4.2. SHORT-WAVE HYDRODYNAMIC MODELS
99
at the middle of the basin. However, offshore of the surf zone, the longshore
currents were quite varied due to the different boundary conditions. Dalrymple, et al. (1977) gave recommendations for basin layouts that would
decrease this laboratory effect.
Correct simulation of wave transformation requires that the bathymetry
be molded in the wave basin with a reasonable degree of accuracy. Furthermore, reflections from side wall boundaries should be minimized to less
than 5% using absorbing beaches and wave absorbers (see Svendsen 1985).
When unidirectional currents are generated along with wave motion, the
currents can influence the direction of wave propagation. This could cause
waves to turn toward a highly reflective boundary, such as a wave guide or
a vertical side wall; and consequently, unwanted reflected waves might be
introduced into the basin study area.
Waves generated in basins can be either long-crested waves propagating
from a single direction or short-crested waves composed of waves coming
from a multitude of directions. Usually snake-type wave boards are used to
produce oblique waves, and the method of generation can produce unwanted
spurious long waves and also waves that traverse along the wavemaker. The
spurious long waves need to be controlled, particularly in harbor resonance
model studies in which irregular short waves create bound long waves that
excite the harbor in a low frequency mode of oscillation. Lateral waves
created by discontinuities between segments of a snake-type wave generator
can be damped by using screens placed in front of the wavemaker.
Scale Effects
Scale effects in short-wave hydrodynamic models result primarily from the
scaling assumption that gravity is the dominant physical force balancing
the inertial forces. Scaling based on this assumption (Froude scaling) incorrectly scales the other physical forces of viscosity, elasticity, surface tension, etc. with the belief that these forces contribute little to the physical
processes. Scale effects in physical models are analogous to decreased accuracy that occurs in numerical models when complex physical processes are
represented by simplified mathematical formulations (Kamphuis 1991).
Costa (1981, 1984, 1990) examined the influence of the “secondary”
forces, and presented tables that showing that the viscous, elastic, and surface tension forces are represented in the model many times greater than in
the prototype. He recommended (Costa 1984) that some cases of unsteady,
nonuniform flow with wide variations in velocity require variable length and
time scales throughout the model. A method was given (Costa 1990) to
calculate modifications to the Froude time scale if the relative magnitude
and direction of the “secondary” forces are known. This method should,
Précédent

- 115/590

Suivant