Trans-dimensional geometrical gravity inversion and posterior analysis: application to undercover exploration in the Boulia region (Queensland, Australia).

Jeremie Giraud and Marc Lindsay and Vitaliy Ogarko and Guillaume Caumon and Leonardo Portes dos Santos and Guillaume Pirot and Mahtab Rashidifard and Paul Cupillard and Julien Herrero. ( 2026 )
in: 2026 {RING} meeting, pages 352--382, ASGA

Abstract

Imaging the subsurface beneath cover remains a fundamental challenge in mineral exploration because direct observation of geological features of interest is often impossible. In this context, geophysical inversion is well-suited to derive subsurface models of these geological features. However, the geophysical solution’s non-uniqueness makes it difficult to estimate uncertainty. To alleviate this, geometrical three-dimensional (3D) trans-dimensional (trans-d) Markov chain Monte Carlo (MCMC) inversion offers a probabilistic framework, as it allows exploring a broad range of scenarios by perturbing geological parameters. The algorithm employs a birth–death process that can add or remove rock units, together with random perturbations of geometry and density, sampling the posterior over models consistent with the data, prior information, and constraints. The first field application using this approach is presented, using gravity data from a prospective exploration area in the Boulia region (Queensland, Australia). Overall, 3 million 3D models of varying complexity are generated. To synthesise this ensemble, an interpretation workflow is presented that combines topological analysis, feature extraction via a 3D convolutional encoder, and dimensionality reduction. Clustering of the resulting features identifies model families, enabling a compact visualisation of the model space. Results indicate the persistence of one or two ‘birthed’ units not present a priori: relatively shallow, N–S-elongated bodies of higher density intruding into horizontal layers. This suggests the presence of relatively shallow buried intrusive units that have otherwise not been recognised until now and demonstrates the usefulness of the proposed methods in undercover scenarios.

Download / Links

BibTeX Reference

@inproceedings{Giraud2026RM,
 abstract = {Imaging the subsurface beneath cover remains a fundamental challenge in mineral exploration because direct observation of geological features of interest is often impossible. In this context, geophysical inversion is well-suited to derive subsurface models of these geological features. However, the geophysical solution’s non-uniqueness makes it difficult to estimate uncertainty. To alleviate this, geometrical three-dimensional (3D) trans-dimensional (trans-d) Markov chain Monte Carlo (MCMC) inversion offers a probabilistic framework, as it allows exploring a broad range of scenarios by perturbing geological parameters. The algorithm employs a birth–death process that can add or remove rock units, together with random perturbations of geometry and density, sampling the posterior over models consistent with the data, prior information, and constraints. The first field application using this approach is presented, using gravity data from a prospective exploration area in the Boulia region (Queensland, Australia). Overall, 3 million 3D models of varying complexity are generated. To synthesise this ensemble, an interpretation workflow is presented that combines topological analysis, feature extraction via a 3D convolutional encoder, and dimensionality reduction. Clustering of the resulting features identifies model families, enabling a compact visualisation of the model space. Results indicate the persistence of one or two ‘birthed’ units not present a priori: relatively shallow, N–S-elongated bodies of higher density intruding into horizontal layers. This suggests the presence of relatively shallow buried intrusive units that have otherwise not been recognised until now and demonstrates the usefulness of the proposed methods in undercover scenarios.},
 author = {Giraud, Jérémie and Lindsay, Marc and Ogarko, Vitaliy and Caumon, Guillaume and Portes dos Santos, Leonardo and Pirot, Guillaume and Rashidifard, Mahtab and Cupillard, Paul and Herrero, Julien},
 booktitle = {2026 {RING} meeting},
 language = {en},
 pages = {352--382},
 publisher = {ASGA},
 title = {Trans-dimensional geometrical gravity inversion and posterior analysis: application to undercover exploration in the Boulia region (Queensland, Australia).},
 year = {2026}
}