Seismic fault interpretation is a critical task for any type of energy industry. Correct fault mapping can be crucial for the success of a project. Common geometric seismic attributes, such as coherence and curvature, are routinely employed to enhance fault visualization in seismic data. However, they can show limitations for subseismic faulting. In this study, we highlight the usefulness of including novel aberrancy attributes for fault identification in multiattribute analysis and unsupervised machine learning (ML) techniques. We compare broadband coherence, curvature, multispectral coherence, and aberrancy when trying to map faults in a potential CO2 storage location. We also compare self-organizing maps and generative topographic mapping techniques when including and excluding aberrancy attributes. Our results show that integrating aberrancy attributes during multiattribute analysis and ML steps considerably enhanced the visualization of lineaments with strikes similar to those of fracture sets seen only with well-log data and that were not clearly captured by the conventional seismic attributes and ML scenarios excluding aberrancy attributes. We demonstrate the potential of these novel geometric seismic attributes to map subseismic faults. We also provide an example that can encourage interpreters to include them in their interpretation workflows.

You do not have access to this content, please speak to your institutional administrator if you feel you should have access.