Attribute-Assisted Automatic Fault Extraction — A Case Study in a Tectonically Complex Area Offshore East Coast of Canada

Abstract

The geological history of sedimentary basins in the Grand Banks is complex due to the influence of several cycles of Mesozoic rifting related to the opening of the North Atlantic. Subsurface evaluations of the recent Harpoon and Bay du Nord discoveries show the reservoir intervals are extensively faulted. While some aspects of the basin development are still debated, the implications for exploration and field development are clear: complex fault networks can significantly affect regional hydrocarbon migration pathways and result in local reservoir compartmentalization. To determine whether the faults are generally sealing or open to hydrocarbon flow, a thorough understanding of the structural fabric is needed. Exploration and reservoir development in highly faulted areas presents a significant challenge for the structural interpreter. Much detailed mapping is usually needed before the tectonic fabric of a complex area is fully understood. Nonetheless, in most cases, a comprehensive fault interpretation is impossible due to time constraints, data quality and, in most cases the very large number of faults. Attribute-assisted fault extraction can identify a complete fault network within a very short timeframe. In this case study we will walk through the workflow that was applied to a 3D post stack data set from the east coast of Canada. The workflow consists of: data conditioning to clean the data to minimize false positives; selecting appropriate attributes to image the full extent of the faults (this is dependent of the fault type, data quality and seismic response); optimizing the combination of attributes by enhancing the fault responses; and extracting the faults. As part of the workflow, limited frequency bandpass volumes are also used. Using reflectivity data with limited frequency content enables faults to be investigated at different scales. The results can be used at exploration scale to gain a thorough understanding of the tectonic history of an area, and at reservoir scale to assist the manual picking of fault sticks for the geocellular model. In conclusion: implementing a data-driven fault extraction workflow into the normal seismic interpretation routine can generate significant time-savings for the seismic interpreter and cost-savings for a subsurface project. The extractions enable a more rapid and robust identification of faults. They can also highlight the structures which have implications for hydrocarbon migration and reservoir production.

AAPG Datapages/Search and Discovery Article #90259 ©2016 AAPG Annual Convention and Exhibition, Calgary, Alberta, Canada, June 19-22, 2016