interaction force during the tests. We observed that
DISCUSSION
the ice sheet buckled many times during a test, and
The functions of a riprap armor are to protect a
that the buckling failure limited the maximum force
sloping embankment against waves or currents in
an advancing ice sheet could exert during the small-
a channel at a minimum-cost balance between ini-
scale tests.
tial construction cost and future maintenance. In
There was very little or no damage to the riprap
cold regions, the design procedures against wave
during ice ride-up events, which occurred on shal-
action or erosive forces of currents remain the same
low-slope (1V:3H) banks. Most of the riprap dam-
as those in warm regions. In addition, Matheson
age occurred when the ice piled up on the riprap,
(1988) discussed that one should consider the
and the incoming ice sheet was forced to go be-
breakdown of rocks due to freeze-thaw action,
tween the riprap and the piled-up ice. We observed
plucking of rocks by rising and falling ice sheets
that some of the rock material was removed from
due to water level changes, and shoving action by
the bed and brought up to the surface of the ice
moving ice sheets.
pile. The most severe damage occurred at or be-
The observations made in relation to dams and
low the water level. We considered the riprap to
rivers (Matheson 1988, Doyle 1988, Wuebben 1995)
have failed if the rock material was displaced by
have been discussed earlier. To avoid plucking of
ice to expose a certain amount of area on the bank.
rocks by rising ice sheets, Matheson (1988) sug-
From the results of these tests, we conclude that
gested that riprap should have a D50 in excess of
the maximum size (D100) of rocks should be twice
the maximum winter ice thickness. Though based
the ice thickness for shallow slopes (1V:3H) to sus-
on a limited number of tests, the results of this
tain no damage by ice shoving, and that the maxi-
study show that the maximum rock size (D100) in
mum rock size (D100) should be about three times
a riprap should be two to three times the ice thick-
the ice thickness for steeper slopes (1V:1.5H).
ness to avoid any damage to riprap by ice shov-
ing. Costs of construction and maintenance for a
given site should also be considered. It may be
LITERATURE CITED
cost-effective to protect those areas of an embank-
Construction Industry Research and Information
ment that are more prone to damage by ice shov-
Association (1991) Manual on the Use of Rock in Coast-
ing, according to the results of this study. Frequen-
al and Shoreline Engineering. Construction Industry
cy of damage by ice shoving is another factor that
Research and Information Association, CIRIA Spe-
needs to be considered in the design of a riprap
cial Publication 83, CUR Report 154. Rotterdam:
protection of a bank, because it may be cost-effec-
A.A. Balkema Publishers.
tive to repair riprap damage that occurs rarely
Collins, J.I. (1988) Large precast concrete armor
(Doyle 1988).
units in the Arctic. In Arctic Coastal Processes and
Slope Protection Design (A.T. Chen and C.B. Leiders-
dorf, Ed.). Technical Council on Cold Regions En-
SUMMARY AND CONCLUSIONS
gineering Monograph, American Society of Civil
A review of literature on ice effects on riprap
Engineers, New York, p. 208215.
revealed practically no guidance available for de-
Coastal Engineering Research Center (1984) Shore
sign of riprap in the cold regions, where the pres-
Protection Manual. U.S. Army Corps of Engineers
ence of moving ice can cause considerable dam-
Waterways Experiment Station, Vicksburg, Missis-
age to a riprapped bank. To generate data on riprap
sippi, Vol. I and II.
damage by ice shoving, we conducted a series of
Croasdale, K.R. (1980) Ice forces on fixed, rigid
small-scale tests in which we pushed a model
structures. In Working Group on Ice Forces on Struc-
riprap-covered embankment against 1.22-m-wide
tures (T. Carstens, Ed.). USA Cold Regions Research
model ice sheets grown in the test basin of our lab-
and Engineering Laboratory, Special Report 80-26,
oratory. During the 35 tests, we changed the slope
p. 34106.
of the model riprap bank, the size and the mix of
Croasdale, K.R., R.W . Marcellus (1978) Ice and
rocks, and the thickness of model ice sheets. Ex-
wave action on artificial islands in the Beaufort Sea.
cept during the first two tests, we moved the car-
Canadian Journal of Civil Engineering, 5(1): 98113.
riage at a constant speed of 40 cm/s. The interac-
Croasdale, K.R., N. Allyn and W . Roggensack (1988)
tion between the model riprap bank and the ice
Arctic slope protection: Considerations for ice. In
mostly resulted in ice pileup, except during two
Arctic Coastal Processes and Slope Protection Design
tests when ice rode up the bank. We measured the
(A.T. Chen and C.B. Leidersdorf, Ed.). Technical
horizontal and the vertical components of the
20
Previous Page
