Reactive transport modeling for supporting climate resilience at groundwater contamination sites

Abstract

Abstract. Climate resilience is an emerging issue at contaminated sites and hazardous waste sites, since projected climate shifts (e.g., increased/decreased precipitation) and extreme events (e.g., flooding, drought) could affect ongoing remediation or closure strategies. In this study, we develop a reactive transport model (Amanzi) for radionuclides (uranium, tritium, and others) and evaluate how different scenarios under climate change will influence the contaminant plume conditions and groundwater well concentrations. We demonstrate our approach using a two-dimensional (2D) reactive transport model for the Savannah River Site F-Area, including mineral reaction and sorption processes. Different recharge scenarios are considered by perturbing the infiltration rate from the base case as well as considering cap-failure and climate projection scenarios. We also evaluate the uranium and nitrate concentration ratios between scenarios and the base case to isolate the sorption effects with changing recharge rates. The modeling results indicate that the competing effects of dilution and remobilization significantly influence pH, thus changing the sorption of uranium. At the maximum concentration on the breakthrough curve, higher aqueous uranium concentration implies that sorption is reduced with lower pH due to remobilization. To better evaluate the climate change impacts in the future, we develop the workflow to include the downscaled CMIP5 (Coupled Model Intercomparison Project) climate projection data in the reactive transport model and evaluate how residual contamination evolves through 2100 under four climate Representative Concentration Pathway (RCP) scenarios. The integration of climate modeling data and hydrogeochemistry models enables us to quantify the climate change impacts, assess which impacts need to be planned for, and therefore assist climate resiliency efforts and help guide site management.

PyLEnM: A Machine Learning Framework for Long-Term Groundwater Contamination Monitoring Strategies

“…This lack of predictive power is the reason why simpler regression methods (such as linear regressions rather than RF) were selected for the tritium concentration estimation (Table 1). Recent advances in physics-informed ML 35 could enable the integration of contaminant transport simulations (e.g., Xu et al 2022) 36 to improve the contaminant concentration estimations in the future.…”

Reactive Transport Modeling for Exploring the Potential of Water Quality Sensors to Estimate Hydrocarbon Levels in Groundwater

“…Additionally, extreme events such as increased precipitation due to climate change can affect the water table by influencing groundwater recharge. Although previous studies have explored how climate change affects contaminant transport (Libera et al., 2019; Xu et al., 2022), our focus in this study did not extend to simulating the potential impacts of climate change.…”

Can machine learning accelerate process understanding and decision‐relevant predictions of river water quality?

“…High‐fidelity mechanistic models can incorporate the tight coupling of physical and biogeochemical processes such as climatic variations, hydrologic fluxes, weathering, and biological interactions on solute transport and mixing at pore to sub‐catchment scales. For example, integrated surface‐subsurface hydrologic models (e.g., Amanzi/ATS) coupled with reactive transport models (RTMs) can attain spatially explicit representations of critical zone components, and provide accurate estimates of geochemical exports to rivers at reach to hillslope scales (Arora, Spycher, et al, 2016; Dwivedi et al, 2018; Steefel et al, 2015; Xu et al, 2021). Despite the demonstrated success of these models, they face important challenges at larger spatial domains in part due to the computational expense from having highly resolved spatial grids and complex biogeochemical processes needed to represent heterogeneous watershed characteristics influencing water quality (Steefel, 2019).…”