In collaboration with Iranian Watershed Management Association

Document Type : Research Paper

Authors

1 Assistant Professor, Maybod Branch, Islamic Azad University, Maybod, Iran

2 MSc Student, Maybod Branch, Islamic Azad University, Maybod, Iran

Abstract

This research aims to study temporal and spatial variations of groundwater quality of Marvast Plain to manage these recourses. For this purpose, total anion and cation, a percentage of sodium, Ca, Cl, EC, HCO3, K, Mg, Na, pH, SRA, So4, TH parameters were prepared and a statistical period of 14 years (1377_1390) was selected using the evaluation of incomplete data. Semivariograms and cross semivariograms were potted to perform geostatistcal methods and the best model was selected with regard to more R2, less RSS and a strong spatial structure, and an exponential model was fitted for most parameters. Then, interpolation methods, such as Radial Basis Function (RBF), Inverse Distance Weighting with powers of 1 and 2 (IDW), Kriging and coKriging were used. The most accurate method was selected for parameters using cross validation method and error estimation indexes such as Mean Absolute Error (MAE), Mean Bias Error (MBE) and Root Mean Square Error (RMSE). The results showed that coKriging is the best method for estimating most parameters (total anion and cation, EC, Cl, Ca, K, Mg, Na, TH). Kriging is defined as the most suitable method for estimating SRA, So4 and pH, and RBF and IDW with power 1 are the most fitted methods for a percentage of Sodium and HCO3, respectively. The results showed that the density of most parameters increases from west to east and south to north of the study area. Also, decreasing underground water quality is observed over the time.

Keywords