72-Hour Rainfall Distribution 1

[DMcGrath]

I will likely soon be asked to analyze a basin and spillway for the 72-hour 1/2 PMF. I can easily find the rainfall amount for that storm, but when it comes to modeling the runoff, where can I find (or create) the 72-hour rainfall distribution for my region (central Virginia)?

The drainage basin is about 800 acres and I am using PondPack (grrrr...) My Tc is less than 2 hours, and I think the 72-hour storm is excessive, but this 72-hour analysis may be the rule in the future. For the 24-hour storm, I am using the NRCS Type II storm event. I was thinking perhaps to use that as a basis somehow to interpolate by adding time intervals or something like that.

Any ideas?

 

[cvg]

I am assuming that this is a dam that you are analyzing. If so, you should be asking for some guidance from your state dam safety department. If not, use HMR-51 for the precip and HMR-52 for application. Chapter 2 in HMR-52 gives a procedure for creating the rainfall distribution.

 

[RWF7437]

CVG,

What is HMR 51?
Where would one find a copy of it ?
I'll Google it but a direct link would help.

rwf7437

 

[cvg]

http://www.nws.noaa.gov/oh/hdsc/studies/pmp.html

HMR 51
Probable Maximum Precipitation Estimates, United States East of 105th Meridian

HMR 52
Application of Probable Maximum Precipitation Estimates - United States East of 105th Meridian

 

[blueoak]

DMcGrath,

I would double check your Tc also, it seems high.

I believe HMR-51 nests PMP events for different durations(though cvg seems familiar with it)

Another good source for distributing the rainfall is the Bureau of Reclamation's Flood Hydrology Manual by Cudworth. A lot of of it is in Design of Small Dams by the Bureau, which is probably the best book I have ever read and relied on. The Bureau typically uses a late weighted scheme that for a single basin should give you the greatest discharge flow rate. The COE uses a center weight. (Like cvg says your dam safety regulator should let you know what they allow).

One thing to note is that for stability analysis it is sometimes unconservative to peak your event too much. It is sometimes good to flatten the hyetograph. Most PMP distributions are concerned with determing maximum capacity and may not be appropriate for stability analysis.

 

[DMcGrath]

Blueoak,

My Tc is just a shade over an hour. I mentioned the Tc because a 72-hour storm analysis seems excessive if the Tc is so short. The dam in question impounds a reasonably small amount of water and is not connected to an extensive river system which could potentially take >24 hours to see peak flows.

The dam safety regulations are intentionally vague and it even says right in the text that "...rigid design flood criteria or standards is not intended." Oh great. Thankfully they are under technical review right now and I hope they get better. Perhaps this is a topic for another day.

I will look into the Design of Small Dams book. I agree it is a wonderful reference.

Thanks to all for the advice.

 

[DMcGrath]

So much for the 72-hour storm distribution.....It turns out that the required analysis will be for 6, 12 and 24-hour storm events, which seems a whole lot more reasonable. It turns out the 72-hour analysis was a miscue from an earlier discussion.

For those interested, for my particular application, the storms in ascending order of runoff were: 12-hr, 6-hr, then 24-hr being largest. Storm was 1/2 PMP.

 

[blueoak]

The 1/2 PMP is my favorite. You take the biggest storm and then say, How about half of that. I understand using the PMP or at least looking at it, but the 1/2 PMP is the oddest regulation I know of. Most areas you can make a guess at the (statistical) 500 or 1000 yr event.

If someone knows where the 1/2 comes from I would love to know.

 

[Sbortz]

I'm working on a similar situation (south/central VA, small drainage area under 2 sq mi) but I don't have access to PondPack. I've been trying to use software which uses TR-55 to generate a hydrograph but I can't seem get a reasonable peak flow number.

My thought was that for such a small basin, I could use the PMP from HMR 51 (figure 20) for 24hr for a 10 sq mi basin and use a 24-yr hyetograph for distribution and use the SCS TR-55 to get generate a hydrogaph.

I know I'm doing something wrong/stupid so any insight would be appreciated.

 

[DMcGrath]

After I computed my flows for the 1/2 PMP event, they too seemed ridiculously large. If it wasn't for some first-hand anecdotal evidence about a recent storm event (Gaston) I would have seriously doubted the results as well. A PMP-level rainfall event packed into a 24-hour distribution is a LOT of water!

 

[blueoak]

Sbortz,
I would use the hyetograph procedure as shown in the HMR. If you use HEC-HMS at least for the hydrograph generation if not for the pond. You can use the SCS methods as desired in HMS but still run the meteorology like the HMR. The SCS stuff sometimes have their hyetograph distributions built into their charts and software.

I cited it earlier, but for large flood events on isolated sheds the Bureau of Reclamation's Flood Hydrology Manual by Cudworth is really a great resource.

 

[cvg]

Sbortz
if you are designing a dam - and I assume you are since you are considering using the PMP, it might be beneficial to meet with your state dam safety engineer to determine what the requirements for hydrology will be. In virginia, you can check on the web at:

http://www.dcr.virginia.gov

http://www.damsafety.org/layout/subsection.aspx?groupid=1&contentid=187

 

[blueoak]

Sbortz,
cvg makes a good point. It might be even better to find out the criteria for a "dam" and stay below it. I remember using an internal 24.9' high criteria for designing dams that the state was happy with since they looked at 25' and higher. Of course camber almost got me in trouble.