FLOOD10216

• County: Statewide
• Basin: Statewide
• Title: Method for Estimating the Magnitude and Frequency of Floods at Ungaged Sites on Unregulated Rural Streams in Iowa
• Date: 1/1/1987
• Prepared For: State of Iowa
• Prepared By: USGS
• Floodplain - Doc Type: Educational/Technical/Reference Information

Re2ional Flood Freauencv Eauations

The historical record of flood peaks at 251 gaged stations in Iowa and
adjacent States were used in the development of regional flood-frequency
equations. All gaging stations were separated into groups representing the
hydrologic regions shown on figure 1. A gaging station was included in the
analysis of only one hydrologic region if more than two-thirds of the
drainage area is in that region. If less than two-thirds of the drainage
area is in any single hydrologic region, the station was used in the data
set for both regions. Within each regional data set, a series of subsets
of the flood discharges for recurrence intervals of 2, 5, la, 25, 50, and
100 years were made. The resulting subsets were used to define the
relation between flood discharge and drainage area for each recurrence
interval in each of the five hydrologic regions.

Least-squares regression analysis was used to
flood discharge and drainage area. The general
developed from this analysis are:

define relations between
form of the equations

Log Qt - bLog A + Log c.

On transformation the equation becomes:

b
Q - cA
t

where Qt

the discharge for the selected recurrence interval, in
cubic feet per second;
the drainage area, in square miles; and
regression equation coefficients.

A
band c

Flood frequency equations for each of the five hydrologic regions are
listed in table 2.

Previous flood reports (Schwob, 1963 and Lara. 1973) indicated that
comparing the magnitude of floods to the drainage area and the slope of the
main channel accounted for a large part of the variance in the magnitude of
flood peaks. This study indicated that within the newly delineated
hydrologic regions, the size of the drainage area and the slope of the main
channel are significantly correlated. Therefore, the channel slope was not
used as an independent variable for these equations.

The flood-frequency equations for each hydrologic region in Iowa
(table 2) were developed using drainage area as the only independent
variable. These equations explain from 7l to 95 percent of the variance in
the flood discharges in all cases tested. An attempt was not made to
identify other independent variables. It was assumed that most of the
unexplained variance is due to spatial-sampling errors, varying lengths of
station record, errors in the measurement of stage and discharge,
uncertainties in the stage-discharge rating curves and the use of drainage
area as the only independent variable. Unexplained variation is quantified
as the standard error of estimate (table 2) and was calculated using
methods described by Hardison (1971, p C231-C232).

6