A Hybrid Neural Network-Particle Swarm Optimization Informed Spatial Interpolation Technique for Groundwater Quality Mapping in a Small Island Province of the Philippines

Comparison of individual and ensemble machine learning models for prediction of sulphate levels in untreated and treated Acid Mine Drainage

Machine learning was used to provide data for further evaluation of potential extraction of octathiocane (S8), a commercially useful by-product, from Acid Mine Drainage (AMD) by predicting sulphate levels in an AMD water quality dataset. Individual ML regressor models, namely: Linear Regression (LR), Least Absolute Shrinkage and Selection Operator (LASSO), Ridge (RD), Elastic Net (EN), K-Nearest Neighbours (KNN), Support Vector Regression (SVR), Decision Tree (DT), Extreme Gradient Boosting (XGBoost), Random Forest (RF), Multi-Layer Perceptron Artificial Neural Network (MLP) and Stacking Ensemble (SE-ML) combinations of these models were successfully used to predict sulphate levels. A SE-ML regressor trained on untreated AMD which stacked seven of the best-performing individual models and fed them to a LR meta-learner model was found to be the best-performing model with a Mean Squared Error (MSE) of 0.000011, Mean Absolute Error (MAE) of 0.002617 and R2 of 0.9997. Temperature (°C), Total Dissolved Solids (mg/L) and, importantly, iron (mg/L) were highly correlated to sulphate (mg/L) with iron showing a strong positive linear correlation that indicated dissolved products from pyrite oxidation. Ensemble learning (bagging, boosting and stacking) outperformed individual methods due to their combined predictive accuracies. Surprisingly, when comparing SE-ML that combined all models with SE-ML that combined only the best-performing models, there was only a slight difference in model accuracies which indicated that including bad-performing models in the stack had no adverse effect on its predictive performance.

Application of machine learning in groundwater quality modeling - A comprehensive review

Groundwater is a crucial resource across agricultural, civil, and industrial sectors. The prediction of groundwater pollution due to various chemical components is vital for planning, policymaking, and management of groundwater resources. In the last two decades, the application of machine learning (ML) techniques for groundwater quality (GWQ) modeling has grown exponentially. This review assesses all supervised, semi-supervised, unsupervised, and ensemble ML models implemented to predict any groundwater quality parameter, making this the most extensive modern review on this topic. Neural networks are the most used ML model in GWQ modeling. Their usage has declined in recent years, giving rise to more accurate or advanced techniques such as deep learning or unsupervised algorithms. Iran and the United States lead the world in areas modeled, with a wealth of historical data available. Nitrate has been modeled most exhaustively, targeted by nearly half of all studies. Advancements in future work will be made with further implementation of deep learning and explainable artificial intelligence or other cutting-edge techniques, application of these techniques for sparsely studied variables, the modeling of new or unique study areas, and the implementation of ML techniques for groundwater quality management.

Low-Cost Internet-of-Things Water-Quality Monitoring System for Rural Areas

Water is a vital source for life and natural environments. This is the reason why water sources should be constantly monitored in order to detect any pollutants that might jeopardize the quality of water. This paper presents a low-cost internet-of-things system that is capable of measuring and reporting the quality of different water sources. It comprises the following components: Arduino UNO board, Bluetooth module BT04, temperature sensor DS18B20, pH sensor-SEN0161, TDS sensor-SEN0244, turbidity sensor-SKU SEN0189. The system will be controlled and managed from a mobile application, which will monitor the actual status of water sources. We propose to monitor and evaluate the quality of water from five different water sources in a rural settlement. The results show that most of the water sources we have monitored are proper for consumption, with a single exception where the TDS values are not within proper limits, as they outperform the maximum accepted value of 500 ppm.

