|Powell, D. - UNIV. OF LEICESTER|
|Brazier, R. - UNIV. OF SHEFFIELD|
|Wainwright, J. - UNIV. OF SHEFFIELD|
|Parsons, A. - UNIV. OF LEICESTER|
Submitted to: Water Resources Research
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: March 22, 2006
Publication Date: August 9, 2006
Citation: Powell, D.M., Brazier, R., Wainwright, J., Parsons, A., Nichols, M.H. 2006. Spatial patterns of scour and fill in dryland sand bed streams. Water Resources Research Vol. 42, W08412,doi:10.1029/2005WR004516. Interpretive Summary: Many channels in the semiarid southwestern US are dry most of the time and only flow in response to thunderstorm rainfall during the summer months. When the channels do flow, the sand on the channel bed can be scoured, and the sediment that is mobilized can either be transported down channel or redeposited in the scoured areas. A field experiment was initiated to measure the amount and patterns of sour and fill during runoff. Small chains, called scour chains, were buried in the channel bed with the length of chain extending up through the sediment and onto the bed. After each flow, the length of exposed chain was measured to determine whether there was net scour or fill. Overall, the depth of scour was related to discharge, and the maximum scour depths traced a sinuous path about the channel centerline. During each flow, compensating fill returned the streambeds to preflow elevations, indicating that the streams were in approximate steady state over the period of study. The data and nterpretations resulting from this experiment are important for understanding sediment transport by flash floods in normally dry channels.
Technical Abstract: Spatial patterns of scour and fill in two dryland ephemeral stream channels with sandy bed material have been measured with dense arrays of scour chains. Although the depth and areal extent of bed activity increased with discharge, active bed reworking at particular locations within the reaches resulted in down stream patterns of alternate shallower and deeper areas of scour. The variation was such that mean scour depths for individual cross sections varied about the mean for the reach by a factor of 2-4 while the locus of maximum scour traced a sinuous path about the channel centerline. The wavelength of the pattern of scour was about seven times the channel width. During each event, compensating fill returned the streambeds to preflow elevations, indicating that the streams were in approximate steady state over the period of study. Although the patterns of periodically enhanced scour along alternate sides of the channels are consistent with models of periodically reversing helical flow, further work is required to identify the causal relationships between patterns of flow and sediment transport in dryland sand bed channels.