Accounting for Detention in Hydrology Modeling 1

[rlondeen]

What are the different approaches to including smaller non-regional detention ponds in your hydrology models?

Typically, if I am doing pre and post analysis, I take the more conservative approach of assuming a pre-developed CN (for NRCS) or pre-developed impervious cover (Clark) when calculating runoff for developments utilizing detention ponds. However, this doesn't accurately reflect the timing of the hydrograph and the increased runoff volume.

When doing flood plain mapping, I assume post-developed CN and impervious cover. While conservative, it is not always accurate.

 

[psmart]

Perhaps I'm missing something in your question, but when modeling a detention pond in a proposed condition you need to use the proposed ground conditions (i.e cover, CN, Tc) and not the pre-existing conditions.

Peter Smart
HydroCAD Software

http://www.hydrocad.net

 

[rlondeen]

I am talking about modeling a large watershed that contains existing developments with detention ponds. Since I do not have the information for each those nor the time to model them, what would be the best way to include that in the watershed characteristics.

 

[SACPEengineer]

It seems you are asking for a modified version of the USGS regression equations that take into account temporary detention storage?

 

[beej67]

I am talking about modeling a large watershed that contains existing developments with detention ponds. Since I do not have the information for each those nor the time to model them, what would be the best way to include that in the watershed characteristics.

It's a tough question, because there's no good way to do it accurately. If you presume the detained sites are undeveloped for your analysis, then you might appropriately characterize their peak flows but would underrepresent their runoff volumes. This would be dangerous, as the volumes all add up. If you take their watersheds as developed and ignore detention, then you get the correct volume but a much higher peak than you would see with detention appropriately accounted for.

In the end I think it depends on your application. If you're sizing a culvert under a road, in my opinion you should take the most conservative case and presume your watershed is developed and undetained. If you're doing a flood study for a very large regional analysis, then chances are pretty good that the USGS regression equations have already taken the effects of local or regional detention regulations into account, because the regression curves are calibrated against historical data.

If you're sizing a regional BMP on-line in a creek with other detention ponds draining into it, then you may have a little more freedom to approximate something within your methodology, depending on what the reviewer will accept. One thing you might consider, is to treat the watershed as a developed condition, undetained, but then tweak the initial abstraction values to account for a certain amount of detention scraping volume out of the runoff hydrograph. It wouldn't be exactly correct, but it could give you a ballpark, and would be within your available toolkit. Obviously document your assumptions and be ready to defend them.


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

http://www.campbellcivil.com