Chronic arsenic (As) exposure continues to be a public health problem of major concern worldwide, affecting hundreds of millions of people. A long-term groundwater quality survey has revealed that 20% of the groundwater in southern Taiwan’s Pingtung Plain is clearly contaminated with a measured As concentration in excess of the maximum level of 10 µg/L recommended by the World Health Organization. The situation is further complicated by the fact that more than half of the inhabitants in this area continue to use groundwater for drinking. Efforts to assess the health risk associated with the ingestion of As from the contaminated drinking water are required in order to determine the priorities for health risk management. The conventional approach to conducting a human health risk assessment may be insufficient for this purpose, so this study adopts a geostatistical Kriging method to perform a spatial analysis of the health risk associated with ingesting As through drinking groundwater in the Pingtung Plain. The health risk is assessed based on the hazard quotient (HQ) and target cancer risk (TR) established by the U.S. Environmental Protection Agency. The results show that most areas where the HQ exceeds 1 are in the southwestern part of the study area. In addition, the high-population density townships of Daliao, Linyuan, Donggang, Linbian, Jiadong, and Fangliao presently have exceedingly high TR values that are two orders of magnitude higher than the acceptable standard. Thus, the use of groundwater for drinking in these townships should be strictly avoided. A map that delineates areas with high TR values and high population densities is provided. The findings broaden the scope of the spatial analysis of human health risk and provide a basis for improving the decision-making process.

Groundwater resources are abundant and widely used in Taiwan’s Lanyang Plain. However, in some places the groundwater arsenic (As) concentrations far exceed the World Health Organization’s standards for drinking water quality. Measurements of the As concentrations in groundwater show considerable spatial variability, which means that the associated risk to human health would also vary from region to region. This study aims to adapt a back-propagation neural network (BPNN) method to carry out more reliable spatial mapping of the As concentrations in the groundwater for comparison with the geostatistical ordinary kriging (OK) method results. Cross validation is performed to evaluate the prediction performance by dividing the As monitoring data into three sets. The cross-validation results show that the average determination coefficients (R²) for the As concentrations obtained with BPNN and OK are 0.55 and 0.49, whereas the average root mean square errors (RMSE) are 0.49 and 0.54, respectively. Given the better prediction performance of the BPNN, it is recommended as a more reliable tool for the spatial mapping of the groundwater As concentration. Subsequently, the As concentrations estimated obtained using the BPNN are applied to develop a spatial map illustrating the risk to human health associated with the ingestion of As-containing groundwater based on the noncarcinogenic hazard quotient (HQ) and carcinogenic target risk (TR) standards established by the U.S. Environmental Protection Agency. Such maps can be used to demarcate the areas where residents are at higher risk due to the ingestion of As-containing groundwater, and prioritize the areas where more intensive monitoring of groundwater quality is required. The spatial mapping of As concentrations from the BPNN was also used to demarcate the regions where the groundwater is suitable for farmland and fishponds based on the water quality standards for As for irrigation and aquaculture.

In this study, we present a semi-analytical model for simulating three-dimensional radioactivity transport of a radionuclide decay chain and assessing the radiological dose impact on the general public. The mechanisms and processes considered in the model include the one-dimensional advection, hydrodynamic dispersion in longitudinal and two lateral directions, linear equilibrium sorption, and first-order radioactive decay reactions. The semi-analytical model is derived for a semi-infinite domain, and the solutions in the Laplace domain for members of the decay chain are first generalized in a compact format. The concentrations in the original domain of each nuclide are independently evaluated with the help of the efficient and robust Laplace numerical inverse algorithms. The accuracy of the derived semi-analytical model is demonstrated by comparison of our developed model with an existing analytical model described in the literature. The results of the verification exercise indicate that the derived semi-analytical model is accurate and robust. The developed semi-analytical model is applied to an illustrative example that simulates the three-dimensional plume migration of a radionuclide decay chain on both the temporal and spatial scales. Moreover, the time histories of the radiological doses at different distances from the inflow source boundary are presented to understand the potential radiological impact on the general public. The developed model facilitates rapid assessment of the radiological impact posed by the presence of radionuclides in the environment because of leakage from a nuclear waste repository or accidental discharges from nuclear facilities.

