|
|
|
|
Braid belt dynamics
The development and evolution of channel and bar forms in the active braid belt are currently being analysed from the video camera images. Preliminary findings show several interesting features.
Typically, we observe rapidly migrating gravel sheets during floods and freshes. During recessions, we see the larger gravel sheets 'freezing', often with very rapid development of 'recession topography' as the falling water surface is draped over the bar topography and incision of the bar front results. However, smaller gravel sheets within the braids continue to migrate downstream for several days after a fresh. At a slower pace during recessions and normal flows, we see the more classical channel change expected with braiding (bank erosion, channel migration, bar growth at diffluences). These features are illustrated in the following two time-lapse video clips. Both are from the downstream-looking video camera.
Clip 1 covers the period 15 June to 25 July 1999. This illustrates the classic braiding process, with bank erosion feeding bar growth at a diffluence and the whole channel pattern becoming unstable and avulsing. Note the rapid erosion of the left bank in the foreground and the concurrent build-up of a bar at the diffluence with a side channel in the middle distance. This deposition closes off the side channel, and the whole process increases the sinuosity of the foreground braid. During the next fresh (on 5 July 1999), this braid avulses to the left. Note also the often rapidly advancing gravel sheets along the right side of the foreground braid after 5 July. These continue to migrate well into the flow recessions. (If nothing or a broken link appears above, please click here (.avi file, 17.6mb size)
Clip 2 covers the period 1 April to 1 May 2000. This also illustrates gravel sheets advancing downstream during freshes and with recession flows. Note the rapidly dissecting front of a 'frozen' gravel sheet after the flood on 22 April, creating the row of chutes and lobes that we term recession topography. (If nothing or a broken link appears above, please click here (.avi file, 13.5mb size)
An important observation from the video imagery is that gravel moves even at normal flows, significantly less than 80 m3/s. While this transport of gravel is intermittent in space and time, it still contributes to the long-term average flux of bedload. Some previous assessments of bedload transport in the Waimakariri have assumed no transport at discharges under 600-800 m3/s, based on Shields-type analyses of the critical tractive shear stress and section-mean hydraulic parameters (Carson and Griffiths, 1987). The videos and field inspection clearly show that the bed material is mobilised by gravity-assisted river scour of braid banks at much lower discharges (as also noted by Carson and Griffiths, 1989).
The upshot is that, particularly in the braid belt where the normal flow tends to concentrate, the river bed morphology sculpted during flood flows is rapidly altered by braiding processes and the development of recession topography. Thus even if the braid belt remains in much the same broad location, the braid-scale topography is very transient. In contrast, the fresh- and flood-developed morphologies on the broad, low relief alternate bars have a greater tendency to be preserved due to the rapid emergence of the bars during recessions and their likely starvation of gravel from upstream during all but the largest floods.