Dynamic Hydraulic Modeling - Green Roofs

[MLasell]

I am new to this forum but have used it many times to find helpful informaiton. I could not find another posts where others have modeled using a a dynamix model like ASSA or XPSWMMM. I am in NY and have worked with the regulators on the best way to treat a green roof or other runoff reducing practice such as a dry swale in a dynamic model. I would like to account for the peak flow rate reduction and also the runoff reduction. If I treat it as a storage node as the regulator sugests, I do not think this helps mimic how water is going into the extensive system and what is being absorbed my the soil. Once saturated or "full" then what is the outlet structure for the green roof in larger storms. I would love to hear from anyone that has modeled a practice similiar to this in ASSA or similiar programs. ASSA speaks about how it allows for modeling of a green roof but I cant find how to execute this. Thank you for the help.

 

[beej67]

No experience with ASSA, extensive experience with both XP-SWMM and green roof BMPs.

Ordinarily, I have simply assigned a curve number to the pervious portions of the green roof, and included that in my weighted CN calculation on the basis that the roof will mimic a ground level hardscape with landscape islands. It's a simple approach, and regulators usually buy it. If you do this, do *not* also account for storage in the node, because if so you'll be double dipping.

I have only accounted for the rooftop storage in a storage node once. It was a significant amount of storage on a grass-topped parking deck with a chamber system in it and a drawdown orifice. I modeled this just as I would a concrete vault with the appropriate outlet control structure geometry.

I have modeled storage within rain gardens on occasion, and I did that by using known porosity of the backfill material and reducing the subsurface volume accordingly, including full volume for detention above the surface.

It's hard to know the best approach without seeing some design plans.

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

http://www.campbellcivil.com

 

[MLasell]

Thank you for reply, I know speaking with the top reviewer in Albany here he gave that option but then said he would rather have us not use that after discussing it with his co-workers. He wanted us to use the storage node approach. Just seems your not accounting for everything.

 

[MLasell]

The project consists of two 300' diameter storage tanks and they are considering put a green roof on top of them. AN extensive system with a small green roof section. The water will be allowed to pond a few inches and then enter the 2-4" of media, a drainage layer exists at he bottom that goes of the roof as well as anything that ponds above the 2" of ponding. It will have to be designed as a big wier draining the storage node. I can assume the start of the storm the first 2" of rain will be contained on the roof then anything more then that will start using the outlet structure of the long weir. Just a bunch of assumptions are going on, and what do you account for whats getting into the drainage layer and leaving the roof.

 

[beej67]

With geometry that complex, I think I'd model it as a two stage pond. Model your green roof as a pond with volume equal to (%porosity)*(media volume), with its underdrain as a primary control and the weir as it's secondary control, both of which dump into the tanks. Then model the tanks as tanks, with whatever control they have. Pay attention to tailwater. Call the rooftop impervious with a minimum Tc, since whatever hits the roof does in fact drain to the greenroof storage.

In actual practice, some of the stormwater is going to cling to the rain garden media, but I'd neglect that to be conservative.

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

http://www.campbellcivil.com

 

[LincolnPE]

Yes, with a Storage Node in SWMM you can model exfiltration. They call it "Infiltration" now, but are changing the name of that property to Exfiltration in the next engine release coming up.