The velocity reversal hypothesis and the implications to the sustainability of pool-riffle bed morphology.

摘要

Pool-riffle sequences are known to be critical habitat for several species of fish and benthic organisms. The morphological complexity of pool-riffle channels provides conditions for spawning, refugia and feeding.;Pool-riffle sequences normally occur in low gradient gravel-bed rivers, and it has been observed that these features are sometimes very resilient despite significant disturbance in the stream or major changes in sediment delivery from the watershed, but in other cases a relatively minor disturbance results in significant loss of pool habitat.;Mechanisms responsible for pool-riffle maintenance are unclear and despite contributions from many researchers, no universal explanation has been developed. The most popular hypothesis for pool-riffle sustainability is the occurrence of velocity reversal, i.e. at low flows the maximum velocities in the channel occur across the riffle, but at higher less frequent flows the area of maximum velocity migrates to the pool. Reversal of velocity may cause corresponding reversal of shear stress and transport capacity that scours sediment previously deposited in the pool, with the larger clasts being deposited on the downstream riffle due to relatively lower competence, thereby providing a mechanism for maintenance of the pool-riffle morphology.;The velocity reversal hypothesis for pool-riffle channels was first proposed by Keller (1971) based on observations made by Gilbert (1914) and raised considerable interest and debate among scientists in the intervening years. A diverse range of opinions about whether this process exists and, if it does, the conditions under which it can be expected to occur had been discussed in the literature (Chapter 1: Pool-Riffle Sustainability and Velocity Reversal).;In this study, conditions potentially responsible for velocity reversal are critically assessed using published data from field studies and supplemented by additional field data collected by the author. An analytical solution for the physical conditions required for velocity reversal is developed. Although this approach uses a one-dimensional approximation to analyze the flow field through pool-riffle sequences, the simple analytical equation correctly predicts whether velocity reversal occurs in all cases cited in the literature, and quantifies the physical characteristics of the channel morphology necessary for reversal to occur. Results show that reversal depends critically on the ratio of riffle-to-pool width, residual pool depth (difference between pool and riffle elevations) and on the depth of flow over the riffle ( Chapter 2: A Unifying Criterion for the Velocity Reversal Hypothesis in Gravel-Bed Rivers).;In addition, three-dimensional numerical modeling was performed to study the effects of different discharge and pool aggradation scenarios on channel hydraulics and the implications for velocity reversal. Two consecutive pool-riffle sequences at the Red River Wildlife Management Area in northern Idaho were chosen as a study site for this investigation. The model allows characterization of the flow structure, and identification of jet formation and dissipation zones, as well as the development of local turbulence features (i.e. vertical and horizontal eddies). The analysis demonstrated a significant influence of the residual pool depth on the flow structure. With pool aggradation and reduced residual depth causing a shift in the orientation of the jet and reduction in the influence of vertical eddies and the size and intensity of horizontal eddies. Based on these detailed observations and simulations of the flow structure through pools, a conceptual model is proposed to explain the sustainability of self-formed pool-riffle sequences in gravel-bed rivers due to jet formation and dissipation zones (Chapter 3: The Flow Structure in Pool-Riffle Sequences).;Further insight to the velocity reversal process and the associated bed shear stress and transport capacity variations were investigated at the site using the three-dimensional model results. Local depth-average, surface and near-bed velocities were evaluated and reversal assessed for all modeled scenarios. Results show that cross-section average velocities, near-bed velocities, shear stress and flux reversal do not all occur at the same discharge. Furthermore, the results corroborate the conclusion of the simple one-dimensional analysis (Chapter 2) and also show the importance of the location of the concentrated jet flow (Chapter 4: A Mechanism for Sustainability of Self-formed Pool-riffle Sequences).;Finally, recommendations for further work investigating the mechanisms of jet dissipation are suggested. The author believes that clarification on this matter will help to explain the spacing between pools and riffles ( Chapter 5: Recommendations for Further Work).