--> 3-D Modeling of Flank Margin Caves, by Richard Labourdette, Ioan Lascu, and John Mylroie; #90052 (2006)
[First Hit]

Datapages, Inc.Print this page

Previous Hit3-DNext Hit Modeling of Flank Margin Caves

Richard Labourdette1, Ioan Lascu2, and John Mylroie2
1 Total S.A, Pau, France
2 Mississippi State University, MS

An important challenge in petroleum modeling is to represent the diagenesis effect “sensus largo” and all associated features such as karst cavities, porosity, and permeability characteristics by describing their geometry and generating processes. The focus of this study was flank margin caves developed on carbonate islands, taking into account numerous controversies concerning possible preservation over geologic time (paleo-karst). Due to the numerous measurements and observations available on Quaternary features (e.g. Bermuda and Bahamian Islands) and their robust hypothesis of genesis, island karst makes a good subject to develop concepts and modeling methods, which can be applied to other diagenetic features. The speleogenesis of flank margin caves was modeled following deterministic and stochastic methods in order to describe their conceptual generating processes.

Using the topography of Long Island, Bahamas, a fresh-water lens surface was built and used in simulating virtual flank margin cave development, in several growth stages. Focussing on preservation potential and considering the area affected, the model was extended to a cave-collapse condition. A cave of large size and characteristic morphology (Salt Pond Cave) was chosen as a reference and reconstructed in 3D from survey data. Finally, comparisons were made using statistics from the cave simulations of Salt Pond Cave and other Bahamian caves, to support our modeling processes and results.

The results reveal great resemblance and statistical Previous HitcoherencyTop between modeled and observed data (including the side effects of collapse) concluding that stochastic modeling is the optimal methodology for diagenetic feature restitution if driven by robust concepts and modeling methods.