of about 0.3 ft/s. The direction of the near-bottom
a larger percentage of the water area in more-
water motion rotated 360 during the event, with
confined channel reaches. The numerical model
velocities in all directions significantly greater than
accompanying their report accounts for both a
the ambient downstream current. Numerous other
change in vessel effects with the presence of ice
data sets for the variation of water level and ve-
and with increasing ice thickness. For one exam-
locity are available in the reports by Alger (1977a,b,
ple given, a 1-ft open-water drawdown would
1978, 1979a,b). Nearshore drawdowns of up to 3
be increased as much as 33% for 18 in. of ice. For
ft have been measured on the connecting chan-
the same ice condition, the percentage increase in
nels (USACE 1974, Wuebben 1978), but the high-
drawdown would increase as ship speed increased.
est recorded values have been for large vessels
approaching or exceeding the speed limits. A field
Analysis of drawdown
study of drawdown and surge on the St. Lawrence
Most analytical and predictive work on draw-
Seaway (Normandeau Associates 1979) docu-
down in the Great Lakes connecting channels has
mented no drawdown events greater than 2 in.
employed a one-dimensional approach. Although
and felt that no reasonable correlation with ves-
a multi-dimensional treatment would provide
sel parameters was possible.
more detail, especially in regard to water veloci-
ties, there are insufficient data to calibrate or vali-
Ice effects
date an expanded treatment. Fortunately field data
In a study of the effects of winter navigation
(Wuebben et al. 1978a) show that the magnitude
on shoreline erosion and dock damage (USACE
of drawdown is relatively constant over most of
1974), continuous recordings of water level fluc-
the channel cross section during vessel passage
tuations were collected at various sites along the
and that a one-dimensional treatment predicts this
St. Marys River. While the ship- and wind-gen-
value within acceptable accuracy (Alger 1977a,
erated waves were well defined in the open-wa-
Wuebben 1983a, Hodek et al. 1986).
ter recordings, they were undetectable for peri-
If the channel cross section is not symmetrical
ods with ice cover, indicating total damping. In
or the ship passes closer to one shore, the one-
contrast, ship-generated drawdown and surge
dimensional results can be improved by assum-
were undamped and apparently even enhanced.
ing that no water crosses the sailing line so that
From data at an individual site, the authors
the section may be split into separate pieces for
were able to get a reasonable correlation between
calculation (Wuebben 1983a, Hodek et al. 1986).
measured drawdown values and an estimate of
For highly non-uniform flow distributions or com-
the drag force on a ship hull (USACE 1974). At
plex channel shapes, empirical cross-section shape
one site on the mainland shore of Lake Nicolet,
factors can also be included, but these are highly
ship passages were monitored during open-
site specific and cannot be reliably transferred else-
water conditions, early ice (0.30.5 ft) and mid-
where (Wuebben 1983a). The distribution of ve-
winter ice (11.3 ft). Although a comparison of
locities and sediment transport potential across
the drawdown and surge among these three con-
a river cross section cannot be directly consid-
ditions is somewhat limited by the relatively few
ered, however. Previous work has generally used
data points (22) and the scatter inherent in mak-
the existing field database to develop shore and
ing such measurements, their parameterization
shore structure damage criteria that can be em-
technique indicates that a drawdown of 0.4 ft
pirically correlated to one-dimensional modeling
during open-water conditions might be increased
(Wuebben 1981b, 1983a, Wuebben et al. 1984,
by about 40% during periods with ice. The data
Hodek et al. 1986).
did not indicate any clear difference with increas-
Wuebben (1981b, 1983a) and Wuebben et al.
ing ice thickness.
(1984) developed such a one-dimensional treat-
Hodek et al. (1986) also examined the effect of
ment to allow an assessment of vessel size on draw-
ice on the magnitude of drawdown. Although the
down and resulting sediment transport potential.
flexure and cracking of an ice cover would dissi-
For the long, parallel mid-body commercial ves-
pate some energy, they felt that the primary ef-
sels common on the Great Lakes, vessel length is
fect of an ice cover would be to decrease the area
relatively insignificant in determining drawdown.
available for flow and thus increase the magni-
A sensitivity analysis demonstrated that ship ve-
tude of drawdown. On that basis, smaller cross
locity is by far the most important variable con-
sections would be more severely affected by ice
trolling the magnitude of drawdown. As shown
since the same thickness of ice would constitute
in Figure 5, a change in vessel speed of 2 ft/s would
6
Previous Page
