ABSTRACT

Since the emergence of time-lapse, TL, seismic technology as a viable element of reservoir management, monitoring production and enhanced oil recovery programs, in carbonates, has been very challenging. Coupled effects of complicated petrophysical and/or lithofacies heterogeneities “difficult to model or image,” low compressibility “high stiffness, ” and thinness of carbonates have made seismic monitoring a high-risk component of EOR programs requiring both highly repeatable 4D-seismic acquisition and weak anomaly-sensitive processing and interpretation approaches. Most important is the need to maximize interdisciplinary synergy among all reservoir management team members. In this study, we present a new non-differencing approach called “Parallel Progressive Blanking” ((PPB) for detecting time-lapse seismic attributes anomalies. This approach is more efficient than differencing TL attributes and/or data in cases where weak anomalies are concealed by non-repeatable noise. Four 4D-seismic datasets, interpreted seismic lineaments, seismic facies, sedimentological models and production data provided a means of testing and validating the PPB approach of interpreting weak TL-seismic attributes of anomalous zones observed during a miscible flood. Using four TL-seismic datasets and applying the PPB approach, we have successfully monitored changes in seismic response related to the miscible bank in the Hall-Gurney Field in Kansas. A predicted “via reservoir simulation” breakthrough in April, 2004 in well 12 and an unpredicted delayed response from well 13 have been in compliance with the interpreted spatial outline of the injected bank. Changes in pump strategies between two seismic surveys resulted in retreating of the plume. Relatively fine (two months) survey-to-survey time spacing has assisted qualitative estimation of both reservoir heterogeneity and robustness of implementing the PPB approach, furthermore, dynamic flood management was aided by TL-findings.

This content is only available via PDF.
You can access this article if you purchase or spend a download.