Submarine channels convey turbidity currents, the primary means for distributing sand and coarser sediments to the deep ocean. In some cases, submarine channels have been shown to braid, in a similar way to rivers. Yet the strength of the analogy between the subaerial and submarine braided channels is incompletely understood. Six experiments with subaqueous density currents and two experiments with subaerial rivers were conducted to quantify: (i) submarine channel kinematics; and (ii) the responses of channel and bar geometry to subaerial versus submarine basin conditions, inlet conditions and the ratio of ‘flow to sediment’ discharge (Qw/Qs). For a range of Qw/Qs values spanning a factor of 2·7, subaqueous braided channels consistently developed, were deeper upstream compared to downstream, and alternated with zones of sheet flow downstream. Topographic analyses included spatial statistics and mapping bars and channels using a reduced-complexity flow model. The ratio of the estimated depth-slope product for the submarine channels versus the subaerial channels was greater than unity, consistent with theoretical predictions, but with downstream variations ranging over a factor of 10. For the same inlet geometry and Qw/Qs, a subaqueous experiment produced deeper, steeper channels with fewer channel threads than its subaerial counterpart. For the subaqueous cases, neither slope, nor braiding index, nor bar aspect ratio varied consistently with Qw/Qs. For the subaqueous channels, the timescale for avulsion was double the time to migrate one channel width, and one-third the time to aggrade one channel depth. The experiments inform a new stratigraphic model for submarine braided channels, wherein sand bodies are more laterally connected and less vertically persistent than those formed by submarine meandering channels.
Bibliographical noteFunding Information:
Jeff Peakall and Associate Editor Jaco Baas provided constructive reviews. The SAFL Industrial Consortium supported this work. Jeff Marr, Rob Gabrielson, and Ben Erickson helped to design the basin, and David Baldus helped conduct the experiments. Charles Nguyen and Patrick Arnold provided computing support. Experiment methods are archived through the Sediment Experimentalist Network (http://www.sedexp.net/). Topography, image and flow model data are available at the Digital Repository for the University of Minnesota (https://doi.org/10.13020/D6R088).
- Braided rivers
- landscape evolution
- turbidity currents