Spatial Dynamics of Gravel Bedload Transport Between a Pool-Exit Slope and the Head of a Point-Bar

Spatial dynamics of gravel transport in mountain streams are not well known. Results compiled from different streams suggest that the path of gravel transport in coarse-bedded streams with a meandering thalweg follows the path described for the coarsest bedload in sand-bedded meandering streams. From the thalweg in the pool exit slope, gravel moves laterally over the downstream riffle to reach the head of the next point bar. Gravel then proceeds across the bar towards the thalweg and into the pool. If proven correct, this finding has important implications for bedload sampling in partially wadable streams. Sampling could be limited to the bar head where equipment requiring wadable flows can be used, while still collecting the majority of gravel transported. To confirm the travel path between the pool exit and the head of the next bar, gravel transport was measured in a Colorado mountain stream in two cross-sections less than 10 m apart: one at a pool exit with the thalweg near the stream center and one at the bar head with the thalweg hugging the left bank. Six bedload traps were installed in each cross-section. The lateral location of maximum gravel transport differed greatly between the two cross-sections. In the pool exit, most of the transport occurred just to the right of the thalweg. In the bar cross-section, transport was absent in the left bank thalweg where flow was deepest and fastest but focused on the bankward side of the bar where flow was much shallower and slower, and with increasing flows the location of maximum gravel transport moved progressively further up the bar. Movement of tracer particles placed at several locations across the pool exit confirmed that most of the particles take a curved path and move onto the bar head and towards the bankward side of the bar. The bedload rating curve on the bar head was found to be better defined and steeper than the curve measured in the pool exit and the curves crossed at about half the measured maximum flow. This indicates that critical flow to initiate marginal gravel transport is higher on the bar than in the pool exit, while during the highest flows, the bar transports more gravel bedload than the pool exit. Differences in critical flow and tightness of the rating relationship are attributed to different sources of gravel bedload in the two cross-sections. The coarse-bottomed pool exit is not a local sediment source but receives most of its gravel supply from further upstream, causing a supply-dependent response of bedload to changes in flow. By contrast, the erodible bed on the bar provides a local bedload source allowing transport rates to respond promptly to changes in flow. In contrast to depth-averaged flow hydraulics, bed material particle sizes conform with the transport pattern in the bar cross-section. Around the thalweg, imbricated cobbles form an erosion resistant bed to locally deep and fast flows. Towards the bankward side of the bar, the bed becomes increasingly more erodible, with less armoring and an increasing percentage of pea gravel, facilitating mobility at relatively low local flows. Bed morphology and gravel transport work together to sustain the described transport pattern which in turn sustains the morphology. A helical flow between the pool exit and the point bar diverts the main body of (surface) water towards the thalweg, while the near-bottom flow moves upward and bankward along the bar. This flow pattern transports gravel over the bar head and along the bankward side of the bar. As the gravel is passing over the bank, it creates an erodible bed moveable by locally low flows.