WSP07730

• File Number: 8062
• Description: Federal Reserved Water Rights
• State: CO
• Basin: Statewide
• Date: 1/1/2000
• Author: USGS
• Title: Analysis of the Magnitude and Frequency of Floods in Colorado - USGS - Water Resources Investigations Report 99-4190
• Water Supply Pro - Doc Type: Report/Study

0);)235

Purpose and Scope

This repon presents the regression equations
and the methods for determining the magnitude
and frequency of floods on unregulmed streams in
Colorado. In addition to data used in previous sludies
(see the "Previous Studies' section), about 2,700 addi-
tional years of record (12-22 years for each gaging
station) and 64 additional gaging stations were avail-
able for the new analysis of flood magnitude and
frequency. Data through water year 1993 were used,
and additional periods of record included water
years 1983-85 when high runoff resulted in peaks
of record at numerous gaging stations west of the
Continental Divide.
The regression equations were developed for
recurrence intervals ranging from 2 to 500 years. The
methods for deternlining peak discharges depended
on whether the site was gaged. was on a stream near a
gaged site, or was ungaged and whether the drainage
area upstream from a site crossed a hydrologic region
boundary.

Previous Studies

Hydrologic data collected from the gaging-
station network in Colorado have been used in
previous studies to develop regionalized flood-
frequency equations, A series of repons in the 1960's
defined flood-frequency relations for the Arkansas
(Patterson. 1964), Rio Grande (Patterson, 1965),
Colorado (Patterson and Somers. 1966). and South
Platte (Matthai, 1968) River Basins. Livingston (1970)
developed equations for a range of flow characteristics
for the mountains of Colorado, Hedman and others
(1972) improved the accuracy of some of Livingston's
(1970) equations by relating measurements of channel
geometry to mean annual and peak flows. McCain
and Jarrett (1976) used improved methods for esti-
mating flood frequency and used additional stream-
flow records to estimate flood characteristics for
unregulated streams in Colorado. Kircher and others
(1985) developed equations for flow characteristics in
Colorado. excluding the plains region, using addi-
tional streamflow records and improved statistical
methods for equation selection. Livingston and
Minges (1987) developed equations for estimating
flood characteristics for small (less than 20 mi2)
rural drainage basins in the plains region of eastern

2 Analysis 0' the Magnitude and Frequency 0' Floods in ColoradO

Colorado. Jarrett and Costa (1988) developed
regression equations for a relatively homogeneous
hydrologic foothill region in the South Platte River
Basin (excluding drainage areas upstream from the
South Platte River at South Platte, Colo.), Each of
these previous flood-frequency studies identified
geographical areas of the State where additional
data would improve the accuracy of the regression
eq uations.

'\.
C'1:
~
e...
e.;
e:
e-
e
e
e
e
e
c
c'
e
e
e.
e
e
e
e
e
e
e
e
e
e
c
e
c
e
c
c
c
c
e
e
e
e
c
e
e
e
e

Approach

This study differs from most of the previous
flood-frequency studies in areal coverage, number
of gaging stations used, and lengths of slreamflow
records used. Results in this study were based on
as much as 12 additional years of streamflow record
at gaging stations used in previous studies and
include 64 gaging stations not used previously. Only
one gaging station per stream was used unless the
drainage area of a downstream gaging station was
greater than about 2.5 times the drainage area of the
upstream gaging station. Drainage areas for the gaging
stations used in the analysis ranged from about 5 mi2
to about 1,000 mi2 (table 3 in the "Supplemental Data"
section at the back of the repon). The regression equa-
tions are based on at least 10 or more years of streanl-
flow records for 328 gaging stations in Colorado and
adjacent States (fig. I). A gaging station was not in the
regression analysis for a region if basin characteristics
were not available,

Flood-Frequency Analysis at Streamflow-Gaging
Stations

Records of annual peak discharges at gaging
stations and drainage-basin characteristics are the data
bases used in this study. The data bases are compiled
and maintained by the USGS. Records throughout the
study area were selected and examined for accuracy
and for the assumptions needed for a valid flood-
frequency analysis. The following assumptions are
used to validate flood-frequency analyses: (I) the data
represent independent, random events, (2) the process
generating the events is stationary with respect to time,
(3) the data are from the same population and are iden-
tically distributed, and (4) the sample is representative
of the entire population. Data through water year 1993
were used in the analyses.

.'