Submarine landslides are a common type of disaster which threaten property and the safety of human life. To effectively prevent and control such disasters, we conduct a series of large-scale physical model tests to determine the mechanism of submarine landslides. First, a large-scale physical model test system is designed and developed, including flume test frame, wave-making system, wave-absorbing system, and data monitoring system. In the tests, we investigate the effect of different sea waves by changing the parameters of the wave-making system and the influence of the slope inclination by constructing different models. Data regarding the wave pressure acting on the slope surface, seepage pressure, and displacement are monitored during the test procedure. The test results show that the seepage pressure in the faults varies cyclically with the sea waves and is lower at internal points than at outcrops. If the wave loading time is sufficiently long, the seepage pressure and displacement deformation in the fault zone will gradually increase. In other words, failures in fault zones precede submarine landslides. The weak fault zone provides the preferred sliding surface, and the sea waves supply the external dynamic energy for submarine landslides. The conclusions provide guidelines for similar engineering and research.
Geodetic measurements from 1963 through 1994 are used to estimate horizontal strain rates across the Red River fault near
Thac Ba, Vietnam. Whether or not this fault system is currently active is a subject of some debate. By combining: (1) triangulation
from 1963, (2) triangulation in 1983, and (3) Global Positioning System (GPS) observations in 1994, horizontal shear strain
rates are estimated without imposing any prior information on fixed stations. The estimated rates of shear strain in ten triangular
subnetworks surrounding the fault trace are not significantly different from zero at 95% confidence. The maximum rate of dextral
shear is less than 0.3 μrad/year in all but one of the triangles. The estimates help bound the slip rate in a simple elastic
dislocation model for a locked, vertical strike-slip fault. By assuming a locking depth of 5–20 km, the most likely values
for the deep slip rate are between 1 and 5 mm/year of right-lateral motion. These values delimit the 23% confidence interval.
At 95% confidence, the slip rate estimate falls between 7 mm/year of left-lateral motion and 15 mm/year of right-lateral motion.
Received: 18 November 1997 / Accepted: 28 January 1999 相似文献
