WSP07063 - Laserfiche WebLink

Home Browse Search

Relation of inversely graded deposits to suspended-sediment grain-size evolution during the 1996 flood experiment in Grand Canyon David M. Rubin u.s. Geological Survey. M.S. 999. Menlo Park. California 94025 Jonathan M. Nelson David J. Topping U.S. Geological Survey, M.S. 413. Lakewood, Colorado 80225 ABSTRACT Before Glen Conyon Dam was completed npstream from Grand Conyon, noods scoured sand from the chonnel bed ond deposited sand on bars within recirculating eddies. ARer com. pletion of Glen Conyon Dam in 1963, peak discharge of the mean annual noods dropped from ahout 2600 to 900 m'ts, and 85% of the sediment supply was eliminated. Under the postdam now regime, sand bars in eddies have degraded. In an experiment to study, in part, the effects of noods in rehuilding these bars, a conlroUed nood was released from Glen Conyon Dam in late March and early April 1996. Although nuvial sequences characteristically fine upward, the deposits of the experimental nood systematically coarsen upward. Measurements of suspended. sediment concentration and grain size and of bed-material grain size suggest that the upward coarsening results from the channel becoming relatively depleted of line.grained sediment duro ing the seven days of the high.now experimenL Predam nood beds of the Colorado River also coarsen upward, indicating that supply-limitation and grain.size evolution are natural processes that do not require the presence of a dam. INTRODUCTION Background Water in the Colorado River passes through Glen Canyon Dam before flowing through lIle 400 km length of Grand Canyon. Unlil Glen Canyon Dam was completed in 1%3. lhe Colo- rado River in Grand Canyon had a mean peak annual discharge of 2600 m'ts. and carried a mean annual sediment load of 3 x 107 m3 (Andrews. 1990); discharge ranged from as linle as 28 m'ts (1000 c(s) 10 more Iban 3400 m'ts (120000 crs) over the cou,,", of a year. The dam drastically reduced these annual discharge fluctuations. but introduced daily fluctuations (in the extreme case having daily minima as low as 28 m3ts [1000 cfs] and max:ima as great as 850 m'/s [30000 crs]) 10 meel electrical power demands. VinuaJly all sediment coming down the river was cut off by the dam. and sediment was supplied to the posrdam river only from trib- utaries downstream from the dam (chiefly the Paria and Little Colorado Rivers); -15% and 10% of the predam total sediment and sand loads. respectively. were supplied by these sources. As a result of the new flow regime and greatly reduced sediment supply. sand bars in the canyon began eroding. necessitating research to deler- mine whether a new operaling scheme for the dam could mitigate this degradation. This paper presents selected resuhs from research conducted during a week-long experimental flood released from Glen Canyon Dam during March-April 1996 (Collier et al.. 1997). This experiment was designed. in part. to [est the hypothesis that new sand bars could be built in Grand Canyon- r,,,,,I,..n. Fl'nnl.lr1. ll)l).\( \" ~A: n.) ~: 11 l)q_ln~. -t n~url'<'. despite [he reduced postdam (10% of predaml supply of sand-by transporting sand from the channel bed 10 channel-margin bars. The hydro- graph of the experimental flood consisted of a rapid increase in discharge from 238 m3js (8400 cfs) 101290 m'ts (454OOcrs) over 5.75 hr, followed by seven days of constant discharge at 1290 m3/s. and then a slow decrease over 3.2 days to 238 m'ts. Supply Limitation Sand bars in the Colorado River in Grand Canyon fonn in recirculating eddies in lateral sepaJation zones (Schmidt and Graf. 1990; Rubin et aJ.. 1990). The fonnation and morphology of these eddy deposits are controlled by main. channel sediment supply and by eddy hydraulics and geometry. Because of the dependence on sediment supply. understanding and predicting the flow structure in lateral separation eddies is insufficient for predicting deposi!ion or erosion in eddy flows; main-channel sediment supply is atleas[ as important as flow patterns. For identi- cal flows. an eddy deposil can either aggrade or erode. depending on the concentration of sedi- men[ in the main-channel flow. In rivers where the main-channel sediment transport IS uniquely determined by flow discharge. {he dependence of eddy deposit morphology on main-ch<lnneltrans- pon can be parameterized in a straightforward manner; this is not the case in GrJ.nd Canyon. The Colorado River in Grand Canyon is cur- rently and historically a supply-limited system. wiLh respect to boLh sand and finer material; we define "supply-limited" to mean that the flow cransports less of cenain grain sizes than it would if more sediment of those sizes were available. Historically. sediment concentration in Grand Canyon decreased through time during spring snowmelt floods. reflecting this supply limitation. as explored in Ibe classic paper by Leopold and Maddock (1953) in which the sediment rating. curve hysteresis at this site was described. Supply limitation is particularly evident in Grand Canyon because sediment-supplying events are typically nOI synchronous with periods of high flows. This situation is characteristic of both predam and postdam conditions. although for somewhat different reasons. Prior to the construction of the dam. the majorilY of high flows occurred during annual snowmelt periods. and the majority of sediment was supplied to and stored in the channel dunng the monsoon season in late summer and early fall. Likewise. in the postdam period. sediment-rransponing flows do nol occur simultaneously with sediment- supplying events. Tnbutary sediment is still added locally during the monsoon season. but hIgh main-channel sediment transpon occurs when discharge is increased to meet power- generation needs or to adjust reservoir levels. Because of the mismatch in timing oftributaJ)' sediment supply and main-channel sediment rransport. the concenrrarion and grain size of sus- pended sediment and the associated bed material evolve during lhe year (finer grJin size and higher sand transpol1 for a given discharge immediately after sediment is introduced. and coarser grain size and lower transpon afler high-flow events have winnowed the bed). In the predam period 99