Aqueous PFAS concentrations fluctuate seasonally and precursor concentrations rapidly decline within the surface-water/groundwater boundary, indicating that transport through this boundary can affect spatial and temporal PFAS composition.

Emerging groundwater contaminants such as per-and polyfluoroalkyl substances (PFAS) may impact surface-water quality and groundwater-dependent ecosystems of gaining streams. Although complex near-surface hydrogeology of stream corridors challenges sampling efforts, recent advances in heat tracing of discharge zones enable efficient and informed data collection. For this study, we used a combination of streambed temperature push-probe and thermal infrared methods to guide a discharge-zone-oriented sample collection along approximately 6 km of a coastal trout stream on Cape Cod, MA. Eight surface-water locations and discharging groundwater from 24 streambed and bank seepages were analysed for dissolved oxygen (DO), specific conductance, stable water isotopes, and a range of PFAS compounds, which are contaminants of emerging concern in aquatic environments. The results indicate a complex system of groundwater discharge source flowpaths, where the sum of concentrations of six PFAS compounds (corresponding to the U.S. Environmental Protection Agency third Unregulated Contaminant Monitoring Rule "UCMR 3") showed a median concentration of 52 ± 331 (SD) ng/L with two higher outliers and three discharges with PFAS concentrations below the quantification limit. Higher PFAS concentration was related (−0.66 Spearman rank, p < .001) to discharging groundwater that showed an evaporative signature (deuterium excess), indicating flow through at least one upgradient kettle lake. Therefore, more regional groundwater flowpaths originating from outside the local river corridor tended to show higher PFAS concentrations as evaluated at their respective discharge zones. Conversely, PFAS concentrations were typically low at discharges that did not indicate evaporation and were adjacent to steep hillslopes and, therefore, were classified as locally recharged groundwater. Previous research at this stream found that the native brook trout spawn at discharge points of groundwater recharged on local hillslopes, likely in response to generally higher levels of DO. Our study shows that by targeting high oxygen discharges the trout may thereby be avoiding emerging contaminants such as PFAS in groundwater recharged farther from the stream.

Integrated hydrologic models solve coupled mathematical equations that represent natural processes, including groundwater, unsaturated, and overland flow. However, these models are computationally expensive. It has been recently shown that machine leaning (ML) and deep learning (DL) in particular could be used to emulate complex physical processes in the earth system. In this study, we demonstrate how a DL model can emulate transient, three-dimensional integrated hydrologic model simulations at a fraction of the computational expense. This emulator is based on a DL model previously used for modeling video dynamics, PredRNN. The emulator is trained based on physical parameters used in the original model, inputs such as hydraulic conductivity and topography, and produces spatially distributed outputs (e.g., pressure head) from which quantities such as streamflow and water table depth can be calculated. Simulation results from the emulator and ParFlow agree well with average relative biases of 0.070, 0.092, and 0.032 for streamflow, water table depth, and total water storage, respectively. Moreover, the emulator is up to 42 times faster than ParFlow. Given this promising proof of concept, our results open the door to future applications of full hydrologic model emulation, particularly at larger scales.

No abstract

The water content in the soil regulates exchanges between soil and atmosphere, impacts plant livelihood, and determines the antecedent condition for several natural hazards. Accurate soil moisture estimates are key to applications such as natural hazard prediction, agriculture, and water management. We explore how to best predict soil moisture at a high resolution in the context of a changing climate. Physics-based hydrological models are promising as they provide distributed soil moisture estimates and allow prediction outside the range of prior observations. This is particularly important considering that the climate is changing, and the available historical records are often too short to capture extreme events. Unfortunately, these models are extremely computationally expensive, which makes their use challenging, especially when dealing with strong uncertainties. These characteristics make them complementary to machine learning approaches, which rely on training data quality/quantity but are typically computationally efficient. We first demonstrate the ability of Convolutional Neural Networks (CNNs) to reproduce soil moisture fields simulated by the hydrological model ParFlow-CLM. Then, we show how these two approaches can be successfully combined to predict future droughts not seen in the historical timeseries. We do this by generating additional ParFlow-CLM simulations with altered forcing mimicking future drought scenarios. Comparing the performance of CNN models trained on historical forcing and CNN models trained also on simulations with altered forcing reveals the potential of combining these two approaches. The CNN can not only reproduce the moisture response to a given forcing but also learn and predict the impact of altered forcing. Given the uncertainties in projected climate change, we can create a limited number of representative ParFlow-CLM simulations (ca. 25 min/water year on 9 CPUs for our case study), train our CNNs, and use them to efficiently (seconds/water-year on 1 CPU) predict additional water years/scenarios and improve our understanding of future drought potential. This framework allows users to explore scenarios beyond past observation and tailor the training data to their application of interest (e.g., wet conditions for flooding, dry conditions for drought, etc…). With the trained ML model they can rely on high resolution soil moisture estimates and explore the impact of uncertainties.