Quantitative Assessment and Spatial Analysis of Metals and Metalloids in Soil Using the Geo-Accumulation Index in the Capital Town of Romblon Province, Philippines

The municipality of Romblon in the Philippines is an island known for its marble industry. The subsurface of the Philippines is known for its limestone. The production of marble into slab, tiles, and novelty items requires heavy equipment to cut rocks and boulders. The finishing of marble requires polishing to smoothen the surface. During the manufacturing process, massive amounts of particulates and slurry are produced, and with a lack of technology and human expertise, the environment can be adversely affected. Hence, this study assessed and monitored the environmental conditions in the municipality of Romblon, particularly the soils and sediments, which were affected due to uncontrolled discharges and particulates deposition. A total of fifty-six soil and twenty-three sediment samples were collected and used to estimate the metal and metalloid (MM) concentrations in the whole area using a neural network-particle swarm optimization inverse distance weighting model (NN-PSO). There were nine MMs; e.g., As, Cr, Ni, Pb, Cu, Ba, Mn, Zn and Fe, with significant concentrations detected in the area in both soils and sediments. The geo-accumulation index was computed to assess the level of contamination in the area, and only the soil exhibited contamination with zinc, while others were still on a safe level. Nemerow's pollution index (NPI) was calculated for the samples collected, and soil was evaluated and seen to have a light pollution level, while sediment was considered as "clean". Furthermore, the single ecological risk (Er) index for both soil and sediment samples was considered to be a low pollution risk because all values of Er were less than 40.

In Situ Measurements of Domestic Water Quality and Health Risks by Elevated Concentration of Heavy Metals and Metalloids Using Monte Carlo and MLGI Methods

The domestic water (DW) quality of an island province in the Philippines that experienced two major mining disasters in the 1990s was assessed and evaluated in 2021 utilizing the heavy metals pollution index (MPI), Nemerow’s pollution index (NPI), and the total carcinogenic risk (TCR) index. The island province sources its DW supply from groundwater (GW), surface water (SW), tap water (TP), and water refilling stations (WRS). This DW supply is used for drinking and cooking by the population. In situ analyses were carried out using an Olympus Vanta X-ray fluorescence spectrometer (XRF) and Accusensing Metals Analysis System (MAS) G1 and the target heavy metals and metalloids (HMM) were arsenic (As), barium (Ba), copper (Cu), iron (Fe), lead (Pb), manganese (Mn), nickel (Ni), and zinc (Zn). The carcinogenic risk was evaluated using the Monte Carlo (MC) method while a machine learning geostatistical interpolation (MLGI) technique was employed to create spatial maps of the metal concentrations and health risk indices. The MPI values calculated at all sampling locations for all water samples indicated a high pollution. Additionally, the NPI values computed at all sampling locations for all DW samples were categorized as “highly polluted”. The results showed that the health quotient indices (HQI) for As and Pb were significantly greater than 1 in all water sources, indicating a probable significant health risk (HR) to the population of the island province. Additionally, As exhibited the highest carcinogenic risk (CR), which was observed in TW samples. This accounted for 89.7% of the total CR observed in TW. Furthermore, all sampling locations exceeded the recommended maximum threshold level of 1.0 × 10⁻⁴ by the USEPA. Spatial distribution maps of the contaminant concentrations and health risks provide valuable information to households and guide local government units as well as regional and national agencies in developing strategic interventions to improve DW quality in the island province.

Geo-Accumulation Index of Manganese in Soils Due to Flooding in Boac and Mogpog Rivers, Marinduque, Philippines with Mining Disaster Exposure

This paper presents the effects of flooding on the accumulation of manganese (Mn) in soils within proximity of the Boac and Mogpog rivers in Marinduque of The Philippines. Marinduque, an island province in the Philippines, experienced two catastrophic tailings storage facility (TSF) failures in the 1990s that released sulfide-rich tailings into the two major rivers. The Philippines experiences 21–23 typhoons every year, 11 of which pass thru Marinduque that causing inundation of floodplain areas in the province. A flood hazard map developed using LiDAR DEM was utilized for the Boac and Mogpog rivers for an accurate representation of flooding events. A portable X-ray fluorescence spectrometer (pXRF) and a Hannah multi-parameter device were used for the on-site analyses of Mn concentration and water physico-chemical properties, respectively. Spatial grid mapping with zonal statistics was employed for a comprehensive analysis of all the data collected and processed. Correlation analysis was carried out on Mn concentrations in soil and surface water, electrical conductivity (EC), total dissolved solids (TDS), pH, temperature, curve number (CN), and flood heights. The curve number indicates the runoff response characteristic of the Mogpog-Boac River basin. The results show that 40% of the total floodplain area of Boac and Mogpog were subjected to high hazards with flood heights above 1.5 m. The Mn content of soils had a statistically significant moderate positive correlation with flood height (r = 0.458) and a moderate negative correlation with pH (r = −0.438). This condition suggested that more extensive flooding promotes Mn contamination of floodplain soils in the two rivers, the source of which includes the mobilization of Mn-bearing silt, sediments, and mine drainage from the abandoned mine pits and TSFs. There is also a strong negative correlation between pH and Mn concentrations in surface water, a relationship attributed to the solubilization of Mn-bearing precipitates based on geochemical modeling results. Using Muller’s geo-accumulation index, 77.5% of the total floodplain of the two rivers was identified as “moderately contaminated” with an average Mn soil content of 3.4% by weight (34,000 mg/kg). The Mn contamination map of floodplain soils in the Mogpog and Boac rivers described in this study could guide relevant regional, national, and local government agencies in planning appropriate intervention, mitigation, remediation, and rehabilitation strategies to limit human exposure to highly contaminated areas.

Neuro-Particle Swarm Optimization Based In-Situ Prediction Model for Heavy Metals Concentration in Groundwater and Surface Water

Limited monitoring activities to assess data on heavy metal (HM) concentration contribute to worldwide concern for the environmental quality and the degree of toxicants in areas where there are elevated metals concentrations. Hence, this study used in-situ physicochemical parameters to the limited data on HM concentration in SW and GW. The site of the study was Marinduque Island Province in the Philippines, which experienced two mining disasters. Prediction model results showed that the SW models during the dry and wet seasons recorded a mean squared error (MSE) ranging from 6 × 10-7 to 0.070276. The GW models recorded a range from 5 × 10-8 to 0.045373, all of which were approaching the ideal MSE value of 0. Kling-Gupta efficiency values of developed models were all greater than 0.95. The developed neural network-particle swarm optimization (NN-PSO) models for SW and GW were compared to linear and support vector machine (SVM) models and previously published deterministic and artificial intelligence (AI) models. The findings indicated that the developed NN-PSO models are superior to the developed linear and SVM models, up to 1.60 and 1.40 times greater than the best model observed created by linear and SVM models for SW and GW, respectively. The developed models were also on par with previously published deterministic and AI-based models considering their prediction capability. Sensitivity analysis using Olden's connection weights approach showed that pH influenced the concentration of HM significantly. Established on the research findings, it can be stated that the NN-PSO is an effective and practical approach in the prediction of HM concentration in water resources that contributes a solution to the limited HM concentration monitored data.

Health Risks Due to Metal Concentrations in Soil and Vegetables from the Six Municipalities of the Island Province in the Philippines

This paper investigated the health risks due to metal concentrations in soil and vegetables from the island province in the Philippines and the potential ecological risks. The concentrations of Cd, Cr, Cu, Fe, Mn, Ni, Pb, and Zn in vegetables and soil in six municipalities of the province were analyzed using the Inductively Coupled Plasma Optical Emission Spectrometry (ICP-OES) Perkin Elmer Optima 8000. It was recorded that all metal concentrations in the soil, except for Cd, exceeded the soil quality standard (SQS). The concentration of Fe and Mn was highest among other metals. The Nemerow synthetical pollution index (Pn) in all soil samples was under Class V which means severe pollution level. Likewise, the risk index (RI) of soil ranged from high to very high pollution risk. Most of the metal concentrations in the vegetables analyzed also exceeded the maximum permissible limit (MPL). All health hazard indices (HHIs) were less than 1, which means potential low non-carcinogenic risk to human population by vegetable consumption. However, it was found that concentration of Cr and Ni in vegetables is a potential health hazard having concentrations exceeding the maximum threshold limit. A 75% temporary consumption reduction of bitter melon, eggplant, sweet potato tops, and string beans produced from two municipalities may be helpful in reducing exposure to target metals. Additional studies are needed to confirm this recommendation. Spatial correlation analysis showed that six out of target metals had datasets that were more spatially clustered than would be expected. The recorded data are useful for creation of research direction, and aid in developing strategies for remediation, tools, and programs for improving environmental and vegetable quality monitoring.