ence jam releases. The jam sites are about 40 to 50 miles (6580 km) upstream of
Lake Sakakawea, a reservoir on the Missouri River. The extent to which Lake
Sakakawea contributes to the incidence of ice jams is unclear. Its possible effects
(channel aggradation, raised groundwater, and altered ice regime) on jam inci-
dence have yet to be fully investigated (Wuebben and Gagnon 1995).
The literature contains several similar accounts of ice jams at the confluence of
two rivers. Andres (1996, 1997, 1998) documents three cases of jams involving ice
runs in a confluent channel blocked by a stationary ice cover breakup on the main-
stem channel. One case occurs in Alberta at the confluence of the Smoky River and
the Peace River, which is larger and has a more northern location (Andres 1996).
The ice cover on the Smoky River breaks up first during spring, but is blocked by
the intact cover on the Peace River. If breakup flows are sufficiently large in the
Smoky River, ice jammed at the confluence may thrust through ice on the Peace
River and produce a subsequent jam a short distance downstream in the Peace
River. Ice on the McLeod River typically breaks up before ice on the Athabasca
River, and it jams in the confluence of the two rivers in Alberta (Andres 1998). In
some years, the jam develops in the Athabasca River at a short distance down-
stream of the confluence. A freezeup ice jam commonly occurs at the confluence of
the Nechako and Fraser Rivers in British Columbia (Andres 1997). The Fraser
River, the larger river, and of flatter slope, typically freezes over first.
Jasek (1997) describes an unusual case, in the Yukon Territory, of an ice jam formed
at the confluence of the Porcupine River and its tributary the Bluefish River, in the
Yukon Territory. The case is unusual because the jam was attributable to the for-
mation of aufeis at the confluence. Aufeis forming in the Bluefish River encroached
over and enveloped the ice cover formed over the Porcupine River. The aufeis
thickened substantially and virtually dammed the Porcupine River.
Prowse (1986) presents the findings of an extensive investigation of ice jams
formed in the Liard River at the confluence of the Liard and Mackenzie Rivers in
the Northwest Territories, Canada. His study, which was conducted over a six-
year period, 1978 through 1984, indicates that two factors led to jam formation in
the Liard River at its confluence with the Mackenzie River. One factor was the
presence of an ice cover in the Mackenzie River. When the Mackenzie is ice-cov-
ered at the confluence, ice cannot pass out of the Liard River. The other factor is
the confluence morphology of the Liard and Mackenzie Rivers. The mouth of the
Liard River opens relatively widely at the confluence and is marked by the pres-
ence of sand bars and islands, the latter having formed from the more permanent
bars. Large ice pieces have difficulty moving through the mouth of the Liard River
without arching or grounding. Arching of ice pieces at the mouth of the Liard may
cause ice to jam in the Liard when open water conditions exist in the Mackenzie
River. Figure 3 summarizes the ice jam conditions in the Liard River at the conflu-
ence for the years studied.
The literature contains accounts of jams formed in channels of complex mor-
phology. White and Kay (1994), for example, describe the recurrent formation of
ice jams in the meandering braided channel of the Loup River in Nebraska. Dur-
ing low flow conditions, such rivers comprise several subchannels that cause wa-
ter flow to diverge and converge within the main channel. Chen (1986) describes
similar difficulties for ice movement in the braided meandering channel of the
lower Yellow River, above Lankao, China. The reach is relatively shallow and broad,
containing numerous sand bars and stream forks. Drifting ice is liable to ground
along the frontal edge of the bars and become clogged in the subchannel forks.
5
Back to contents page
Previous Page
