Storm‐induced turbidity currents on a sediment‐starved shelf: Insight from direct monitoring and repeat seabed mapping of upslope migrating bedforms

Abstract

The monitoring of turbidity currents enables accurate internal structure and timing of these flows to be understood. Without monitoring, triggers of turbidity currents often remain hypothetical and are inferred from sedimentary structures of deposits and their age. In this study, the bottom currents within 20 m of the seabed in one of the Pointe-des-Monts (Gulf of St. Lawrence, eastern Canada) submarine canyons were monitored for two consecutive years using Acoustic Doppler Current Profilers. In addition, multibeam bathymetric surveys were carried out during deployment of the Acoustic Doppler Current Profilers and recovery operations. These new surveys, along with previous multibeam surveys carried out over the last decade, revealed that crescentic bedforms have migrated upslope by about 20 to 40 m since 2007, despite the limited supply of sediment on the shelf or river inflow in the region. During the winter of 2017, two turbidity currents with velocities reaching 0Á5 m sec À1 and 2Á0 m sec À1 , respectively, were recorded and were responsible for the rapid (<1 min) upstream migration of crescentic bedforms measured between the autumn surveys of 2016 and 2017. The 200 kg (in water) mooring was also displaced 10 m down-canyon, up the stoss side of a bedform, suggesting that a dense basal layer could be driving the flow during the first minute of the event. Two other weaker turbidity currents with speeds <0Á5 m sec À1 occurred, but did not lead to any significant change on the seabed. These four turbidity currents coincided with strong and sustained wind speed >60 km h À1 and higher than normal wave heights. Repeat seabed mapping suggests that the turbidity currents cannot be attributed to a canyon-wall slope failure. Rather, sustained windstorms triggered turbidity currents either by remobilizing limited volumes of sediment on the shelf or by resuspending sediment in the canyon head. Turbidity currents can thus be triggered when the sediment volume available is limited, likely by eroding and incorporating canyon thalweg sediment in the flow, thereby igniting the flow. This process appears to be particularly important for the generation 1045 of turbidity currents capable of eroding the lee side of upslope migrating bedforms in sediment-starved environments and might have wider implications for the activity of submarine canyons worldwide. In addition, this study suggests that a large external trigger (in this case storms) is required to initiate turbidity currents in sediment-starved environments, which contrasts with supply-dominated environments where turbidity currents are sometimes recorded without a clear triggering mechanism.

The depositional signature of cyclic steps: A late Quaternary analogue compared to modern active delta slopes

“…The quasi‐draping bed conformation and/or restricted lee‐side reworking would have been, in this hypothesis, the result of cyclic‐step instabilities with flow properties precisely and recurrently adjusted to the pre‐existing cyclic‐step morphology. Because such a consistency is unlikely, it is reasonable to assume that the bedforms, apparently linked to the well‐bedded suites, do not chart the proper hydrodynamic signature of a parent flow but rather the response of an interaction between a train of pre‐existing bedforms and a superimposed flow (see also Normandeau et al ., 2020). It is contended that the currents originating the bulk of the well‐bedded suites of the Portneuf‐Forestville delta‐slope deposits do not correspond to flow dynamics envisioned from numerical or conceptual models for cyclic steps (e.g.…”

“…Here, the increasing development of rippled sand rather suggests flow deceleration and increasing sedimentation rates over an upslope dipping depositional surface rather than the expected general flow acceleration over the stoss side of a cyclic step. The fact that flows travelling over cyclically‐stepped surfaces are not necessarily cyclic steps was also suggested on the basis of flow monitoring in some other case studies (Hughes Clarke, 2016; Normandeau et al ., 2020).…”

“…Vellinga et al ., 2018) and inferred from stratigraphies reconstructed from repeat swath bathymetry (e.g. Hage et al ., 2018; Vendettuoli et al ., 2019; Normandeau et al ., 2020). It should be noted, however, that the backstepping conformation might be subtle without vertical exaggeration (Figs 5 and 12).…”

“…The repeat swath bathymetry from 2013 and 2014 clearly shows that sediments are not preserved on the lee side of the bedforms, which are zones of net erosion. It indicates an upslope migration of the crescentic bedforms, similar to those observed in other active submarine canyons and channels (Smith et al ., 2005; Paull et al ., 2010; Normandeau et al ., 2014, 2020; Hughes Clarke, 2016; Hage et a l ., 2018). Their crescentic shape and upslope migration suggest that the bedforms are formed by cyclic steps (Cartigny et al ., 2011; Hughes Clarke, 2016; Normandeau et al ., 2019).…”

“…The resulting bedforms are asymmetrical, often crescent‐shaped in confined settings, and have been observed to migrate upslope, both in numerical and flume‐tank experiments (Sequeiros et al ., 2009; Spinewine et al ., 2009; Yokokawa et al ., 2009, 2011; Kostic, 2011; Vellinga et al ., 2018). In natural case studies, related bedforms are inferred to migrate upslope from seismic stratigraphy (Covault et al ., 2014; Tubau et al ., 2015; Zhong et al ., 2015; Li & Gong, 2018; Normandeau et al ., 2019), as corroborated by repeat swath bathymetry mapping (Hughes Clarke et al ., 2014; Normandeau et al ., 2014, 2020; Hage et al ., 2018; Vendettuoli et al ., 2019). Asymmetrical bedforms and their upslope migration originate from continuous or successive sediment‐laden flow events combining: (i) lee‐side erosion linked to a Froude‐supercritical regime over the relatively steep slopes; (ii) hydraulic jump processes in the trough; and (iii) deposition from sediment density flows under the subcritical regime characterizing the bedform stoss sides (Cartigny et al ., 2014; Postma & Cartigny, 2014; Hughes Clarke, 2016; Hage et al ., 2018; Vellinga et al ., 2018; Slootman & Cartigny, 2020).…”