WSP05649 - Laserfiche WebLink

Home Browse Search

560 TOPPING ET AL: COLORADO RIVER SEDIMENT TRANSPORT. 2 during periods of little or no tributary activity but greatly un- derestimates sand transpon when the bed is enriched in fines. Therefore the magnitudes of sand accumulation in the Colo- rado River predicted by a stable rating curve calibrated to depleted bed conditions (as done hy U.S. Department of the Interior [1995]) are probably too high. In addition to the average increase in suspended-sand con- centration after the increase in tributary activity on about Sep- tember 2, shorter-term changes in suspended-sand concentra- tion coupled to changes in suspended-sand grain size are evident after some of the tributary floods. After each tribntary flood, peaks in the measured concentration of suspended sand occur at the Lower Marble Canyon gage (Figure 12b). The greatest measured increase in suspended-sand concentration was between September 15 and 16, when the concentration of suspended Band increased by a factor of 6.3. This occurred in response to a large increase in the discharge of the Paria River the day before (Figure 12a). Because the mouth of the Paria River is located 97 kin upstream from the Lower Marble Can- yon gage (Figure I), this suggests that, like in the 1983 exam- ple, some portion of the sand input during a tributary flood travels downstream relatively quickly (at nearly the mean ve- locity of the water). Substantial changes in suspended-sand grain size were also measured at the Lower Marble Canyon gage after two of the four tributary floods (Figure 120). Fol- lowing the rapid increase in suspended-sand concentration af- ter the arrival of water from an ungaged tributary flood on September 2, the median size of the suspended sand increased from about 0.10-0.11 to 0.15-0.17 mm as the concentrations started to decrease. Also, after the September 7 Paria River flood, the median size of the suspended sand at the Lower Marble Canyon gage de<:reased from about 0.15-0.17 to 0.10- 0.12 mm as the concentrations increased. 5,6. Methods: Collection and Processing or the 1998 Data To monitor seasonal changes in the grain size of the bed throughout Marble Canyon following large inputs of sand from the Paria River during the summer thunderstorm season (mid- July through early-October), bed samples were colle<:ted in March 1998, September 1998, and May 1999. Bed samples were collected with a pipe dredge in the center of the channel throughout the 99-kIn length of Marble Canyon (Figures 13 and 14). Lucltily, the Paria River cooperated with this sampling program, and on September 5, 1998, a flood with a peak dis- charge of approximately 200 m'fs, the largest flood in 18 years, occurred on the Paria River. During this flood and a second flood (with a peak discharge of about 110 m'fs) on September 12 a total of 1.2 :t 0.2 million t of sand and 1.5 :t 0.8 million t of silt and clay [after Topping, 1997] were transported from the Paria River into Colorado River. This second flood oc. curred 3 days prior to the day on which the September 1998 sampling trip was launched. 5,7, Results: 1998 The bed of the upper 80% of Marble Canyon fined both considerably and quickly as a result of the Paria River floods on September 5 and 12, 1998, as indicated by oomparison of the grain-size data colle<:ted in March and September 1998 (Figures 13, 14, and 15). This newly input sediment was ob- served on the bed as far downstream as river mile 50. This sediment was observed to be in the form of a sediment wave, with an upstream portion that blanketed the preexisting coarser hed and a downstream portion that occurred as a secondary mode mixed with the preexisting coarser bed. The upstream portion of the sediment wave blanketed the preexisting coarser bed sediment, such that from the mouth of the Paria River to river mile 6.8, the bed was unimodal and very fine. 11tis portion of the sediment wave fined in the down- stream direction. From the mouth of the Paria River to river mile 6.8 the median size of the fine sediment on the bed decreased from 0.30 to 0.11 nun (Figure 13a), the fraction of the fine sediment on the bed composed of 0.0625- to 0.1 25-mm sand increased from 0.97 to 88% (Figure 13b), and the fraction of the fine sedimenl on the bed composed of sill and clay increased from 0.094 to 2.2% (Figure 13c). As a result of the 1998 Paria River floods, the amount of fine sediment on the bed composed of the finer sand sizes increased by about a factor of 1000-1500, and the amount of fine sediment on the bed composed of silt and clay increased by a factor of 71 at river mile 6.8 (Figures 15a). At both river miles 5.7 and 6.8 the median size of the fine sediment measured on the bed of the Colorado River was 0.11 mm (Figure 13a and 14), roughly equal to the 0.11- to 0.15-mm median size of the suspended sand measured in the Paria River during floods [Topping, 1997]. The front of this upstream portion of the sediment wave occurred between river miles 6.8 and 7.2, Presumably, the coarser upstream tail of this portion of the wave (from the mooth of the Poria River to river mile 6.8) resulted from winnowing of the finer grain sizes from the bed after cessation of the upstream supply of sediment from the Paria River. The downstream ponion of the sediment wave (from about river miles 7.2 through 50.1) occurred as a secondary finer mode mixed with the preexisting coarser bed sediment (Figure 14). 11uoughoutthis portion of Marble Canyon the primary mode was composed of the coarser grain-size distribution mea- sured in March 1998. The secondary mode was composed of the newly input finer sediment. Though no suhstantial de<:rease in the median size of the fine sediment was observed from river mile 7.2 through 50.1 (Figure 13a), fining of the bed due to the growth of this secondary mode resulted in an increase in the silt and clay and an increase in the 0.0625- to O.I25-mm sand relative to the amounts composing the fine sediment on the bed in March 1998 (Figure 15a). The secondary mode de<:reased in magnitude from about river miles 7.2 through 50.1 and decreased in grain size from about 0.1 nun at river mile 7.2 to about 0.07 mm at river mile 50.1 (Figure 14). Also, immediately downstream from the front of the upstream unimodal portion of the sediment wave (Le., from river miles 6.8 through 7.2), enrichment of the fine sed- iment on the bed decreased faster with increasing grain size (Figure 15a). Between river miles 6.8 and 7.2 the September 1998 to March 1998 ratio of the amount of silt and clay com- posing the fine sediment on the bed decreased by only a factor of 4.4. Over this same section of the river the comparable rati?s for the 0.0625- to O.077-mm sand, 0.077- to O.088-mm sand, 0.088- to 0.105-mm sand, and 0.105- to O.l25-mm sand de- creased by factors of 6.8,6.9, 13.9, and 69, respectively (Figure 15a). These observations suggest that (as previously suggested by the suspended-sand data in Figure 12d) following a tribu- tary flood, finer grain sizes travel downstream faster than the coarser sizes. Between September 1998 and May 1999 the grain size of the fine sediment on the bed in Marble Canyon evolved substan- tially, and the bimodality observed in September 1998 disap- peared. Depletion of the finer sand on the bed during this