Investigating geostatistical techniques in predicting some groundwater quality parameters, case study: Arsanjan plain

Increasing and preserving water quality requires sufficient data of the location, value as well as spatial distribution of the chemical factors. Preparation of the updated maps of groundwater salinity variation and salt concentration is an important step for appropriate management and its exploitation. Selection an accurate method for zonation and mapping groundwater quality parameters variation depends on regional condition and availability of data which is an important stage for groundwater management. The purpose of this research is evaluation of geostatistical techniques accuracy for determination of EC, NO3, pH and TDS of groundwater and mapping them using geographic information systems in Arsanjan plain, northwest of Fars province. For this aim, different techniques of geostatistics including Ordinary Kriging (OK) and Simple Kriging (SK) as well as deterministic techniques such as Inverse Distance Weights (IDW), Radial Basis Function (RBF), local polynomial interpolation (LPI) and Global Polynomial Interpolation (GPI) were used. The result based on lowest root mean squared error (RMSe) and Mean Absolute Error (MAE) showed that Simple Kriging (SK) technique is the most appropriate technique for salinity (EC) mapping (RMSe=724.8 and MAE=513.41) and TDS mapping (RMSe=463.9 and MAE=328.58) among all techniques in the study area. Also the results revealed that OK with RMSe=0.197 and MAE=0.139 for pH parameter and RMSe=7.92 and MAE=6.29 for nitrate parameter has the highest priority among other techniques and is selected for pH and nitrate variation mapping. Finally, the results of this research indicated that Geostatistical techniques have higher priority than deterministic techniques and thus are selected as suitable methods in Arsanjan plain.