Probabilistic human health risk assessment of degradation‐related chemical mixtures in heterogeneous aquifers: Risk statistics, hot spots, and preferential channels

Abstract

The increasing presence of toxic chemicals released in the subsurface has led to a rapid growth of social concerns and the need to develop and employ models that can predict the impact of groundwater contamination on human health risk under uncertainty. Monitored natural attenuation is a common remediation action in many contamination cases. However, natural attenuation can lead to the production of daughter species of distinct toxicity that may pose challenges in pollution management strategies. The actual threat that these contaminants pose to human health depends on the interplay between the complex structure of the geological media and the toxicity of each pollutant byproduct. This work addresses human health risk for chemical mixtures resulting from the sequential degradation of a contaminant (such as a chlorinated solvent) under uncertainty through high-resolution three-dimensional numerical simulations. We systematically investigate the interaction between aquifer heterogeneity, flow connectivity, contaminant injection model, and chemical toxicity in the probabilistic characterization of health risk. We illustrate how chemical-specific travel times control the regime of the expected risk and its corresponding uncertainties. Results indicate conditions where preferential flow paths can favor the reduction of the overall risk of the chemical mixture. The overall human risk response to aquifer connectivity is shown to be nontrivial for multispecies transport. This nontriviality is a result of the interaction between aquifer heterogeneity and chemical toxicity. To quantify the joint effect of connectivity and toxicity in health risk, we propose a toxicity-based Damk€ ohler number. Furthermore, we provide a statistical characterization in terms of low-order moments and the probability density function of the individual and total risks.

Radial solute transport in highly heterogeneous aquifers: Modeling and experimental comparison

“…We observed that increasing , τ 50 decreases in case of FP injection and increases in case of RC injection. Given that the risk is influenced by the ratio between the travel time and the characteristic time of reaction [ Henri et al ., ], different injection modes may lead to different impacts on the degradation of reactive substances, thereby producing different toxic daughter compounds at the extracting well, in case for instance of chlorinated solvents. Furthermore, τ 90 grows dramatically with in case of RC injection, reaching a difference of 1 order of magnitude with the FP case when .…”

“…As shown in Henri et al . [], the solute injection mode has a significant impact on the probability of the extreme events (i.e., probability of exceeding a high concentration threshold).…”

Debates—Stochastic subsurface hydrology from theory to practice: Why stochastic modeling has not yet permeated into practitioners?

“…Direct upscaling is typically challenged in real field applications by the presence of hydraulically connected features [ Trinchero et al ., ; Pedretti et al ., ], often exceeding the size of the model representative volume. The spatial distribution of highly permeable persistent geological bodies that concentrate solutes in connected channels controls not only the arrival of toxic concentrations and its subsequent risk to human life or ecosystems [ Henri et al ., ; Fiori et al ., ] but also the occurrence of biochemical reactions [ Rubol et al ., ; Sanchez‐Vila et al ., ], as they provide most of the nutrients that are vital to ecological systems. The representation of connected features in stochastic theories is still a major challenge.…”

Adaptive POD model reduction for solute transport in heterogeneous porous media

“…As an additional and challenging test case, we then consider the performance of our reduced order modeling strategy to model chemical transport in a heterogeneous porous medium. In this context, modern probabilistic approaches to environmental risk quantification [4,6,22,46] rely on the assessment of the probability that concentration of a given chemical species at a target environmental location exceeds a given threshold value, which is typically linked to human health or to environmental constraints. Other environmental performance metrics of interest in the context of the environmental risk assessment associated with solute migration in the subsurface include, e.g., peak values of concentrations, and/or early/late times of arrival of solute at sensitive locations in the aquifer [3].…”