GSA 2020 Connects Online

Paper No. 253-5
Presentation Time: 11:20 AM

UNCERTAINTIES IN HYPORHEIC ZONE SOLUTE TRANSPORT ESTIMATED FROM ELECTRICAL RESISTIVITY INVERSIONS: A NUMERICAL STUDY


SAWYER, Audrey H., School of Earth Sciences, Ohio State University, 125 Oval Dr. S, Columbus, OH 43210 and SINGHA, Kamini, Hydrologic Science and Engineering Program, Colorado School of Mines, Golden, CO 80401

Hyporheic zones retain solutes and influence stream water quality, a process that has traditionally been studied with solute tracer injections. Over the last decade, a growing number of injections have been paired with time-lapse resistivity imaging to map the arrival and flushing of solute within the hyporheic zone. Inversion results like pixel breakthrough curves can extend our understanding of hyporheic processes but are also influenced by inversion artifacts. Here, we use a synthetic modeling approach to assess how changes in tracer concentration and tracer-impacted hyporheic area are estimated by electrical resistivity imaging. For a synthetic channel cross-section, we create a coupled fluid flow and solute transport model to simulate solute movement during an in-stream tracer test and solve for electrical flow and voltage readings during a corresponding time-lapse electrical resistivity survey. We then invert the synthetic resistance data to obtain electrical conductivity images for comparison with simulated solute distributions. We test different stream dimensions, solute plateau concentrations, and injection periods. During the early part of the detection period, the area of the hyporheic zone that contains the salt tracer is overestimated due to smoothing in the inversion, but the tracer-impacted area rapidly diminishes and becomes underestimated as the tracer plume spreads within the hyporheic zone and becomes a more diffuse target. The sensitivity and accuracy of bulk electrical conductivity at inversion pixels varies substantially with channel size, injection design, and pixel location. Unlike inversions, the time-series of average apparent bulk electrical conductivity provides a high-fidelity representation of lingering salt in the stream system but is not spatially explicit. We caution against quantitative analysis of time-lapse inversions for hyporheic area and recommend injecting low concentrations of salt for longer time periods to minimize inversion artifacts.