Mapping of Clastic Reservoirs With RGB-Blending Techniques — A Review From the Maria Discovery in the UK North Sea

Abstract

A number of different RGB blending techniques are currently available to the geoscience community to image geological features using seismic data. The application of frequency decomposition methods to seismic data has routinely been used as a technique to image structural and stratigraphic features. RGB blending of different frequencies offers a route to simultaneous analysis of multiple data sets at a larger dynamic range than available from standard multi-trace seismic attributes. In the presence of an AVO anomaly RGB blending of angle stacks can help to image depositional features and also highlight hydrocarbon accumulations. A widely used method uses RGB blending of the seismic envelope attribute of near, mid and far stacks. In addition to the well-established RGB blending of frequency and offset attributes, a new method based on image processing has become available to the industry. This method uses RGB blending of seismic depth slices in a similar way in which satellite imagery is processed. It then extracts structural boundaries from the images and outputs both a structural data cube as well as sharpened RGB color cubes. A comparison of different types of RGB blending techniques has been run on a 3D seismic data set across the Maria Discovery in the UK North Sea. The study area is part of the prominent Upper Paleocene aged Forties Sandstone turbidite channel complex in the Sele Formation, that trends NW-SE in the prolific Witch Ground Graben. A 20 ft gas and 44 ft oil column was encountered in a 85 ft high net-to-gross and high porosity turbidite channel sandstone. The channel overlays the Upper Paleocene Balmoral Sandstone deepwater slope system in the Lista Formation. The Forties sandstone turbidite system exhibits a Class II AVO response with a hydrocarbon fill. While traditional RGB blending input derives seismic attributes from the characteristics of the seismic trace, this new method processes the seismic data as images. As a result, this technique achieves enhanced seismic sample resolution compared to frequency and offset RGB blending techniques. The structurally sharpened RGB images show highlighted details of the reservoir and the surrounding depositional systems that have not previously been imaged using traditional seismic attribute blending techniques. The new results are currently being integrated in the ongoing subsurface evaluation of the Maria Discovery.

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