Active width of gravel-bed braided rivers.

Previous analyses have identified the active width of braided rivers, the bed area over which bed load flux and short-term morphological change occurs, as an important element of braiding dynamics and predictions of bed load flux. Here we compare theoretical predictions of active width in gravel-bed braided rivers with observations from Sunwapta River, and from a generic physical model of gravel braided rivers, to provide general observations of the variation in active width, and to develop an understanding of the causes of variation. Bed topography was surveyed daily along a 150 m reach of the pro-glacial Sunwapta River for a total of four weeks during summer when flow was above threshold for morphological activity. In the laboratory, detailed digital elevation models (DEMs) were derived from photogrammetric survey at regular intervals during a constant discharge run. From the field and flume observations there is considerable local and circumstantial variation in active width, but also a general trend in average active width with increasing discharge. There is also a clear relationship of active width with active braiding index (number of active branches in the braided channel network), and with dimensionless stream power, which appears to be consistent across the range of data from field and physical models. Thus there is a link between active width and the river morphology and dynamics, and the possibility of a general relationship for estimating active width from channel pattern properties or reach-scale stream power values, from which approximate bedload flux calculations may be made. The analysis also raises questions about differences between hydraulically-based numerical model computations of instantaneous active width and observation of time-integrated morphological active width. Understanding these differences can give insight into the nature of bedload transport in braided rivers and the relationship to morphological processes of braiding.