Objectives
The primary objectives of the workshop are to discuss several critical issues concerning the accuracy of long-wave runup models.
| 1. |
Calculations of the moving shoreline. This is certainly not a
new issue. The accuracy of the numerical algorithm used in locating the
shoreline has direct consequence in estimating the maximum tsunami
runup. The moving nature of the shoreline boundary makes the problem
nonlinear. Various kinds of numerical treatments have been invented and
some of them have caused numerical instability or dissipation. In some
cases, artificial damping is introduced into the numerical algorithm,
masking the instability and clouding the accuracy of solutions.
However, surprisingly, these issues have not been the focus of
discussions in the past workshops. With the development of higher-order
models for tsunami propagation and runup, such as the Boussinesq
equations, and the inclusion of more physical processes, such as wave
breaking, and bottom friction, it is important to create a forum to
discuss this difficult, but critical issue openly.
|
| 2. |
Modeling of bathymetry and topography. It is obvious that the
predictability of any numerical model depends largely on the source
region information (i.e., the initial condition) and the correct
representation of bathymetry and topography. There are two levels of
issues to be discussed as far as the bathymetry and topography data are
concerned. The first is the availability and the resolution of
bathymetry and topography data. The second concerns the grid resolution
requirement in a numerical model and its relationship to the resolution
of the bathymetry and topography data. The common practice for deciding
the grid resolution has been based on the characteristic wavelength.
However, there is very little understanding on the grid resolution
requirement in terms of bathymetric and topographical variations; i.e.,
slope and curvature of the seafloor and land surface. Furthermore,
different numerical schemes can be used to interpolate and/or
extrapolate a given set of bathymetry and topography data to obtain
approximated data at the grid points of a numerical model. The
differences in the resulting slope and curvature could affect the
numerical solutions for wave heights and the direction of propagation,
especially in the case of long distance propagation.
|
| 3. |
Landslide generated tsunami. The characteristics of landslide
generated tsunamis could be different from those generated by tectonic
movements. For example, the wavelength could be shorter and the
dispersion effects might be important. The traditional shallow water
equation model might not be adequate. Moreover, the duration of a
landslide is usually in the order of magnitude of a few minutes or
longer. Therefore, it is hard to justify not considering the landslide
motion in the tsunami generation process. On the other hand, our
knowledge of the dynamics of sub-aerial and sub-marine landslides is
still incomplete. The strategy for parameterizing the slide movement to
provide adequate initial conditions for tsunami propagation models is
an important topic to be discussed.
|
| 4. |
Tsunami forces on a nearshore structure. When tsunamis impinge a
structure on land, the flow system is very complex. Quite often the
waves are breaking and the flow filed is three-dimensional. The
depth-integrated hydrodynamic models are no longer suitable for
describing this complex flow. Instead, the Navier-Stokes Equations with
suitable turbulence model are required. The important modeling issues
include the selection of free surface tracking technique, the selection
of appropriate turbulent model for wave breaking generated and boundary
layer generated turbulence, and the interface between 3D model and the
depth-integrated hydrodynamic model. Although there are only a few
researchers are involved in this type of modeling efforts, it is a
challenge that should be brought to the attention of the entire coastal
and tsunami community. | |
i
