[S-SS10] Fault Rheology and Earthquake Physics

[SSS10-24] Evaluation of fault behavior with a complex structure using dual-slip-plane friction experiment

*Wataru Tanikawa^1,3, Osamu Kuwano¹, Yohei Hamada¹, Takehiro Hirose¹, Takahiro Suzuki² (1.Japan Agency for Marine-Earth Science and Technology, X-star, Kochi Institute for Core Sample Research, 2.MARINE WORKS JAPAN LTD., 3.Kochi University)

Keywords:fault, friction test, shortening, compounded slip surface

Earthquake process is frequently discussed in a simplistic manner, assuming that it occurs as the movement of a single fault plane. However, in reality, natural faults are comprised of multiple faults or weak slip planes of various sizes, and in some cases, they cannot be simplified as a single fault plane. Moreover, the interaction between multiple faults in close proximity has not been evaluated experimentally. In this study, we developed a system to accurately measure the movement of each fault plane and the amount of shortening of each fault plane when shear stress is applied to two parallel fault planes in close proximity at the same time. We attached inductive angle encoders to each fault plane to observe the slip behavior. We also used multiple laser displacement sensors to measure the amount of shortening in the axial load direction of each fault plane. In this study, we report on the results of tests using silicon dioxide powder as a fault simulation material. Two types of friction experiments were conducted: (1) constant speed control (rotational speed 0.2 rpm = 0.18 mm/s) and (2) torque increase control (maximum speed 0.2 rpm). In addition, a uniaxial shear experiment was conducted using the same sample, and the differences between this and the biaxial shear experiment were compared. In both experiments, the maximum slip amount was set at 0.1 m, and a vertical load of 4 MPa was applied.
From the initiation of the slide to a sliding displacement of 1 cm, the two fault planes slid either concurrently or in minor, alternating increments. Subsequently, the movement of the fault on one side ceased, and only the other side began to slip. Following an initial slip of 1 to 2 centimeters, the slippage halted, and the fault on the opposing side initiated slip. Subsequent observation of the slip surface replacement behavior was recorded at intervals of 1 to 2 centimeters of slip displacement. The time variation in the friction coefficient can be interpreted as a smooth change, irrespective of the slip surface being replaced. However, upon examination of the relationship between the friction coefficient and the amount of slip for each fault, it was found that the friction coefficient of the lower fault, which exhibited a greater amount of slip, was lower than that of the upper fault when the same displacement was compared. Furthermore, it was determined that the amount of shortening per unit displacement was greater for the lower fault. The series of same slip behavior was observed in both the torque control experiment and the constant velocity experiment. A comparison of the single-slip-plane test and the dual-slip-plane test revealed that the latter exhibited a more rapid increase in the friction coefficient and a lower steady-state friction. Furthermore, the dual-slip-plane exhibited a more pronounced shortening rate during slip. The findings from these experiments offer significant insights that inform the ongoing discourse surrounding the discrepancy between earthquake fault behavior as derived from seismic observations and geodesy, and earthquake behavior estimated from material science analysis.