Self‐potential (SP) monitoring is increasingly used for subsurface flow characterization due to its sensitivity to hydrogeological and geochemical processes. However, SP inversion remains challenging due to its ill‐posed nature, sparse data coverage, and strong transient noise. This study proposes a hybrid framework to image hyporheic exchange using a time‐lapse SP data set monitored from a streamflow site in Oak Ridge, Tennessee. Dipole moment tomography grids generated from the physics‐informed numerical inversion is first used to train a Vision Transformer (ViT) model that maps surface SP sequences to 2D source distributions. While the numerical method is more responsive to transient signals, the ViT model better captures persistent spatial structures. Their complementary outputs are jointly analyzed in the spatiotemporal domain to isolate dynamic hyporheic exchange zones and distinguish transient from steady state subsurface flow features. This approach integrates physical inversion and deep learning to enhance interpretability, generalization, and temporal awareness in SP analysis.

In south-central Texas the lower Guadalupe River has incised into the outcrop of the Carrizo-Wilcox aquifer. The river and the aquifer are hydraulically connected across the outcrop, although the connectivity is obscured at the surface by alluvium and surface-water and groundwater exchange dynamics are currently poorly understood. To investigate surface-water and groundwater exchange dynamics between the lower Guadalupe River and the Carrizo-Wilcox aquifer, a geophysical study was completed along a 14.86 km reach of the river by using water-borne gradient self-potential (SP) profiling and two-dimensional direct-current electric resistivity tomography. This paper explores the applicability of these water-borne geoelectric methods in delineating gaining and losing channel reaches, and demonstrates that geoelectric signals in the form of total electric field strength can be logged with an electric dipole and decomposed into component SP signals depicting regional and local groundwater flow patterns attributable to regional and localized hydraulic gradients. Localized SP anomalies of several tens of millivolts, indicative of hyporheic exchange flows, are observed and superimposed upon a 124 mV regional SP anomaly indicative of ambient groundwater exchange flows between the river and the aquifer. The observed SP signals are interpreted through two-dimensional finite-element modeling of streaming potentials attributable to ambient groundwater exchange and hyporheic exchange flow patterns. Variables of the channel environment such as temperature and concentration gradients, depth, and velocity are considered and subsequently eliminated as alternative sources of the SP signals that are presented.

Groundwater-flow models require the spatial distribution of the hydraulic conductivity parameter. One approach to defining this spatial distribution in groundwater-flow model grids is to map the electrical resistivity distribution by airborne electromagnetic (AEM) survey and establish a petrophysical relation between mean resistivity calculated as a nonlinear function of the resistivity layering and thicknesses of the layers and aquifer transmissivity compiled from historical aquifer tests completed within the AEM survey area. The petrophysical relation is used to transform AEM resistivity to transmissivity and to hydraulic conductivity over areas where the saturated thickness of the aquifer is known. The US Geological Survey applied this approach to a gain better understanding of the aquifer properties of the Mississippi River Valley alluvial aquifer. Alluvial-aquifer transmissivity data, compiled from 160 historical aquifer tests in the Mississippi Alluvial Plain (MAP), were correlated to mean resistivity calculated from 16,816 line-kilometers (km) of inverted resistivity soundings produced from a frequency-domain AEM survey of 95,000 km² of the MAP. Correlated data were used to define petrophysical relations between transmissivity and mean resistivity by omitting from the correlations the aquifer-test and AEM sounding data that were separated by distances greater than 1 km and manually calibrating the relation coefficients to slug-test data. The petrophysical relation yielding the minimum residual error between simulated and slug-test data was applied to 2,364 line-km of AEM soundings in the 1,000-km² Shellmound (Mississippi) study area to calculate hydraulic property distributions of the alluvial aquifer for use in future groundwater-flow models.

A multimethod geoelectric survey was implemented between January and March 2022 along a 220‐m long reach of the bedrock‐lined streambed of East Fork Poplar Creek in Oak Ridge, Tennessee to identify locations of surface‐water and groundwater exchange and characterize the subsurface flow paths that convey water between the stream and flood plain. A waterborne self‐potential (WaSP) survey was completed in January 2022 to measure the electric streaming‐potential field in the stream. Electric resistivity tomography (ERT) was performed in March 2022 on the flood plain adjacent to the WaSP survey reach to map the electric resistivity distribution and characterize the hydrogeology and subsurface flow paths that facilitate surface‐water and groundwater exchange in the bedrock‐lined stream. The combination of WaSP and ERT data support the qualitative interpretation that surface‐water and groundwater exchange likely occurs along fractures in outcropping bedrock and along two fault lines that intersect the limestone creek bed.

A self‐potential (SP) data‐inversion algorithm was developed and tested on an analytical model of electrical‐potential profile data attributed to single and multiple polarized electrical sources. The developed algorithm was then validated by an application to SP‐monitoring field data measured on the floodplain of East Fork Poplar Creek, Oak Ridge, Tennessee, to image electrical sources in areas conducive to preferential flow into the flood plain from the bedrock‐lined riverbed. The algorithm combined stochastic source‐localization by particle‐swarm‐optimization (PSO) of electrical sources characterized by simplified geometries with source tomography by regularized weighted least‐squares minimization of a quadratic objective function. Prior information was incorporated by preconditioning the tomography algorithm by PSO results. Variable percentages of random noise were added to analytical‐model data to evaluate the algorithm performance. Results indicated that true parameters of single‐source models were inverted and approximated with small residual error, whereas inversion of analytical‐model data representing multiple electrical sources accurately approximated the locations of the sources but miscalculated some parameters because of the non‐uniqueness of the inverse‐model solution. Source tomography applied to analytical model data during testing produced a spatially continuous parameter field that identified the locations of point‐scale synthetic dipole sources of electrical current flow with varying degrees of accuracy depending on the prior information incorporated into the tomography. When applied to SP‐monitoring field data, the algorithm imaged electrical sources within a known fault that intersects the bedrock riverbed and flood plain of East Fork Poplar Creek and depicted dynamic electrical conditions attributed to hyporheic exchange.