--> Integrated Carbonate Research for Reservoir Characterization and Simulation; A Valuable Tool, But Do Not Lose Sight of the Fundamentals, by James H. Anderson; #90029 (2004)
[First Hit]

Datapages, Inc.Print this page

Integrated Carbonate Research for Reservoir Characterization and Simulation; A Valuable Tool, But Do Not Lose Sight of the Fundamentals

James H. Anderson
ExxonMobil Upstream Research Company

 

Following a decline in the level of oil industry carbonate research during the late 90's, there has been an upsurge in the past few years. This has resulted from both business drivers and advances in technology. It is a dynamic and exciting time for carbonate research.

The combination of major new field developments and discoveries in the North Caspian, Middle East and Southeast Asia, and the continued challenge of maintaining economic production volumes in existing fields has put pressure on industry to develop more accurate reservoir predictions and to provide alternative scenarios for development and production. In response to these drivers, interdisciplinary approaches to solve carbonate reservoir problems are being developed. Integrated teams that include stratigraphers, modelers, structural geologists, geophysicists, and engineers can facilitate step changes in our understanding and predictive capabilities of carbonate reservoirs. 

This talk will focus on three key technical areas that are yielding results now, and have the potential to yield further advances: 1) volume interpretation and visualization, 2) integration of structure, stratigraphy and geophysics, and 3) integration of geologic and engineering data onto the same Previous HitcomputerNext Hit platform. In each of these areas, interdisciplinary integration is a key to success. Another key factor is a sound basis in geologic and engineering fundamentals. 

Sequence stratigraphy has become a common, even routine, tool for developing reservoir frameworks and has been shown in numerous case studies to provide a more accurate geologic model for simulation. Visualization tools and volume interpretation techniques are significantly advancing our ability to image and directly extract facies and rock properties from seismic data and logs to populate geologic models with quantitative spatial data. Purely stochastic approaches are becoming outdated. These advances generate a need to continue fundamental research in carbonate stratigraphy and depositional environments to validate the models.

Increased seismic resolution has improved our ability to make more precise structural interpretations. Combining mechanical stratigraphy with litho- and chrono-stratigraphy, we can map more accurately detailed fault systems and understand the relationships between faults and the associated fracture networks. Geophysicists working with structural geologists are analyzing prestack seismic data to evaluate the relationships between azimuthal variations in seismic velocity and fracture systems. By integrating all of these approaches, we can develop models that are better constrained.

Corendering geologic and engineering data in a visualization environment provides significant insights for improved reservoir optimization. Further advances likely will occur when simpler, more efficient techniques are developed to rapidly display, analyze and integrate data on the same Previous HitcomputerNext Hit platform. Again the key is putting the right people and the right disciplines together in the same place and providing them time to think.

However, all of this advanced quantitative technology development must be within the context of sound geologic and engineering fundamentals. We must constantly ask ourselves "do the interpretations make sense?" Is there a sound technical basis, grounded in fundamentals, for the algorithms and images we produce? We must also not lose sight of the value of multiple hypotheses. Technology is neither a substitute for good fundamental Previous HitscienceTop and engineering nor should it be used as an excuse to avoid considering alternative scenarios.