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

Table 2.--Summary of statistics describing observed annual maximum

precipitation values for selected durations

Duration

Mean

Standard
deviation

Skewness

Arithmetic data

5 minutes
10 minutes
15 minutes
30 minutes
1 hour
2 hours
24 hours

0.311
.476
.589
.749
.864
1.00
1.55

0.130
.198
.245
.369
.494
.668
.753

1.42
.984
1.06
1.90
2.91
3.41
2.92

Logarithmically transformed data
5 minutes -.542 .177 -.355
10 minutes -.359 .183 -.401
15 minutes -.265 .179 -.207
30 minutes -.168 .188 .303
1 hour -.112 .198 .595
2 hours -.055 .205 .995
24 hours .153 .169 .655

A second analysis was carried out--this time based on the assumption
that the logarithms of the data were described by a Gumbel distribution.
Again, the method of the National Weather Service was used to fit a straight
line to the data. Skewnesses of the logarithmically transformed data ranged
from -0.359 to 0.995. The logarithmically transformed data for durations of
1 hour or less have skewnesses substantially smaller than the skewness of
1.14 assumed in the Gumbel distribution. Therefore, a straight line based
on the log-Gumbel distribution fitted to this data may overpredict the
magnitude of the 100-year storm.

Because the calculated skewness of both the data and the logarithms of
the data for certain durations was substantially different from the value of
1.14 assumed by the Gumbel distribution, another distribution was examined
to see if it would better describe the data. The log-Pearson Type III
distribution was chosen because the log-Pearson Type III distribution,
unlike the Gumbel distribution, is based on three parameters: Mean,
standard deviation, and skewness. The analysis utilized Geological Survey
computer program J407 (Lepkin and others, 1979).

13