Water resources are considered as one of the basic foundations of sustainable development, and in addition to quantity, water quality is also considered as one of the important parameters.Underground water sources are one of the best and in some cases the only solution to the problem of providing drinking and agricultural water in arid and semi-arid areas such Iran. In recent years, the sharp drop in the water table level due to the increase in the exploitation of underground water sources, the increase in salinity and the decrease in the quality of underground water due to the penetration and mixing of pollutants, which are generally aggravated by human activities and industrial growth, have become one of the problems in the world (i.e. serious environment challenges). In this regard, water quality parameters are among the components that must be accurately predicted and simulated in planning. By estimating and modeling the qualitative parameters of underground water, in addition to managing the exploitation of water resources, natural disasters such as drought can also be controlled. Electrical conductivity parameter (EC) is also one of the main parameters in water quality monitoring in terms of drinking and agriculture. This parameter has a direct relationship with water salinity, sodium absorption and drinking water quality, and is of particular importance in soil management and stability. Qorveh- Dehgolan plain in the east of Kurdistan province is one of the most important sources of water supply for various parts of its neighboring areas. It is necessary to estimate and model the groundwater quality of this region. Nowadays, data driven methods with adaptability shown satisfactory results in modeling complex nonlinear systems in water resources management issues. Therefore, the aim of the current research is to apply GEP and M5Tree methods and compare its results with the artificial neural network method in estimating EC groundwater quality parameters, as well as using this method for managerial decisions, the reliability of monitoring results, and the reduction of related costs.
