SCS Storms 2

[Ryb01]

Hello,

I was just wodering if there was anyone who had experience with a 12-hour SCS or 6-hour SCS Storm? I wasn't aware, until recently that these two storms (if they can be created) would still be appicable to the TR-55. I am aware of a "24-hour SCS" storm but was not aware that a shorter duration storm could still be applied to the same methodology. Is there a reason why you would want to look at a shorter duration storm? Any help on this matter would be greatly appreciated. Thanks.

Ryb01

 

[psmart]

Do you have a specific application or regulation that calls for a 12-hour or 6-hour storm?


Peter Smart
HydroCAD Software

http://www.hydrocad.net

 

[Ryb01]

The regulation/situation that has come up has been to look at what particular design storm produces the highest runoff for a particular catchment.

I've been told that depending on the catchment size the 6-hour and/or 12-hour SCS storm may govern. However, this still brings me back to the same question......Can you create a 6 or 12-hour SCS storm and apply the methodology provided by the TR-55. I had never heard of this concept until recently. Any thoughts?

 

[psmart]

As you know, the SCS/NRCS method almost always uses a 24-hour storm. Although these procedures can be used with a shorter storm, there generally isn't any benefit to doing so, since the 24-hour disribution contains all the shorter storms. In effect, the 6-hour storm is contained at the center of the 24-hour storm. This means that the 24-hour storm will have exactly the same peak intensity as the corresponding 6-hour storm. And since the 24-hour storm will also have a greater volume, there is no situation under which the 6-hour distribution will produce a higher peak flow or volume.

BTW, the wording of your question makes more sense when using the Rational method, where it is necessary to determine the critical duration. But the SCS/NRCS procedure addresses this issue by effectively analyzing all possible durations (up to 24 hours) using a single synthetic rainfall distribution.




Peter Smart
HydroCAD Software

http://www.hydrocad.net

 

[DMcGrath]

The new Virginia Impounding Structure Regulations require analysis of the 6, 12 and 24-hour storm events, although there is no guidance on how. Wide open for a little "Voodoo Hydrology".

(Stormwater Magazine)

http://stormh2o.com/july-august-2006/urban-hydrology-methods.aspx

 

[Ryb01]

Thank you psmart and DmcGrath for your help on this matter. I took some time to read through the Voodoo Hydrology Article and found it very interesting.

One final question regarding the storm distributions. After reading through the Voodoo Hydrology article, I got to thinking....If you were to look at the 24-hour center portion of the distribution(containing the 6-hour storm for example), the intensities of the storm when comparing the 6-hour and the 24-hour would be the same, however, if you were only looking at a 6-hour duration, would your time of concentration paly a role in increasing your peak runoff values??? For example if you were to calcualte a Tc of say 2hrs for a rural type area the caculated time to peak (approx) Tp= 0.6 x Tc= 1.2hrs. When looking at the distributions for the 24 hour and say the 6-hour, wouldn't there be a significant more amount of rainfall with the 6-hour storm at the time to peak given that it's a shorter duration of storm??? Just thinking that your Tc would be closer to the peak of the storm. Any additional thoughts on this matter would be greatly appreciated. Again thank you for your valuable insight and guidance.

 

[RWF7437]

"Any additional thoughts on this matter would be greatly appreciated."

The SCS storm distributions are synthetic; that is to say, they are "made up" based on some statistical analysis. For my area of the U.S. ( the Pacific Northwest) they do not resemble actual storm events, at all. For that reason, and several others, it seems likely the best way to model runoff is to try calibrate the runoff model of your choice using a range of known rainfall and runoff measurements from actual storms. This has been done in some areas of the country with modest success. Modest success means that the error between the actual runoff and the predicted runoff is reduced from +/- 200% to about +/- 50%.

Such calibration efforts also confirm what we already knew; that there is nothing magic about 24 hour storms. Many years ago, when the SCS methods were devised, the 24 hour storm was chosen mainly because precipitation records were only available from many weather stations for that time interval. Few stations had 15 minute data. Many had missing data, some had only monthly data. So, the SCS did the best they could with what they had.

Similarly, there is nothing magic about the "100 year storm". It was a legislative compromise reached in the U.S. Congress in the 1960s. It is NOT the largest possible storm. It is not even the storm which will occur once every 100 years. Indeed, the probability that the "100 Year storm" will occur in 100 years is about 63%. In other words, it is more likely to occur, than not.

Voodoo remains Voodoo but without real world data it can become doodoo.

Good luck

 

[psmart]

Rby01: Although the intensity distribution will be the same during the 6-hour period, you are correct that the runoff with be greater from the 24-hour event. This is partly because the total rainfall is greater, but also because of the non-linear nature of the SCS runoff equation. Since the early rainfall will satisfy some (or all) of the intial abstraction, a higher precentage of the peak rainfall will appear as runoff.


Peter Smart
HydroCAD Software

http://www.hydrocad.net