Channel stability in threshold channels
Although the term channel stability has been employed in various contexts, the overarching definition of a stable channel is one in which the pattern remains constant, although not necessarily fixed, over a period of time; channel location may shift over time but channel geometry remains relatively constant, a state which has been termed dynamic or quasi-equilibrium. Channel instability is thus associated with processes that significantly alter channel form or pattern, such as significant bank erosion or avulsions. Given that channel form will remain unchanged, or stable, so long as there is no sediment moving, channel stability is seemingly linked to particle mobility. As such, an alluvial channel in which there is no sediment transport occurring provides an end case for channel stability (i.e. completely stable channel). However, channels do not immediately become unstable once the first particles become mobile. The question then becomes: is there a threshold that can be used to predict when a channel will be destabilized and is this threshold the same or different than that of particle mobility?
Prolonged high precipitation in June 2013 resulted in catastrophic flooding in many parts of the southwestern Rocky Mountains. While high flows resulted in the inundation of low lying areas in the valleys, steep alluvial channels, such as those in the vicinity of Canmore, AB, experienced dramatic channel change, driven by high rates and volumes of sediment transport, bank erosion, and the formation of new channels. This rapid change affected communities and infrastructure in the area and resulted in billions of dollars worth of damage to the surrounding area (Figure 2). While this event provided a unique case study into how such systems respond to large flood events, it also highlighted our lack of understanding in regards to the dynamics that control channel stability in these environments.
Although stream tables provide a unique opportunity to study channel morphodynamics of fully alluvial channels, to date their use has been limited in comparison to that of straight-walled flumes. Stream tables are unique in that they are reach-scale models that encompass both the channel as well the immediate floodplain, which allows for the study of both lateral and vertical processes.
Experiments will be conducted on the Adjustible-Boundary Experimental System (A-BES) a 12.2 m long and 1.5 wide tilting stream table located in the BioGeoMorphic eXperimental Laboratory (BGMX Lab) at the University of British Columbia (Figure 3). A-BES is equipped with an advanced set of instruments and systems facilitate the capture of high spatial and temporal resolution data. Spatial data is collected using a motorized instrument cart complete with a laser scanning system that creates 2 mm x 2 mm resolution digital elevation models (DEMs) of the bed surface, and down-looking camera that captures 1 mm x 1 mm resolution images of the bed. Flow is introduced to the system and monitored using a variable speed pump with an in-line flow meter. Sediment is fed into the system using a rotating feeder and captured at the base of the stream table using a mesh-lined bucket trap.