Differences in SCS and Modified Rational Storage Numbers

[Woody2185]

This is not intended to be a SCS vs. Modified Rational thread.

My question concerns the vastly different storage numbers I am getting using the two different methods. My site is a 6.5 acre watershed that will be almost compeltely developed. I am using PondPack to perform the calculations.

Modified Rational:
Pre-developed Peak: 7.32 cfs
Post-developed Peak: 10.31 cfs
Estimated Storage: 2.22 ac-ft

SCS:
Pre-developed Peak: 19.70 cfs
Post-developed Peak: 43.97 cfs
Estimated Storage: 1.15 ac-ft

I understand where the differences in flows in coming, but the difference in storage is confusing me. Which method is more accurate with this size development, and why?

 

[RWF7437]

Not a bit surprising.

First, if what you refer to as the Modified Rational Method is the simplistic notion that a longer duration storm of uniform intensity will produce a trapezoidal hydrograph then you cannot get any meaningful comparison to a SCS ( now NRCS) hydrograph based on one of the 4 storm types described in TR-55.

The runoff hydrographs size and shape depend heavily on the size and shape of the rainfall pattern that produced the runoff. The Rational Method rainfall pattern is assumed to be of constant intensity. This is a hopelessly unrealistic assumption. Such storms never have, and probably never will, occur.

The NRCS method also assumes a rainfall pattern or patterns. While these too are artificial, they are somewhat more realistic. They are based in a statistical analysis of historical storms. But they too are subject to wide variation in the real world.

If you are familiar with the Los Angeles County Modified Rational Method you can get better agreement between the NRCS and LA Mod Rat methods. Plus or minus 50%, in both peak flow and volume, would be considered good agreement.


But the best course of action would be to select a method and then CALIBRATE it using real storms over a wide range of durations and intensities.

Often this isn't possible or affordable. Then you are stuck with simply meeting some arbitrary local "standard" to prove "compliance". Even if you can't calibrate your hydrology model you can always look beyond the "standards" and satisfy yourself that you have done a careful and conservative design.

good luck

 

[cvg]

if you are calculating retention volume, just take the rainfall depth times the watershed area times the estimated runoff coefficient C to get the total runoff that should be retained.

 

[RWF7437]

http://www.pdhonline.org/courses/h131/h131content.pdf

 

[RWF7437]

Let’s see. If we multiply a guess, times a statistic, times an estimate do we get a certainty ?

Suppose our guess is plus or minus 30%.
And suppose our statistic is plus or minus 50%.
And suppose our estimate is plus or minus 5%
Then our certainty is:

1.30 x 1.50 x 1.05 = 2.05, rather than than 1.00

Our certainty is now plus or minus 105% but only for that particular combination of guess, statistic and estimate.