ارزیابی آسیب‌پذیری آبخوان دشت بیلوردی بر مبنای ترکیب روش‌های DRASTIC و SINTACS

Groundwater is the main source of fresh water in Bilverdi plain aquifer. The groundwater level declination in this aquifer has increased salinity and recently, more withdrawals have been forbidden. This aquifer exposes to high pollution risk due to presence of Vallilu arsenic mine. In this study the DRASTIC and SINTACS methods combined to assess Bilverdi aquifer vulnerability, The correlation coefficient between nitrate concentration in groundwater and vulnerability index  results showed that although SINTACS method have relatively better performance due to high correlation coefficient rather than drastic method, but selection of one methods as a best model do not seem reasonable due to similarity of the performances. Therefore, this study reaps advantages of both DRASTIC and SINTACS methods to evaluate vulnerability of Bilverdi plain aquifer. Based on this proposed method 36% of the area in plain center is located in low vulnerability area, 20% in the northeastern and southwest regions have high vulnerability and 43% of the area covered medium vulnerability. 
Introduction
Groundwater pollution due to increasing population, industrial and agricultural development need treatment which is costly and problematic. Therefore, The best and most important action is to identify vulnerable areas. Infiltration and release of pollutants from ground surface to groundwater can be defined as the concept of vulnerability. Bilverdi plain is one of the most active agricultural, livestock and industry sectors in which agricultural fertilizers are used. In addition, occurrence of Valilu arsenic mine and the probability of the impact of mine drainage on groundwater resources show the necessity of the aquifer vulnerability assessment. Therefore, the aim of this study was to investigate the vulnerability of Bilverdi plain aquifer to determine areas with high potential of groundwater contamination using DRASTIC and SINTACS methods. combination method was suggested to reap the advantages of both methods to assess the vulnerability of this aquifer.
Methodology
The Bilverdi sub-basin, 289 km2 in area, is located approx. 65 km of Tabriz city, Eest Azerbaijan, Iran. In this study, Bilverdi plain vulnerability was considered with a new combination approach. DRASTIC method is a quantitative model that is designed to estimate the potential of groundwater system pollution in a regional scale based on seven hydrological parameters including groundwater depth (D), recharge (R), aquifer media (A), soil (S), topography (T), impact of unsaturated zone (I) and the hydraulic conductivity (C). These layers are overlaid by weighting based on Aller et al. (1987) to generate vulnerability map.
The type and the number of parameters of the SINTACS method is also the same as to the DRASTIC method, but the rating and weighting of the parameters are different. Both methods have high predictive capability despite the similar correlation index. Choosing each of the methods as a superior method increases the uncertainty in the results. Therefore, in this study, for the simultaneous used efficiency of the both model, the combined model was presented to obtain accurate results.
 
Results and Discussion 
The vulnerability maps were obtained by overlaying seven parameters of SINTACS and DRASTIC methods in the Arc GIS environment. The DRASTIC index for Bilverdi plain is numerically between 53 and 160. According to Aller classification, three groups of vulnerability including low (53-97), medium (98-141) and high (160-414) corresponded to 13%, 82% and 5% of the area, respectively The SINTACS index is numerically between 70 and 169, which is divided into three groups: low (116-70), medium (117-164) and high (169-164) vulnerability and they are corresponded to 30%, 67% and 3% of the area, respectively.
Correlation index (CI) of both methods show the similarity of them. Therefore, normalized vulnerability index of both methods are weighted by their CI values to generate combination method. The vulnerability index of combination method was obtained from 0. 00005 to 0. 96, which was classified into three groups: low vulnerability (0.52-0.00005), medium (0.69-0.50) and high (0.96-0.96) which was corresponded to 36.5%, 43.5% and 20% of the area, respectively.
 
Conclusion
In this study, SINTACS and DRASTIC methods developed to assess groundwater vulnerability maps. According to results similarity of these methods, combination method adopted to reap the advantages of both methods. Presented method was indicated better results. In the Northeast and Southwest region of the Bilverdi plain, high vulnerability area located in the low groundwater depths areas and in the central areas of the plain, the vulnerability is low due to the presence of fine silt and clay grains. Advantages of both methods are adopted by combination method to cover disadvantages of each method to assess groundwater vulnerability in the Bilverdi plain aquifer.