Hairpin/Horseshoe Bend in HEC-RAS

[OHbridgeguy]

All,

I have a unique situation that I am trying to model. Looking at a culvert replacement that has a sharp bend downstream with most of the inner point bar region in the floodplain. This results in flows going over the floodplain during large storms as well as in the meander bend. In modeling this the flow from each side of the stream will contribute to flooding and inundation in the floodplain and create something like a reservoir effect. My question is can HEC-RAS handle this properly or is a 2 dimensional model (say in FLO-2D) required? Perhaps an in-line weir approach? If so how does the additive flow from the same stream work in HEC-RAS?

I've attached an aerial image to help clarify the issue.

Thanks

 

[SMIAH]

Short answer : 2D modeling.

Good luck with FlO-2D.

With we could do more stream relocalisation... or Expropriation.

 

[beej67]

I don't see why HEC-RAS couldn't handle this case as long as you cut your sections properly, and tightly enough together. I'd only worry about FLO2D if there was a clearly defined overflow channel that's only activated during the flood, and conveys water along a shorter path than the main channel.

Sucks to be that guy who's house is out on the point bar, after you guys remove that culvert and blow the flow through your new bridge. :/

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

http://www.campbellcivil.com

 

[Rye1]

I agree with beej67. You should be fine with RAS but cross-section placement will be critical. If the house is in the FEMA floodplain, you cannot upsize the culvert and increase the flooding on the house.



Robert Billings

http://www.newrivereng.com

 

[OHbridgeguy]

I agree with the flooding comments, and as part of the scope the replacement is required to maintain or lower the flood elevations. It already floods at a 10 year storm which is really his own fault for building in the flood plain. It did give me some fairly exact storm/water elevation measurements to calibrate the model flows though, which was nice.

I have been successful in getting this into HEC-RAS, and used ineffective flow limits to roughly approximate the eddy pools. I may do a quick sensitivity analysis with the ineffective flow settings to bound my results. Thanks for the comments all

 

[SMIAH]

Why use HEC RAS if FLO2D can handle it?

Sure you'll need more data but it might be easy to model a short river bend like this one.

As long as it's not an overkill for the project (available money, data).

 

[SMIAH]

Don't say why use Flo2D is HEC RAS can handle it

 

[OHbridgeguy]

I've never used FLO-2D but have seen a few example setups, so prefer HEC-RAS from a familiarity aspect. Also, the state DOT wants files in HEC-RAS format as a deliverable for their use.

 

[beej67]

My opinion of FLO2D, after seeing several very impressive demos and sample projects, is "Gee, that looks really cool, I wonder who will pay me to use it."

So far nobody.

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

http://www.campbellcivil.com

 

[SMIAH]

Did one floodplain study with it... I had extensive data though (LIDAR + bathymetric survey, Unsteady flow with recorded data 2-3 km downstream and upstream).
Grid cells 15 m x 15 m and the program was running for 2 days though.
It's a good alternative to MIKE2D IMO.

Who will pay you for it if you don't need it