WSP11681

• File Number: 8220.101.17
• Description: Glen Canyon Dam/Lake Powell
• State: AZ
• Basin: Colorado Mainstem
• Date: 4/1/1995
• Title: Measured and Predicted Velocity and Longitudinal Dispersion at Steady and Unsteady Flow / Colorado River / Glen Canyon Dam to Lake Mead
• Water Supply Pro - Doc Type: Report/Study

Graf

TABLE 2. Statistics of the Time-Concentration Curves, Glen Canyon Reach, 1989 and 1991.
[Average velocity WBI!l computed all velocity of the centroid of the time-concentration curve.]

D10tance
From
Injection
(kiIometero)

Discharge
(cubic
meters per
lI8Oond)

M8.][imum
Concentration
(micrograDUI
per liter)

Time After Injection
(bouro)
Peak Centroid

Time
Variance
(bouro
oquored)

Average
Velocity
Coefficient (meter.
of Skew per oecornl)

Measurement, 1989

61.2 1.12 1.35 0.115 1.238
5.78 20.2 21.8 10.4 1.225 0.33
Steady-Flow Measurement, 1991
2.27 9.70 9.84 1.34 .450 .72
Unsteady-Flow Measurement, 1991
1.98 6.60 7.07 0.708 .560 1.0

1.5
25.9

144
139

25.6

425

25.6

651

Steady and Unsteady Flow, Grand Canyon
Reach. The time-concentration curVes generated
from samples collected at steady flow are atypical in
that although the curves have a slight positive skew,
they do not have the long tails typical of natural
streams (Figure 4 and Table 3). The time-concentra-
tion curves for unsteady flow are similar to those for
steady flow in that they do not have long tails, but the
shapes of curve8 at individual sites are strongly influ-
enced by discharge changes in the reach as the dye
passed (Figures 2 and 5). For example, dye curves for
sites at which the dye passed during decreasing flow
are positively skewed, whereas dye curves for sites
where the dye passed on increasing flow are negative-
ly skewed (Figure 2 and Table 3). The dye curve at the
site above the Little Colorado River, where the dye
passed on the trough of the daily hydrograph, is near-
ly symmetrical.
For the steady-flow measurement, velocity varied
slightly from subreach to subreach (Table 3). The low-
est velocity (0.75 m/s) was measured in the subreach
between Nautiloid Canyon and the sample site above
Little Colorado River - the Lower Marble Canyon
reach (Schmidt and Graf, 1990, Table 2, p. 55). The
highest velocity (1.1 mls) was measured between the
site above the Little Colorado River and the site below
Nevill's Rapid (Furnace Flats reach) and the two sub-
reaches between Mile 118 Camp and Pumpkin
Springs (Middle Granite and Muav Gorges). Velocity
is not significantly correlated with any of the channel
geometry characteristics given in Table 1. Poor corre-
lation of velocity with slope and channel geometry
probably is caused by the inadequate characterization
of the channel by the 199 measured cross sections.
Velocity of flow in individual subreaches during
unsteady flow ranged from 0.61 mls in the Lower

WATER RESOURCES BULLETIN

Marble Canyon reach to 1.3 m/s in the subreach
between the site below Nevill's Rapid and the site at
Mile 118 Camp (Granite Gorge). For unsteady flow,
differences in velocity through individual subreaches
were more strongly influenced by discharge in the
reach as the dye passed than by the geometry of the
subreach. Velocity was highest in the subreaches in
which the dye cloud traveled near the peak discharge
of the daily hydrograph - from Lees Ferry to Nau-
tiloid Canyon, below Nevill's Rapid to Mile 118 Camp,
and National Canyon to Gneiss Canyon (Figure 2 and
Table 3). The lowest velocity was measured in the
reach between Nautiloid Canyon and the Little Col-
orado River, a reach where the dye cloud traveled
with the trough of the daily hydrograph (Figure 2 and
Table 3).
Traveltime of the dye-cloud centroid increased lin-
early with distance traveled for both steady and
unsteady flow. Although average velocity varied from
subreach to subreach during both measurements,
velocity differences were not great enough to signifi-
cantly affect the linear traveltime-distance relation
(Figure 6). Traveltime was slightly less during
unsteady flow than during steady flow, but average
velocity in the entire measured reach was not signifi-
cantly different - 0.98 m/s for steady flow and 1.0 mls
for unsteady flow.
Downstream changes in peak concentration and
dye-cloud variance and duration are all measures of
the longitudinal dispersion. For steady flow, peak con-
centration decreased as the square root of traveltime.
Peak concentration was 12.5 J.lg/l at the first sampling
site, 57.7 km downstream from the injection, and 5.3
J.lg/l at the last site, 380 km from the injection (Figure
7). Nonlinear regression was used to relate the peak

272