In the Pingtung Plain of southern Taiwan, elevated levels of NO₃⁻-N in groundwater have been reported. Therefore, efforts for assessing groundwater vulnerability are required as part of the critical steps to prevent and control groundwater pollution. This study makes a groundwater vulnerability assessment for the Pingtung Plain using an improved overlay and index-based DRASTIC model. The improvement of the DRASTIC model is achieved by reassigning the weighting coefficients of the factors in this model with the help of a discriminant analysis statistical method. The analytical results obtained from the improved DRASTIC model provide a reliable prediction for use in groundwater vulnerability assessment to nitrate pollution and can correctly identify the groundwater protection zones in the Pingtung Plain. Moreover, the results of the sensitivity analysis conducted for the seven parameters in the improved DRASTIC model demonstrate that the aquifer media (A) is the most sensitive factor when the nitrate-N concentration is below 2.5 mg/L. For the cases where the nitrate-N concentration is above 2.5 mg/L, the aquifer media (A) and net recharge (R) are the two most important factors.

Although the average municipal water coverage in Taiwan is quite high, at 93.91%, only around half of the residents in the Pingtung Plain use tap water originating from the Taiwan Water Corporation to meet their needs. This means the exploitation of a substantial amount of groundwater as a source of water to meet drinking, agriculture, aquaculture, and industry requirements. Long-term groundwater quality surveys in Taiwan have revealed obvious contamination of the groundwater in several locations in the Pingtung Plain, with measured concentration levels of some groundwater quality parameters in excess of the permissible levels specified by the Taiwan Environmental Protection Administration. Clearly, establishing a sound plan for groundwater quality protection in this area is imperative for maximizing the protection of human health. The inappropriate use of hazardous chemicals and poor management of land use have allowed pollutants to permeate through unsaturated soil and ultimately reach the underlying shallow unconfined groundwater system. Thus, the quality of the water stored in shallow aquifers has been significantly affected by land use. This study is designed to characterize the relationship between groundwater quality and land use in the Pingtung Plain. This goal is achieved by the application of factor analysis to characterize the measured concentrations of 14 groundwater quality parameters sampled from 46 observation wells, the area percentages for nine land use categories in the neighborhood of these 46 observation wells, and the thicknesses of four unsaturated types of soil based on core samples obtained during the establishment of 46 observation wells. The results show that a four-factor model can explain 56% of the total variance. Factor 1 (seawater salinization), which includes the groundwater quality parameters of EC, SO₄²⁻, Cl⁻, Ca²⁺, Mg²⁺, Na⁺, and K⁺, shows a moderate correlation to land used for water conservation. Factor 2 (nitrate pollution), which includes the groundwater quality parameters of NO₃⁻-N and HCO₃⁻, shows a strong correlation to land used for fruit farming and a moderate correlation to the thickness of the gravel comprising unsaturated soil. Factor 3 (arsenic pollution), which is composed of groundwater quality parameters of total organic carbon (TOC) and As, is very weakly affected by land use. Factor 4 (reductive dissolution of Fe³⁺ and Mn²⁺), which involves Mn²⁺ and Fe³⁺, is weakly impacted by land use. Based on a geographic visualization of the scores for the four different factors and the patterns for land use, we can demarcate the areas where the groundwater in shallow unconfined aquifers is more vulnerable to being polluted by specific contaminants. We can then prioritize the areas where more intensive monitoring might be required, evaluate current land use practices, and adopt new measures to better prevent or control groundwater pollution.

Traditional numerical models have been widely employed to simulate the transport of multispecies reactive contaminants in groundwater systems; however, their high computational cost limits their applicability in real-time or large-scale scenarios. Recent advances in artificial intelligence (AI) offer promising alternatives, particularly data-driven machine learning techniques, for accelerating such simulations. This study presents the development of a surrogate model based on artificial neural networks (ANNs) to simulate the transport and decay of interacting multispecies contaminants in groundwater. High-fidelity training datasets are generated through finite difference-based reactive transport simulations across a wide range of environmental and geochemical conditions. The ANN model is trained to learn the complex nonlinear relationships governing the multispecies transport and transformation processes. Model validation reveals that the ANN surrogate accurately reproduces the spatial–temporal concentration profiles of both original and degradation species, capturing key dynamic behaviors with high precision. Notably, the ANN model achieves up to a 100-fold reduction in computational time compared to traditional analytical or semi-analytical solutions. These results highlight the ANN’s potential as an efficient and accurate surrogate modeling tool for groundwater contamination assessment, offering a valuable advancement for decision-making in environmental risk analysis and remediation planning.