Determining the Most Appropriate Geostatistical Method for Groundwater Salinity Mapping (Case Study: Shiramin Plain, East Azerbaijan Province)

In all of the world and including our country, groundwater is the most important resource for water supply needed. Determination of the quality of water has specific importance in the management of resources and its monitoring should be considered as an important principle in planning. The aim of this research is of the determination of the most appropriate interpolation method for spatial analysis of electric conductivity and sodium absorption ratio changes of groundwater of Shiramin plain located in East Azerbaijan province. The related information are obtained from analysis of the samples involving 30 deep and semi-deep wells based on the sampling of the year 1390 by regional water organization of East Azerbaijan province. The geostatistics methods such as simple, ordinary and discrete Kriging and also CoKriging were used for doing this research. For investigation of the spatial correlation of data, the experimental variograms of each variables and its reciprocal variogram were calculated and plotted. The correlation coefficient of two variables were calculated at 0.93 according to their related variogram. The spherical model was fitted as the most appropriate model for both of the EC and SAR parameters based on the minimum of RSS. The reciprocal estimating method with root mean square error criterion and also correlation coefficient (R) between the observed and estimated values was used for assessment of the reliabality of methods. The results of research showed that because of higher R and lower RMSE, the simple Kriging method was the most appropriate method within the others for preparation of the map of SAR and EC changes.