JpGU-AGU Joint Meeting 2020

Presentation information

[E] Poster

S (Solid Earth Sciences ) » S-CG Complex & General

[S-CG57] New perspectives of subduction zone earthquake dynamics through experiments across-scales

convener:Masataka Kinoshita(Earthquake Research Institute, University of Tokyo), Yoshio Kono(Geodynamics Research Center, Ehime University), Eiichiro Araki(Japan Agency for Marine-Earth Science and Technology), Hiroko Kitajima(Texas A&M University College Station)

[SCG57-P02] Changes in frictional properties of mud gouge with depth in the Nankai Trough accretionary prism

*Kyuichi Kanagawa1, Takuma Ogawa2, Tomoya Funaki2, Tomoya Nakanishi3, Sayumi Sagano3, Junya Fujimori3, Michiyo Sawai1 (1.School of Science, Chiba University, 2.Faculty of Science, Chiba University, 3.Graduate School of Science and Engineering, Chiba University)

Keywords:frictional properties, mud gouge, Nankai Trough accretionary prism

In order to investigate the changes in frictional properties of mud gouge with depth in the Nankai Trough accretionary prism, we conducted triaxial friction experiments on gouge of a mud sample cored from 2183.6 mbsf (meters below seafloor) at IODP Site C0002, at three conditions supposed at depths of 1000, 3000 and 5000 mbsf there. We conducted two series of friction experiments; slide-hold-slide experiments for static friction data and rate-stepping experiments for dynamic friction data. In the former series of friction experiments, sliding at an axial displacement rate of 1 μm/s was held for periods ranging from 10s to 104s, after which sliding was resumed at the same rate. In the latter series of friction experiments, axial displacement rate was changed stepwise among 0.1, 1 and 10 μm/s.

The slide-hold-slide experiments show that the coefficient of static friction μs decreases from 0.37–0.38 at the 1000 mbsf condition to 0.30–0.31 at the 3000 mbsf condition, while it increases to 0.33–0.36 at the 5000 mbsf condition. Healing Δμ (increase in friction coefficient upon resuming sliding after a holding time th) was not noticeable at th ≦ 1000s during the 3000 mbsf condition experiment, while Δμ was recognized at th ≧ 100s during the other experiments. Δμ increases logarithmically with th, which was more prominent at the 5000 mbsf condition. Dehydration of smectite at the 3000 mbsf condition where temperature was 100°C possibly increased pore pressure in the impermeable gouge layer, which was responsible for the minimum μs and non-healing effect at th ≦ 1000s at this condition. While lithification of gouge particles by pressure solution was possibly responsible for increasing μs and Δμ with th at the 5000 mbsf condition where temperature was 150°C.

The rate-stepping experiments show that the steady-state friction coefficient μss also decreases from 0.32–0.35 at the 1000 mbsf condition to 0.29–0.31 at the 3000 mbsf condition, while it increases to 0.32–0.33 at the 5000 mbsf condition. The minimum μss at the 3000 mbsf condition is also attributable to dehydration of smectite as in the case of μs. ab value (rate dependence of μss) decreases with depth from 0.0018–0.0062 at the 1000 mbsf condition to –0.0005–0.0017 at the 5000 mbsf condition, suggesting that a transition from aseismic faulting to seismic faulting occurs around the depth of 5000 mbsf at IODP Site C0002. Thus the megathrust supposed to be present at ≈5200 mbsf there is likely at a seismogenic condition. Dependence of ab value not only on temperature but also on displacement rate suggests that thermally activated pressure solution is relevant to this change in frictional stability.