2.5D crosshole GPR full-waveform inversion including water-filled boreholes and finite-length antennae in subgrids

Crosshole ground penetrating radar (GPR) full-waveform inversion (FWI) provides high resolution estimates of dielectric permittivity and electrical conductivity images, with a wide range of applications. Existing implementations predominantly rely on 2D forward modeling with point sources and receivers, requiring a 3D-to-2D data transformation and neglecting the presence of water-filled boreholes. Artifacts caused by unmodeled water-filled boreholes have led to the common exclusion of high-angle data beyond 45° relative to horizontal, sacrificing vertical resolution that these data could otherwise provide. Together, these simplifications and exclusions introduce systematic modeling errors that propagate into the final inversion results. A time-domain 2.5D GPR FWI framework is presented that extends a conventional 2D inversion scheme with 3D forward modeling. Water-filled boreholes and finite-length antennas are incorporated efficiently via subgridding, such that the known components of the acquisition setup are correctly modeled and the main unknowns are the subsurface properties between the boreholes. This eliminates the need for a 3D-to-2D transformation and removes the approximations responsible for systematic modeling errors. The framework is validated using synthetic data generated for a realistic model derived from a well-characterized test site, incorporating high-contrast permittivity and conductivity structures. Systematic comparison of conventional 2D FWI, 2.5D FWI, and full 2.5D FWI including boreholes and finite-length antennas demonstrates clear improvements in data fit. Including water-filled boreholes prevents upward bias in reconstructed permittivity caused by smearing of borehole fluid properties into the inversion domain, while incorporating finite-length antennas significantly improves conductivity reconstruction. High-angle data can be included without introducing artifacts. The proposed framework establishes a more rigorous foundation for crosshole GPR FWI, advancing the reliability of subsurface imaging for critical applications including groundwater characterization, contaminant monitoring, and subsurface site investigation.

Authors

Institutions

Publication Details

Journal
Geophysics
Published
2026-11-18
DOI
https://doi.org/10.1190/geo-2026-1965
Primary Topic
Geophysical Methods and Applications
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

2.5D crosshole GPR full-waveform inversion including water-filled boreholes and finite-length antennae in subgrids

Dominik Hoven, Craig Warren, Harry Vereecken, Anja Klotzsche et al.
1 links
Geophysics
Geophysical Methods and Applications
article

2.5D crosshole GPR full-waveform inversion including water-filled boreholes and finite-length antennae in subgrids

Dominik Hoven, Craig Warren, Harry Vereecken, Anja Klotzsche, Jan van der Kruk
article en

Abstract

Crosshole ground penetrating radar (GPR) full-waveform inversion (FWI) provides high resolution estimates of dielectric permittivity and electrical conductivity images, with a wide range of applications. Existing implementations predominantly rely on 2D forward modeling with point sources and receivers, requiring a 3D-to-2D data transformation and neglecting the presence of water-filled boreholes. Artifacts caused by unmodeled water-filled boreholes have led to the common exclusion of high-angle data beyond 45° relative to horizontal, sacrificing vertical resolution that these data could otherwise provide. Together, these simplifications and exclusions introduce systematic modeling errors that propagate into the final inversion results. A time-domain 2.5D GPR FWI framework is presented that extends a conventional 2D inversion scheme with 3D forward modeling. Water-filled boreholes and finite-length antennas are incorporated efficiently via subgridding, such that the known components of the acquisition setup are correctly modeled and the main unknowns are the subsurface properties between the boreholes. This eliminates the need for a 3D-to-2D transformation and removes the approximations responsible for systematic modeling errors. The framework is validated using synthetic data generated for a realistic model derived from a well-characterized test site, incorporating high-contrast permittivity and conductivity structures. Systematic comparison of conventional 2D FWI, 2.5D FWI, and full 2.5D FWI including boreholes and finite-length antennas demonstrates clear improvements in data fit. Including water-filled boreholes prevents upward bias in reconstructed permittivity caused by smearing of borehole fluid properties into the inversion domain, while incorporating finite-length antennas significantly improves conductivity reconstruction. High-angle data can be included without introducing artifacts. The proposed framework establishes a more rigorous foundation for crosshole GPR FWI, advancing the reliability of subsurface imaging for critical applications including groundwater characterization, contaminant monitoring, and subsurface site investigation.

Geophysics
Forschungszentrum Jülich (DE), Northumbria University (GB)
Geophysical Methods and Applications
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon