Masterclass with Jeremie Giraud

17 February 2023,
Institute of Advanced Studies, UWA
Postgraduate Students, Early Career Researchers, Academics, Professional Researchers

Masterclass - methods for the integration of geological modelling and geophysical inversion

The masterclass will be facilitated jointly between Dr Jeremie Giraud and Professor Mark Jessell's team, who will be presenting recent research work. The goal is to bring together a group of interested geoscientists and specialists to showcase and discuss recent advances in numerical geosciences, integrated geological and geophysical modelling.

The focus will be on methods for the integration of geological modelling and geophysical inversion as part of work carried out in Jeremie Giraud's current project and around the Australia-based Loop3D collaborative consortium.

The first part of the masterclass will present recent developments of the Tomofast-x open-source inversion platform, with an application example using MT and magnetic data from the Cloncurry district (QLD) for the characterisation of the depth-to-basement interface. The second part of the masterclass will be focusing on the integration of implicit geological modelling into geophysical inversion, with an extension to the consideration of features unseen by geology and to the effect of the geophysical inversion problem's non-uniqueness.


13:00 - 13:05  Introduction to Masterclass (Jeremie Giraud)

13:05 - 13:15  Presentation of participants (All)

13:15-13:45    Integrated inversion platform Tomofast-x (Jeremie Giraud)

13:45- 14:15  MT-mag for depth to basement inversion (Jeremie Giraud)

14:15-14:35   Latest developments and outlook for Tomofast-x (Vitaliy Ogarko)

14:35-14:45   Group discussion (All)

14:45-15:00   Coffe/Tea break (All)

15:00-15:30   Introduction to geometrical inversion with application to Yerrida Basin (Jeremie Giraud)

15:30-15:50   Automated geophysics-driven insertion of rock units in an existing model (Jeremie Giraud)

15:50 - 16:10  Null space shuttles to explore different scenarios (Jeremie Giraud)

16:10 - 16:30  Current work and outlook for geometrical inversion using level-sets (Jeremie Giraud)

16:30-16:45    Feature engineering of the physical response for classification using the NoddyVerse dataset (Leonardo Portes)

16:45-17:00   Group discussion (All)

About the presenters:

Jeremie Giraud will be the main facilitator of the masterclass. He is a Marie Sklodowska-Curie Fellow at the Universite de Lorraine in France and an Adjunct Research Fellow at The University of Western Australia.

After obtaining his MSc. Eng. in geophysics (Uni. of Strasbourg, France), he spent a few years at Schlumberger's WesternGeco Center of Excellence for Integrated EM in Milan, Italy, where he learned about joint inversion in the R&D team and applied it to integrated reservoir modelling and went through Schlumberger?s training program. Jeremie then started his PhD in 2015 at the Centre for Exploration Targeting (School of Earth Sciences), University of Western Australia, which he completed in 2018. His project focused on the integration of geological and petrophysical information in geophysical inversion. He then continued his research on geophysical integration methods at the Centre for Exploration Targeting as a Research Fellow. Since August 2021, working on the integration of implicit methods into geophysical inversion as the recipient of Marie Sklodowska-Curie fellowship in the 'Research for Integrative Numerical Geology' (RING) team in Nancy, France.

As part of his collaboration with the Loop consortium, Jeremie is leading 'Work Package 4 - Geophysical Integration', around the development of new approaches for the automated integration of geological modelling and geophysical inversion. He also collaborates with scientists from CSIRO and the Data Analytics for Resources and the Environment (DARE). Through these collaborations, Jeremie is also involved in the supervision of Australia-based PhD students.

Vitaliy Ogarko is an applied mathematician who holds a PhD degree in Computational Physics. His research interests include design and development of high-performance computational algorithms in application to geophysical inversion, and computational geology. In 2021, Vitaliy joined the Centre for Exploration Targeting to work on the MinEx project Recovery of stratigraphic data from drillhole databases. He is also a main developer of an open-source geophysical inversion code Tomofast-x.

Leonardo Porte is a Senior Research Fellow at the Centre for Exploration Targeting at The University of Western Australia. With a background in Physics, he completed his MSc and PhD, followed by two postdoctoral positions, at the University of Minas Gerais (UFMG, Brazil). He joined the University of Western Australia in 2018 as a Research Fellow at the Dept of Mathematics and Statistics.

He was recently granted the Australian Global Talent Visa in Data Science/DigiTech in recognition for his scientific carrier and industry engagement. His expertise could be more broadly described as statistics, statistical physics and data science.

Leonardo Portes' scientific interests revolve around developing and applying analytical tools and concepts from Complex System Science to real-world problems - ranging from genetics, human motor behavior and medical data to resources industry data. By blending Network Science, Dynamical Systems theory, Nonlinear Time Series Analysis and Machine Learning/Statistics, his methods have and are being applied and deployed by Leonardo and his UWA collaborators within the resources sector in Australia and with medical science and technology. Most notably, he has completed a substantial research project with Woodside (2019), entitled "Subsurface Big Data Visualisation and Analysis: Oil/Gas Reservoirs", to apply his techniques to better understand modelling, drilling within oil extraction campaigns and uncertainty quantification. Separate to this he has completed a world first benchmarking project with Anglo American (2021), allowing for state-of-the-art layer detection and segmentation of geological record from drill core data via novel nonlinear time series methods.