Transforming 24-hour SCS rainfall distributions to shorter durations

[JVon1]

Hello,
Some time ago I came across the attached spreadsheet on the USDA's website.
I would like to know why a transformation procedure has to be applied to use the 24-hour rainfall distributions for shorter durations?
Wouldn't it be sufficient to squeeze in the 24-hour graph horizontally to find out the distribution for shorter durations?
By that I mean shouldn't the rainfall fraction at the 12th hour of a 24 hour storm coincide with the rainfall fraction in the 6th hour of a 12 hour storm?
Similarly shouldn't the rainfall fraction at the 6th hour of a 24-hour storm be the same as the rainfall fraction in the 3rd hour of a 12 hour storm?
If so, there should be no need for a transformation procedure and it would be sufficient to sample the 24 hour storm accordingly.
I've however noticed that in applying this transformation, the SCS Type Curves for 6, 12 and 24 hours differ from each other when plotted as a dimensionless graph, which is not what I'm expecting based on my above assumptions.
Grateful for any clarification as to the difference in the distributions for the same curve type at different storm durations.

 

[BridgeSmith]

The rainfall intensities in in/hr increase non-linearly as the duration decreases. The heaviest rainfall intensities typically don't last more than a few minutes (except maybe during a hurricane), so the 5 minute intensity is somewhat higher than the 15 minute intensity, which is a little higher than the 1 hr, which is only slightly higher than the 24 hr. Over a 24 hr period, the total actual rainfall expected is not much more than what's expected in 15 minutes, so while the 5 minute intensity may be 5 in/hr, the 15 minute could be 3.5 in/hr, the 1 hour intensity 1.5 in/hr, and the 24 hour, could be maybe .25 in/hr (which would still be 6 inches total over the 24 hours). It may be possible to fit an exponential decay curve to time vs. intensity points, but I'm not sure that would even work.

Rod Smith, P.E., The artist formerly known as HotRod10

 

[1503-44]

Recently discussed here

https://www.eng-tips.com/viewthread.cfm?qid=500868

Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[psmart]

These are "nested" rainfall distributions, with all the shorter storms contained within the 24-hour storm. To obtain a shorter storm you should basically "zoom in" to the desired duration, dropping the beginning and end of the original storm. This procedure correctly preserves the peak intensity. If you rescale the entire 24-hours of data, as suggested in the OP, you will exaggerate the peak intensity.

For details see

https://www.hydrocad.net/rainfall/duration.htm

Peter Smart
HydroCAD Software

http://www.hydrocad.net

 

[JVon1]

Thanks for the much needed clarification and references All.