Role of Physical Experiments and Near-Seafloor Reservoir Analogues in Modeling Deepwater Reservoirs

Carlos Pirmez¹, Mark Barton¹, Ciaran O'Byrne¹, Zoltan Sylvester¹, Frans van der Vlugt¹, Mark Deptuck¹, Bradford Prather¹, David Mohrig², Jasim Imran³, Gary Parker⁴, James Buttles², Kyle Straub², Alessandro Cantelli⁵, Heqing Huang³, Brandon McElroy², and Sara Johnson^5
1 Shell International E & P, Houston, TX
² Massachusetts Institute of Technology, Cambridge, MA
³ University of South Carolina, Columbia, SC
⁴ University of Illinois, Urbana, IL
⁵ University of Minnesota, Minneapolis, MN

The scale of heterogeneities (flow baffles and barriers) in hydrocarbon reservoirs is such that despite major advances in 3d imaging and detailed outcrop studies a wide range of uncertainty needs to be included in static model scenarios. Understanding of physical processes of erosion and deposition, and the resulting stratal architecture can be gained from detailed outcrop analogue studies, but these are largely limited to two-dimensional exposures. Depositional systems at the seafloor offer the ability to study reservoir analogues at very high-resolution, but sampling at the scale of outcrop sections is still limited. A complementary approach is to generate synthetic deposits in the laboratory, where one can directly link the flow processes and the resultant deposits. We have generated synthetic channel fills that are scalable to natural depositional systems. Such laboratory-scale reservoir models, coupled with numerical models of flow, erosion and deposition, provide the ability to enhance our knowledge of how reservoir heterogeneities are distributed in three-dimensions, and how they evolve through time. The flume tank and numerical models close the 3d gap in our knowledge base, and serve as a testing ground to challenge our conceptual models based on outcrop observations, or on limited subsurface measurements. In addition, the resulting synthetic deposits can be visualized in great detail using three-dimensional imaging software and quantified in the same way we study outcrop exposures, thus augmenting our quantitative databases of reservoir heterogeneities. This new breed of reservoir analogue tools has the potential to reduce uncertainties and enhance our ability to forecast production.