A Two-Dimensional, High-Resolution Model of Submarine Channel Sedimentation: Implications for Heterogeneity

Abstract

Submarine channels often form reservoirs with a complex three-dimensional architecture and significant heterogeneity that involves rapid changes from a thick-bedded and amalgamated channel axis succession to lower N/G, thinner-bedded margin, terrace, and levee deposits. The channel-related heterolithic facies can hold a significant percentage of the hydrocarbon reserves, yet the temporal and physical link between the thick-bedded axial channel deposits that record a small number of flows and the much larger number of thin-bedded turbidites forming terrace and levee deposits is poorly understood. We have developed a simple two-dimensional model that, given a few input flow parameters (mean velocity, grain size, duration of deposition, flow thickness), predicts the thickness and composition of the turbidite that is left behind in the channel and in the overbank areas. The model is based on a Rouse-type suspended sediment concentration profile and the Garcia-Parker entrainment function. In the vertical direction, turbidites tend to rapidly become thinner and finer-grained with height above thalweg, due to decreasing concentration. High near-thalweg concentrations result in thick axial beds. However, an increase in flow velocity can result in high entrainment and no deposition at the bottom of the channel, yet a thin layer of sand and mud is still deposited higher up on the channel bank. If channel thalwegs are largely in a bypass condition, relatively minor velocity fluctuations result in a few occasionally preserved thick beds in the axis, and numerous thin turbidites, and a more complete record, on the channel banks. We use near-seafloor data from the Niger Delta slope and an optimization algorithm to show how our model can be used to invert for likely flow parameters and match the bed thickness and grain size of ~100 turbidites observed in a core taken from a slope channel terrace. The same approach can be used to predict the distribution of reservoir facies and properties given an arbitrary cross-sectional profile.

AAPG Datapages/Search and Discovery Article #90323 ©2018 AAPG Annual Convention and Exhibition, Salt Lake City, Utah, May 20-23, 2018