Regional variability of slope stability: Application to the Eel margin, California

Relative values of downslope driving forces and sediment resisting forces determine the locations of submarine slope failures. Both of these vary regionally, and their impact can be addressed when the data are organized in a Geographic Information System (GIS). The study area on the continental margin near the Eel River provides an excellent opportunity to apply GIS spatial analysis techniques for evaluation of slope stability. In this area, swath bathymetric mapping shows seafloor morphology and distribution of slope steepness in fine detail, and sediment analysis of over 70 box cores delineates the variability of sediment density near the seafloor surface. Based on the results of ten geotechnical studies of submarine study areas, we developed an algorithm that relates surface sediment density to the shear strength appropriate to the type of cyclic loading produced by an earthquake. Strength and stress normalization procedures provide results that are conceptually independent of subbottom depth. Results at depth are rigorously applicable if sediment lithology does not vary significantly and consolidation state can be estimated. Otherwise, the method applies only to shallow-seated slope failure. Regional density, slope, and level of anticipated seismic shaking information were combined in a GIS framework to yield a map that illustrates the relative stability of slopes in the face of seismically induced failure. When a measure of predicted relative slope stability is draped on an oblique view of swath bathymetry, a variation in this slope stability is observed on an otherwise smooth slope along the mid-slope region north of a plunging anticline. The section of slope containing diffuse, pockmarked gullies has a lower measure of stability than a separate section containing gullies that have sharper boundaries and somewhat steeper sides. Such an association suggests that our slope-stability analysis relates to the stability of the gully sides. The remainder of the study area shows few obvious indications of slope instability except for a feature that has become known as the 'Humboldt Slide,' but it is too deep-seated to be amenable to the slope-stability-prediction techniques presented herein. In general, few slope failures have been mapped in the Eel margin study area despite the high level of seismicity, the relatively high rates of sediment accumulation, and the extent of gas charging observed by others.