Evaluation of Support Vector Machine, and Linear and Quadratic Discriminant Analysis for Groundwater Level Variations in Shahrekord Plain

In recent years, due to the increasing rate of water demand and severe droughts, groundwater resources are considered as the most important sources of fresh water. Accordingly, a comprehensive strategy along with a long term plan is needed for preventing groundwater destruction. Variations in aquifer water level, are amongst the main factors which provide correct judgment about groundwater status and govern the watershed management projects. In the present study, monthly data (1999 to 2009) from 33 observational wells in Shahrekord Plain have been used for simulating the groundwater level. The relationship among the Shahrekord Plain coordinates and the groundwater level variations, for 1, 3, 5 and 10 year period, were investigated using Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), and Support Vector Machine (SVM). The results showed that the SVM is superior to the other two models due to its lowest average relative error in 1 and 3 year periods, and its acceptable precision in 5 and 10 year periods.