007. Moreover, the cross-validation results indicated that LPI, with RMSE equal to 6.94, 5.87 and 8.65 m, respectively, for May 2003, 2007 and 2012, and 6.86, 6.54 and 8.68 m, respectively, for October 2003, 2007 and 2012 is the best method of interpolation among others. The generated maps of groundwater depth revealed a drop in depth to groundwater; therefore, an occurrence of water crisis over the study region during the recent years. Therefore, it is necessary to consider some management scenarios including exploitation control and alteration of crop pattern and irrigation systems for an optimum use of water resources and achieving a sustainable agriculture across the region.