Micro-mechanics of weak layers: key role of sediment structure and composition

Abstract

Submarine landslides are gravity-driven mass movements that occur in underwater slope settings worldwide. They are one of the volumetrically most important processes for transporting sediments from the continental margin into the deep ocean. Despite the hazard they pose to coastal communities and critical seafloor infrastructure, many aspects of submarine landslides remain poorly understood. Our understanding of submarine landslides is often based on hypotheses that are hard to test, and we tend to infer landslide behaviour rather than understand the reason behind their formation. Sufficient information regarding the internal structure and composition, i.e. from sediment cores and in-situ measurements is often missing. Therefore, some key questions still remain unanswered, which include why some areas fail while adjacent slopes do not, or how submarine landslides can fail on low angle slops (<2°). Many studies proposed that these phenomena and the large areal extend of submarine landslides may be explained by laterally-extensive weak layers within the slope stratigraphy. Our knowledge regarding weak layers, in particular their compositional and structural characteristics, as well as the processes that control and form them, however, is still very limited. This thesis makes use of a variety of datasets at different scales and resolution in order to both qualitatively and quantitatively investigate the role of sediment structure and composition on weak layer and submarine landslide formation. Furthermore, the role of the environmental setting on the formation of weak layers, and their control on the triggering mechanism are investigated. Establishing such a relation is crucial to identify conditions (i.e. failure mechanism) under which slope failure may occur. Part of this thesis is a comprehensive literature review of published submarine landslide studies that examine the failure planes and apparent weak layers of historic and ancient submarine landslides, to evaluate what types of sediment are capable of forming weak layers and to understand their global distribution. The results show that failure planes usually form in the vicinity of an interface between distinct lithologies that together comprise a weak layer. The review further demonstrates that different types of weak layers show an affinity to specific geographical and physiographical locations. These include contourite or turbidite systems that can create siliciclastic sediment sequences, areas of high productivity or upwelling where biogenic sediments may dominate, or regions that experience repeated ash deposition from proximal or distal volcanic sources. Weak layers are further investivated by means of two selected case studies, a cohesive submarine landslide that occurred in a low angle sheeted contourite drift (namely the AFEN Slide) and a coastal retrogressive submarine landside that initiated along a regional turbidite event bed (namely the Finneidfjord Slide). The AFEN Slide is investigated using a combination of geophysical, sedimentological, geochemical, and geotechnical data. These data reveal abrupt lithological contrasts characterised by distinct changes in physical, geochemical and geotechnical properties. The findings indicate that failure likely initiated along this distinct climatically-controlled lithological contrast, which marks the boundary between a sandy contourite and underlying softer mud-rich sediments. Whether climate change played a role in triggering slope failure remains unclear, however, the data demonstrate its role in dictating the location of the failure plane. Furthermore, the results highlight the necessity to integrate high-resolution sediment core analyses and information about the regional setting to identify potential weak layers over the depth range of stratigraphy. The second case study, the Finneidfjord Slide offshore Norway, is investigated by means of high-resolution 3D micro-Computed Tomography imaging. The results reveal clear compositional and structural differences between individual sub-units of the weak layer, as well as the background sediment. The pore space distribution is highly spatially variable. Such high variability may be masked by bulk porosity measurements. Bulk-porosity measurements work on a centimetre-scale, while the observed changes are found on a millimetre-scale. Such differences, however, may be crucial for the formation of weak layers as they appear to dictate the location of the failure plane. These findings have important implications for understanding how weak layers are formed and their influence on failure plane formation. The results further enable a better constraint on the relation between environmental setting and weak layer distribution, as well as triggering and failure mechanisms.