WSP00684

• File Number: 8220.101.16
• Description: Glen Canyon Dam/Lake Powell
• State: AZ
• Basin: Colorado Mainstem
• Date: 1/1/2000
• Author: Webb et al.
• Title: Sediment Delivery by Ungaged Tributaries of the Colorado River in Grand Canyon Arizona
• Water Supply Pro - Doc Type: Report/Study

Table 14. Debris flows that have had significantly changed debris fans and rapids during the last century in
Grand Canyon (continued).

Year(s) or year range of
Tributary name Name of rapid River mile Side debris flow(s) Method useef
Monument Creek tOranite 93.5 L 1960s 2
Monument Creek tGranitc 93.5 L 1984,1996 I
Hermit Creek tHennit 95.5 L 1996 I
Boucher Creek Boucher 96.7 L 1951 to 1952 1,2
Crystal Creek Crystal 98.2 R 1966 1.2,3
Waltenberg Canyon Waltenberg 112.2 R 1938 to 1987 2
Unnamed canyon New rapid 127.6 L 1989 I
I 28-Mile Creek 128-Mile 128.5 R 1890 to 1923 2
Specter Chasm tSpecter 129.0 L 1989 1
Bedrock Canyon tBedrock \30.5 R 1989 I
Unnamed canyon Unnamed riffle 133.0 L 1890 to 1923 2
Kanab Canyon Kanab 143.5 R 1923101942 2
Unnamed canyon #New rapid 160.8 R 1993 I
Prospect Canyon Lava Falls 179.4 L 1939 2
Prospect Canyon Lava Falls 179.4 L 1954.1955.1963.1966 2,3
Prospect Canyon tLaw Falls 179.4 L 1995 I
194-Mile Canyon tUnnarned riffle 194.5 L 1998 I
205-Mile Canyon 205-Mile 205.5 L 1937101956 2
205-Mile Canyon t205-Mile 205.5 L 1998 I
Unnamed canyon Unnamed riffle 222.6 L 1890 to 1990 2
Diamond Creek Diamond Creek 225.8 L 1984 I
Unnamed canyon 231-Mile 231.0 R 1890 to 1990 2

.1 = Direct observation; 2:::: Repeat photography; 3:::: Aerial photography.
tEnlarged an existing rime or rapid after 1984.
#Created a new rime or rapid after 1984.

With logistic regression, the probability that an
event will occur, 1t(x), is:

1t(x) = eg(x) (7)
1 + eg(x)

where

g(x) = /30 + /3;Xi + ... + /3"x.. j = 1..... n, (8)

and Xi are the variables, /3i are the modeled variable
coefficients, and /30 is the y-axis intercept. Each
variable was chosen based on the statistical
significance of its contribution to the model
(Griffiths and others, 1996). Owing to the high
degree of spatial variability in tributary variables,
Grand Canyon tributaries were grouped into eastern
and western sets at Hermit Creek basin (mile 95).
For both the eastern and western Grand Canyon
models, n = 5 variables (table 15). Noteworthy
among the significant variables are several terms

that reflect the topographic relations of shale-
bearing formations to the Colorado River as well as
the aspect of the river corridor, which affects how
storms interact with canyon walls (Griffiths and
others, 1996). Fitted models explain about 74
percent of observed debris flow occurrences.
These logistic-regression probabilities derive
from a cumulative-distribution function that can be
related to the cumulative-distribution function for
the binomial distribution. As such, these
probabilities must be converted into a more useful
form that reflects the magnitude-frequency relation
for debris flow in Grand Canyon. For large numbers
of data, such as the Colorado River tributaries, in
Grand Canyon the cumulative-binomial
distribution can be described by a
cumulative-distribution function for the normal
distribution (Haan, 1977). Because lognormally

DEBRIS-FLOW SEDIMENT YIELD 25