Total vulnerability estimation for the Tabriz aquifer (Iran) by combining a new model with DRASTIC

Groundwater risk assessment based on optimization framework using DRASTIC method

“…This layer was generated using DEM of the area. Impact of vadose zone is a layer located between groundwater table and unsaturated soil, which controls the passage of contaminant into the groundwater (Vaezihir and Tabarmayeh 2015). With respect to many physical and chemical reactions occurring in this zone such as mechanical filtration, volatilization, dispersion, biodegradation, and neutralization, this layer is potential for filtering the contaminants (Wu et al 2016).…”

“…Hydraulic conductivity controls contaminant migration and dispersion within the aquifer media and consequently, the concentration of the plume in the groundwater (Vaezihir and Tabarmayeh 2015). The hydraulic conductivity was calculated by the following equation: k = T b −1 , where k (m/d) is the hydraulic conductivity, T (m 2 /d) and b −1 (m) are transmissivity and thickness of the aquifer, respectively.…”

“…Process-based methods use simulation models to estimate contaminant migration, but these models are often restricted by inaccessible data and/or computational difficulties (Barbash and Resek 1996). Statistical methods as their name implies are based on statistics to determine associations and relationships between spatial variables and actual occurrence of pollutants in groundwater, which are accompanied by limitations such as insufficient observation data and necessity of careful selection of spatial variables (Vaezihir and Tabarmayeh 2015). Moreover, they are reliable only for specific regions and not suitable for the transfer to the other regions (Babiker et al 2005;Victorine Neh et al 2014).…”

“…For instance, SadatNoori and Ebrahimi (2015) used statistical methods for optimizing rating scores of the DRASTIC model based on point data chloride concentrations to reduce the subjectivity associated with assigning rates and weights to the DRASTIC model. The vulnerability indices have been correlated with nitrate concentrations and/or controlled with sensitivity analysis in most of the previous studies (Rupert 2001;Huan et al 2012;Neshat et al 2014a;Pacheco et al 2014;Ahmed et al 2015;Vaezihir and Tabarmayeh 2015). Recent researchers made alteration to rating scales to adjust the model by data interpolation (kriging method) using ArcGIS® (Ahmed et al 2015;Vaezihir and Tabarmayeh 2015;Sadat-Noori and Ebrahimi 2015).…”

“…The vulnerability indices have been correlated with nitrate concentrations and/or controlled with sensitivity analysis in most of the previous studies (Rupert 2001;Huan et al 2012;Neshat et al 2014a;Pacheco et al 2014;Ahmed et al 2015;Vaezihir and Tabarmayeh 2015). Recent researchers made alteration to rating scales to adjust the model by data interpolation (kriging method) using ArcGIS® (Ahmed et al 2015;Vaezihir and Tabarmayeh 2015;Sadat-Noori and Ebrahimi 2015). In addition, modifying the model with omitting or adding layers such as land use pattern to estimate contamination risk to pollutants mostly nitrate (Panagopoulos et al 2006;Bai et al 2012;Huan et al 2012;Neshat et al 2014a;Ahmed et al 2015;Kazakis and Voudouris 2015;Vaezihir and Tabarmayeh 2015) or applying analytic hierarchy process (AHP), a multiple criteria decision-making (MCDM) technique, to adjust weights of the DRASTIC parameters to local settings have become prevalent (Sener and Davraz 2013;Neshat and Pradhan 2014;Neshat et al 2014a).…”