13th International Conference on Fracture June 16–21, 2013, Beijing, China -7- 4. Preliminary experimental results Next, in order to try to confirm the numerical results, we perform laboratory photoelastic fracture experiments using a birefringent linear elastic material as well as a Q-switched Nd:YAG laser system or a projectile launched by a gun. We prepare a pre-cut interface in a polyurethane plate (50 mm × 25 mm × 12 mm), which is essentially under no static stresses. Dynamic fracture may be induced upon initiation of irradiation of laser pulses or incidence of a projectile (at a speed of 55.5 m/s) and propagated along the interface that is vertical or inclined at an angle of 45 degrees. We record the development of dynamic wave field with a high speed digital video camera system at a frame rate of 100,000 fps (e.g., Fig. 5 for incidence of a projectile), and at the same time, we monitor the particle motions on the free surface using a laser displacement meter (at positions 6 mm away from the surface-breaking point). The observations show that the experimental dynamic isochromatic fringe patterns are similar to the wave patterns generated by the numerical simulations, and if the inclined fault rupture reaches the free surface, the maximum vertical surface displacement on the hanging wall is about 4.6 times as large as that on the footwall (Fig. 6). On the contrary, as is expected from Fig. 3, when the fracture of the interface is arrested well below the free surface (arrested depth equal to a half of the plate height), the vertical motion on the footwall becomes about 1.2 times as large as that on the hanging wall. 5. Conclusions We have studied the dynamic model of the seismic wave field radiated by rupture of a dip-slip fault located near Earth’s free surface. The results of the numerical simulations and preliminary experiments suggest that the dynamic stresses and surface particle motions in the hanging wall, induced by nonvertical dip-slip faulting, may become larger or smaller than those in the footwall, depending on the depth of the arresting point of the fault rupture. We have also indicated that if the fault rupture starting at some depth approaches the free surface, four Rayleigh-type pulses (waves) may be produced: two moving along the free surface and the other two propagating back downwards along the ruptured interface. The downward interface pulses may considerably affect the stopping phase of the dynamic fracture process, and indeed, they might have governed the seismic rupture associated with the 2011 off the Pacific coast of Tohoku earthquake in Japan. If the fault plane is inclined, the interface pulse may interact with the Rayleigh pulse to generate a strong corner shear wave in the hanging wall. In the footwall, the interaction between the Rayleigh and interface waves is much smaller and the ground (free surface) motion is dominated simply by a weaker Rayleigh wave propagating along the free surface. The existence of downward interface waves and corner wave, which is not expected from a conventional seismological analyses where a fault ruptures only at depth, may play a crucial role in understanding the effect of the geometrical asymmetry on the strong motion induced by shallow dip-slip faulting. As stated above, the seismological recordings of recent earthquakes seem to support this idea, and there is certainly a need for careful consideration of dynamic fracture process along a shallow dip-slip fault plane. Although the model employed here is simple, they may provide a dynamic physical explanation of the observations associated with shallow dip-slip earthquakes. Figure 3 (continued). The τmax stress field induced by the (a) vertical and (b) nonvertical crack-like dip-slip faulting. Rupture starts at depth at time t = 0 and moves upward at a constant speed to suddenly arrest well below the free surface. The rupture propagation speed is assumed to be below the Rayleigh wave speed. In both cases, the rupture front wave, a strong disturbance in the proximity of the propagating rupture front, can be identified. In (b), the amplitude of the shear wave is larger in the footwall than in the hanging wall, because much of the energy carried by the rupture front wave in the hanging wall is diffracted at the tip of the broken fault segment and flows into the footwall.
RkJQdWJsaXNoZXIy MjM0NDE=