Can you creat IDF Curve for specific storm types?

[justinspace]

I am an IDF noob and was wondering how you would go about creating an IDF curve for a specific storm type and not a point location (i.e. for a gauge). Where does the area of the precipitation come into play, and would the precip. just be an areal average (i.e. an average depth over some area).


Thanks for the tips! I'll take any advice on how to go about it or where to look for further learning.

 

[RWF7437]

You will find agreat deal of good information at this website:

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

I don't know why you would want to create IDF curves for specific storm types. The IDF curves are useful only for Rational Method calc's which are "valid", if at all, only for small urban basins. Try learning other methods such as Regional Regression Equations, the Santa Barbara Urban hydrograph and the methods described in TR-55 and the National flood Frequency program. These are likely to yield more reasonable estimates for larger drainage basins.

good luck

 

[justinspace]

thanks again!

The reason I asked about the specific storm type was because I was under the impression that IDF curves were for a specific point location, so there would be a probability of occurrence tied to a specific intensity and duration rainfall event over that location. But it seems that you should be able arrange the information so that for a given storm (or really a spectrum of storm types), you could tie a probability of a specific I-D with the average occurrence of each storm type. For example, say a hypothetical storm type "A" tracks across the Upper Midwest an average of 50 times a year. Take the descriptives from that avg. 50 and compute average rainfall depth over average duration of each occurrence. This will provide you with a suitable range to arbitrarily select for the y and x-axis, respectively. Choose a method for IDF calculation. And ultimately, be able to provide a statistic saying for example, the return period for 1-hr. 4" rain produced from storm type "A" in the Upper Midwest is 2 years.

Does that make sense?

 

[blueoak]

justinspace,
The HEC-HMS tech. ref. manual references the World Meteorological Organization that point values should be used without areal reduction up to 9.6 sq miles. This is consistent with the NWS HDR's which at least a few cite no reduction for 10 sq miles or smaller even for local events.

As for "storm types", what do you mean? I normally use that for duration and intensity spectrum. Are you talking meteorologically?

Rather than creating IDF curves, the depth duration frequency curves from NWS are nice to allow you to run a hydrograph that nests events of different durations of one frequency. It also answers your 1 hr 4" rain question except for storm type.

 

[bltseattle]

You might check out the procedures used in Colorado (CUHP), as I recall there are agency provided spreadsheets to perform those calculations, linking watershed travel time to IDF's etc. Maybe you can adapt to your region....

http://www.udfcd.org/downloads/down_technical.htm

 

[env21tech]

(1)IDF data is based on raincloud precipitation and is independent of watershed travel time (what is happening on the ground).
(2)Hydro-35 precipitation data appears to be based on sampling durations of 5, 15, and 60 minute.
Average rainfall intensity = depth collected /duration of sampling interval.
(3)For $32 you can buy current IDF curves from the NOAA.
(4) Based on BlueOak the smallest raincloud that can be represented by sampling would be 9.6 sq miles = a raincloud about 4 miles long.
Assuming this represents the 5-minute duration, the raincloud velocity is 48 mph.