AAPG ANNUAL CONFERENCE AND EXHIBITION
Making the Next Giant Leap in Geosciences
April 10-13, 2011, Houston, Texas, USA
Source Rock Prediction from Seismic. Part III: Application in Basin Modelling
(1) Statoil ASA, Trondheim, Norway.
(2) Statoil ASA, Stavanger, Norway.
Source rock properties extracted from seismic data can be utilised directly in hydrocarbon generation modelling to improve risking of the petroleum system. Seismic data can thus reduce the uncertainty in the source rock parameters used in basin modelling and thereby improve ranking of basins, plays and prospects. A case study from the Norwegian Continental Shelf illustrates such work in practice.
A workflow presented here for the first time allows us to confidently map organic rich claystones on seismic data. The workflow is based on established relationships between organic content and acoustic - elastic properties of source rocks and how organic content influences seismic responses. These relationships enable us to predict presence, extent, thickness and richness of the source rock from seismic data.
The data used in the study were acquired in the Norwegian Sea where many oil- and gas fields prove the existence of a working petroleum system. A 3D seismic survey with near and far offset stacks was available for the study, as well as a standard suite of well logs and organic geochemical data from two exploration wells. Extent, thickness and richness of the source rock were interpreted from the seismic amplitudes and the resulting maps were input to hydrocarbon generation modelling. We compare results from traditional basin modelling (parameters predicted from interpolated well data) with two different scenarios taking seismic amplitudes into account; 1) seismically derived thickness of the source rock and organic content derived from interpolated well data and 2) seismically derived thickness and organic content.
The traditional basin modelling gave significantly larger generated hydrocarbon volumes than the models using seismically derived source rock parameters. We conclude that seismically derived source rock parameters are better constrained than parameters traditionally used in basin modelling and thus give more realistic modelling results.