Graham Baines

Halliburton (United Kingdom)

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

David Giles

Flinders University

Guillaume Backé

University of Adelaide

Peter Betts

Monash University

Benjamin Gréselle

Halliburton (United Kingdom)

Andrew Davies

Halliburton (United Kingdom)

W.V. Preiss

University of Adelaide

M. J. Cheadle

University of Wisconsin–Madison

N.J.S. Stacy

South Australia Pathology

D. Possee

Halliburton (United Kingdom)

Alan M. Haywood

University of Exeter

Cooperative Institutions

Halliburton (United Kingdom)

California Institute of Technology

Commonwealth Scientific and Industrial Research Organisation

University of Adelaide

Jet Propulsion Laboratory

Astronomy and Space

Geoscience Australia

Australian National University

Australia Telescope National Facility

United States Geological Survey

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Potential field analysis images Paleoproterozoic terrane boundaries in the unexposed Northern Gawler Craton, Australia

AGUFM (2008)

Graham Baines David Giles Peter Betts Guillaume Backé

Source

Cite

Citations (1)

Geochronology and Hf isotopes of the bimodal mafic–felsic high heat producing igneous suite from Mt Painter Province, South Australia

Gondwana Research (2013)

Kamonporn Kromkhun John Foden S. B. Hore Graham Baines

Felsic

Geochronology

Anatexis

10.1016/j.gr.2013.01.011

Cite

Citations (14)

A Machine Learning Enabled Workflow for Efficient Channel Detection from Seismic Data

M. Servais Graham Baines

Summary Sedimentary channel systems can form significant hydrocarbon reservoirs, and reliably detecting the presence of channels in a seismic volume can be a time-consuming process, often requiring input from expert geoscientists. Supervised machine learning approaches to channel detection in seismic data require labelled data, which can be a challenge for applying such techniques to subsurface data. We present an approach to streamlining the data labelling process by making use of an assisted horizon interpretation process. This allows the extraction of horizons in the seismic volume. Once enough representative channels have been labelled (for training and validation datasets), they can be used to train a machine learning model specific to the seismic volume. This model can then be used to predict the presence of channels throughout the volume. Using real-world seismic data, we show that this approach offers a useful pipeline for training reliable machine learning models for channel prediction from seismic data.

10.3997/2214-4609.202310189

Cite

Citations (0)

Intraplate deformation on north-dipping basement structures in the Northern Gawler Craton, Australia: reactivation of original terrane boundaries or later intra-cratonic thrusts?

AGUFM (2007)

Graham Baines David Giles Peter Betts Guillaume Backé

Basement

Source

Cite

Citations (0)

Assessing the impact of aquifer-eustasy on short-term Cretaceous sea-level

Cretaceous Research (2020)

Andrew Davies Benjamin Gréselle Stephen J. Hunter Graham Baines Chris Robson

10.1016/j.cretres.2020.104445

Cite

Citations (72)

Guided strategies for improving machine learning models applied to geophysical problems

Satyan Singh Konstantin Osypov Graham Baines Mark E. Willis

10.1190/image2024-4101010.1

Cite

Citations (0)

Crustal boundaries of the marginal terranes of the northern Gawler Craton

Peter Betts Robin Armit Graham Baines David Giles Bruce F. Schaefer

Source

Cite

Citations (0)

Data for: Assessing the Impact of Aquifer-Eustasy on Short-Term Cretaceous Sea Level

Andrew Davies Mike Simmons Christopher Robson Alan M. Haywood Stephen J. Hunter

10.17632/bxcfny6xck.1

Cite

Citations (0)

A Global Evaluation of Source Rock Potential Since the Ediacaran

Proceedings (2013)

Jean‐Christophe Wrobel‐Daveau Graham Baines James L. Etienne

Key factors influencing source rock potential are the extent and nature of primary deposition and post-depositional thermal, tectonic and geodynamic history, which will affect the subsequent preservation and maturity of the source rock. We present a global evaluation of source rock potential and risk from the Ediacaran and Phanerozoic. This evaluation is made by combining 52 palinspastically restored gross depositional environment maps with a global plate model for the Phanerozoic and Ediacaran taken from the Neftex Global Earth Model product suite. The 52 palinspastic gross depositional environment maps were produced by taking reconstructed observations and interpreting them regionally based on both the tectonic setting and a predictive sequence stratigraphic framework. When combined with predictions of paleoclimate and paleo-ocean circulation, this framework allows us to predict zones of organic enrichment and source rock deposition through time. Once deposited, the global plate model allows us to identify regions where geodynamic processes (e.g. collision zones, large igneous provinces, rifts) are likely to have influenced source rock preservation and maturity. By combining these data with the observed geology and sediment thickness at the present day, we generate a global risk map of source rock potential that can be used to identify new exploration opportunities,

Deposition

Paleoclimatology

10.3997/2214-4609.20130097

Cite

Citations (1)

Rapid Assisted Horizon Interpretation Using Multi-Scale Optimization

83rd EAGE Annual Conference & Exhibition (2022)

D. Possee Graham Baines Yuanqi Liu

Summary The interpretation of seismic horizons is a critical step in characterizing subsurface geological structures. However, it remains a time-consuming process using traditional tools due to complex structures and noise found within seismic data. Automatic horizon extraction methods have continued to improve, particularly with the adoption of deep-learning architectures, but may still fail in any dataset for many different reasons. We present a workflow that focuses on assisting the seismic interpreter to extract high-quality seismic horizons rather than being purely a predictive process. The algorithm utilizes a multi-scale constrained optimization framework that enables the global structure of a horizon to be interpreted in near-to-real time. Since the process is extremely fast, it enables a highly streamlined and interactive workflow where the interpreter can continue to add additional constraints to refine the horizon until satisfactory. We demonstrate the benefits of this workflow on two case study datasets, the F3 and Poseidon Seismic volumes. In both, a dense stack of horizons was interactively interpreted which accurately captures the stratigraphic and structural detail in each dataset. Such a workflow may take weeks to complete using traditional horizon picking tools but can be completed in less than an hour using the assisted methodology presented here.

10.3997/2214-4609.202210717

Cite

Citations (0)