Zoning Electrical Conductivity and Acidity of Groundwater through Using Geo-statistical Method: A Case Study in Semirom Plain, Esfahan Province

The groundwater quality research is one of the important and its pollution control was included insome research literatures. Ground water quality has spatial and temporal variation so classical statisticscould not account these variations at the regional scale researches. This study usedgeo-statisticalmethodsto optimize an interpolation method in order to estimate the spatial distribution of pH andelectrical conductivity in ground water. The geo-statistical methods which used in this procedure,includedkriging, ordinary kriging, simple kriging, disjunctive kriging, inverse distance weighting andradial. Cross validation was used to evaluate fault detection, root mean square error for statisticalcomparisons,and the geo-statistical analysis was performed in ArcGIS9x software environment. Thecase study was Semirom plain, Esfahan and historical data was collected from 386 springs in yearsfrom 2006 to 2007.The resultsof model validations showed that the variogram spherical model has thebest fit to the spatial data structure of the electrical conductivity and pH. The analysis of RMSE andMAE showed that inverse distance method (with raised to the power of one); RMSE = 0.065; MAE=0.041) and radial function (with RMSE = 3.57; MAE=2.27) were more statistically accurate or havelower RMSE and MAE in comparison to the other methods. Both of these methods have optimumspatial distribution of the pH and electrical conductivity of ground water inSemirom plain. Finally,spatial distributions maps of pH and electrical conductivity was created using ArcGIS9.x (orgeographical information systems). The resultsof maps showed reduction of both variables from theeast to west, in Semirom plain.