Assessing the Accuracy of a Production Forecast: West Africa Field Case History

Abstract The subject field is located in the deep water off West Africa. Over its 7-year life, the field reached a peak production of 90 thousand BOPD and produced nearly 100 million barrels of oil from a high-quality stacked Lower Miocene deep-water channel complex. This field provides a unique opportunity to perform a look back on how well the reservoir performance predictions made at project funding matched with the final actual field performance. The West Africa field was a subsea development tied back to a FPSO (Floating Production Storage and Offloading vessel). The production forecast used for project funding was generated using a full field simulation model and optimized depletion plan. The original depletion plan called for crestal gas storage and peripheral water flood. Model properties were derived from a single exploration well and high resolution 3D seismic data. The production forecast developed at project funding did a very accurate job of predicting average reservoir behavior such as most-likely ultimate recovery and production plateau. However, a more detailed comparison of the pre-production depletion plan and the actual field performance shows significant differences. In particular, both gas and water breakthrough and build up were faster than expected. These factors were offset by higher well productivity and larger in place oil volumes. 4D seismic acquired after 3 years of production was particularly effective in illuminating gas and water flow pathways in the reservoir that had not been modeled or predicted. In addition, changes were made to the original depletion plan to increase recovery and in reaction to operational issues. The factors that drove the evolution in the depletion plan are reviewed and an assessment is made on the accuracy of the original production forecast. SPE is sponsoring a series of Global Integrated workshops on production forecasting with the objective of assembling guidelines. This paper presents the results of a comparison of a field's production forecast from project inception to abandonment. This example demonstrates that the assessment needs to involve much more than the numerical comparison of predicted volumes versus actual. During the course of field operation, enhancements are constantly identified to increase field recovery as our understanding of the reservoir improves.