WSP05649 - Laserfiche WebLink

Home Browse Search

522 TOPPING ET AL: COLORADO RIVER SEDlMEl'IT TRANSPORT, I Canyon and Lees Ferry gages suggests that though the river in Grand Canyon may have been annually supply-limited with respect to finer (0.0625-0.25 mm) sand, it was not supply- limited with respect to this size class of sand in aU seasons. Because the concentrations of both finer (0.0625-0.25 mm) and coarser (>0.25 mm) sand were lower at low nows in Grand Canyon than they were in Glen Canyon, sand did accumulate in Marble and Grand Canyons when flows were sufficiently low, However, the fact that hysteresis in sand concentration and grain size occurred even in years with small snowmelt floods, like 1954 (Figure 2), suggests that the seasonal sand- storage potential was relatively small in Marble Canyon and upper Grand Canyon. As shown by (1) the higher concentrations of suspended co"""r (>0.25 mm) sand during the entire snowmelt flood and (2) the higher concentrations of finer (0.0625-0.25 mm) sand during the initial portion of the snowmelt flood (when flows first exceeded 400-500 m'/s), the river in Grand Canyon had the capacity to transport more sand than was supplied to it from Glen Canyon. Therefore, as required by conservation of mass, the sand stored in Marble Canyon and upper Grand Canyon would have been depleted at these higher flows until the concentration of suspended sand at the Grand Canyon gage decreased to a value equal to that upstream at the Lees Ferry gage. By virtue of their lower settling velocities the finer .izes of sand would have had higher transport rates. Thus the finer sizes of sand stored in Marble Canyon and upper Grand Canyon would bave been depleted more quickly than the co"""r sizes during these higher flows. This effect led to the observed decrease in the concenlration of finer (0.0625-0.25 mm) sand at the Grand Canyon gage to match the lower upstream supply of this size class of sand at the Lees Ferry gage during the latter portion of the snowmelt flood. Furthenoore, this effect led to the lack of any substantial decrease in the concentration of suspended coarser (>0.25 mm) sand at the Grand Canyon gage, even though in flows in excess of about 400-500 m'/s the upstream supply of this size class of sand was less at the Lees Ferry gage. 5.3. Dilference. in Coupled Changes in Suspeaded-Sand Concentration, Suspended-Sand Grain Size, Bed Grain Size, and Bed Elevation Calendar year 1956 was the only predam year in which suspended.sand concentration, suspended-sand grain size, and bed-sediment grain size were measured at both the Grand Canyon and Lees Ferry gages. Therefore comparison of data collected at these two gages during this year serves to further illustrate some of the predam sediment-related differences be- tween Grand and Glen Canyons. During the 1956 snowmelt flood at the Grand Canyon gage, hysteresis existed in suspend- ed-sand concentration and grain size, with the concentration of suspended sand decreasing and the suspended sand coarsening during the latter part of the rising limb of the nood (Figure 5a). In contrast, no significant hysteresis in either suspended-sand concentration or grain size existed at the Lees Ferry gage, and the suspended-sand grain size increased with concentration such that the maximum sand concentration and grain size occurred together at the peak of the nood (Figure 5b). During the initial part of the rising limb of the 1956 snowmelt nood, the bed at the Grand Canyon cableway aggraded, while the bed at the upper Lees Ferry cableway scoured (Figures 5a and 5b). Maximum bed elevation at the Grand Canyon cableway led the nood peak by about 4 weeks and occLlrred at about the same time as the maximum suspended-sand concentration. In con- trast, at the upper Lees Ferry cableway, minimum bed eleva- tion occurred at about the same rime as the flood peak and the maximum suspended-sand concentration. During rhe rising limb of the 1956 snowmelt flood the fine sediment on the bed at the Grand Canyon cableway coarsened, whereas the fine sediment on the bed at the upper Lees Ferry cableway fined slightly (Figures 5a and 5b). At the Grand Canyon cableway, coarsening of the bed was associated with coarsening of the suspended sand (Figure 5a). This style of coupled bed and suspended-sand coarsening was similar to that observed by Rubin et al. [1998], Topping et at. [1999], and Toppinger al. [this issue J in the postdam river. To investigate the degree to which the observed systematic changes in bed elevation in 1956 reflected general changes in the upstream sediment supply during an average predam year in Grand and Glen Canyons, a new methodology was devel- oped and applied to the predam periods of suspended-sand record at the Grand Canyon and Lees Ferry gages. The first goal of this analysis was to identify the degree to which a lag existed between the annual flood peak and either maximum or minimum bed elevation during an average year. The second goal of this analysis was to detenoine the degree to which sand-supply depletion was coupled to bed scour during floods. At both gages the mean relationship hetween bed elevation and the flow during the average predam year was determined for the periods of suspended-sand record. This was done by binning and averaging the measurements of water-surface stage, mean bed stage, and minimum bed stage [from Burkham, 1986] (Figures 6a and 6b). A bin size of 2 weeks was found to be optimum to ensure that enough data were included in each bin. Both mean and minimum bed stage were included in this analysis to determine if a change in bed elevation across tbe cross section was real and not an artifact of cross-section geometry. First, mean bed stage provides infonoation on the change in bed elevation over the entire width of a cross section, whereas minimum bed stage provides information only on the magnitude of the deepest scour in the cross section. Second, a slight apparent increase in mean bed stage will occur with an increase in water-surface stage simply because of the trapezoi- dal cross-sectional shape of a river channel. To quantify the seasonal style and importance of changes in the upstream sand supply, we developed a method based on Figure 5. (Opposite) Water-surface stage, bed stage, grain size (in suspension and on the bed), and suspended-sand con- centration at the Grand Canyon and upper Lees Ferry cable- ways during calendar year 1956. Greater hysteresis was present in all of these quantities at the Grand Canyon cableway, sug- gesting a greater depletion of the upstream supply of sand in Grand Canyon than in Glen Canyon. (a) Water-surface stage, mean bed stage, median size of the suspended sand, median size of the fine sediment (i.e., sand and finer material) on the bed, and suspended-sand concentration at the Grand Canyon cableway. Stage is that measured at the lower gage (measured relative to the same datum as at the upper gage). (b) Water- surface stage, mean bed stage, median size of the suspended sand, median size of the fine sediment on the bed, and sus- pended-sand concentration at the upper Lees Ferry cableway. Cableway stage is measured relative to a different datum than stage at the recording gage. Data in Figures 5a and 5b are from unpublished U.S. Geological Survey (USGS) discharge- measurement field notes.