published accounts of ice jams in the vicinity of confluences. The second set of
experiments focused on the confluence of the Missouri and Mississippi Rivers. It
entailed a relatively large hydraulic model built to facilitate detailed measure-
ment of flow field and ice movement velocities in the confluence before, during,
and after the implementation of bendway weirs.
A numerical simulation study reported by Lui and Shen (1998), and Lui et al.
(1998) augments the hydraulic model investigation reported here. The simulation
confirms the findings from the hydraulic model, and it demonstrates the
potential utility of numerical simulation for investigating ice accumulation pro-
cesses at channel confluences.
LITERATURE ON CONFLUENCE ICE JAMS
The literature on ice jams contains quite a few case study articles concerning ice
jam formation in river confluences. Andres (1996, 1997, 1998), for instance,
describes three situations in western Canada where jams result in recurrent flood-
ing problems for towns located near confluences. In most situations, the principal
mechanism leading to jam formation is straightforward: a stationary ice cover in
the mainstem channel blocks ice conveyed by the confluent channel. There are,
however, more complicated jam processes within a confluence.
From the literature, several factors clearly influence ice jam formation in conflu-
ences. The relative importance of each factor varies from confluence to conflu-
ence. A prominent factor is the confluence function of concentrating water flow
and ice from higher reaches of a watershed. Ice quantities may increase abruptly
at a confluence, possibly exceeding the capacity of its mainstem channel. A sta-
tionary ice cover in the mainstem channel may simply block ice on the confluent
channel. Additionally, a simultaneous ice run from confluent channels may have
difficulty merging in an ice run in the mainstem channel, to the extent that ice
jams in one or both of the channels.
Confluence morphology may trigger ice jams. It usually is relatively complex,
being characterized by marked variations in flow depth, flow velocity, slope, and
the presence of depositional alluvial features. Alluvial bars typically form at loca-
tions where the flow's capacity to move bed sediment locally diminishes within a
confluence. For instance, deltaic bars may form at the mouth of a channel conflu-
ent with a larger and more sluggish water body. Also, point bars may form within
flow separation zones that develop when confluent flows merge within the curved
planform of a confluence. Alluvial bars reduce surface area and depth of flow,
thereby potentially congesting ice movement through a confluence.
An interesting side issue concerns the possible influences that ice jams may exert
on confluence morphology. An ice jam constricts flow, possibly locally scouring an
alluvial bed beneath the jam, especially at the jam toe (Wuebben 1988). In most
situations, such scouring likely would be transitory and only alter bed morphol-
ogy locally. Yet, in some situations, jam-induced scouring may conceivably activate
substantial changes in confluence morphology. The overall subject of jam-induced
scour requires further investigation, and it is beyond the scope of this report.
This section presents a literature review on ice jams for which river confluences
played a significant role. To understand how the various factors lead to ice jam
formation in the vicinity of a confluence, the principal features of water flow
through confluences must be considered and how they modify confluence mor-
phology. Those factors are considered in the subsequent section.
3
Back to contents page
Previous Page
