Submarine landslides and debris flows are widely distributed in modern and ancient deep-water sedimentary basins, and can damage marine infrastructures and generate destructive tsunamis. However, the mechanisms that initiate slope failures and facilitate long runout of flows on gentle slopes remain poorly understand. This thesis integrates seismic reflection data and cores from Exmouth Plateau, offshore NW Australia, with subaqueous granular-flow experiments, to investigate flow-substrate interactions that account for submarine failure initiation and flow evolution processes. The results show that large-scale submarine landslides on gentle carbonate slopes are fundamentally governed by sediment composition. Clay-capped foraminifera-rich successions lead to excess water preservation within chambered foraminifera shells, causing the formation of weak layers. Under contractive shearing, basal weak layers release pore water through crushing of foram shells, causing elevated pore pressure and generating a lubrication layer that facilitates runout of the overriding slide. The study demonstrates that submarine debris flows can become self-confined through a combination of lateral depositional relief and substrate incision. The self-confining process establishes a positive feedback, in which flow confinement drives substrate incision that in turn increases flow confinement, thereby lengthening flow runout distances even on gentle slopes.
Experimental subaqueous granular flows form coarse-enriched lateral levees and fines-enriched conduit deposits through particle segregation, like their subaerial counterparts, but are more mobile due to interstitial fluids. In subaqueous multiple-flow experiments, the presence of lateral levees increases the velocity of subsequent flows. This results in substrate erosion and downslope particle remobilisation that increases the flow volume, substrate incision, flow confinement, and runout distance. Together, these findings provide a new process-based model for how subaqueous landslides initiate, evolve, and reach long runout distances even on gentle slopes. This thesis highlights the importance of dynamic flow-substrate interactions (i.e., lubrication, deposition, or erosion), with implications for elucidating flow dynamics and sedimentary processes and assessing submarine geohazards.
