Flood Mitigation - Modeling Existing Developed Areas (With / Without Detention) 2

[rlondeen]

I am doing a mitigation study for a proposed development that will be outfalling into a creek. Using the SCS method to developed hydrographs for the creek and the development, I am trying to prove that the new development will not create an increase in peak flow for the creek due to the storms peaking at different times.

When modeling storm runoff for large drainage areas, how do you treat currently developed with and without detention ponds? The creek has a 400 acre drainage area at the point of analysis and contains several commercial sites. For pre-developed flows for the creek I am using pre-developed surface conditions for developments with detention ponds (low CN). For developments with no apparent flood mitigation, I am using their current surface conditions (high CN).

Is this the correct approach for handling both developments detaining and not detaining?

 

[francesca]

It certainly isn't how I'd do it or anything I'd approve.

 

[gbam]

If you can show that the detention basins do maintain a predeveloped condition, then maybe. Have you reviewed all the development drainage reports? If not, I recommend that you do. You can even include the basin(s) within your hydrologic model. It is only a 400-ac watershed; it shouldn't take long to model it.

 

[rlondeen]

There are about 5 different developments with detention ponds in this drainage area. Tracking down the drainage reports for these developments and then modeling them would be difficult. I could model as developed, but wouldn't modeling with pre-existing conditions be conservative? Perhaps tracking down the reports for those ponds will be necessary though.

 

[beej67]

The cleanest approach is to track all the other reports down, and enter their discharge hydrographs manually into nodes in your hydrology software. Then route those hydrographs through reaches that describe how their discharge travels to the point of analysis. That approach preserves the work done prior by other engineers, and saves you time, but it will only really work properly if all the other prior engineers used the same method. (e.g. SCS) Don't mix hydrographs generated by different methods, or your results will be meaningless.

Most of the time I'm forced to study a partially developed region in which the other hydrology studies for other developments aren't available, I treat them as fully developed and undetained. There are benefits and drawbacks to that approach, but most regulatory agencies I encounter consider it to be the most conservative approach, particularly provided you are in fact detaining your own site and meeting allowable discharges within your own developed area.

I avoid timing analyses that try to prove an undetained development will have no impact on a watershed, unless I'm waaaay down the watershed, and adjacent to the floodplain, and the watershed is a lot bigger than 400 acres. The truth about timing analyses is that you could show there's no increase for a 24 hour storm distribution but in fact there still might be one for an 8 hour storm distribution, or a 4 hour or 2 hour, and your 24 hour analysis might get you past the reviewers but might not stand in court. 24 hour distributions are used as a standard for detention design in part because when detaining, they characterize the behavior of the watershed well. But if you're trying the old "our flow gets out before the flood wave gets here" thing, and you want to do it responsibly, you really need to test storm distributions of different durations.

In FDOT work, for example, they have a half dozen(ish) storm distributions they want you to run for any hydrology study. You start by doing an analysis to determine which duration storm creates the highest discharge, and then you use that for your proposed model.



Hydrology, Drainage Analysis, Flood Studies, and Complex Stormwater Litigation for Atlanta and the South East -

http://www.campbellcivil.com