RIVER ICE INFLUENCES ON FORT PECK REACH, MISSOURI RIVER
27
Bankfast ice weakening of riverbanks likely is significant for steep river-
banks, especially those riverbanks containing enough clay to be termed cohesive.
It also likely is significant for riverbanks where the water table declines in eleva-
tion away from the flow elevation in a channel, because the bankfast ice is less
securely anchored into the riverbank. This erosion mechanism seems not to have
been investigated but had been observed along the Fort Peck reach (e.g., Simon
et al. 1999).
Gouging and abrasion of riverbanks
During heavy ice runs resulting from ice cover breakup or ice jam release,
large pieces of ice may gouge and abrade channel riverbanks, substantially
affecting the channel riverbank morphology (Marusenko 1956, Smith 1979,
Hamelin 1979, Uunila 1997, Doyle 1988, Wuebben 1995). The affected channels
are usually relatively steep and convey high-velocity flows. Their ice covers
typically break up dramatically in concert with a sudden rise in flow, as might be
caused by rapid snowmelt and/or rain. The resulting ice rubble comprises hard,
angular blocks of ice.
One study of 24 rivers in Alberta (Smith 1979) led to the intriguing hypothe-
sis that ice runs enlarge channel cross sections at riverbank-full stage by as much
as 2.63 times those of rivers of comparable flow but not subject to ice runs. The
hypothesis is based on a comparison of the recurrence interval of riverbank-full
flows in the 24 rivers and an empirical relationship between the cross-section
area and flow rate for riverbank-full flow. The concept is plausible, but the extent
of widening seems overly large and requires confirmation. Kellerhals and Church
(1980) argued against Smith's hypothesis, suggesting that other factors led to the
apparent widening of the channels analyzed by Smith, including recent
entrenchment of major rivers in Alberta and ice-jam effects of flow levels. More-
over, it is possible that the riverbanks are somewhat protected by a band of ice
that forms a shear wall flanking the riverbanks. It is interesting to contrast
Smith's hypothesis with the idea that ice jams may promote channel narrowing
by causing overbank flow (e.g., Uunila 1997). For channels with the dominant
channel-forming flow coinciding with ice cover breakup, the overbank loss of
flow reduces the flow rate that has to be accommodated by the channel.
In many situations, notably those in which an ice run is sluggish, a shear wall
of broken ice may prevent moving ice from contacting the riverbank. The shear
wall usually is smooth-faced and helps the river pass the ice. Running ice, if
sufficiently thick, may still gouge the lower portion of a riverbank.
Ice gouging and abrasion, though, can be severe for channel features and riv-
erbank stabilization structures protruding into the flow. In addition, channel
Previous Page
