FLOOD00005

• County: Statewide
• Title: Precipitation Records and Flood-Producing Storms in Cheyenne Wyoming
• Date: 10/23/2003
• Prepared For: Cheyenne
• Floodplain - Doc Type: Educational/Technical/Reference Information

SUMMARY

Annual maximum precipitation data for Cheyenne, Wyoming, have been
collected for durations of 5, 10, 15, and 30 minutes, and 1 and 2 hours
since 1902 and continuing to the present. Annual maximum precipitation data
for 24-hour duration have been collected beginning as early as 1871. Data
were plotted and analyzed using three different assumptions: (1) The data
could be described by an arithmetic Gumbel distribution; (2) the logarithms
of the data could be described by a Gumbel distribution; and (3) the
logarithms of the data could be described by a Pearson Type III
distribution. For each of these distributions, the data were examined
twice. First, the "previously definedll precipitation-freqency curve was
developed, us ing data collected before 1985. Updated curves were then
developed, using all data collected through 1986, in order to examine the
effect that the large thunderstorm of August 1, 1985, had on the predicted
magnitude of storms of long recurrence interval. The inclusion of the storm
of August 1, 1985, had the most noticeable effect on the prediction of the
amount of precipitation expected in the 1- and 2-hour durations, and the
least noticeable effect on the prediction of the amount of precipitation
expected in the 10- and 15-minute durations, regardless of which
distribution was assumed to describe the data. The predicted magnitude of
storms was similar (within 10 percent) for all distributions for storms of
less than 25-year recurrence interval. Predicted magnitude of storms with
extremely long recurrence intervals varied greatly depending on which
distribution was assumed to describe the data. The skewness both of the
arithmetic and of the logarithmically transformed data was different than
the value of 1.14 assumed in the Gumbel distribution, which caused the
Gumbel distribution to underpredict--and the log-Gumbel distribution to
overpredict--the amount of precipitation expected for a given recurrence
interval. Therefore, the log-Pearson Type III distribution was considered
the most suitable for describing the Cheyenne preCipitation data.

Four different methods of analysis were
intensity-duration curves for the 100-year storm.
pared to the observed curve for the August 1, 1985,
greater than 30 minutes, the observed curve indicates
do five of the seven calculated curves.

used to develop seven
These curves were com-
storm. For durations
greater, intensity than

Dimensionless hyetographs were developed for 10 flood-producing storms
that have occurred in the Cheyenne area since 1903. Pattern index (integral
of the dimensionless hyetograph curve) was calculated, and the index for the
storm of August 1, 1985, is three standard deviations lower than the mean of
the pattern indices for the other nine storms, indicating that the distribu-
tion of precipitation over time for the August 1 storm was outside the
normal range for Cheyenne.

43