9500

• UCREFRP Catalog Number: 9500
• Author: Pitlick, J. and R. Cress
• Title: Downstream Changes in the Channel Geometry of a Large Gravel Bed River
• USFW Year: 2002
• USFW - Doc Type: American Geophysical Union
• Copyright Material: YES

34 - to PITLICK AND CRESS: DOWNSTREAM CHANGES IN CHANNEL GEOMETRY
Figure 10. Downstream trends in the bank-full dimensionless shear stress of the Colorado River.
Symbols are the same as in Figure 7.
higher. Both data sets support the hypothesis that Tb is
independent of Q*, and the average values of Tb are
identical (0.049). This last result has interesting implications
for bed load transport. Typical transport equations express
the dimensionless unit bed load transport rate, qb, as a
function of the excess shear stress, T* - T,, raised to a
power >1. The results above imply that if Tb is constant,
then qb is also constant (assuming that T*, does not also
vary). By definition, the volumetric bed load transport rate
per unit width is qb = qb RgD3/2, and the total transport
rate is Qb = Bqb. Combining these relations, we get Qb =
Bqb RgD3/c, or more generally, Qb a BD312. Thus, if q*b 1
indeed constant, then the downstream change in total bed
load transport depends on the change in B relative to the
change in D3 2. Our data show that the grain size decreases
more rapidly than the width increases (Table 3), suggesting
that the total volume of bed load carried by the Colorado
River is decreasing downstream. This interpretation is
supported to a limited extent by the comparison of surface
and subsurface sediment (Figure 6), which shows a pro-
gressive coarsening of the surface layer downstream. The
surface layer regulates the transport of different sizes, and
for particular combinations of discharge, slope and sediment
supply, the surface texture may become finer or coarser
downstream [Parker, 1990]. The development of a static
armor downstream of a dam represents the extreme case of
supply limitation, while the absence of a surface layer may
indicate high sediment supply [Lisle and Madej, 1992]. The
Colorado River lies somewhere in the middle of this
spectrum, with hillslopes and ephemeral tributaries main-
taining the supply of coarse material, but apparently at rates
that diminish downstream.
6. Conclusions
type. In contrast to the trends observed in many other rivers,
the bank-full width of the Colorado River increases slowly
downstream in comparison to the bank-full depth. The
differences in bank-full width and depth appear to be driven
by competing effects of slope and grain size. In comparison
to other rivers, the slope of the Colorado River decreases
more rapidly downstream, whereas the grain size decreases
more slowly. Continuity requires that the river form a deeper
104 103
r ?.. ?
B.
•
103 + * 10
M • ` B- = 39.3()'`0.32 H*
H- = 0.070_1-0.53
102 101
S S = 0.780"-0.50 100
v
10-3 cr:,: c
10'1
•
•
[3o] The segment of the Colorado River examined in this
study exhibits morphologic characteristics that are, in some
aspects, very different from other gravel bed rivers, but in
other aspects quite similar. We find, for example, that
channel properties (bank-full width, depth, slope and bed
material size) change systematically downstream, and that
transitions in reach type caused by local changes in bedrock
geology often have little effect on overall trends. Partitioning
of cross section data from alluvial and quasi-alluvial reaches
shows that downstrcam relations for bank-full width and
bank-full depth cannot be distinguished on the basis of reach
10-41 0.049
1° ( 10-2
105 106 107
Q*
T*
Figure 11. Downstream hydraulic geometry relations for
dimensionless width B*, depth H*, slope S, and Shields
stress T*.