Probability mapping of groundwater contamination by hydrocarbon from the deep oil reservoirs using GIS-based machine-learning algorithms: a case study of the Dammam aquifer (middle of Iraq).

The Dammam Formation in the southern and western deserts of Iraq is an important aquifer because it contains a huge groundwater reserve suitable for various uses. In the Karbala-Najaf plateau and the neighboring areas of the middle of Iraq, the drilling of groundwater wells usually fails due to the contamination of this aquifer with hydrocarbon from the deep oil reservoirs. This work suggests a method for the spatial delineation of groundwater contamination in this aquifer. Three machine learning classifiers, backpropagation multi-layer perceptron artificial neural networks (ANN), support vector machine with radial basis function (SVM-radial), and random forest (RF) with GIS, were used to map the probability of contamination in this aquifer. An inventory map of 139 groundwater boreholes (contaminated and non-contaminated) was utilized for building the models with seven factors that are considered to control contamination: fault density, distance to faults in general and the Abu Jir fault in particular, groundwater depth, hydraulic conductivity, aquifer saturated thickness, and land-surface elevation. The Relief-F feature selection method indicated that all factors were relevant. Five statistical measures were used for comparing the model performance: accuracy, sensitivity, specificity, kappa, and the area under the receiver operating characteristics curve (AUC). Applying the models using the R statistical package indicated that all models had excellent goodness-of-fit (accuracy > 90%), but the ANN (accuracy = 97%, sensitivity = 1.00%, specificity = 96%, kappa = 0.93, and AUC = 0.97) and RF (accuracy = 95%, sensitivity = 1.00%, specificity = 93%, kappa = 0.88, and AUC = 0.98) outperformed SVM-radial (accuracy = 92%, sensitivity = 1.00%, specificity = 90%, kappa = 0.82, and AUC = 0.95). The contamination probability values produced by these three models were categorized into different contamination zones range from very low to very high. The finding of this analysis may be used as a guide for drilling uncontaminated wells of groundwater.

A comparison between index of entropy and catastrophe theory methods for mapping groundwater potential in an arid region.

In this study, index of entropy and catastrophe theory methods were used for demarcating groundwater potential in an arid region using weighted linear combination techniques in geographical information system (GIS) environment. A case study from Badra area in the eastern part of central of Iraq was analyzed and discussed. Six factors believed to have influence on groundwater occurrence namely elevation, slope, aquifer transmissivity and storativity, soil, and distance to fault were prepared as raster thematic layers to facility integration into GIS environment. The factors were chosen based on the availability of data and local conditions of the study area. Both techniques were used for computing weights and assigning ranks vital for applying weighted linear combination approach. The results of application of both modes indicated that the most influential groundwater occurrence factors were slope and elevation. The other factors have relatively smaller values of weights implying that these factors have a minor role in groundwater occurrence conditions. The groundwater potential index (GPI) values for both models were classified using natural break classification scheme into five categories: very low, low, moderate, high, and very high. For validation of generated GPI, the relative operating characteristic (ROC) curves were used. According to the obtained area under the curve, the catastrophe model with 78 % prediction accuracy was found to perform better than entropy model with 77 % prediction accuracy. The overall results indicated that both models have good capability for predicting groundwater potential zones.

Prediction of groundwater flowing well zone at An-Najif Province, central Iraq using evidential belief functions model and GIS.

The objective of this study is to delineate groundwater flowing well zone potential in An-Najif Province of Iraq in a data-driven evidential belief function model developed in a geographical information system (GIS) environment. An inventory map of 68 groundwater flowing wells was prepared through field survey. Seventy percent or 43 wells were used for training the evidential belief functions model and the reset 30 % or 19 wells were used for validation of the model. Seven groundwater conditioning factors mostly derived from RS were used, namely elevation, slope angle, curvature, topographic wetness index, stream power index, lithological units, and distance to the Euphrates River in this study. The relationship between training flowing well locations and the conditioning factors were investigated using evidential belief functions technique in a GIS environment. The integrated belief values were classified into five categories using natural break classification scheme to predict spatial zoning of groundwater flowing well, namely very low (0.17-0.34), low (0.34-0.46), moderate (0.46-0.58), high (0.58-0.80), and very high (0.80-0.99). The results show that very low and low zones cover 72 % (19,282 km(2)) of the study area mostly clustered in the central part, the moderate zone concentrated in the west part covers 13 % (3481 km(2)), and the high and very high zones extended over the northern part cover 15 % (3977 km(2)) of the study area. The vast spatial extension of very low and low zones indicates that groundwater flowing wells potential in the study area is low. The performance of the evidential belief functions spatial model was validated using the receiver operating characteristic curve. A success rate of 0.95 and a prediction rate of 0.94 were estimated from the area under relative operating characteristics curves, which indicate that the developed model has excellent capability to predict groundwater flowing well zones. The produced map of groundwater flowing well zones could be used to identify new wells and manage groundwater storage in a sustainable manner.