A supervised committee machine artificial intelligent for improving DRASTIC method to assess groundwater contamination risk: a case study from Tabriz plain aquifer, Iran

Abstract

Vulnerability maps are designed to show areas of greatest potential for groundwater contamination on the basis of hydrogeological conditions and human impacts. The objective of this research is (1) to assess the groundwater vulnerability using DRASTIC method and (2) to improve the DRASTIC method for evaluation of groundwater contamination risk using AI methods, such as ANN, SFL, MFL, NF and SCMAI approaches. This optimization method is illustrated using a case study. For this purpose, DRASTIC model is developed using seven parameters. For validating the contamination risk assessment, a total of 243 groundwater samples were collected from different aquifer types of the study area to analyze NO À 3 concentration. To develop AI and CMAI models, 243 data points are divided in two sets; training and validation based on cross validation approach. The calculated vulnerability indices from the DRASTIC method are corrected by the NO À 3 data used in the training step. The input data of the AI models include seven parameters of DRASTIC method. However, the output is the corrected vulnerability index using NO À 3 concentration data from the study area, which is called groundwater contamination risk. In other words, there is some target value (known output) which is estimated by some formula from DRASTIC vulnerability and NO À 3 concentration values. After model training, the AI models are verified by the second NO À 3 concentration dataset. The results revealed that NF and SFL produced acceptable performance while ANN and MFL had poor prediction. A supervised committee machine artificial intelligent (SCMAI), which combines the results of individual AI models using a supervised artificial neural network, was developed for better prediction of vulnerability. The performance of SCMAI was also compared to those of the simple averaging and weighted averaging committee machine intelligent (CMI) methods. As a result, the SCMAI model produced reliable estimates of groundwater contamination risk.

Modification of the DRASTIC Framework for Mapping Groundwater Vulnerability Zones

“… where r and w represent the rating and weight of layers, respectively, and the variables D , R , A , S , T , I , C correspond the seven parameters explained above. A high DRASTIC index (DI) value indicates high vulnerability (Barzegar et al ).…”

“…Some studies have used external parameters, such as land use and irrigation type and intensity, to improve the DRASTIC method (Secunda et al ; McLay et al ). Others have used the analytical hierarchy process (AHP), statistical methods, the Wilcoxon test, genetic algorithm (GA), and artificial intelligence (AI) (Panagopoulos et al ; Ahn et al ; Huan et al ; Barzegar et al ; Jafari and Nikoo ; Jang et al ; Kihumba et al ; Yang et al ) to overcome the limitations of the DRASTIC framework. In a study similar to the one presented here, Barzegar et al () used a multi‐model ensemble of machine learning algorithms in the Marand plain of northwest Iran to investigate vulnerability to contamination in multiple aquifers (e.g., unconfined, semiconfined, and confined).…”

“…Existing studies indicate that modifying the ratings and weights of DRASTIC parameters in relation to measured NO 3 concentrations can be useful for achieving optimal agreement between the vulnerability indices and NO 3 concentrations (Barzegar et al ). Through the preparation of vulnerability maps of the study area, the current study aimed to compare different linear (i.e., Wilcoxon test and statistical approaches) and nonlinear (GA) modifications for calibration of the DRASTIC framework using NO 3 concentrations as a basis.…”

Hydrogeochemical features of groundwater resources in Tabriz plain, northwest of Iran

“…The formation consists of red and green andesite tuff admixed with large quantities of blocks, boulders, gravel and sand of volcanic and alluvial origin with thickness of up to 500 meters and it has good quality groundwater resources (Barzegar et al 2015(Barzegar et al , 2016b.…”

“…The presence of high nitrate concentration in the drinking water increases the incidence of gastric cancer and other potential hazards to infants and pregnant women (Nagireddi Srinivasa Rao 2006). Barzegar et al (2016b) studied the intrinsic vulnerability of Tabriz plain aquifer using the DRASTIC method. They indicated that the least vulnerable area was located in the west where the aquifer system is confined, while the most vulnerable areas were located in the east and south where the aquifer is unconfined and the recharge is high.…”