278
CHAPTER 6. SEDIMENT TRANSPORT MODELS
Distorted Models of Tidal Inlets
A number of geometrically distorted movable-bed model studies have been
conducted in the past to examine such inlet problems as shoaling and deposition, channel infilling, channel meandering, and scour. One of the earliest
such studies was a model of the Columbia River entrance conducted at
Berkeley during 1932 - 1936. O’Brien (1972) summarized this model study
and provided some insight into what it was like conduct a study during the
formative years of coastal physical modeling.
The purpose of the Columbia River model was to investigate evolution
of estuarine and bar channels under the influence of tidal currents and
waves. After preliminary testing, the horizontal scale was selected to be
Nx = 3600 and the vertical scale was Nz = 64. Model sand having d50 =
0.2 mm was used to simulate prototype sand of about the same size. Results
from the movable-bed model did not correspond well with the prototype
entrance; and consequently, further investigations were undertaken using
tracer material in a fixed-bed model. O’Brien (1972) stated in retrospect,
“The movable-bed model may have been more reliable than it was believed to
be at the time, and this approach may have been abandoned prematurely.”
Giese, Harten, and Vollmers (1974) described their experiences in conducting a geometrically distorted movable-bed model study of the Elbe
estuary. The model had a horizontal scale of Nx = 800 and a vertical scale
of Nz = 100. Lightweight polystyrene grains with a 2-mm diameter were
used as the model sediment. The morphological time scale for the model
was determined to be Ntm = 705, as compared to the Froude time scale,
Nt = a/100. This meant that morphological changes in the lightweight
sediment model occurred about 70 times more rapidly than predicted by
the Froude time scale.
The effectiveness of movable-bed tidal inlet models was investigated by
Jain (1982)9. He compared prototype measurements to predictions from a
geometrically distorted, movable-bed model of Galveston Harbor entrance
having Nx = 500 and Nz = 100. Model sediment was crushed coal with
d50 = 1.4 mm and a specific gravity of ps/p — 1.4. Both waves and tidal
currents were simulated in the model; however, model currents had to be
increased about 35% above the speed called for by the Froude velocity scale.
9 Also see the earlier paper by Jain and Kennedy (1976) which provided suggested
similitude requirements and proposed a method for quantifying the movable-bed model
performance.
Jain judged the model calibration unsatisfactory in terms of bed change
reproduction, and the model shoaling rates and spatial distributions in
the navigation channel were not in agreement with the prototype. The
differences between model and prototype were attributed to four causes:
CHAPTER 6. SEDIMENT TRANSPORT MODELS
Distorted Models of Tidal Inlets
A number of geometrically distorted movable-bed model studies have been
conducted in the past to examine such inlet problems as shoaling and deposition, channel infilling, channel meandering, and scour. One of the earliest
such studies was a model of the Columbia River entrance conducted at
Berkeley during 1932 - 1936. O’Brien (1972) summarized this model study
and provided some insight into what it was like conduct a study during the
formative years of coastal physical modeling.
The purpose of the Columbia River model was to investigate evolution
of estuarine and bar channels under the influence of tidal currents and
waves. After preliminary testing, the horizontal scale was selected to be
Nx = 3600 and the vertical scale was Nz = 64. Model sand having d50 =
0.2 mm was used to simulate prototype sand of about the same size. Results
from the movable-bed model did not correspond well with the prototype
entrance; and consequently, further investigations were undertaken using
tracer material in a fixed-bed model. O’Brien (1972) stated in retrospect,
“The movable-bed model may have been more reliable than it was believed to
be at the time, and this approach may have been abandoned prematurely.”
Giese, Harten, and Vollmers (1974) described their experiences in conducting a geometrically distorted movable-bed model study of the Elbe
estuary. The model had a horizontal scale of Nx = 800 and a vertical scale
of Nz = 100. Lightweight polystyrene grains with a 2-mm diameter were
used as the model sediment. The morphological time scale for the model
was determined to be Ntm = 705, as compared to the Froude time scale,
Nt = a/100. This meant that morphological changes in the lightweight
sediment model occurred about 70 times more rapidly than predicted by
the Froude time scale.
The effectiveness of movable-bed tidal inlet models was investigated by
Jain (1982)9. He compared prototype measurements to predictions from a
geometrically distorted, movable-bed model of Galveston Harbor entrance
having Nx = 500 and Nz = 100. Model sediment was crushed coal with
d50 = 1.4 mm and a specific gravity of ps/p — 1.4. Both waves and tidal
currents were simulated in the model; however, model currents had to be
increased about 35% above the speed called for by the Froude velocity scale.
9 Also see the earlier paper by Jain and Kennedy (1976) which provided suggested
similitude requirements and proposed a method for quantifying the movable-bed model
performance.
Jain judged the model calibration unsatisfactory in terms of bed change
reproduction, and the model shoaling rates and spatial distributions in
the navigation channel were not in agreement with the prototype. The
differences between model and prototype were attributed to four causes: