Controls on grain-size variability in the Holocene fill of the Indus Submarine Canyon

Abstract

What processes control grain size and bed thickness in submarine canyon deposits? Erosive, shelf-cutting canyons contrast with accretionary basin-floor submarine fan accretionary channels because the former tightly constrain turbidity flows in deep channels. This study addresses such a deep-water depositional system in the Indus Submarine Canyon using a series of cores collected along the canyon. Grain-size analysis was conducted for turbidite and hemipelagic sediment deposited in the Holocene Indus Submarine Canyon mostly by diffuse, fine-grained turbidity currents and hemipelagic hypopycnal plumes. We investigate the links between sedimentary grain size, bedding thickness, facies, and canyon morphology. Well-sorted silt in layers mostly < 2 cm thick dominates the canyon. Core sites in the canyon located downstream of knickpoints have coarser, less well sorted sediments because of current acceleration in these areas and then the slowing of flows downslope. Sediments fine with increasing height above the canyon thalweg, implying deposition from a turbulent plume head. The great depth of the canyon, caused by the exceptionally wide shelf and steep slope, prevents channel overspill which controls sedimentation and channel form in submarine fans. Thalweg sediment fines down-canyon into the mid canyon, where sediment bypassing is inferred. The thickest turbidites are found in the sinuous lower canyon where the gradient shallows from ∼ 0.7° to 0.3°. However, canyon gradient has little impact on mean grain size, but does correlate with bed thickness. The active canyon channel, located in a channel belt gradually becomes less steep, more meandering, and narrower farther downstream. Sinuosity is an influence on turbidite bedding thickness but does not control grain size, in contrast to the situation in submarine-fan channel–levee complexes. Compared to the well-known, more proximal Monterey Canyon of California the grain sizes are much finer, although both systems show evidence of > 200 m plume heads.