Computational assessment of groundwater salinity distribution within coastal multi-aquifers of Bangladesh

Groundwater salinity modeling and mapping using machine learning approaches: a case study in Sidi Okba region, Algeria

The groundwater salinization process complexity and the lack of data on its controlling factors are the main challenges for accurate predictions and mapping of aquifer salinity. For this purpose, effective machine learning (ML) methodologies are employed for effective modeling and mapping of groundwater salinity (GWS) in the Mio-Pliocene aquifer in the Sidi Okba region, Algeria, based on limited dataset of electrical conductivity (EC) measurements and readily available digital elevation model (DEM) derivatives. The dataset was randomly split into training (70%) and testing (30%) sets, and three wrapper selection methods, recursive feature elimination (RFE), forward feature selection (FFS), and backward feature selection (BFS) are applied to train the data. The resulting combinations are used as inputs for five ML models, namely random forest (RF), hybrid neuro-fuzzy inference system (HyFIS), K-nearest neighbors (KNN), cubist regression model (CRM), and support vector machine (SVM). The best-performing model is identified and applied to predict and map GWS across the entire study area. It is highlighted that the applied methods yield input variation combinations as critical factors that are often overlocked by many researchers, which substantially impacts the models' accuracy. Among different alternatives the RF model emerged as the most effective for predicting and mapping GWS in the study area, which led to the high performance in both the training (RMSE = 1.016, R = 0.854, and MAE = 0.759) and testing (RMSE = 1.069, R = 0.831, and MAE = 0.921) phases. The generated digital map highlighted the alarming situation regarding excessive GWS levels in the study area, particularly in zones of low elevations and far from the Foum Elgherza dam and Elbiraz wadi. Overall, this study represents a significant advancement over previous approaches, offering enhanced predictive performance for GWS with the minimum number of input variables.

Multi-step ahead forecasting of electrical conductivity in rivers by using a hybrid Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) model enhanced by Boruta-XGBoost feature selection algorithm

Electrical conductivity (EC) is widely recognized as one of the most essential water quality metrics for predicting salinity and mineralization. In the current research, the EC of two Australian rivers (Albert River and Barratta Creek) was forecasted for up to 10 days using a novel deep learning algorithm (Convolutional Neural Network combined with Long Short-Term Memory Model, CNN-LSTM). The Boruta-XGBoost feature selection method was used to determine the significant inputs (time series lagged data) to the model. To compare the performance of Boruta-XGB-CNN-LSTM models, three machine learning approaches-multi-layer perceptron neural network (MLP), K-nearest neighbour (KNN), and extreme gradient boosting (XGBoost) were used. Different statistical metrics, such as correlation coefficient (R), root mean square error (RMSE), and mean absolute percentage error, were used to assess the models' performance. From 10 years of data in both rivers, 7 years (2012-2018) were used as a training set, and 3 years (2019-2021) were used for testing the models. Application of the Boruta-XGB-CNN-LSTM model in forecasting one day ahead of EC showed that in both stations, Boruta-XGB-CNN-LSTM can forecast the EC parameter better than other machine learning models for the test dataset (R = 0.9429, RMSE = 45.6896, MAPE = 5.9749 for Albert River, and R = 0.9215, RMSE = 43.8315, MAPE = 7.6029 for Barratta Creek). Considering the better performance of the Boruta-XGB-CNN-LSTM model in both rivers, this model was used to forecast 3-10 days ahead of EC. The results showed that the Boruta-XGB-CNN-LSTM model is very capable of forecasting the EC for the next 10 days. The results showed that by increasing the forecasting horizon from 3 to 10 days, the performance of the Boruta-XGB-CNN-LSTM model slightly decreased. The results of this study show that the Boruta-XGB-CNN-LSTM model can be used as a good soft computing method for accurately predicting how the EC will change in rivers.

