Assessing the accuracy of algebraic and geostatistical techniques to determine the spatial variations of groundwater quality in Boroojen Plain

Abstract

Population growth and increasing use of water resources for agriculture, industry, and sanitary sectors are the main causes of water degradation and groundwater pollution. Due to over-extraction of groundwater resources in the fertile plain of Borujen in recent years, this region has been recognized as a prohibitive plain. The main objective of this study is investigation of groundwater quality in Borujen plain aquifer using Algebraic analysis and Geostatistical methods. In this regard, the data from 19 sampling wells were used to study the groundwater quality of Borujen Plain during the period of 1989-2013. Spatial variations of water quality parameters were evaluated using algebraic methods (IDW and RBF) and Kriging. The ME and RMSE statistical measures were used to evaluate the spatial interpolation results using cross validation method, and the appropriate method was selected for zoning the different ground water quality parameters. The results of spatial interpolation of groundwater quality parameters showed that the Kriging method was the appropriate approach (RMSE = 30.72 and ME = 0.09) to estimate the Total Hardness parameter (TH) in the Borujen Plain. Also, the pH, SAR, Cl and SO4 parameters are well estimated using the RBF method with RMSE, ME of 0.05, 0.33, 0.27 and 0.4. While the IDW recognized as the best interpolation method for EC and TDS parameter with 84.62 and 56.36 RMSE values. In conclusion, the comparison of prepared maps of TH, EC, Chlorine ion concentration, Sulfate ion concentration and SAR, parameters showed a similar pattern and the minimum and maximum values of all parameters is located in in the common areas of the area under study.

Keywords

• groundwater
• quality changes
• geostatistical methods
• algebriac methods
• spatial changes pattern