Reservoir characterization is a difficult task, particularly with limited well data. For example, modeling a field with good vertical resolution can present a problem if there are few wells, even with a 3D seismic survey. This does not necessarily change the demand for immediate detailed information because of imposed deadlines for field development, limited time-windows of rig availability, seasonal constraints, or foreign government requirements. In the face of imposed deadlines, it is common to find the geologic model poorly constructed or even bypassed in lieu of a petrophysical model, the primary input to a reservoir simulator. Many pitfalls lie in a hasty approach, and resulting models are often less than satisfactory and the cause of a potentially costly mistake.

Key to modeling sparse data is the judicious use of geostatistics, blending both qualitative and quantitative information. For example, quantitative information from wells can be used to supply vertical detail while seismic data and/or conceptual geologic models can supply information about horizontal continuity, geologic trends, or sequence stratigraphic constraints. The result is a detailed 3D computer model that can be used for constructing reservoir fluid flow studies and a plan for development. While integrating the geological detail into the model takes time, it is more efficient and less costly than running a corrected second model. In this presentation, different methodologies are demonstrated from actual case histories.

End_of_Record - Last_Page 13---------------