9500 - Laserfiche WebLink

Home Browse Search

. 2002 R+iYeK 4- cress WATER RESOURCES RESEARCH, VOL. 38, NO. 10, 1216, doi:10.1029/2001 WR000898, 2002 Downstream changes in the channel geometry of a large gravel bed river John Pitlick and Robert Cress Department of Geography, University of Colorado, Boulder, Colorado, USA Received 20 August 2001; revised 14 May 2002, accepted 24 May 2002, published 30 October 2002. [r] Field data obtained on a nearly contiguous segment of the Colorado River in western Colorado and eastern Utah are used to examine the mechanisms driving downstream changes in channel geometry. Measurements characterizing the bank-full hydraulic geometry, bed material grain size, and average channel gradient were made at closely spaced intervals in 10 alluvial and quasi-alluvial reaches covering 260 km of the river. These data indicate that changes in surface and subsurface grain sizes are small in relation to the change in channel slope: over the full length of the study area, the median grain size of the surface sediment decreases by a factor of a little more than 2, whereas the average channel slope decreases by a factor of about 5. The decreases in slope and median grain size are offset by a large increase in bank-full depth relative to width, such that the bank-full Shields stress, Tb, is constant downstream. For the reach as a whole, Tb averages 0.049, which is roughly 50% higher than the threshold for bed load transport. INDEX TERMS: 1815 Hydrology: Erosion and sedimentation; 1824 Hydrology: Geomorphology (1625); KEYWORDS: channel geometry, downstream fining, Shields stress Citation: Pitlick, J., and R. Cress, Downstream changes in the channel geometry of a large gravel bed river, Water Resour. Res.. 38(10), 1216, doi:10.1029/2001WR000898, 2002. 1. Introduction [2] The problem of predicting downstream trends in channel geometry (width, depth, and slope) remains central to the understanding and modeling of river behavior. Changes in channel geometry are determined both by changes in water discharge [Leopold and Maddock, 1953], and by changes in sediment load and bank properties [Schumm, 19601. The problem therefore is to predict the equilibrium width, depth and slope of a river given repre- sentative values of discharge, sediment load and grain size. Approaches to this problem have been largely empirical, with a heavy emphasis on discharge as the primary control- ling factor [Ferguson, 1986; Knighton, 1987; Church, 19921. Discharge clearly governs sediment transport capacity; how- ever, differences in sediment supply, as well as processes of downstream fining and bank erosion can force longitudinal changes in channel properties without much change in discharge [Schumm, 1960; Brierly and Hickin, 1985; Fergu- son and Ashworth, 1991]. Thus, in addition to the equations for continuity, flow resistance, and sediment transport capacity, the solution to the problem of channel geometry requires a criterion for bank erosion to model the change in width [Henderson, 1966; Parker, 1979; Ferguson, 1986; Chang, 19881, or a "rule" that governs the downstream change in grain size to model the change in sediment load [Pizzuto, 1992]. [3] This paper focuses on channel adjustments in a gravel bed river where the downstream change in discharge is small in relation to the change in sediment supply. The study objectives and approach draw on theory developed by Copyright 2002 by the American Geophysical Union. 0043-1397/02/2001 WR000898S09.00 Parker [1978, 1979] to explain the equilibrium morphology of rivers with mobile beds and stable banks. Parker [1978, 1979] reasoned that gravel bed rivers will adjust their bank- full width and depth to provide a boundary shear stress, T, that is higher than the threshold for bed load transport, T,., but not so high as to cause bank erosion and widening. The geometry of a gravel bed river should thus bear a consistent relation to the excess shear stress, T/-r,, and the long-term bed load flux, which scales with T/-r,. In the downstream direction, changes in width, depth, slope and/or grain size should nudge the channel in a direction that maintains equilibrium transport. This approach is fundamentally dif- ferent from that taken by Leopold and Maddock [1953] and many others, because flow and sediment transport are coupled directly to the processes of erosion and deposition. The theory has been refined and tested in a number of laboratory experiments [Ikeda et al., 1988; Diplas, 1990; Cao and Knight, 1998; Macky, 1999], and in a few field studies [Andrews, 1984; Pitlick and Van Steeter, 1998]. [4] The data presented in this paper provide a relatively rigorous test of the hypothesis that gravel bed rivers adjust their bank-full channel geometry to carry bed load at shear stresses not far above critical. We test this hypothesis with closely spaced measurements of width, depth, slope, and bed-material grain size along a 260-km reach of the Colo- rado River in and regions of western Colorado and eastern Utah. The study segment is characterized by repeated transitions in valley fonm related to hard and soft sedimen- tary rocks, and it includes two major tributaries: the Gunni- son River and the Dolores River. In contrast to basins in humid and temperate regions, these tributaries and the surrounding area provide proportionally much more sedi- ment than water. Most of this sediment is fine grained (silt and sand); however, the supply of coarse material to the 34 -1 q50©