Semi-Automatic Seismic Interpretation Through the Extraction of Unconformities and Faults

Abstract

In this study, we present two independent methods for semi-automatic seismic interpretation through the extraction of three-dimensional unconformities and faults from seismic data and seismic attributes. Unconformities are detected and extracted from seismic sequences that are defined based on seismic patterns and acoustic impedance. Independently, the faults are extracted from filtered fault likelihood data where each fault is considered a binary object. Our methods use basic functions found in well-known image processing libraries such as MATLAB's Image Processing Toolbox© or python's scikit-image package.

For either unconformities or faults, each extracted feature is stored in a 3D matrix and can be exported to different seismic software platforms, such as Petrel©, Kingdom© or DecisionSpace©. Our methods can be used to enhance and speed-up workflows in seismic interpretation which is generally a very time-consuming manual process.

We have evaluated our methods using a case study on the Polhem Sub-platform off the Barents Sea part of the Norwegian Continental Shelf. The Polhem Sub-platform is part of the well-explored Loppa High area: a Triassic carbonate platform which has been subsequently experienced a sequence of faulting, erosional and subsidence events since the Jurassic. In the study area, the methods are successful in extracting two unconformities: the base Pleistocene and the base Cretaceous, and the dominating faults on the platform: major Jurassic normal faults with N-S or NW-SE structural trends.

AAPG Datapages/Search and Discovery Article #90323 ©2018 AAPG Annual Convention and Exhibition, Salt Lake City, Utah, May 20-23, 2018