6.3. BEDLOAD-DOMINATED TRANSPORT MODELS
259
From the above relationships, we see that the Best Model (assuming
water in the model and a flat bed) allows the experimenter to select either
the model sediment diameter (smaller than prototype) or the length scale.
All other scale ratios then become fixed. The sediment must be the same
density as the prototype sediment, and all the hydrodynamics are specified
according to the geometrically undistorted Froude criterion in terms of the
length scale.
Best Model Scale Effects
A major limitation on the practical use of the Best Model is the requirement
that model sediment be reduced as the model length scale. In many cases
of importance in coastal engineering, this would result in model sediment
having diameters typical of clay; and this introduces a whole new set of
problems. So in essence the Best Model is only good for cases when the
prototype grain size is on the order of several millimeters or greater in
diameter.
Relaxing the grain size Reynolds number criterion in the Best Model
introduces possible scale effects related to viscosity, so it is necessary to
assume viscous forces related to sediment transport are small. This will
be reasonably satisfied so long as the boundary layer in the model remains
rough turbulent. Prototype grain size Reynolds numbers will be quite large,
and even with the large scale factor derived, the model Reynolds numbers
should also stay in the turbulent range. The exception will be at low flow
velocities or during flow reversals when the flow around the sediment particles becomes viscous earlier in the model than it would in the prototype.
Nevertheless, Kamphuis (1975, 1985) concluded that viscous scale effects
are minimal in the Best Model.
The ratio of sediment fall speed to shear velocity is less in the Best
Model than in the prototype, which means that scaling the sediment by
the length scale results in model sediment that takes too long to fall out of
suspension relative to the shear stress “driving force”. However, this scale
effect may not be a problem because the Best Model is practical only when
prototype grain sizes are relatively large, so suspended sediment transport
is probably minimal (Kamphuis 1991).
Derivation of the shear stress similitude assumed a flat bed. In reality
there will be ripples formed, and unless bottom ripple heights and lengths
appear reduced to the depth scale, the Best Model contains a distortion
in roughness. This may impact models of offshore regions (Motta 1986).
Kamphuis acknowledged that presence of ripple bedforms in the model and
prototype complicates the scaling process, but he maintained that scales
derived for flat bed motion are still approximately correct.
259
From the above relationships, we see that the Best Model (assuming
water in the model and a flat bed) allows the experimenter to select either
the model sediment diameter (smaller than prototype) or the length scale.
All other scale ratios then become fixed. The sediment must be the same
density as the prototype sediment, and all the hydrodynamics are specified
according to the geometrically undistorted Froude criterion in terms of the
length scale.
Best Model Scale Effects
A major limitation on the practical use of the Best Model is the requirement
that model sediment be reduced as the model length scale. In many cases
of importance in coastal engineering, this would result in model sediment
having diameters typical of clay; and this introduces a whole new set of
problems. So in essence the Best Model is only good for cases when the
prototype grain size is on the order of several millimeters or greater in
diameter.
Relaxing the grain size Reynolds number criterion in the Best Model
introduces possible scale effects related to viscosity, so it is necessary to
assume viscous forces related to sediment transport are small. This will
be reasonably satisfied so long as the boundary layer in the model remains
rough turbulent. Prototype grain size Reynolds numbers will be quite large,
and even with the large scale factor derived, the model Reynolds numbers
should also stay in the turbulent range. The exception will be at low flow
velocities or during flow reversals when the flow around the sediment particles becomes viscous earlier in the model than it would in the prototype.
Nevertheless, Kamphuis (1975, 1985) concluded that viscous scale effects
are minimal in the Best Model.
The ratio of sediment fall speed to shear velocity is less in the Best
Model than in the prototype, which means that scaling the sediment by
the length scale results in model sediment that takes too long to fall out of
suspension relative to the shear stress “driving force”. However, this scale
effect may not be a problem because the Best Model is practical only when
prototype grain sizes are relatively large, so suspended sediment transport
is probably minimal (Kamphuis 1991).
Derivation of the shear stress similitude assumed a flat bed. In reality
there will be ripples formed, and unless bottom ripple heights and lengths
appear reduced to the depth scale, the Best Model contains a distortion
in roughness. This may impact models of offshore regions (Motta 1986).
Kamphuis acknowledged that presence of ripple bedforms in the model and
prototype complicates the scaling process, but he maintained that scales
derived for flat bed motion are still approximately correct.