WSP05071

• File Number: 8146.400
• Description: Pueblo Dam - Reports
• State: CO
• Basin: Arkansas
• Water Division: 2
• Date: 10/1/1994
• Author: US Geological Survey
• Title: Use of Frequency Analysis and the Extended Streamflow Prediction Procedure to Estimate Evacuation Dates for the Joint-Use Pool of Pueblo Reservoir Colorado
• Water Supply Pro - Doc Type: Report/Study

OS'?>
'3 was computed for each day during April and May. The
April-May RDM discharges for the near-Pueblo station
for water years 1895-1975 then were adjusted for the
effects of diversion by adding the 75th percentiles of
the daily April-May diversions. Although the actual
diversion amounts could have been used for water
years 1949-75, the 75th percentiles were used for con-
sistency. Use of the 75th percentiles for the discharge
adjustment is subject to some error; however, use of the
75th percentile results in a long-term over-adjustment
of the discharge record at the near-Pueblo station. The
over-adjustment would ensure that the 0.0 I EP dis-
charges would not be underestimated in the frequency
analysis. The 0.01 EP April-May volume for the
near-Pueblo station (79-year record) was about
253,000 acre-ft before adjustment (fig. 9); when the
adjusted discharges were used in the frequency
analysis, the 0.01 EP April-May volume Was about
270,000 acre-ft.
The record for the near-Pueblo station could
be extended for water years 1976-90 by two methods:
(I) By discharge correlation with the upstream
at-Portland station, or (2) by adjusting the record for
the above-Pueblo station for the effects of diversion
and regulation by Pueblo Reservoir. Both methods
were evaluated.
The at-Portland and near-Pueblo stations have
concurrent discharge record for 13 years-water years
1940-52 (table I). A least-squares, linear regression
between the RDM discharges for the at-Portland sta-
tion and the diversion-adjusted discharges for the near-
Pueblo station was computed. For the regression
model, the coefficient of determination was 0.89, the
standard error of the estimate was 312, and 793 data
pairs were used. The model was used to estimate daily
discharges during April and May for the near-Pueblo
station for water years 1976-90.
Reliable daily contents data, and hence, change-
in-storage data are available for Pueblo Reservoir;
therefore, nonregulated discharge can be estimated for
the above-Pueblo station. Recorded discharge at this
station also is affected by diversion, but only by the
Bessemer Ditch; the estimated nonreguJated daily
mean discharge for the above-Pueblo station also was
adjusted for the effects of diversion by the Bessemer
Ditch. In this case, however, the recorded daily diver-
sion amounts were used for the adjustment, except that
the 75th percentiles were used for water years
1988-90.

Overall, there was about a 12-percent difference
in the discharges estimated for the near-Pueblo station
by the two methods. Because it was not possible to
detennine which method provided the most accurate
estimates of discharge, the results from both methods

were averaged to provide the estimated daily mean dis-
charges for the near-Pueblo station for the 1976-90
record extension. The record extension resulted in a
94-year period of record. A frequency curve for the
April-May volume for the 94-year record was devel-
oped for comparison to the frequency curve for the
79-year record. The characteristics of the curve that
includes the adjusted and extended record (fig. II) are
similar to the characteristics of the curve for the un-
adjusted 79-year record (fig. 9). The 0.01 EP April-
May volume from figure II is about 29 I ,000 acre-ft.
The extended record for the near Pueblo station was
considered satisfactory for use in the PRIUP study.

Frequency Analysis Results

The frequency analyses of RDM discharges to
derive the 0.0 I EP discharges flowing into Pueblo Res-
ervoir and the om EP discharges on Fountain Creek
and the St. Charles River were made using a computer
program (Kirby, 198 I) that incorporates all of the LP3
techniques described by the U.S. Interagency Advisory
Committee on Water Data (1982); this computer pro-
gram also was used in the preliminary analyses
described in the previous two report sections.
The April-May om EP discharges for the near-
Pueblo station are shown in figure 12. The 95-percent
upper and lower confidence limits (fig. 12) indicate the
range in which the om EP discharges would be esti-
mated 95 percent of the time from different random
samples of daily mean discharge (all sample sizes are
the same as in this analysis). The discharges in the
95-percent upper confidence limit generally are about
20 to 25 percent larger and the discharges in the
95-percentlower confidence limit generally are about
15 to 20 percent smaller than the om EP discharges.
The 0.01 EPdischarges for April15-May 14 (the
actual period of analysis for the PRIUP study) for the
near-Pueblo station are listed in table 4. The maximum
RDM discharge (adjusted for diversion) for each day
and the year of these maxima also are listed in table 4.
All. except two, of the maximum RDM discharges for
the 30 days were recorded during water year 1942.
These maximum discharges in 1942 resulted from
substantial precipitation that was coupled with low-
elevation snowmelt (Follansbee and Sawyer, 1948,
p. 105-108).
The 0.0 I EP discharges for Fountain Creek and
the St. Charles River are shown in figure 13. The range
of the confidence limits is considerably larger for these
two streams than the range for the near-Pueblo station
(fig. 12); the difference is largely explained by greater
variability in daily discharge because of rapid runoff

22 Use 01 Frequency Analysis and the Extanded Streamllow Prediction Procedure to Estimate Evacuation Dates lor the
Jotnt-Usa Pool 01 Pueblo Reservoir, Colorado