Simulation of Ground Water Salinity by Using Artificial Neural Network(ANN) on the Mazandaran province coasts

Groundwater is one of the most important water resources and its qualitative study is very important for water resources protection and planning. Also, qualitative parameters measuring of the ground water is costly and time consumer. Models will be reduced the cost of water qualitative estimations and it will be provided a data bank to manage water resources. In this study, Artificial Neural Network (ANN) has been used to simulate ground water salinity on the Caspian southern coasts (MazandaranProvince). So, a comprehensive data bank of water qualitative experiments has been provided and then quantitative values of the effective factors on ground water salinity, such as: Water conductivity of aquifer formation, surface water salinity, slope and elevation (topography) and the distance from Caspian Sea were estimated. The efficiency of Artificial Neural Network (ANN) has been considered through two parameters: Median Root of Square of the Error (RMSE) and co efficiency between the actual and desirable outputs (R). The results showed that co efficiency between the actual and desirable outputs (R) is 0.75 in the best network structure. surface water salinity, water conductivity of aquifer formation and distance from Sea are the best inputs for simulating ground water salinity and those are the most important factors on ground water salinity in Mazandaran Province Coasts.