Link between the Nankai underthrust turbidites and shallow slow earthquakes.

Trench sediments such as pelagic clay or terrigenous turbidites have long been invoked to explain the seismogenic behavior of the megathrust fault (i.e., décollement). Recent numerous studies suggest that slow earthquakes may be associated with huge megathrust earthquake; however, controls on the slow earthquake occurrence remain poorly understood. We investigate seismic reflection data along the Nankai Trough subduction zone to understand the correlations between the spatial distribution of the broad turbidites and along-strike variations in shallow slow earthquakes and slip-deficit rates. This report presents a unique map of regional distribution of the three discrete Miocene turbidites that underthrust apparently along the décollement beneath the Nankai accretionary prism. A comparison of distributions of the Nankai underthrust turbidites, shallow slow earthquakes, and slip-deficit rates enables us to infer that the underthrust turbidites may cause primarily low pore-fluid overpressures and high effective vertical stresses across the décollement, leading to potentially inhibiting the slow earthquake occurrence. Our findings provide a new insight into potential role of the underthrust turbidites for shallow slow earthquakes at subduction zone.

Connection between high pore-fluid pressure and frictional instability at tsunamigenic plate boundary fault of 2011 Tohoku-Oki earthquake.

The 2011 Tohoku-Oki earthquake (M 9.0) rupture propagated along a shallow plate boundary thrust fault (i.e. decollement) to the trench, displaced the seafloor, and triggered a devastating tsunami. Physical properties of the underthrust sediments which control the rupture propagation are yet poorly known. We use a 2D seismic dataset to build velocity model for imaging and apply reverse time migration. We then calculate pore-fluid pressure along the decollement as the top boundary of underthrust sediments, and along the backstop interface as the boundary between undeformed structures in the continental plate and the severely deformed sediments in the accretionary prism. The results show that within horizontal distance of 40-22 km toward the trench, pore-fluid pressure is 82-60% higher than the hydrostatic pressure for both decollement and backstop interface. It then reduces to hydrostatic level for the backstop interface but remains 60-40% higher than hydrostatic level for the decollement, causing frictional instability in favor of fault rupture along the decollement. We report for the first time, by our knowledge, detailed seismic images of fluid-rich trapped bucket sediments, quantitative stress states, and fluid drainage conditions at shallow tsunamigenic portion of the Japan Trench, which are consistent with the seafloor and borehole observations.

Mantle-derived helium released through the Japan trench bend-faults.

Plate bending-related normal faults (i.e. bend-faults) develop at the outer trench-slope of the oceanic plate incoming into the subduction zone. Numerous geophysical studies and numerical simulations suggest that bend-faults play a key role by providing pathways for seawater to flow into the oceanic crust and the upper mantle, thereby promoting hydration of the oceanic plate. However, deep penetration of seawater along bend-faults remains controversial because fluids that have percolated down into the mantle are difficult to detect. This report presents anomalously high helium isotope (3He/4He) ratios in sediment pore water and seismic reflection data which suggest fluid infiltration into the upper mantle and subsequent outflow through bend-faults across the outer slope of the Japan trench. The 3He/4He and 4He/20Ne ratios at sites near-trench bend-faults, which are close to the isotopic ratios of bottom seawater, are almost constant with depth, supporting local seawater inflow. Our findings provide the first reported evidence for a potentially large-scale active hydrothermal circulation system through bend-faults across the Moho (crust-mantle boundary) in and out of the oceanic lithospheric mantle.