Datapages, Inc.Print this page

Quantifying the Impact of Uncertainty in Seismic Interpretation

Bond, Clare E.1; Butler, Robert 2; Jones, Serena 1; Muir, Roddy 1; Gibbs, Alan D.1
1 Midland Valley Exploration, Glasgow, United Kingdom.
2 Geology and Petroleum Geology, Aberdeen University, Aberdeen, United Kingdom.

Recent work has shown the range of conceptual models applied to an interpretation of a single 2D synthetic seismic line (Bond et al., 2007). The variety of concepts, the ‘conceptual uncertainty’, applied by geoscientists to the seismic image ranged from the main tectonic settings (i.e. compressional, extensional, strike-slip…) through to salt tectonics, coral reefs and mega-sequences. The exercise was an under-constrained problem, common across the geosciences and in particular for seismic interpretation. Further work by Bond et al. (2008) showed the importance of specific techniques and workflows to optimise interpretational ability and decrease uncertainty in the interpretation process. However, for under-constrained problems there are no unique solutions. Simply, from the data available there is not a right answer.

Sharing interpretations, knowledge and information, is important for understanding the range of conceptual uncertainty for a given dataset. The virtual seismic atlas (VSA), a not for profit resource, has been set-up to share interpretations of seismic data. The resource can be searched for concepts and models which can be used as analogues for similar datasets. The ‘steep’ limb of thrust structures are often poorly imaged and can be difficult to interpret. Using examples of thrust structures from the VSA, we have created multiple conceptual models for thrust fault generation i.e. a fault propagation fold versus fault bend fold mechanism. For these models it is possible to draw ‘uncertainty sausages’ that define the X,Y,Z space of uncertainty in fault location and steep limb angle, but how useful is this information?

The models created span a range of possible concepts that affect not only fault position, but also the mechanics of the structural evolution. This combination of fault position and mechanical evolution impact key parameters important for exploration decisions. Here we have forward-modelled the end member concepts for poorly imaged thrust structures and looked at the impact of their structural evolution on reservoir compartmentalisation, volume and permeability. We argue that the question is not whether we can quantify the uncertainty in the interpretation, but the impact of this uncertainty on the key decision making parameters.


AAPG Search and Discovery Article #90090©2009 AAPG Annual Convention and Exhibition, Denver, Colorado, June 7-10, 2009