Limitations of Fluvial Analogs for the Dynamical Interpretation of Submarine Channel Systems: a Physical Modeling Case Study of Leveed Channel Formation.

Inferring process from morphology and architectures is commonly used in the interpretation of the dynamics of depositional submarine systems based on seismic and outcrop observations. The linking form to process often requires that an observed morphology is produced by a unique set of dynamics. Accordingly, submarine channel dynamics are often inferred from observed channel form through analogy with subaerial fluvial systems. For example, prograding deltaic channels bear striking morphological resemblance to leveed submarine channels that traverse continental slopes and submarine basins. However, given the different physical circumstances of these two environments, it is unclear how analogous are the relationships between channel form and flow dynamics in these two systems. To test whether the jet morphodynamics responsible for deltaic channel formation were applicable to submarine settings, we conducted a series of physical experiments in which sediment-laden flows that were considerably denser than the surrounding fluids were fully submerged in the less dense fluid. Given these dynamical conditions, we were unable to find a set of experimental conditions under which a pair of bounding levees could develop from a non-eroding sediment- laden density current undergoing sudden unconfinement. At supercritical bulk Richardson numbers, high rates of entrainment of the surrounding fluid into the sediment-laden flow led to rapid current deceleration and deposition of sediment along the axis of the flow. As bulk Richardson numbers increased, fluid entrainment decreased, allowing the current to flow further into the basin. Simultaneously, however, the rate at which the current collapsed laterally increased and as a result depositional patterns transitioned from one in which sediment was deposited parallel to the margins of the flow outlet to one in which sediment was radially distributed. As the denser flows collapsed, sediment was more broadly distributed, and consequently, relief of the deposited material was lower than observed for deposits produced by low Richardson number flows. Depositional patterns produced by dense flows differ significantly from those produced by the jet-like processes that govern deposition by deltaic channels, highlighting the fact that while such forms within subaerial and submarine environments may be similar, the underlying dynamics may be quite different. Our inability to produce self-organizing channel-levee systems for dense, depositional flows leads us to speculate that construction of leveed channels in the submarine environment may first require bed incision to provide partial confinement of the current.