Experimental Study of Tsunami Generation by Three-Dimensional Rigid Underwater Landslides

Abstract

Large scale, three-dimensional, laboratory experiments are performed to study tsunami generation by rigid underwater landslides. The main purpose of these experiments is to both gain insight into landslide tsunami generation processes and provide data for subsequent validation of a three-dimensional numerical model. In each experiment a smooth and streamlined rigid body slides down a plane slope, starting from different initial submergence depths, and generates surface waves. Different conditions of wave nonlinearity and dispersion are generated by varying the model slide initial submergence depth. Surface elevations are measured with capacitance gauges. Runup is measured at the tank axis using a video camera. Landslide acceleration is measured with a microaccelerometer embedded within the model slide, and its time of passage is further recorded at three locations down the slope. The repeatability of experiments is very good. Landslide kinematics is inferred from these measurements and an analytic law of motion is derived, based on which the slide added mass and drag coefficients are computed. Characteristic distance and time of slide motion, as well as a characteristic tsunami wavelength, are parameters derived from these analyses. Measured wave elevations yield characteristic tsunami amplitudes, which are found to be well predicted by empirical equations derived in earlier work, based on two-dimensional numerical computations. The strongly dispersive nature and directionality of tsunamis generated by underwater landslides is confirmed by wave measurements at gauges. Measured coastal runup is analyzed and found to correlate well with initial slide submergence depth or characteristic tsunami amplitude.

The Papua New Guinea tsunami of 17 July 1998: anatomy of a catastrophic event

“…4d) are based on the revised slump architecture presented above. The new modelling also benefits from recent advances made in understanding tsunami generation from SMFs, that have now clarified and/or validated assumptions made previously Watts et al, 2005a;Enet and Grilli, 2007). We have also refined the grid spacing, and in place of the previous 100×100 m grid we have now used a uniform 50×50 m grid.…”

“…The initial water depth over the middle of the slump is 1420 m. These results are the tsunami source parameters. In our simulations, the transformation of a 2-D into a 3-D tsunami source uses mass conservation considerations along the third dimension (Watts et al 2005a;Grilli et al, 2002;Enet et al, 2005Enet et al, , 2007. In the most recent simulations, both the free surface shape and horizontal velocities of the 3-D slump tsunami source calculated at time t=t o are specified as initial conditions in FUNWAVE.…”

“…These shear and friction values are now validated for slumps Watts et al, 2005a) even though the values are greater by one to two orders of magnitude than those of most translational landslides. The equation of SMF motion previously applied (see Watts et al, 1999Watts et al, , 2002 was fully derived, experimentally validated, and the secondary effect of SMF deformation on tsunami generation quantified Watts et al, 2005a;Enet and Grilli, 2007). The transformation of a 2-D to a 3-D tsunami source, using mass conservation considerations along the third dimension, was fully supported by 3-D numerical experiments, and validated by large-scale 3-D laboratory experiments (Watts et al 2005a;Grilli et al, 2002;Enet et al, 2003Enet et al, , 2005Enet et al, , 2007.…”

“…The slump centroid is located at longitude and latitude (142.2582 E, 2.8791 S) and, based on experimental results (e.g. Enet and Grilli, 2007), the deepest trough of the tsunami source is usually positioned above this location. We also ran simulations with the deepest trough located at (142.2546 E, 2.8610 S), and (142.2509 E, 2.8429 S), respectively downslope at, 2 km or one-quarter wavelength, and 4 km or a half wavelength.…”

Frictional heating lubrication for submarine landslide

“…Submarine landslides are often modeled as (1) an incompressible Newtonian viscous laminar flow Leblond 1992, 1994;Heinrich et al 2000), (2) a non-Newtonian mud flow (Jiang and Leblond 1993;Assier-Rzadkiewicz et al 1997), or (3) a sliding or slumping rigid block (Harbitz 1992;Watts 1999, 2005;Hürlimann et al 2000;Ward 2001;Grilli et al 2002;Liu 2002, 2005;Liu et al 2005;Enet and Grilli 2007;Hornbach et al 2007;De Blasio 2011;Wang et al 2011;Gargani et al 2014). As was described in De Blasio (2011, p. 319), it is often impossible to assess the rheology of submarine landslides.…”