Assessment of groundwater vulnerability and nitrate contamination risk using GIS-based drastic model with hybrid statistical and probabilistic techniques

Groundwater pollution is one of the most significant environmental problems today. It is caused by human activities, especially agricultural activities. Agricultural systemsdeveloped from traditional methods to modern applications, resulting in an overuse of chemical fertilizers that increase the amount of pollutants. Fertilizers such as nitrates play a significant role in water and soil pollution because of their special characteristics. Most of these fertilizers enter the groundwater through surplus water and create high-risk groundwater resources. Therefore, identifying and diagnosing the amount of pollutants using a groundwater risk map in the future can largely prevent more pollution in groundwater resources. Efficient preventive programs, such as risk management, should be implemented to reduce the risks of groundwater pollution. In this research, the DRASTIC approach based on a geographic information system (GIS) was applied to evaluate groundwater vulnerability in Kerman Plain (Iran), an arid and semi-arid region that encounters intensive agricultural activities and over exploitation of land that has resulted in groundwater contamination. DRASTIC model includes seven parameters of depth to water (D), net recharge (R), aquifer media (A),soil media (S), topography (T), impact of vadose zone (I), and hydraulic conductivity (C) of the Kerman Plain. The original DRASTIC model was applied and integrated using original rates and weights. The generation of groundwater vulnerability map was performed by optimizing the rates and weights of DRASTIC model using GIS modeling techniques. The models used were analytical hierarchy process (AHP),single parameter sensitivity analysis (SPSA), frequency ratio (FR), and Wilcoxon nonparametric model. The optimized rates and weights were computed using each model.The Wilcoxon non-parametric test and FR analysis were applied to optimize the rates of DRASTIC model. AHP method was also used to optimize both the rates and the weights of DRASTIC model, and sensitivity analysis was conducted to optimize only the weights of DRASTIC model. These methods were assigned to DRASTIC model and integrated to produce hybrid methods. So far, some of the generated hybrid methods using the abovementioned models have not been applied in other studies. The most proper optimization of the vulnerability map was determined using Pearson’s coefficient correlation. The Pearson’s correlation value of each modified DRASTIC model used in this study was calculated. The regression coefficients showed the relationship between each vulnerability index and the nitrate concentration. The regression coefficient of DRASTIC model indicated a correlation of 0.37. The combination of Wilcoxon non-parametric test for rates and the sensitivity analysis for weights revealed the highest correlation of 0.87 among all applied hybrid models The most appropriate groundwater vulnerability map with the highest validity and accuracy was selected and combined with the nitrate pollution map that indicates theamount of damage in the Kerman Plain. Then, Dempster–Shafer theory (DST) was applied to develop a new methodology for assessing pollution risk. DST method provides a major advantage by dealing with the varying levels of precision related to information and the more generalized form of probability theory. The combination of the damage map and the pollution occurrence probability map through DST method produces a novel method that can determine the groundwater risk map for the nitrate parameter. The application of risk assessment method is recommended if the objective is to develop a risk map of areas that are vulnerable to pollution. Aside from nitrate, other pollutants can also be identified in other regions. Therefore, analyses are urged to search for other factors that lead to the pollution of groundwater resources.

Preview PDF FK 2014 30R.pdf Download (946kB) | Preview

Actions (login required)

View Item