Projection of mechanical properties from shallow to greater depths seaward of the Nankai accretionary prism

Deformation processes in sediments at accretionary prisms are directly controlled by the state of effective in situ stress, the mechanical-, physical- and geochemical properties of the materials of the fault zone and surrounding wall rocks, as well as time. Measurements of these properties and their evolution in space and time, are therefore needed for a full understanding of the process of earthquake generation within subduction zones.Reconsolidation tests have been carried out on Ocean Drilling Program cores collected from a reference site seaward of the active Nankai décollement zone off the southeast coast of Japan. The reconsolidation stress path subjects the samples to uniaxial strain deformation, which mimics their stress history, however at much higher loading rates than in the natural system. We have conducted two tests each from two mudstone samples within Lower Shikoku Basin. The samples were collected at 361 and 476 m below seafloor, on either side of the protodécollement horizon.The objectives for mechanical testing are to probe the yield- and failure surfaces of these shallow sediments (< 0.5 km depth), and to project their mechanical properties from shallow to greater depths (several kilometers depth). This information is useful for making predictions about sediment response to accretion, underplating, and slip along the décollement. Because the tests were executed with a stress path that may approximate the stress history of the test samples, an additional objective is to estimate the effective in situ vertical stress, and to constrain the pore-fluid pressure at sample depth.Considering their large scale behavior, our tests show that the samples collected above the protodécollement have higher strength than those below. We propose that cementation, microfabric and mineralogy of the sediments above the protodécollement result in a higher effective yield stress than predicted from effective in situ vertical stress at hydrostatic pore pressures. Sediments below the protodécollement, in contrast, are slightly underconsolidated, and provide an upper constraint on the magnitude of effective in situ vertical stress and pore-fluid pressure. We also used the test results to make initial predictions for the yield surface in 2D and 3D for sub-décollement samples across the margin. The construction of the 2D yield surface is the first attempt to quantify the model of sediment deformation proposed by Morgan et al. (2007). These results hint that the presence of cement has a strong, and increasing, influence on sediment behavior. Further testing is needed to verify these findings.