WSP05649 - Laserfiche WebLink

Home Browse Search

TOPPING ET AL: COLORADO RfVER SEDIMENT TRANSPORT, 2 1.5 1.4 " 1.3 ~ 2 u 1.2 .~ ;; E 1.1 ~ ~ I 1.0 c .. 0.9 0 .... 0 0.8 z .. <II 0.7 0 w .... 0.6 en :> ..., 0 0.5 .. w > 0.4 ;:: :s :> 0.3 ::; :> 0 0.2 0.1 b) EARUEST MASS BALANCE AT GRANO CANYON GAGE OCTOBER 557 DECEMBER Figure 11. (continued) As observed at the Grand Canyon gage, the amount of 0.0625- to O.I25-mm sand on the bed at the National Canyon and Above Diamond Creek gages also increased from trace amounts to about 5% of the line sediment on the bed during the Uttle Colorado River flood. However, unlike at the Grand Canyon gage, this increase in finer sand occurred without any substantial decrease in the median size of the fine sediment on the bed, Also, unlike at the Grand Canyon gage, following this initial increase in the amount of finer sand, no measurable change in the grain size of the fine sediment on the bed oc- curred at these two downstream sites. 5.3. A Sand Budget for the 1983 Little Colorado River F100d To evaluate the error associated with the assumption of a time-invariant grain-size distribution on the bed, a sand budget (with uncertainties) was constructed using the measured sand loads during and following the 1983 Uttle Colorado River flood. Cumulative sand loads, computed using the data in Fig- ures lOa and lOb and the predictions of Randle and Pemberton (1987], for the period from the beginning of the Lillie Colo- rado River flood to the ends of the sampling periods are shown in Figure lla. For the reasons discussed by Topping et al. [this issue], 5% uncertainties were assigned to the measured Colo- rado River sand loads, and 20% uncertainties were assigned to the measured Little Colorado River sand loads. As shown in Figure l1a, because their sand-transport algorithm was derived for a coarser bed grain-size distribution than existed at the Grand Canyon gage during the Little Colorado River flood, Randle and Pemberton [1987] underestimate the sand loads at the Grand Canyon gage (and there.fore overestimate the up- stream accumulation of sand) by about 30% (05 million t) during this 3-month period. Interestingly, the magnitude of this error is comparable to the predicted long-tenn accumulation rate between the Uttle Colorado River and the Grand Canyon gage given by U.S. Department of the Inlenor [1995]. The sand budget calculated using the information in Figure lla suggests that the residence time in Grand Canyon of the sand supplied by the 1983 Ullle Colorado River flood was quite short (Figure lib). The Colorado River downstream from the Ullle Colorado River receives sand from both the Little Colorado River and the Colorado River in Marble Can- yon. Therefore, to oonstruct a budget for only the sand sup- plied during the Little Colorado River flood, the cumulative measured and predicted loads at the sites downstream from the mouth of the Little Colorado River were adjusted by sub- tracting the cumulative supply of sand measured passing the Lower Marble Canyon gage. The uncertainties in the loads were propagated through this step to result in the error enve- lopes in Figure lib. In Figure lib the earliest mass balance (given the uncertainties) occurs at a given site when the upper bound of the error envelope intersects the lower bound of the error envelope associated with the sand supplied during the Little Colorado River flood (the cross-hatched region). Uke- wise, the latest mass balance occurs at a given site when the lower bound of the error envelope \ntersects the upper bound of the error envelope associated with the sand supplied during the Little Colorado River flood. Therefore the earliest an amount of sand equivalent to that supplied during the little Colorado River flood could have passed (1) the Grand Canyon gage was on about October 10 (i.e., only to days after the flood peak passed the LCR near Cameron gage), (2) the National Canyon gage was on about October 30 (i.e., only 30 days after the flood peak passed the LCR near Cameron gage), and, by extrapolation, (3) the Above Diamond Creek gage was in