Introduction
Nowadays, the continuous reduction of GWQ is considered as one of the main challenges in the world. In developing countries - especially which located in arid and semi-arid regions - like Iran, the increase in emerging pollutants, rapid population growth and development activities, have led to increase in water demand. Due to the complexity of protecting GW resources for future generations, while meeting the needs of many economic activities, GWQ has become an important issue. Therefore, because the quality of consumed water is directly related to the quality of life, it is very important to study the quality of GW resources, as the main source of water supply. Water quality assessment leads to facilitating and improving the management of water resources for various uses. In recent years, the Salmas aquifer has seen a significant increase in water demand due to the increase in agriculture and the development of orchards, the increase in the extraction of GW resources, climate changes and etc. In this study, an attempt was made to determine the quality of the Salmas aquifer and mapping the areas with pollution potential by using Shannon's entropy.
Methodology
To carry out the current research, the data was received from the regional water company of West Azerbaijan province (RWCWA). Then, some statistical checks were done. In next step, the trend of water level of Salmas aquifer (from 1975-76 to 2020-2021) were assessed. The hydrograph and quality indices of this aquifer (salinity hazard, SAR, magnesium hazard, Residual Sodium Carbonate, Na%, Kelly Index, permeability Index, as well as Piper, Schuler, Durov, Stiff diagrams and WQI) were also calculated (data from 2001 to 2021) by using Excel and AQQA. The environmental layers were also prepared by using ArcGIS 10.8. Finally, Shannon's entropy model was used in the MaxENT software to mapping the areas with pollution potential. The results of maximum entropy were evaluated by ROC curve.
Results and Discussion
According to the results, the general trend of water level changes in Salmas aquifer is decreasing and the annual average of these changes is 38 cm loss and the average change of aquifer volume is estimated to be 4.54 mcm loss. The water type of this aquifer was estimated as Mg-HCO3, with high salinity hazard and magnesium hazard, along with a good quality according to PI, KI, Na% ESR and RSC indices. According to the results of WQI, the groundwater quality of this area was identified as good quality at 81.3%. Based on the sensitivity analysis of the factors affecting the quality of the Salmas aquifer by using the Maxent, NDVI, plan curvature and distance from the river showed the greatest impact and soil texture showed the least impact on vulnerability. Based on the results of the maximum entropy, 23.48% of the aquifers were estimated to have moderate to very high vulnerability potential, which includes rangelands, Agri farms, and orchards that are located on Qt2 and Qf sediments and have a flat surfaces or low slope with clay and clay-loam texture. According to the ROC curve, the accuracy of this estimate was 80% and at the "very good" level.
Conclusion
It is recommended to prevent the creation or increase of pollutants (such as the development of industries, municipal waste disposal, sewage wells, etc.) in areas where the probability of vulnerability is high and very high. In order to increase the accuracy of the results, must be attention in determining the sampling places and environmental layers, because the model is trained from what is defined as input (occurrence/presence data) and determines the prediction algorithm and makes the prediction. It is recommended that this research be investigated using more detailed information such as nitrate and nitrite, and also with other models.