Strathspey Vertical-Cable Seismic Survey: A North Sea First

Abstract

Abstract In 1995-96, the Strathspey Field Group (Texaco, Shell, Esso and Oryx) acquired and processed the first vertical-cable seismic survey in the North Sea. Strathspey is a tilted, eroded fault block, and production is from the Brent and Statfjord reservoirs. The eastern flank of Strathspey is marked by low-angle footwall degradation complexes in both reservoirs. Historically, these features have been very difficult to image seismically, both on Strathspey and on neighbouring fields. Vertical-cable seismic, with its uniform distribution of azimuths and its economical 3-D pre-stack depth migration, was judged to have the best potential to image these features. The survey was acquired in six swaths in the summers of 1995 and 1996. Processing of the pre-stack depth-migrated data volume was completed in October of 1996. The dataset shows remarkable improvement over the previous 3-D dataset (vintage 1985), and compares favourably with the Brent Field dataset which was acquired immediately to the north of Strathspey in 1995. Improved data quality has allowed a more confident interpretation of the low-angle fault scarp features, as well as a more detailed structural model of both reservoirs. In addition, the superior data resulted in an improved assessment of the remaining exploration potential in the Strathspey area. The information gained from the interpretation of the vertical-cable seismic data has given the field partners more confidence in siting wells, and in the reservoir simulation models. Regional Setting and History of Strathspey The Strathspey Field is located in the East Shetland Basin in Block 3/4a of the U.K. sector of the North Sea, approximately 470 km. north-east of Aberdeen (Fig. 1). The water depth is approximately 440 feet. Geologically, the field is a tilted, eroded fault block producing from two reservoirs: the middle Jurassic Brent Group, which contains volatile oil, and the lower Jurassic - upper Triassic Banks Group (Statfjord and Nansen Formations), which contains condensate-rich gas with an oil rim (Figs. 2 & 3). These two reservoirs are separated by the Dunlin Group shale, which is approximately 800 feet thick in the Strathspey area. Expected ultimate recoverable hydrocarbons are 89 million barrels of liquids, and 332 billion cubic feet of gas. The field is produced via a subsea manifold which is controlled remotely from the Ninian Central platform. Partners are Texaco (67% and operator), Shell (13.25%), Esso (13.25%), and Oryx (6.5%). First oil was produced in December of 1993. The Brent Field, immediately to the north, is structurally analogous to Strathspey. By early 1993, extensive drilling on Brent had convinced the operator, Shell, that the erosional crest of the field was far more structurally complex than had been previously believed - in particular, that low-angle fault scarp degradation surfaces (informally referred to as "slumps") featured heavily on the eastern flank of both the Brent reservoir and the Statfjord reservoir. By analogy, these features were highly likely to be present on Strathspey; however, in the absence of significant well control, they would have to be delineated using seismic data alone. Historically, this had been very difficult, both on Strathspey Field and on Brent Field, due to several factors:the low angle of the faults causes them to offset very few seismic reflectors,the material within the slumps themselves is comprised of narrow pieces of strata which have been rotated into fairly steep bed dips (up to 400 in Brent Field), andthe slump complexes are partially overlain by a wedge of onlapping lower Cretaceous material of a high seismic velocity. These last two factors result in an invalidation of some of the assumptions which are made in conventional seismic data processing, and therefore, conventional processing results in mediocre data. The best processing technique for imaging geologic cases such as these is 3-D pre-stack depth migration (3DPSDM), but this process uses tremendous amounts of computer time, and is therefore very expensive to run on conventional marine seismic data. P. 333^