Abstract This study examined mountain-front stream recharge to a karst aquifer in the Denver Basin (USA) from 2016 to 2023, a period of minimal water use, using a combination of existing data (geologic maps, borehole drilling and logging records, geophysical surveys, and hydrochemical stream and groundwater data) and new monitoring efforts (stream gauging, near-stream drilling, and water level (WL) instrumentation at stream gauges and monitoring wells). Stream and karst aquifer WLs were jointly analyzed in the context of all data. In a northern subwatershed, stream and aquifer WLs followed annual snowmelt cycles by rising in spring and falling in summer. Where the streambed intersects the aquifer, the stream recharges the aquifer through fractures and conduits at sinks, as supported by drilling data and similarity of hydrochemical composition and electrical conductivity of stream water at a sink and groundwater in wells located > 1 km downstream. Moreover, during a 2017–2022 drought that reduced streamflow, the karst aquifer received less recharge and its WLs declined. In two southern subwatersheds, recharge to the confined portion of the karst aquifer likely comes from both stream losses and subsurface inflow, whereas recharge to an unconfined portion of the aquifer could not be determined from the available data. Despite regional precipitation data suggesting that the drought ended in 2023, below-average mountain precipitation that year may have caused continued streamflow and aquifer depletion at the mountain front, suggesting the reliance of the Denver Basin water supply on mountain snowmelt runoff.

Astract Coastal aquifers are under increasing threat due to excessive groundwater withdrawal, resulting in freshwater depletion, seawater intrusion, and land subsidence. To address these issues, the Sustainable Water Initiative for Tomorrow (SWIFT) Research Center in Suffolk, Virginia (USA), has been testing managed aquifer recharge (MAR) in the Potomac Aquifer of southeast Virginia since 2018. The Potomac Aquifer is hydraulically connected to the underlying crystalline basement, which may allow injection-induced pressure transients from MAR operations to propagate into the underlying basement rock, potentially inducing a seismic response. To begin understanding the magnitude and extent of basement pressure propagation from MAR operations in Virginia, this study implements ensemble simulation methods to reproduce the pilot-scale SWIFT aquifer recharge program from 2018 to 2022, where the a priori spatial distribution of basement permeability is unknown. The simulation ensemble quantifies basement fluid pressure propagation and corresponding uncertainty for five injection scenarios that are based on the measured injection volume during the pilot-scale recharge program, as well as the permitted rate of 3.79 × 10 3 m 3 /day, and the anticipated injection rate (7.57 × 10 3 m 3 /day) for a single injector when the SWIFT project becomes fully operational. Results show that pressure change of ~ $$\Delta$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mi>Δ</mml:mi> </mml:math> 40 kPa may reach the basement at injection rates of 7.57 × 10 3 m 3 /day, which is the anticipated injection rate for full-scale SWIFT injection wells. Moreover, uncertainty analysis reveals that this basement pressure build-up is localized to a 2-km radius of the injector within the upper 200 m of the underlying basement.