9568 - Laserfiche WebLink

Home Browse Search

numbers of 100,000 or greater, while Wiberg and Smith found such conditions to be critical at particle diameters greater than 5 em. To provide insight on flow conditions as a function of particle Reynolds number, F'tglIl"es 2.2 a..c were prepared. Implicit in this calculation are the assumptions that flow is uniform and that the channels are hydraulically wide (for more detailed discussion on assumptions see Appendix A.) These assumptions are not critical, as the primary purpose of the figures are to illustrate that in steeper gravel bed rivers, it is not uncommon to have particle Reynolds numbers greater than 100,000 for the larger sizes of the armor bed. Also note that it is typical that particle Reynolds numbers are in excess of the 200-500 range, establishing fully developed turbulent flow conditions (i.e., the laminar and transition range where critical dimensionless shear stress is not constant, as in the original Shields curve, are not of great practical significance). The implication of the observation that particle Reynolds numbers can exceed 100,000 in a gravel bed river is that beyond a certain set of flow conditions movement of the larger sizes of the armor layer may require a dramatic increase in discharge. These particles tend to protrude or rest on the surface of the bed and under less severe flow conditions may be relatively easily moved, but under more severe flow conditions may actually become much more difficult to move. In other words, a lower discharge may result in disruption of the armor layer, while higher discharges do not. 25 FlII~hiru~ Flow literature Review Flushing flows are often defmed as those flows required to remove finer material in the pores of a gravel bed river. Therefore, flushing flows may be considered a specific application of incipient motion, but literature that directly considers flushing flows is limited. As suggested by Milhous and Bradley (1986), none of the five methods they reviewed, and considered to be state of the art for quantifying flushing flow requirements, are considered satisfactory. The approach advocated by Milhous and Bradley is a basic application of incipient motion concepts. Their specific contribution is to suggest that a dimensionless shear stress (or Beta movement parameter as they prefer to call it) of 0.035 is adequate for depth flushing of fine material, although this value is admittedly based on limited research. Other available literature indicates that flushing of fme sediments from a coarse bed, without disruption of the coarser material, is limited to a depth equal to about the median coarse particle diameter (Berry 1985). Assuming that this flushing occurs from the action of turbulence near the bed, it is reasonable to assume that greater flushing might occur when larger particles (causing greater turbulent intensity) are present in the surface layer; however, there is no research directly supporting this theory. Based on Einstein's experiments, deposition of silt-sized particles (wash load) is independent of velocity. Thus, one would expect that a gravel bed channel with significant wash load will contain fine material at depth in the gravel bed regardless of flow velocity maintained. The limited research by Everts (1973) and that by Fisher, Sill, and Clark (1983) were the only studies found that provided predictive equations for flushing flow conditions. However, both studies considered limited ranges of particle sizes, neither of which approximates coarse gravel-cobble type systems. Therefore, more research would be required to extend these results to other conditions. 2.6 Conclusions The literature review of regional areas of interest indicates that extensive research and study of the Yampa, Little Snake and Green River systems has been completed in the last 10 years. Much of this information will contribute to this study, both to verify results and conclusions and to supplement the analysis completed. 2-4 I I I I I I I I I I I I I I I I I I I