PREDICTING RESERVOIR ARCHITECTURE OF TURBIDITE CHANNEL COMPLEXES: A GENERAL MODEL ADAPTABLE TO SPECIFIC SITUATIONS

Observations from numerous examples of turbidite channel systems in diverse settings and multiple basins have been summarized as a series of rules. These rules provide a useful basis for constructing predictive, detailed, 3-dimensional, event-based models of turbidite reservoirs. Interaction of the various rules allows for the development of a wide range of possible channel architectures, but the succession of architectures tend to follow a recurring pattern with 4 stages: 1) System-scale erosion; 2) Amalgamation of channel elements with lateral offset during a low rate of aggradation; 3) Disorganized stacking of channel elements during a moderate rate of aggradation; and 4) Organized stacking of channel elements during a high rate of aggradation. Depending on the proportion of sand in the system, some stages may be diminished in volume relative to the other stages, or even absent. Sand-rich systems tend to be dominated by stages 1 and 2 whereas stages 3 and 4 are more prominent in mud-rich systems. Similarly, gradient appears to influence the proportion of the 4 stages with stages 1 through 4 predominating in proximal areas of high gradient whereas stages 2 through 3 predominate in distal areas of low gradient. The 4 stages can develop with or without the presence of outer levees, but prominent outer levees imply abundant mud and high rates of aggradation which favor the development of stage 4 – organized stacking. Lateral accretion can occur in a variety of settings but is particularly common in mud-rich systems during stage 4. Uncertainty and variability can be addressed through multiple realizations of event-based models that account for uncertainties in the prevalence of stages or architectures and rules within each stage. Also, these event-based models can be conditioned to constraining data such as wells or seismic images. This has been demonstrated to work well specifically in sparse data settings common to deepwater reservoirs.