Abstract
Groundwater resources, unlike surface water, are more vulnerable to disturbances and contaminations, as they take a very long time and significant cost to recover. So, predictive modeling and prevention strategies can empower policymakers for efficient groundwater governance through informed decisions and recommendations. Due to the importance of groundwater quality modeling, the hardness susceptibility mapping using machine learning (ML) models has not been explored. For the first time, the current research aimed to predict groundwater hardness susceptibility using the ML models. The performance of two ensemble models of boosted regression trees (BRT) and random forest (RF) is investigated through the arrangement of a comparative study with multivariate discriminant analysis (MDA). According to the hardness values in 135 groundwater quality monitoring wells, the hard and soft water are determined; then, 11 predictor variables including distance from the sea (DFS), land use, elevation, distance from the river (DFR), depth to groundwater (DTGW), pH, precipitation (PCP), evaporation (E), groundwater level (GWL), curvature, and lithology are used for predicting the groundwater hardness susceptibility map. Results indicated that the variables of DFR, DTGW, elevation, and DFS had a higher contribution to the modeling process. So, the high harness areas are mostly related to low elevations, low DTGW, and proximity to river and sea, which facilitate the percolation conditions for minerals containing calcium or magnesium into groundwater.
Highlights
Groundwater is the fundamental source of drinking water and agriculture irrigation for at least half of the world’s population [1]
Machine learning models were applied to model hardness susceptibility of groundwater, which are described as follows: Boosted Regression Trees (BRT): The boosted regression trees (BRT) model was firstly introduced by Freund and Schapire [74]
It is necessary to note that the important assumptions of the multivariate discriminant analysis (MDA) are: (i) the independent parameters are in careful multivariate normal distribution; (ii) the predictors should not be strongly correlated, and their average and variance have not been accounted for; (iii) two predictors should have a stable correlation between groups [81]
Summary
Groundwater is the fundamental source of drinking water and agriculture irrigation for at least half of the world’s population [1]. Among the prediction techniques for groundwater quality modeling, the susceptibility mapping methodologies have been well received within the hydrology research community as well as environmental management and policymakers [13,14,15,16,17,18]. Various aspects of groundwater quality, e.g., nitrates; heavy metals; hardness; acidity; conductivity; turbidity; minerals; and a wide range of physical, chemical, and biological pollutants, are yet to be modeled using novel machine learning methods. The research gap in the applicability of ensemble machine learning methods in groundwater quality modeling has been evident. The contribution of this paper is to explore the performance of two ensemble models of boosted regression trees (BRT) and random forest (RF) for susceptibility mapping of groundwater quality. Despite the importance of hardness susceptibility mapping, there is a research gap in the advancement of novel models [50].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
