Applying ANN and GIS for estimation of effective parameters in determination of plant pattern (Case Study: Nahavand City)

One of the most important issues in irrigated agriculture is determination of optimum lant pattern. Therefore، estimation of effective parameters in quality and quantity of available water is significant and is one of the most important components in adoption of management decisions in development of sustainable agriculture. In this study، Artificial Neural Networks technique has been used for estimation of piezometer wells water level and also effective factors for water quality used in agriculture (EC، SAR). For this purpose، monthly recorded data for piezometer wells water level during a seven year and data related with water quality during a four years period in Nahavand plain were used. Also، a groundwater level in Nahavand in year of 1385-86 was drawn. Efficiency of model was evaluated by statistical criteria including coefficient of determination (R2)، root mean square error (RMSE) and mean absolute error (MAE). The derived results showed that R2 value for estimation of piezometer wells water level is 0. 98 and for SAR and EC is 0. 991 and 0. 990 respectively. The above results indicated the appropriate ability of Artificial Neural Networks as superior technique for simulation of effective quality and quantity parameters in determination of plant pattern. Also the results from spatial drowning of groundwater level by Geographic Information System indicated the shortage of water resource in this region