Gravity inversion constrained by probabilistic magnetotellurics models: methodology and application

Giraud, Dr Jeremie Eugene Cyril1,2, Seillé, Dr Hoël3, Ciolczyk,Damien4, Grose, Dr Lachlan5, Visser, Dr Gerhard3, Lindsay, Dr Mark1,2, Jessell, Dr Mark1,2, Ogarko, Dr Vitaliy6

1Centre For Exploration Targeting, School Of Earth Sciences, University of Western Australia, Crawley, Australia, 2Mineral Exploration Cooperative Research Centre, School Of Earth Sciences, University of Western Australia, Crawley, Australia, 3CSIRO Deep Earth Imaging Future Science Platform, Kensington, Australia, 4Université de Strasbourg, School and Observatory of Earth Sciences (EOST), Strasbourg, France, 5School of Earth, Atmosphere and Environment, Monash University, Clayton, Australia, 6The International Centre for Radio Astronomy Research, The University of Western Australia, Crawley, Australia

In this contribution we introduce an inversion workflow where we integrate magnetotelluric (MT) data, which are primarily sensitive to horizontal resistivity interfaces, together with gravity  measurements, which are well suited for the recovery of lateral density variations. We connect these data using petrophysical prior information and geological principles. The approach we
present relies on a flexible, cooperative workflow where the different datasets are modelled using standalone algorithms. The workflow consists of the following three steps.

First, we perform the inversion of MT data in a 1D probabilistic fashion. For each MT site, the results include an ensemble of plausible 1D models, from which the probability of having an
interface between rock units of varying electrical conductivity is calculated. Second, these probabilities are interpolated to the whole study area using the implicit geological simulator LoopStructural. During this process, geological rules and principles (stratigraphy, superposition and cross-cutting relationships) are used to ensure that the models resulting from such probabilities are geologically plausible. Finally, domains corresponding to different rock units are derived using interfaces’ probabilities and are used in combination with petrophysical information to define the range of density contrast values allowed at every location in the model. These ranges are used to constrain deterministic gravity inversion using the alternating direction method of multipliers as implemented in the Tomofast-x engine.

We first summarise the methodology and present the proof-of-concept using a realistic synthetic model built using geological data from the Mansfield area (Victoria, Australia). Results demonstrate that the proposed workflow can effectively leverage the complementarity between geophysical methods relying on different physical phenomena such as MT and gravity and improve imaging. We then show preliminary results modelling existing real-world data that crosses the Eucla Basin (Western Australia), which is a region that is receiving increasing attention for its undercover mineral potential and where prior geological knowledge suggests complementarity between MT and gravity.

We acknowledge the support of the MinEx CRC and the Loop: Enabling Stochastic 3D Geological Modelling (LP170100985) consortia, and of the CSIRO Deep Earth Imaging Future Science Platform. The work has been supported, in part, by the Mineral Exploration Cooperative Research Centre whose activities are funded by the Australian Government’s Cooperative Research Centre Programme. This is MinEx CRC Document 2020/xx


Biography

Jeremie Giraud is a research fellow at the Centre for Exploration Targeting. His research efforts focus on the integration of geophysics and geology and on multi-physics integration.

About the GSA

The Geological Society of Australia was established as a non-profit organisation in 1952 to promote, advance and support Earth sciences in Australia.

As a broadly based professional society that aims to represent all Earth Science disciplines, the GSA attracts a wide diversity of members working in a similarly broad range of industries.