The thin oil rim of the Gannet-A turbidite field in the Central North Sea has been produced since 1993 though 11 long horizontal wells supported by a strong aquifer and sizeable gas cap. The original zero development strategy called for zero net-voidage of gas through re-injection of produced gas into the gas cap.
A full gas cap blowdown is currently being planned to commence in 2008. In order to predict field behaviour in this high production rate environment, and maximise oil and gas recovery, significant effort has gone into the static and dynamic re-modelling of the reservoir.
This effort has demonstrated that over the production life, complexity of the high quality reservoir has been underestimated, leading to inadequate historical data acquisition and integration efforts, particularly in the low oil price world in the late 90s.
This paper describes how all available subsurface data have been (re-)analysed and integrated, resulting in a range of realistic dynamic reservoir models and how the remaining uncertainty was dealt with during the selection of the optimum further field development plan.
The key to success in this study has been the seamless cross discipline integration that enabled maximum extraction of information from the available data. The same discipline integration led to a thorough understanding of the critical remaining uncertainties in the project and a tailor made surveillance plan that aims at rapidly reducing these.