Accelerated 3D Electrical Resistivity Tomography with a Scalable Jacobian-free Approach

TL;DR

This paper introduces a scalable, Jacobian-free inversion method for 3D electrical resistivity tomography that enables high-resolution subsurface imaging without intrusive code modifications, significantly improving computational efficiency.

Contribution

The paper applies the Principal Component Geostatistical Approach (PCGA) to 3D ERT, demonstrating a non-intrusive, scalable inversion method linked with existing software, facilitating practical high-resolution subsurface imaging.

Findings

High-resolution 3D subsurface characterization is computationally feasible.

The method is scalable and does not require intrusive code changes.

Implementation is available in a public repository.

Abstract

A Jacobian-free inversion method is presented to accelerate Electrical Resistivity Tomography (ERT) for shallow aquifer characterization. The ERT problem typically implements the adjoint state method to efficiently compute Jacobian during the inversion. However, the adjoint state method needs intrusive forward model code changes and may not be computationally scalable with many observations especially when one performs 3D ERT surveys with dense multi-electrode arrays. Here the Principal Component Geostatistical Approach (PCGA), a fast and scalable Jacobian-free inverse modeling method, is applied to solve a high dimensional data-intensive ERT problem. The PNNL's ERT simulation software E4D was linked to the python interface pyPCGA without intrusive code change and the example code is upload in a public repository. The result in this study shows that high-resolution 3D subsurface characterization is computationally feasible, which would have a great potential for implementations in practice.