Numerical assessments of recharge-dominated groundwater flow and transport in the nearshore reclamation area in western Taiwan.

This study employed experimental and numerical methods to assess the behavior of conservative solute transport for a selected temporary solid waste site in a reclamation area in western Taiwan. Calibrating a site-specific numerical model, finite element model of water flow through saturated-unsaturated media (FEMWATER), relies on observations from field- and laboratory-scale hydraulic tests and spatial-temporal monitoring. The field-scale experiment used a modified hydraulic tomography survey (MHTS) to identify near surface aquifer stratifications and estimate the distribution of saturated hydraulic conductivity. The pressure plate experiments provided parameters for the van Genuchten soil characteristic model. Sensitivity analyses were then conducted based on varied recharge rates and dispersivities applied to the calibrated model. Observations of groundwater levels and salinity in the wells indicated that the regional groundwater flow was from southeast to northwest. In addition, a shallow freshwater layer was noted in the study area. The tidal-induced amplitudes for water level fluctuation in the wells ranged from 2 to 20 cm, depending on their distance from the seawater body. MHTS showed clear stratification, similar to that of well loggings at the storage site. The hydraulic conductivity at the test site ranged from 8 to 10 m/day, which is close to that obtained from the laboratory falling head test. The results of particle-tracking modeling showed that the critical recharge rate for the site needed to enhance plume traveling is 1000 mm/year. The increase in dispersivity values induced a decrease in plume travel time of up to 1000 days from the site to the coastal line. A special case for pulse releasing solute at the site shows that the key factor in controlling plume migration is the recharge rate. This is due to the low natural head gradient in the reclamation area. The results therefore suggest that a land drainage system near the site can play an important role in contaminant transport in the reclamation area.

Predicting Heavy Metal Concentrations in Shallow Aquifer Systems Based on Low-Cost Physiochemical Parameters Using Machine Learning Techniques.

Monitoring ex-situ water parameters, namely heavy metals, needs time and laboratory work for water sampling and analytical processes, which can retard the response to ongoing pollution events. Previous studies have successfully applied fast modeling techniques such as artificial intelligence algorithms to predict heavy metals. However, neither low-cost feature predictability nor explainability assessments have been considered in the modeling process. This study proposes a reliable and explainable framework to find an effective model and feature set to predict heavy metals in groundwater. The integrated assessment framework has four steps: model selection uncertainty, feature selection uncertainty, predictive uncertainty, and model interpretability. The results show that Random Forest is the most suitable model, and quick-measure parameters can be used as predictors for arsenic (As), iron (Fe), and manganese (Mn). Although the model performance is auspicious, it likely produces significant uncertainties. The findings also demonstrate that arsenic is related to nutrients and spatial distribution, while Fe and Mn are affected by spatial distribution and salinity. Some limitations and suggestions are also discussed to improve the prediction accuracy and interpretability.