Prediction of weighted arithmetic water quality index for urban water quality using ensemble machine learning model

Urban water quality index (WQI) is an important factor for assessment quality of groundwater in the urban and rural area. In this research, the Weighted Arithmetic Water Quality Index (WA-WQI) was estimated for understanding the groundwater quality. Four machine learning (ML) models were developed including artificial neural network (ANN), support vector machine (SVM), random forest (RF), and extreme gradient boosting (XG-Boost) in addition to multiple linear regression (MLR) for WA-WQI prediction at the Ujjain city of Madhya Pradesh in India. Groundwater quality samples were collected from 54 wards under the urban area, the main eight different physiochemical parameters were selected for WA-WQI prediction. The different input parameters data were analysed and calculated for the relationships of their ability to predict the results of WA-WQI. The ML models performance were calculated using three statistical metrics such as determination coefficient (R2), mean absolute error (MAE), and root mean square error (RMSE). In this research shown the XG-Boost model is better results other than other ML models. The best modelling results over the training phase revealed R2 = 0.969, RMSE = 2.169, MAE = 2.013 and over the testing phase R2 = 0.987, RMSE = 3.273, MAE = 2.727). All the ML models results were validated using receiver operating characteristic (ROC) curve for the best models selection. The results of best model area under curve (AUC) was 0.9048. Hence, XG-Boost model was given the accurate prediction of WA-WQI in the urban area. Based on the graphical presentation evaluation, XG-Boost model showed similar results of superiority. The obtained modelling results emphasis the utility of computer aid models for better planning and essential information for decision-makers, and water experts. The implement agency can adopt the procedures of water quality to decrease pollution and safe and healthy water provide to entire Ujjain city.

Using unsupervised machine learning models to drive groundwater chemistry and associated health risks in Indo-Bangla Sundarban region

Groundwater is an essential resource in the Sundarban regions of India and Bangladesh, but its quality is deteriorating due to anthropogenic impacts. However, the integrated factors affecting groundwater chemistry, source distribution, and health risk are poorly understood along the Indo-Bangla coastal border. The goal of this study is to assess groundwater chemistry, associated driving factors, source contributions, and potential non-carcinogenic health risks (PN-CHR) using unsupervised machine learning models such as a self-organizing map (SOM), positive matrix factorization (PMF), ion ratios, and Monte Carlo simulation. For the Sundarban part of Bangladesh, the SOM clustering approach yielded six clusters, while it yielded five for the Indian Sundarbans. The SOM results showed high correlations among Ca2+, Mg2+, and K+, indicating a common origin. In the Bangladesh Sundarbans, mixed water predominated in all clusters except for cluster 3, whereas in the Indian Sundarbans, Cl--Na+ and mixed water dominated in clusters 1 and 2, and both water types dominated the remaining clusters. Coupling of SOM, PMF, and ionic ratios identified rock weathering as a driving factor for groundwater chemistry. Clusters 1 and 3 were found to be influenced by mineral dissolution and geogenic inputs (overall contribution of 47.7%), while agricultural and industrial effluents dominated clusters 4 and 5 (contribution of 52.7%) in the Bangladesh Sundarbans. Industrial effluents and agricultural activities were associated with clusters 3, 4, and 5 (contributions of 29.5% and 25.4%, respectively) and geogenic sources (contributions of 23 and 22.1% in clusters 1 and 2) in Indian Sundarbans. The probabilistic health risk assessment showed that NO3- poses a higher PN-CHR risk to human health than F- and As, and that potential risk to children is more evident in the Bangladesh Sundarban area than in the Indian Sundarbans. Local authorities must take urgent action to control NO3- emissions in the Indo-Bangla Sundarbans region.

