Machine-learning models to map pH and redox conditions in groundwater in a layered aquifer system, Northern Atlantic Coastal Plain, eastern USA

Leslie A. DeSimone; Jason P. Pope; Katherine Marie Ransom

doi:10.1016/j.ejrh.2020.100697

Home/Archive/USGS

USGS · 70217072

Machine-learning models to map pH and redox conditions in groundwater in a layered aquifer system, Northern Atlantic Coastal Plain, eastern USA

Leslie A. DeSimone·Jason P. Pope·Katherine Marie Ransom

View original source ↗Export metadata ↓

Abstract

Study region The study was conducted in the Northern Atlantic Coastal Plain aquifer system, in the eastern USA. Study focus Groundwater pH and redox conditions are fundamental chemical characteristics controlling the distribution of many contaminants of concern for drinking water or the ecological health of receiving waters. In this study, pH and redox conditions were modeled and mapped in a complex, layered aquifer system. Machine-learning methods (boosted regression trees) were applied to data from 3000 to 5000 wells. Predicted pH and the probability of anoxic conditions, defined by three thresholds of dissolved oxygen (0.5, 1, and 2 mg/L), were mapped at the 1-km 2 scale for each of 10 regional aquifer layers. New Hydrological Insights for the Region Maps depict the extent of acidic groundwater and oxic conditions in the shallow, unconfined surficial aquifer and in unconfined, recharge-proximal areas of underlying aquifers, in contrast to alkaline and anoxic groundwater elsewhere. Geographic patterns and influential predictors–including elevation, overlying confining-units thickness, and simulated groundwater age and flux–are consistent with prior understanding of the processes controlling pH and redox in the aquifer system. The model-based maps support robust estimates of aquifer proportions, either areal or volumetric, likely to contain groundwater of a specified quality or be vulnerable to specific pH- or redox-sensitive contaminants. The machine-learning methods were an effective tool to map groundwater quality at the regional scale.

Explore related subjects

Journal of Hydrology: Regional Studies Article

Source-reported bounding extent: 36.35052700542763° to 40.3130432088809° latitude; -76.6845703125° to -74.0478515625° longitude. This indicates report coverage, not an exact sampling location. View area on OpenStreetMap.

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Leslie A. DeSimone, Jason P. Pope, Katherine Marie Ransom. 2020. Machine-learning models to map pH and redox conditions in groundwater in a layered aquifer system, Northern Atlantic Coastal Plain, eastern USA. https://doi.org/10.1016/j.ejrh.2020.100697

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related USGS reports

USGS2026

Assessing future hydrologic extremes using an integrated hydrology and river operations model in the Russian River watershed

Study region The Russian River watershed, situated in coastal, northern California, experiences hydrologic extremes, including periodic droughts and flooding. Water managers are working to maintain sustainable water supplies and environmental flows, while mitigating flood risks. Study focus This paper introduces an integrated hydrology and river operations model for the Russian River watershed. This model is distinct from models in previous studies because it represents surface-groundwater interactions and uses climate forcings to estimate dynamic water use demands that are superimposed onto both reservoir operations and water supply constraints. The model was used to examine three historical (1990–2015) and eight future (2016–2099) water use and climate change scenarios. New hydrological insights for the region The direct connection between streams and aquifers facilitated both annual aquifer replenishment by high winter streamflows and streamflow depletion by groundwater wells (19 % of pumped groundwater in alluvial aquifers from stream leakage) during critical low flow periods. Simulated streamflow changes included 59 % longer and 54 % more severe streamflow droughts, 26 % lower seasonal low streamflows, and up to 125 % higher peak streamflows, averaged over future climate and water use scenarios, suggesting increased future flood and water availability risks. Results showed the importance of reservoir operations for mitigating the impacts of increased hydroclimatic volatility, despite a decrease in reservoir reliability at Lake Mendocino, suggesting that reservoir management may be used to decrease future risks.

