Predicting the effects of climate change on groundwater resources using artificial intelligence methods (Case study: Talesh plain)

Due to the increase in greenhouse gases and numerous water and climate crises, accurate prediction of  the changes groundwater levels is very important and vital in the  water resources management. Therefore, in this paper,the  climate changes of Talesh plain is studied under RCP scenarios using Lars-WG and its water sources from SVR and ANN models. Also,aquifer pumping parameters, evapotranspiration potential, minimum and maximum temperature and  precipitation are used from (2021-2030). The results of the mean minimum and maximum temperature changes under RCP scenarios indicate the temperature increase by 0.9 and 0.69 °C. Also,studying  the accuracy of SVR and ANN models shows that the AUC in the training and testing phase in the ANN model, the maximum AUC values ​​were calculated as 0.876 and 0.769, while the SVR model, the maximum values ​​were equal to 0.867 and 0.819.Thus SVR has better predictive accuracy.In addition to that  during the time period (2005-2019) the groundwater level has decreased by 10 cm and in the SVR and ANN models by an average nine and six cm respectively more ever during in the time period(2021-2030) ground water levels have decreased in by 18, 20 and 21 cm, 20, 21 and 23 cm under the scenarios RCP2.6, RCP4.5 and RCP8 5 in SVR and ANN models,respectively.Therefor it is suggested that in Talesh plain considering the cultivation pattern appropriate to water resources in different parts of the plain should be the  priority for agricultural planners.