The Importance of Fault Characterization for Safe Carbon Sequestration and How Machine Learning Reduces Risks
Thursday, 31 August 2023 at 10:45 US CDT (UTC-5)
Carbon Management Pavilion | IMAGE 2023
Dr. Carolan (Carrie) Laudon, Sr. Geophysical Consultant for Geophysical Insights
Continuing the company’s demonstrated history of thought leadership, Dr. Carolan (Carrie) Laudon, Sr. Geophysical Consultant with Geophysical Insights, will present a joint paper describing how deep learning technologies are applied to automate seismic fault identification for candidate Carbon Capture Utilization and Storage (CCUS) reservoirs. Scheduled on Thursday, 31 August at 10:45 AM, this talk is a must-see for anyone interested in CCUS.
Contact us for more information and click here for more on the talks at Carbon Management Pavilion.
Carbon Capture, Utilization and Storage (CCUS) will draw on many technologies from upstream oil and gas production and seismic data will undoubtedly play a key role in site characterization and monitoring. Whether the storage reservoir is a depleted oil and/or gas field or a saline reservoir, safety dictates that CO2 must be stored without leakage into either overlying rocks where a catastrophic release could cause severe harm or, into faults and underlying basement where it could induce earthquakes. Understanding each and every fault which has potential to interact with the reservoir is even more essential than for hydrocarbon production.
Interpreting geophysicists have been manually picking faults for as long as there have been seismic lines to pick, however, anyone who has tried to interpret in a complex fault regime also knows that three-dimensional fault geometries can be time consuming and difficult to interpret.
Recent advances in Machine Learning fault detection using Convolution Neural Networks takes advantage of 3D synthetic models to produce perfectly labeled fault volumes (Wu et al., 2019, Zhao, 2019, Laudon et al., 2021). This has several benefits:
- Reduce, by many hours and even days, the manual interpretation that is required to capture every fault which intersects the storage reservoir.
- Removes interpreters’ bias from the fault interpretation.
- Produces volumes which accurately predict complex fault geometries in 3 dimensions.
This will be illustrated on data from several potential target geologic settings including salt dome and the Permian Basin.
Wu X.M., Liang L.M., Shi Y.Z., Fomel S., 2019, FaultSeg3D, using synthetic data sets to train an end-to-end convolutional neural network for 3D seismic fault segmentation, Geophysics, 84, IM35–IM45.
Zhao T., 2019, 3D convolutional neural networks for efficient fault detection and orientation estimation, SEG Technical Program Expanded Abstracts 2019, Society of Exploration Geophysicists, 2418–2422.
Laudon, C., Qi, J., Rondon, A., Rouis, L., Kabazi, H., 2021, An enhanced fault detection workflow combining machine learning and seismic attributes yields an improved fault model for Caspian Sea asset, First Break, Volume 39, Issue 10, p. 53 – 60.
About Dr. Carolan Laudon,
Carolan Laudon holds a Ph.D. in Geophysics from the University of Minnesota and a BS in Geology from the University of Wisconsin Eau Claire. Carolan has worked as a Senior Geophysical Consultant with Geophysical Insights since 2017, applying the Paradise® machine learning workbench. Prior roles include Vice President of Consulting Services and Microseismic Technology for Global Geophysical Services and 17 years with Schlumberger in technical, management, and sales roles. Her work with Schlumberger included offices in Alaska, Aberdeen, Scotland, Houston, TX, Denver, CO, and Reading, England. Carolan spent the first five years after graduate school with ARCO Alaska in the exploration team as a seismic interpreter.