Although widely adopted and unquestionably successful in improving production of sandstone reservoirs, water injection still raises controversy when applied to carbonate reservoirs. In general, the carbonate reservoir response to water injection is generally very unpredictable due to its complex origin and diagenetic processes. Carbonate reservoirs are commonly characterized by high heterogeneity, strong anisotropy and complex petrophysics. Moreover, structural processes can make these reservoirs highly faulted, compartmentalized and even fractured. These features add uncertainties upon fluid displacement efficiency and pressure maintenance on carbonate reservoir undergoing water injection. In this paper, we present the assessment of a water injection pilot project in a mature Albian carbonate reservoir showing low permeability, in the Campos basin, offshore Brazil. This heterogeneous, layered reservoir is not overall fractured, though is likely to present few fractures close to faulted regions.
The water injection project in this reservoir encompasses a pair producer-injector, both multi-fractured horizontal wells. Water injection assessment is the first and fundamental step aimed at guiding a revitalization study. After five years since its commencement, water injection seems to result in good sweep efficiency without channeling or recycling as suggested by production data. However, injection profile logs pointed out preferential regions in which almost the totality of water has been injected. Therefore, an important question still remained: how water injection in a so heterogeneous reservoir with a strong pressure contrast between layers could result in such apparently high sweep efficiency? As repeated seismic data were available, a time-lapse study was conducted to shed some light on this question.
The usual low sensitivity to saturation and pressure variations due to carbonate high incompressibility is known to reduce the chances of detecting time-lapse seismic anomalies. To overcome this difficulty we develop and apply a comprehensive methodology that includes a large feasibility study based on modeling at different scales and a meticulous 4D simultaneous pre-stack inversion. We apply this methodology to 4D processed seismic data, acquired in 2002 (streamer) and 2010 (ocean bottom cable).
The results overcome our expectation. Using time-lapse seismic and production data, a multidisciplinary interpretation team succeeds to relate 4D seismic anomalies to water and pressure changes occurring in the entire reservoir. Although being expected to directly reach the producer, injected water appears to be moving upward to the most permeable, depleted layer. Preferential directions of the injected water are now better understood.
In conclusion, 4D seismic data supported a better understanding of injected water paths, allowing future revitalization decisions to be made under stronger premises. The proposed 4D seismic methodology, which is demonstrated to work for this particular carbonate reservoir, will naturally be expanded to other similar Albian and even stiffer carbonates. We suggest that water injection projects on carbonates should also be monitored using time-lapse seismic data, once feasibility studies are favorable. With the recent huge new discoveries in carbonate reservoirs offshore Brazil, the perspective of being able to successfully use 4D seismic on these plays, in order to optimize hydrocarbon recovery, should have a huge economic impact.