FLOOD10401

• County: El Paso
• Community: Colorado Springs
• Stream Name: Fountain Creek
• Basin: Arkansas
• Title: Fountain Creek Drainage Basin Planning Study - Volume 1
• Date: 7/1/1994
• Prepared For: Colorado Springs
• Prepared By: Muller Engineering Company, Inc.
• Floodplain - Doc Type: Floodplain Report/Masterplan

I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

inference from this oplnlon is that the flows downstream of the confluence are
also in question and possibly too high. The primary value of these numbers is,
therefore, the consistency they provide on regulatory issues. These values and
that of the paleohydrologic evaluation are judged to have the additional value
of defining the upper end of the envelope encompassing the field of reasonable
flow values.

The streamflow statistical analysis, for all three gaging stations, both with and
without approximate values for the 1935 flood, is distinguished by the absence
of high flow events. This is believed to be an inaccurate representation of the
true distribution of flood events. This is partly indicated by the wide variance
in numbers bracketing the 95% confidence interval (i.e. low confidence) of
projected la-year and lOa-year flows. The only reliable value developed from the
statistical analysis is the mean. The projected flows for the la-year and more
infrequent events are judged to be unreliable and probably represent the bottom
end of the envelope encompassing the field of reasonable flow values. In
addition, paleohydrologic evidence confirms that high intensity, high peak flood
events have occurred historically in most individual basins, but not on a basis
which is widespread enough to cause major flooding on Fountain Creek. The issue
as to whether or not extreme meteorologic events can occur over a large enough
area of highly variable topography remains, unfortunately, unresolved.

Regional analyses by Muller (unpublished) and Pitlick (1988) and individual
experienced extreme hydrologic events occur within the envelope of values between
the current adopted flow values and the statistically generated values plotted
on Figures 4.8-1, 4.8-2 and 4.9-1.

Design storm methods in general are used either where gaging information is
inadequate and/or where theoretical future flood peaks caused by urbanization are
of interest. Typically they are not used, without cal ibration to "real" events,
to determine design flow values for major drainageways such as Fountain Creek.
Unfortunately, the absence of known frequency flood values places a significant
limitation on performing such calibration accurately. Without calibration,
design storm methods characteristically produce flows which are on the high side
of the true values (in those rare instances where comparisons can confidently be
made). This is due to the tendency of hydrologists to estimate (even by a small
amount) the values of individual parameters "on the conservative side" to
compensate for inadequate i nformat i on or other uncerta i nt i es. Some of these
design storm methods even have this conservatism embedded in the methods in order
to produce numbers "on the safe side. II There are many opportunities in the
choice of variable values and algorithms (storm pattern, rainfall amounts, ground
cover interpretation, eN estimation, routing parameters, antecedent moisture
assumption, soils interpretation, etc.) for this uncertainty to become compounded
with the result that numbers on the high side are produced. Even with these
limitations, a design storm method (the ses unit graph procedure in this case),
can produce reasonabl e results if parameter val ues are chosen with object ive care
(including varying values for varying frequencies), reflecting as accurately as
possible the physical realities involved. This was the approach taken herein.
Figure 4.9-2 graphically shows the design future peak flows for the study reach
by stream distance according to discharge.

4.0-27

.
.
.
.
.

I
I

I
I
I
I
I
I

I
I
I