Groundwater Salinity Across India: Predicting Occurrences and Controls by Field-Observations and Machine Learning Modeling

Elevated groundwater salinity is unsuitable for drinking and harmful to crop production. Thus, it is crucial to determine groundwater salinity distribution, especially where drinking and agricultural water requirements are largely supported by groundwater. This study used field observation (n = 20,994)-based machine learning models to determine the probabilistic distribution of elevated groundwater salinity (electrical conductivity as a proxy, >2000 μS/cm) at 1 km2 across parts of India for near groundwater-table conditions. The final predictions were made by using the best-performing random forest model. The validation performance also demonstrated the robustness of the model (with 77% accuracy). About 29% of the study area (including 25% of entire cropland areas) was estimated to have elevated salinity, dominantly in northwestern and peninsular India. Also, parts of the northwestern and southeastern coasts, adjoining the Arabian Sea and the Bay of Bengal, were assessed with elevated salinity. The climate was delineated as the dominant factor influencing groundwater salinity occurrence, followed by distance from the coast, geology (lithology), and depth of groundwater. Consequently, ∼330 million people, including ∼109 million coastal populations, were estimated to be potentially exposed to elevated groundwater salinity through groundwater-sourced drinking water, thus substantially limiting clean water access.

Groundwater quality characterization using an integrated water quality index and multivariate statistical techniques

This study attempts to characterize and interpret the groundwater quality (GWQ) using a GIS environment and multivariate statistical approach (MSA) for the Jakham River Basin (JRB) in Southern Rajasthan. In this paper, analysis of various statistical indicators such as the Water Quality Index (WQI) and multivariate statistical methods, i.e., principal component analysis and correspondence analysis (PCA and CA), were implemented on the pre and post-monsoon water quality datasets. All these methods help identify the most critical factor in controlling GWQ for potable water. In pre-monsoon (PRM) and post-monsoon (POM) seasons, the computed value of WQI has ranged between 28.28 to 116.74 and from 29.49 to 111.98, respectively. As per the GIS-based WQI findings, 63.42 percent of the groundwater samples during the PRM season and 42.02 percent during the POM were classed as 'good' and could be consumed for drinking. The Principal component analysis (PCA) is a suitable tool for simplification of the evaluation process in water quality analysis. The PCA correlation matrix defines the relation among the water quality parameters, which helps to detect the natural or anthropogenic influence on sub-surface water. The finding of PCA's factor analysis shows the impact of geological and human intervention, as increased levels of EC, TDS, Na+, Cl-, HCO3-, F-, and SO42- on potable water. In this study, hierarchical cluster analysis (HCA) was used to categories the WQ parameters for PRM and POR seasons using the Ward technique. The research outcomes of this study can be used as baseline data for GWQ development activities and protect human health from water-borne diseases in the southern region of Rajasthan.

Advanced machine learning-based kharif maize evapotranspiration estimation in semi-arid climate

Accurate Crop Evapotranspiration (ETc) estimation is crucial for understanding hydrological and agrometeorological processes, yet it's challenged by multiple parameters, data variations, and lack of continuity. These limitations restrict numerical methods application. To address this, the study aims to develop and assess ML models for daily maize ETc in semi-arid areas, utilizing varied weather inputs. Five ML models viz., Category Boosting (CB), Linear Regression (LR), Support Vector Machine (SVM), Artificial Neural Network (ANN), and Stochastic Gradient Descent (SGD) were developed and validated for the ICAR-IARI, New Delhi, Research Station. Penman-Monteith (PM) model estimated ETc values are used as the standard for comparing the performance of the ML model values. Results revealed that the SVM model achieved the highest coefficient of determination (R2) among all models, with a value of 0.987. Furthermore, the SVM model exhibited the lowest model errors (MAE = 0.121 mm day-1, RMSE = 0.172 mm day-1, and MAPE = 4.37%) compared to other models. The ANN model also demonstrated promising results, comparable to the SVM model. Notably, the wind speed parameter was found most influential input parameter. In conclusion, SVM or ANN could be considered reliable alternative methods for the accurate estimation of kharif maize crop ETc in the semi-arid climate.

