Implications for megathrust slip behavior and pore pressure at the shallow northern Cascadia subduction zone from laboratory friction experiments

Abstract

Geological records of past megathrust earthquakes on the Cascadia subduction zone give a valid reason to assume that another event will occur in the future. Both the magnitude and tsunamigenic potential of such an earthquake depend on the extent of seismic rupture and slip on the shallow megathrust, which are still under debate for Cascadia. The absence of shallow slow slip events at northern Cascadia has recently been interpreted to result from a megathrust that is locked and potentially seismogenic all the way to the trench. A crucial factor controlling both the nucleation and propagation of an earthquake is the frictional slip behavior of the material in the fault system, of which little is currently known for this setting. Here, we report on laboratory friction experiments on all major lithologies of Juan de Fuca plate sediments being input to the northern Cascadia subduction zone. The material was obtained from cores recovered during Integrated Ocean Drilling Program Expedition 301 at Site U1301 in the depth range of 65 to 260 meters below the seafloor. From tests on both intact and powdered samples under in-situ effective normal stress and room temperature, we observe a significant decrease in friction coefficient from 0.76 to 0.19 with increasing depth. A frictionally weak hemipelagic illite-rich unit located directly above the basement is expected to host the décollement upon subduction. This unit reveals predominantly velocity-strengthening frictional behavior at slip velocities of ∼10−9 to 10−4 m/s, and exhibits no slow slip events in the laboratory, indicating a frictionally stable material. Coulomb wedge modeling based on our friction measurements and interpretation of décollement stratigraphic location suggests significant overpressure on the décollement near the trench. Our data offer a possible explanation for the lack of shallow slow slip activity on the northern Cascadia plate-boundary fault, and raise the possibility that the megathrust is not locked near the trench.