2. Test and refine the 1996 provisional toe berm design if the initial model results
indicated that a toe berm is needed to protect the leeside armor layer from
undermining.
The region of Ventura Harbor entrance shown by the dashed line on Figure 3 was
constructed in a large wave facility at the Waterways Experiment Station in Vicks-
burg, Mississippi. The model was constructed at a prototype-to-model length scale of
NL = 25 on a flat-bottom portion of the flume that features a rectangular movable-
bed section. Flows representing longshore currents were generated by pumping water
through a wide current manifold, and vertical guide walls helped direct the flow to-
ward the gap between the north jetty spur and detached breakwater. Additional
description of the physical model configuration and operation is given in Hughes and
Schwichtenberg (1998).
Movable-Bed Modeling Criterion
The main difficulty with movable-bed models is obtaining correct similitude between
the prototype and model sediments. Ideally, for situations were sediment is mov-
ing primarily by bedload, the model sediment should scale the same as the length
scale, whereas suspended sediment transport appears to scale by sediment fall speed
(Hughes 1993). The mean sediment grain-size diameter in the Ventura Harbor area
is about 0.19 mm, and the model sand had a mean diameter of 0.13 mm, giving a
prototype-to-model sediment diameter scale of Nde = 1.46. Thus, model similarity
by strict similitude considerations for either bedload or suspended sediment transport
was impossible.
An alternative method for achieving model similarity with the prototype is to re-
produce a distinctive characteristic of the physical process. For the case of equilibrium
scour caused by steady flow, a logical choice is to maintain similarity of the equilib-
rium discharge relationship that balances the flow boundary layer with the critical
shear stress of the sediment.
First rearrange the form of the equilibrium discharge relationship given by Eqn. 12
in the Appendix into
V
= C onstant
(2)
3/8
1/8
)1/2
(gγi
de
he
where
s - ρw
ρ
(
γi =
3)
ρw
Similarity requires that the constant on the right-hand side of the Eqn. 2 must be
the same in the model as in the prototype. Taking the prototype-to-model ratio of
Eqn. 2, and expressing the result in terms of scale factors yields
3/8
NV = (Ng Nγi )1/2 Nde Nz /8
1
(4)
This scaling criterion is identical to that proposed by Kamphuis (1975) and derived
by similar boundary layer considerations.
8
Hughes/Schwichtenberg
Previous Page
