During 10 years of production, crestal fluid pressures in the Magnus Field (UKCS) have dropped from 6650 psi to under 3500 psi leading to a two- to three-fold increase in the effective stress on the rock fabric. Petroacoustic measurements on cores indicate that this leads to a 12% increase in the acoustic impedance which, in theory, should be detectable using time-lapse seismic data. We analyse the difference between two 3D surveys shot over Magnus and relate this to dynamic changes in fluid pressure and saturation through time. The results demonstrate that, under certain conditions, fluid pressure changes may be detected in the reservoir and that fluid transmissibilities across faults can be deduced. Further, 4D seismic data may detect other dynamic processes, including thermal effects and cold water fracturing around injectors, and stress relaxation and fluid compositional changes around producers.

Ccpyrght199S S=)etyd Pefrokum Englneem ThIsp er waspr~ared lorpresentatlon atthe EurWean 3Dmsewo,r wdelitng conference % hddm tavanyer Norway 16-17 Aixd 1996 Thm paper was selecfed for presentation by the SPE programm conmuftee Iollow!ng revmu of mlormatmn contatned In an abstract s.bmmed by the authors Contents of the paper ap rrnented have not b6enrev,ewed by[he Smmtyof Petroleum Engineers andaresubject!o correctlonby fheauihors The matenalaspresentec dcms nd necrnsanlyrelbct anyposdmn ofihe Socmtyof Petroleum Engrmemortsrratirs P erspfesentedat SPE nmet(ngsar. t subpd to p. bi,catto" mvmw by Editorui Comnnt!ea o{ t e Sce,edy of Petroleum Eng(neem %rnmsto" t.~y is restrc!ed toanabstractol not nmrnihat330 words Illustralcms may not Lwcopmd The stractshould cwta!ncom cuousackcowled ententol where and by whom J' 8 the aperwasprasented Wrte Lbrarmn S E PO 8333636 !chardson Texas 75J833836 USi lax 01.214-952.9435

This paper concerns an example of cross-disciplinary integration: the use of seismic modelling based on a shared earth model to validate reservoir simulation. Whole-field simulator models and finer-scale sector models can be generated from the same detailed geological description in a shared earth model, ensuring that they are consistent. However, it is possible for some aspects of the simulation to depart from the geology, making predictions unreliable. For seismic validation, synthetic seismic volumes are produced from a shared earth model containing fluid data fed back from the simulator. Features in the synthetic which are caused by fluid movements, are then looked for on the real 3D seismic data. If the seismic data confirm the simulator's results in the spaces between history-matched wells, greater confidence can be had in its forward predictions. The methodology is illustrated using the Magnus oil field in the UK North Sea.

Efficient field management and appraisal rely on a good understanding of reservoir geology and recovery efficiency. This requires co-operation and good communication between different disciplines. This paper presents and demonstrates the shared earth model approach, to encourage integration and co-operation between disciplines producing more reliable reservoir characterisation and performance prediction. A North Sea field is used as an example to illustrate the development and use of the shared earth model. The way in which it brings together the subsurface disciplines is shown by dynamic data feedback and seismic validation of simulator predictions. Introduction In the present state of the oil market, subsurface teams are under more pressure than ever before. Field appraisal and development time are being sharply decreased; waterflood and pressure maintenance schemes are designed very early in field life; and infill drilling programmes have to be operated within tight financial constraints. If subsurface teams are to make sound interpretations and predictions in little time, and also deliver the high recover factors which are now expected, the members need to work together, communicate and integrate their information much more thoroughly. One way to achieve this is through the shared ownership of a numerical description of the reservoir and its properties: this is the shared earth model concept which is discussed in detail below. In order to reach the degree of interdisciplinary collaboration which is increasingly needed in our subsurface teams, the lines of demarcation between subsurface technical disciplines must become indistinct. By doing this opportunities are created that did not exist before. Seismic validation of reservoir simulation is an example of this. This paper is in two parts: the first part discusses the shared earth model concept in subsurface integration; and the second part describes an application of this concept to a field in the North Sea, showing the advantage to be gained by integrating seismic modelling with reservoir simulation in the planning of an infill drilling programme. GEOSCIENCE AND ENGINEERING RESERVOIR MODELS Everyone in a subsurface team has a model of the reservoir on which they are working. Some models are numerical, such as a reservoir simulation model; others are graphical, such as a contour map or a well correlation diagram; and some models are conceptual, existing only inside the heads of geoscientists. P. 383

This paper was prepared for presentation at the 1999 SPE Reservoir Simulation Symposium held in Houston, Texas, 14-17 February 1999.