We consider the transient electrokinetic response of an aquifer-aquitard system to groundwater abstraction from the aquifer. The system was instrumented with 18 non-polarizable copper/copper sulphate electrodes installed at three different depths in the aquitard above the aquifer. The sensing electrodes were installed at depth of 1, 2, and 4 m below ground surface along three overlapping transects. The differential voltages relative to a single permanent electrode were measured with a Campbell Scientific CR1000 datalogger with a single multiplexer. Additionally, six piezometers screened in the top 1.5 m of the confined aquifer, were installed by direct-push. All the piezometers were instrumented with pressure transducers to measure directly the hydraulic response of the aquifer. The vertical variation of resistivity in the aquitard was measured on sediment cores recovered from one of the boreholes used to install the deepest piezometer. The resistivity distribution at antecedent sediment wetness (moisture content) was measured using the MC Miller resistivity boxes and meter, with wetness measured gravimetrically. Previous exploratory drilling and sampling activities at the site indicate that the aquifer is fractured greywacke sandstone overlain with clayey aquitard of semi-consolidated alluvial sediment, with the aquifer-aquitard contact at a depth of 10.3 m below the ground surface. We report the results of the site instrumentation, monitoring, characterization, hydraulic testing, and compare the results of parameter estimation using streaming potential and hydraulic data separately and jointly. We explore the effect of depth of installation of the electrodes in the aquitard on signal strength and quality and compare this to model predicted behavior using semi-analytical models from the literature. The results suggest the need for deep sensing of electrokinetic signals generated by groundwater flow to improve signal-to-noise ratios and the usefulness of self-potential data for hydrogeophysical characterization of aquifer-aquitard systems.

To close the water use budget in irrigated agricultural fields in flood plains with substantial riparian corridors, it is necessary to understand groundwater usage by dominant phreatophyte vegetation, particularly when the primary source of the water for irrigation comes from groundwater abstraction. We report here results of measurements of sap flow in phreatophyte vegetation in a riparian corridor, which is part of a watershed located along the coast in Santa Cruz County, California. The riparian corridor is within a study area of 75 to 140 meters wide in the lower portion of Scotts Creek watershed, which is bounded to the west by the Pacific Ocean. Canopy cover in the study area often approaches 100 percent, with dominant trees being red alder (Alnus rubra), arroyo willow (Salix lasiolepis), and pacific willow (Salix lasiandra var. lasiandra). Other trees include boxelder (Acer negundo), big leaf maple (Acer macrophyllum), and California bay laurel (Umbellularia californica). Common understory vegetation includes California blackberry (Rubus ursinus), stinging nettle (Urtica dioica subsp. gracilis), poison hemlock (Conium maculatum), Cape ivy (Delairea odorata), and Italian thistle (Carduus pycnocephalus). For the study reported here, only the two most dominant phreatophyte species, namely red alders and arroyo willows, were instrumented with sap flow sensors. In addition to diurnal fluctuations, sap flow data collected hitherto also shows expected seasonal variation with summer maxima and winter minima, with transition fall and spring periods. Sap flow measurements from the study area are projected across the entire riparian forest using sampled tree sapwood area and used to estimate forest evapotranspiration (ET). The ET is then used in a groundwater flow model to more accurately predict observed groundwater fluctuations and usage by riparian vegetation.

We present an approach to uncoupling the pair of transient governing equations used in electrokinetics (i.e., streaming potential and electroosmosis). This approach allows for the solution of two uncoupled “intermediate” equations, then the physical solution is found by recombination of these intermediate potentials through a matrix multiplication. We present numerically stable expressions for the coefficients, and an example showing electrokinetics arising from pumping a fully penetrating well in a confined aquifer, surrounded by insulating aquicludes. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525. (SAND2019-8712 A)

We present preliminary results from an electrokinetic laboratory rock characterization approach, which uses electroosmosis and streaming potential to estimate permeability. The device has two modes. In its first mode it measures differential pressure across a sample while applying electric current or acoustic pressure, while in its second mode it measures voltage across the sample when applying electric current or acoustic pressure. The two modes result in estimates of the electrokinetic coupling coefficients, which can jointly be used to estimate the permeability of the sample. Our collaboration includes analytical and numerical approaches simulating aspects of the experiments being carried out in the laboratory at Sandia. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Mathematical models for stream depletion typically use the constant-head Dirichlet boundary condition or the general Robin boundary condition at the stream. Both approaches fix stream stage as constant during pumping. Fixed the stream stage implies the stream acts as an infinite water source with depletion affecting stream discharge but having no impact on stream stage. We refer to this depletion without drawdown as the “depletion paradox.” It is a glaring model limitation, ignoring the most observable adverse effect of long-term groundwater abstraction near a stream – dry streambeds. Our field data demonstrate that stream stage responds to pumping near the stream. This motivates the development of a model considering transient stream drawdown using the concepts of finite stream storage and mass continuity at the stream-aquifer interface. The models include the cases for fully- and non-penetrating the stream. First-order mass transfer is also assumed across the streambed. The proposed model reduces to the fixed-stage model as stream storage becomes infinitely large and the confined flow case with a no-flow boundary at the streambed when stream storage vanishes. Sensitivity analysis for hydraulic properties of the stream-aquifer system is also included. Our results suggest that fixed-stage models (a) underestimate late-time aquifer drawdown to pumping adjacent to a stream and (b) overestimate the available groundwater supply from streams to pumping wells because of the infinite stream storage assumption. This can have significant implications for the sustainable management of water resources in interacting stream-aquifer systems with heavy groundwater abstraction.