MORPHOLOGIC AND TEMPORAL DISTRIBUTION OF SUBMARINE LANDSLIDES ALONG THE CASCADIA MARGIN

While turbidite deposits are central to our understanding of Cascadia Subduction Zone earthquake recurrence, less work has been conducted on landslide sources and deposits along the entire margin. We present a newly mapped inventory of 190 large (> 0.1 km²) submarine landslides along a ~70,000 km² swath of the outer deformation wedge of the Cascadia Subduction Zone, including landslide type, mobility, and deposit extents. Variations in landslide type, scale, and mobility are found to track with underlying morphotectonic regions of the Cascadia margin (Watt and Brothers, 2020). This correlation with mapped regions of the margin may suggest control on observed landslide patterns may be driven by variations in source geology and topography. We find high rates of more mobile, and smaller, landslides in northern Cascadia (British Columbia and Washington) relative to the less frequent submarine landslides of southern Cascadia (southern Oregon and northern California). Furthermore, the total volume of landslide deposits in southern Cascadia are found to be more than double that of northern Cascadia. Using 30 m resolution bathymetry data, we compute seafloor roughness of the mapped landslide deposits via standard deviation of slope and 2D continuous wavelet transforms at multiple scales. As in sub-aerial landslide studies, deposit roughness is found to successfully predict cross-cutting relationships, where rougher (inferred younger) deposits intersect smoother (older) ones. Adopting terrestrial landslide methods for estimating landslide age from roughness, we then develop an uncalibrated-age distribution of landslides along the Cascadia margin. Uncalibrated landslide ages decrease from northern to southern Cascadia, which may be consistent with changes in sediment, slope geometry, or inferred rates of Cascadia earthquakes. Further studies on the susceptibility of these slopes to failure, and refinement of age estimates via seascape evolution modeling and dating of landslide deposits are needed to disambiguate the relationship between earthquake shaking and landslide susceptibility in the submarine landslide inventory.