California↗

USGS2026

Snow simulations predict future changes in rain-on-snow events across the upper Gallatin River watershed, a Greater Yellowstone Ecosystem headwater system

Study region: The upper Gallatin River watershed, an alpine headwater system in the Greater Yellowstone Ecosystem, in Wyoming and Montana. Study focus: As global and regional air temperatures rise, mountain headwaters across the Greater Yellowstone Ecosystem (GYE) are projected to see more precipitation falling as rain. While the hydrologic effects of this snow-to-rain transition depends on a variety of factors, it can lead to an increased occurrence of rain-on-snow (RoS) events. To investigate these changes, we used high-resolution (30 m) SnowModel simulations of the upper Gallatin River watershed. Simulations were run for 2001-2013 using two scenarios: (1) historical meteorology as control and (2) pseudo global warming (PGW) where control air temperature and precipitation conditions were perturbed to represent mean end-of-century conditions under a high-emissions scenario. New hydrological insights for the region: SnowModel outputs show that changes in PGW precipitation and snow accumulation varied with elevation. Warmer air temperatures at low elevations (< 2,500 m) led to less snow accumulation and less precipitation falling as snow. Colder baseline air temperatures for elevations above 2,500 meters (m) resulted in minor reductions in winter snowfall fraction. For PGW simulations, spring (April-June) months were rainier, and elevations above 2,500 m experienced more RoS events. Snowpacks between 2,500-3,100 m generated more snowmelt during RoS events, which was reflected in the watershed average. More high-intensity melt events can affect aquatic habitat, water quality, and the accuracy of streamflow forecasts across the region.

Montana, Wyoming↗

USGS2026

VegET evapotranspiration for Africa: Continental-scale simulation, multi-product evaluation, and drought assessment

Study region Continental Africa, encompassing diverse climatic zones—tropical, arid, and temperate—and spanning major transboundary river basins such as the Nile, Niger, Congo, Volta, and Zambezi River Basins. The region exhibits pronounced hydroclimatic gradients and heterogeneous land use systems ranging from rainfed croplands and rangelands to dense tropical forests and irrigated schemes. Study focus Actual evapotranspiration (ETa) is a central component of the terrestrial water balance, governing the redistribution of water and energy between the land surface and the atmosphere. Accurate estimation of ETa at continental scale is critical for hydrological monitoring, water resource management, and climate adaptation, as well as for quantifying water, energy, and carbon fluxes that underpin sustainable development. In this study, we applied the agro-hydrologic VegET v2 model to simulate a new, high-resolution, continental-scale ETa dataset for Africa (2000–2021). The model results were benchmarked against four widely used remote sensing-based products—MODIS16 v6.1, SSEBop v6.1, WaPOR v3, and GLEAM v4.1a—across major climate zones, land use types, and River Basins, providing a comprehensive multi-product evaluation of evapotranspiration dynamics across the continent. New hydrological insights for the region Validation against eddy covariance flux tower observations at eight representative sites confirmed that VegET v2 accurately reproduces the seasonal dynamics of observed ETa, achieving a correlation (r) of 0.8 and an RMSE of 25 mm month⁻¹ —accuracy that is comparable to or higher than accuracies of satellite-based products MODIS16, SSEBop, and GLEAM. This study represents one of the first Africa-wide hydrological simulations of ETa, extending the VegET model beyond basin-scale applications. Intercomparisons reveal that VegET aligns closely with MODIS16, SSEBop, and GLEAM in humid and tropical regions (r = 0.80–0.90; RMSE < 20 mm month⁻¹), while greater discrepancies appear in arid and semi-arid zones, where WaPOR tends to overestimate ETa (RMSE ≥ 28 mm month⁻¹). Despite these differences, VegET effectively captures spatial and temporal ETa variability across rainfed croplands, forests, and savannas, supporting its utility in regional water balance assessments, water accounting, and drought monitoring. A key application of VegET v2 is the Evapotranspiration Deficit Index (ETDI), derived by integrating VegET-based ETa with potential evapotranspiration (PET) to quantify water stress. ETDI successfully captured major drought episodes across Africa, including persistent Sahelian and southern African dry spells, the 2020–2021 winter drought in the Maghreb, and the 2018–2019 austral summer drought in southern Africa, while identifying positive anomalies over central Africa indicative of recurrent wetness. These results underscore VegET’s capability as a hydrologically consistent, operational tool for continental ETa monitoring and drought assessment, offering support for basin-scale water balance studies, food security planning, and climate resilience across Africa’s diverse hydrological environments.

Journal of Hydrology: Regional Studies↗