Quantitative 4-D Seismic as an Additional Constraint in an Automated History Matching Process: An Example from the Harding Field, United Kingdom North Sea

Walker, Greg J., Peter M. Allan, BP Exploration, Aberdeen, United Kingdom

Use of 4D seismic surveys for reservoir monitoring has become common practice in the North Sea, providing information on what is happening to the fluid distribution between wells, with the expectation of a better prediction. This paper explores how the quantitative constraints from multiple 4D seismic surveys interact with well surveillance, and the impli­cations for history matching routines and well planning.

Earlier studies using the uncorrelated well and 2000 4D survey constraints proved there was a small area of intersection, implying that the combination of constraints reduces resid­ual uncertainty, and hence that the 4D seismic is providing a large amount of new informa­tion on the reservoir.

Inclusion of the 2002 and 2005 surveys will allow us to explore the interactions of mul­tiple surveys. In this work we are looking at the shape of the residual uncertainty: are there still multiple reservoir descriptions that are equally valid, and what shape constraint should we be searching for to reduce uncertainty further?

By understanding the uncertainty envelope, we can make informed decisions on acqui­sition and determine the impact of our surveillance program. Our automated and quantita­tive workflow takes 30 minutes to assess a model, allowing fast integration of seismic con­straints and testing of different attributes. The Harding team is proceeding from a survey scheduled to complete at the end of August 2005 to drilling in November 2005 by using the new seismic information to select from a library of history matched reservoir descriptions with their individual depletion options.