A Comprehensive Survey on Aquila Optimizer

Aquila Optimizer (AO) is a well-known nature-inspired optimization algorithm (NIOA) that was created in 2021 based on the prey grabbing behavior of Aquila. AO is a population-based NIOA that has demonstrated its effectiveness in the field of complex and nonlinear optimization in a short period of time. As a result, the purpose of this study is to provide an updated survey on the topic. This survey accurately reports on the designed enhanced AO variations and their applications. In order to properly assess AO, a rigorous comparison between AO and its peer NIOAs is conducted over mathematical benchmark functions. The experimental results show the AO provides competitive outcomes.

Modeling stochastic saline groundwater occurrence in coastal aquifers

The issue of freshwater salinization in coastal areas has grown in importance with the increase of the demand of groundwater supply and the more frequent droughts. However, the spatial patterns of salinity contamination are not easy to be understood, as well as their numerical modeling is subject to various kinds of uncertainty. This paper offers a robust, flexible, and reliable geostatistical methodology to provide a stochastic assessment of salinity distribution in alluvial coastal areas. The methodology is applied to a coastal aquifer in Campania (Italy), where 83 monitoring wells provided depth-averaged salinity data. A Monte Carlo (MC) framework was implemented to simulate depth-averaged groundwater salinity fields. Both MC stochastic fields and the mean across MC simulations enabled to the delineation of which areas are subject to high salinity. Then, a probabilistic approach was developed setting up salinity thresholds for agricultural use to delineate the areas with unsuitable groundwater for irrigation purposes. Furthermore, steady spatial patterns of saline wedge lengths were unveiled through uncertainty estimates of seawater ingression at the Volturno River mouth. The results were compared versus a calibrated numerical model with remarkable model fit (R²=0.96) and versus an analytical solution, obtaining similar wedge lengths. The results pointed out that the high groundwater salinities found inland (more than 2 km from the coastline) could be ascribed to trapped paleo-seawater rather than to actual seawater intrusion. In fact, the inland high salinities were in correspondence of thick peaty layers, which can store trapped saline waters because of their high porosity and low permeability. Furthermore, these results are consistent with the recognition of depositional environments and the position of ancient lagoon alluvial sediments, located in the same areas where are the highest (simulated) salinity fields. This robust probabilistic approach could be applied to similar alluvial coastal areas to understand spatial patterns of present salinization, to disentangle actual from paleo-seawater intrusion, and more in general to delineate zones with unsuitable salinity for irrigation purposes.

The deterioration of groundwater quality by seawater intrusion in the Chao Phraya River Basin, Thailand

The Chao Phraya River Deltaic Plain is the largest basin in Thailand and the second largest one in Southeast Asia after the Mekong River Delta. In recent decades, the groundwater quality in the Lower Chao Phraya River Basin in Thailand has deteriorated due to salinization caused by seawater intrusion. In the present study, hydrogeochemical and statistical methods were employed to determine the hydrochemical characteristics of the groundwater and to investigate the possible sources of salinity in the study region for the years 2008 and 2020. In addition, samples were taken from precipitation, sea water, and river water to analyze their hydrochemical properties. Then, they were used as input in the "Simmr" code in the R programming language to model the hydrochemical conditions of the study area and their evolution over time. The results indicated that in the non-coastal regions, water-rock interaction (mineral weathering and ion exchange), and brine/connate water infiltration affected the quality of the groundwater. However, the seawater intrusion was limited only to the coastal regions. Furthermore, the groundwater quality deteriorated from 2008 to 2020. Finally, using stepwise regression in the R language, the salinity of the groundwater was simulated and compared with the measured salinity data. The results obtained by the stepwise model were in close agreement with those obtained from the hydrochemical studies. This study confirmed seawater intrusion in the coastal aquifer as well as the deterioration of groundwater quality over time. To slow down this process and to achieve sustainable conditions, groundwater extraction should be reduced in the study region.