234-10 Time-Domain Spectral Induced Polarization: A Full-Waveform Approach Applied to Determining Hydrologic and Geotechnical Material Properties

Session: Advance Ground Surface Modeling for Hydrological and Environmental Applications

Presenting Author:

Jason Greenwood

Authors:

Greenwood, Jason¹, Lagmanson, Markus M.², Kiflai, Michael³, Benavides-Iglesias, Alfonso⁴

Abstract:

Spectral Induced Polarization (SIP) extends classical resistivity and IP methods by measuring the frequency-dependent polarization response of the subsurface, offering sensitivity to lithological, mineralogical, hydrological, and contaminant properties. While most SIP applications are performed in the frequency domain and interpreted using the empirical Cole-Cole model, an equivalent and physically intuitive representation exists in the time domain as a sum of exponential decays. These multi-exponential decays capture a spectrum of relaxation processes governed by a material’s electrochemical and pore-scale properties, enabling detailed characterization when coupled with robust signal acquisition and processing. In this study, we implement a time-domain SIP approach using digitally oversampled voltage decay curves processed through waveform analysis to enhance signal-to-noise ratios. We developed a new laboratory test cell that accommodates heterogeneous mixtures of soils and fluids while minimizing electrode polarization effects. Decay curves are fitted using nonlinear least squares with automatic model selection based on the Bayesian Information Criterion (BIC), supported by Tikhonov regularization to stabilize the estimation of decay-rate spectra. This method yields a reliable time-domain representation and can be transformed into frequency-domain Cole-Cole parameters. Laboratory tests were conducted on soil samples of varying textures, moisture contents, and compaction levels, as well as on materials contaminated with different concentrations of DNAPL and LNAPL. The high-resolution transient decay curves captured using our signal processing approach enabled discrimination of material-specific relaxation behaviors and chargeability patterns. The resulting decay-rate spectra proved effective in resolving subtle contrasts between lithologic units and identifying clean versus contaminated zones. These results suggest that time-domain spectral IP provides a forward-compatible framework for enhanced geoelectrical imaging in 2D, 3D, and time-lapse surveys. The technique holds significant promise for environmental assessments, groundwater investigations, and resource exploration, especially where frequency-dependent polarization effects convey valuable information about subsurface composition and fluid content.

Geological Society of America Abstracts with Program. Vol. 57, No. 6, 2025

doi: 10.1130/abs/2025AM-4836

© Copyright 2025 The Geological Society of America (GSA), all rights reserved.