Storm Duration 1

[nzo]

i'm doing some modelling using the MIDUSS program and am not sure what to use for storm duration. i'm fairly new to the concepts of SWM design and modelling so i'm not exactly sure what is a reasonable and acceptable value for storm duration. i've seen reports using 180 or 360 minutes typically. why use one over the other? i take it using 180 min would yeild a more conservative design then a larger duration but is it justified to use? does the value depend on local rainfall characteristics mostly? any answers are very much appreciated, thanks in advance.

 

[francesca]

Actually a shorter duration yields a larger peak flow and a smaller volume, so depending on whether you're doing pipe design or detention/retention design, you're more conservative with the shorter or longer, respectively.

The best guide to learning what's appropriate to use is your local drainage ordinance. Clearly the ordinance isn't a substitute for understanding, but it'll give you a yard stick to measure with.

I've never done design with a 180-min or 360-min storm. When I was designing subdivisions, the standard was to use a duration equal to the time of concentration for pipes and a 24-hour storm for detention/retention. If you're modeling a complex urban storm sewer system, with a combination of pipe size issues and interim storage issues, I could see how you'd be interested in a different storm duration. The trick is to pick the worst case scenario for your conditions.

 

[LCruiser]

Actually, it depends on the drainage basin. The longer the basin, the longer the most destructive storm. Just one more irrational "assumption" we used to make before computational aids that has stuck with us.

 

[RWF7437]

There is nothing "magic" about a duration equal to Tc or 24 hours. Examine as many durations as you can with the software you have. Depending on what you're trying to design, select the duration that yields either the largest peak flow or the largest runoff volume.

By the way, what is MIDUSS ? And what is it you're trying to do ?

 

[LCruiser]

I agree with RWF7437. Tc is a construct - it's not the actual time of concentration of a storm, which will vary by storm direction and velocity, distribution of intensity and intensity itself.

 

[nzo]

thanks for the feedback everyone.

RWF7437, MIDUSS stands for Micro Interactive Design of Urban Stormwater Systems.

what i'm doing is modelling a 100 year storm over a 90 acre watershed because the client wants only subsurface conveyance of the runoff, no surface, therefore all the runoff has to be transfered through pipes.

i've run the model using both durations and the 360min duration yields peak flows about 10% more than 180min so maybe i'll go with that even though it isn't too much more. this design is for a small town that doesn't have any standards for design so local data is limited, especially for a major storm design.

 

[francesca]

Your 180 min storm should have higher a higher intensity and higher peak flows in short pipes. In longer pipes, I still can't see it taking 6 hours for all points of the 90 acre drainage basin to contribute to flow, so I'm somewhat surprised that this duration results in higher peak flows.

Assuming a perfectly circular urban area of 90 acres, your radius is a little over 1,100 ft. Doubling the radius to account for a dendritic drainage pattern, and assuming 2 fps in the pipe, you get a pipe drainage time of less than 20 minutes. For an urban area, allowing 10 minutes for water to reach the storm sewer, you're still only at 30 minutes. A 30 minute storm has a much higher intensity, therefore peak flow, than a 6-hour storm. As all points of the drainage basin can potentially contribute to flow at the outlet after only 30 minutes of rainfall, this is likely to be your worst case storm duration...

... unless you have a lot of temporary storage.

 

[RWF7437]

Thanks nzo. I Googled MIDUSS and found the answer too.

For 90 acre watershed, my suggestion would be to use the NRCS 484 unit hydrograph method or regional regression equations. The latter are pretty conservative. Whatever you do please don't misuse the Rational Method for this basin.
As for your client, let him understand that the streets of his development are surface conveyances and such surface flow cannot be avoided. You might be able to assure him that the system can be designed so that you feel 95% certain that its capacity will not be exceeded in say the next 50 years. To achieve this degree of "certainty" will require you to design for a very large storm. You can never, in hydrology, be 100% certain.

Finally, even though this is a small town, surely it is in a County which has some standards or in a State whose DOT has some standards ?

good luck

 

[RWF7437]

A possibly better way of viewing these events, and speaking about them, is to refer to them by their annual probability of occurrence. The so called 100 Year storm is, by definition, the storm which has a 1 percent probability of occurring in any one year. If we want to know how frequently such a storm might occur over a longer period of time, that probability can be calculated by:

Px = 1 - ( 1-1/N)^x

Where: Px is the probability of occurrence in x number of years

1/N= the Probability of Occurrence in any one year

For example, if we want to calculate the probability of occurrence of the 100 year flood over 100 years the calculations would be:

Px = 1 - ( 1-1/100)100

Px = 1 - ( 1- .01) 100

Px = 1 - ( 0.99)100

Px = 1- 0.366

Px = 0.634

In other words, there is a 63 percent probability that the 1 Percent storm will occur one or more times over the next 100 years.

This kind of calculation can be done for any selected range of frequencies and time periods.

 

[Ron]

Storm events are generally "location oriented". The selected design storm, as Francesca noted, is usually dictated by ordinance, based on a variety of factors including surface profiles, expected rainfall intensity, runoff coefficients, and many other variables.

The storm event is also based on probabilities of occurrence, such as designing for a 24-hour storm that occurs once every 25 years, or an 8-hour storm that occurs every 100 years.

 

[blueoak]

Just a comment on the duration versus intensity comments suggesting a shorter storm has a higher peak.

I have rarely seen good data on expected peak storm intensities for different storm durations and split up for different frequencies. I think Texas has some data, but from what I have looked at it is like a lot of hydrographic data where there is a line in the middle of scatter. The data I use most often see recommends nesting of intensities especially for freqeunt events such as the 100 yr and smaller. Despite this data showing a 15 min duration which indicates a higher intensity it is often the case that this is supposed to be nested within storms of longer durations. This is very well shown in Atlas 14 data and was what the NRCS attempted with the Type 2 event. This event wasn't a 24 hr duration but rather a nesting which is why some areas have alterred the Type 2 based on flooding histories.

By nesting storms your peak shouldn't decrease, but flatten until you get multiple storm events mixed in the duration.

 

[blueoak]

nzo,
Make sure you have your answers checked and provide overland release. All pipes plug and all inlets plug.
Having a street or two for discharge during some bizarre event, freeze/leaves(grass cuttings will plug most inlets)/tumbleweeds or what have you will help. I remember swimming in the low streets during big events when I was a kid. It may have been undersized drains, but there was an awful lot of trash floating in it too.

 

[francesca]

blueoak by your chosen profession I take it you didn't like swimming in the streets in the rain?

 

[RWF7437]

Blue Oak,

For some "good data" visit:

http://www.weather.gov/oh/hdsc/

Click on any one of the blue states.

Good night and good data

 

[blueoak]

RWF7437,

Atlas 14 and the other NOAA data is excellent and I do like the reports. However, if you read through many of the reports the point rainfall for each duration is not for a storm of that duration as seemingly suggested by francesca's comments about short intense versus long gentle storm comments. Rather as utilized by the HEC-HMS update for frequent events, e.g. 100 yr or 20 yr, nesting the storm intensities appears to be the preferred method rather than looking at each data point independently.

The "good data" I meant to refer to was referenced data about storm distibution based not only on frequency but also distribution. This is much more limited.