Here in Colorado USA there is still a massive effort going on to model roadway culvert crossing failures and detention pond breaches from the 2013 flooding for analysis and repair. Currently in May 2015 here we have already been under various Flood Warnings due to the continued rain. Yesterday there was a late Winter Storm warning for 20" of snow in the mountains. I fear we may have substantial flooding again here in the next few weeks due to the snowmelt event combined with the rain.

There are pressing needs in SWMM for:

• We need capability to model our culvert/roadway systems with their roadway overtopping flow scenarios per industry standard calculation procedures.
• We need to be able to model our detention pond emergency spillways, berm breaching, and other broad crested weir scenarios over gravel and paved surfaces.

*We are currently this week beginning to model a large pond breach failure with various players involved.

Our industry standard, is to design emergency spillways and roadway/culvert overtopping to operate within the broad crested weir flow regime, where there is essentially only a hydrostatic pressure component, for public safety reasons.

1. Please add an Irregular Weir type option. -The current irregular channel Transect editor is the exact format we need for an editor. This will allows us to model the roadway sag vertical curve conditions over the culverts.
2. Please add a toggle check box option to choose Broad Crested weir calculation methodology per industry standard HDS 5 (FHWA –NHI- 01-020 HDS No. 5) Pages 38-40, for both the new Irregular Weir option, and the currently available Trapezoidal Weir option.

We can use this new Irregular Weir option with the Broad Crested weir toggle OFF, to model compound thin plate weir configurations on our detention pond outlets using the default sharp crested weir formula.

Those page numbers were for the HDS-5 2nd Edition. The current HDS-5 3rd Edition pages are 3.15 through 3.17.

More colorful pictures in the 3rd Edition.

I believe that SWMM already has the ability to model roadway overtopping using the approach in FWHA-HDS-5. It does require more work on the user's part to set up than the built-in new feature you suggest. Here's how it would work:

1. To model a roadway cross-section divided into N segments, use N weirs in parallel, each of whose crest height equals the height of the roadway segment above the common upstream node's invert.
2. To apply the HDS-5 discharge coefficients one has to create a SWMM curve of weir setting (0 to 1) versus weir water depth that is equivalent to the discharge coefficient (0 to 3.1) versus water depth curve in Fig. 3.11 of HDS-5. The points on the SWMM curve would depend on the width of the roadway crossing and its surface material. The SWMM weir discharge coefficient would be set to the upper limit of 3.1 from the HDS-5 curve. (SWMM already accounts for submergence using the Villemonte equation.)
3. Add a Control Rule for each of the roadway weirs that looks as follows:

If Link RWj Depth > 0
Then Weir Rwj Setting = Curve HDS5_Curve

RWj is the name for each weir (e.g., RW1, RW2, etc.) and HDS5_Curve is the name of the curve developed in step 2.

Thanks for taking the time to lay this out here. I will have to try this when I get the time..

As you can see the Irregular Weir object absolutely begs to be automated in SWMM, especially when the GUI for the Transect (sta/elev) editor is already available..

This should have been done when the HDS 5 inlet control methodology was first implemented. The culvert/roadway, culvert/gravel embankment systems always go hand in hand all over our Civil projects, and these systems need to be modeled with the industry standard HDS 5 method in the USA, just like the setup in FHWA HY8. We also use HY8 across our projects, and we need SWMM to be able to correlate to that output as we then route our site flows in SWMM to our detention ponds, outlet channels, etc...

The workaround you offer is unnecessarily problematic regarding everyday implementation. We would have to setup, compute average elevations for each N segment, troubleshoot and continuously manage the multiple N segments after that. The setup would be readily susceptible to error by busts on the average elevations. This would quickly create designer fatigue...

All this could be easily eliminated by implementing the current Transect editor, where we can simply plot out the transect for quick easy inspection, and subsequently harness the computer for all the segment calcs.

Can you please elaborate a little more on the curve development in Step 2?

I have never controlled a weir before. The Action Clause says the control raises the weir crest.....as in an inflatable dam.

The weir crest elevations represent our road elevations. Will this technique still give us an accurate backwater response?

It would be nice if we could modulate the weir coefficient.

I have to take back my suggestion -- it won't work for the reason you cited (a weir control raises the crest but doesn't affect the discharge coefficient). Sorry about that. I think I was confusing weirs with orifices.

For the next update I'll try to add the weir coefficient to the variables that can be modified with controls. Then you would be able to follow the procedure I suggested for modeling roadway overtopping with HDS-5. Speaking of which, do you know if the X-axis of Figure 3.11(a) in the document is mislabeled? It says HWr, but shouldn't it be HWr/Lr?

Ok that makes physical sense to me then.

It seems like you're right about Figure A...

Now you're saying that Kt can be set to unity because of the recent Villemonte weir implementation in SWMM, correct?

Yes. Since there's no way to "turn off" the Villemonte adjustment, modifying your weir coefficient by Kt would be double counting. (BTW, the Villemonte submergence adjustment isn't new -- it's been used since at least SWMM 4.4).

Villemonte Weir Submergence was added to SWMM in 1995 as this parameter on the Weir G1 line Chuck Moore and CDM in 1995. I eventually found it in bound.for - it was hard to find even with my long years of past experience! One of the fantastic results of the SWMM 5 redevelopment project was getting clean and beautiful SWMM 5 C code instead of the messier looking Fortran code of SWMM 3/4. You can only really understand the code difference if you look at both sets of code at the same time. The CHI Visual Code viewer mentioned by Rob James is also a really neat tool for viewing the current code.

*     ISUBEQ  : Submergence equation.
*                   = 0 Submergence equation used in previous versions of
*                       SWMM.
*                   = 1 Villemonte equation. (ref.  Handbook of Hydraulics
*                       Brater and King, 1976)

Thanks Lew and Michael for your dedication in developing and maintaing the SWMM 5 code and GUI.

Maybe I was thinking there was something going on with the Villemonte implementation recently when the weir/orifice capability was going back and forth..?

Lew,

This is not going to work for a broad crested weir and the needed HDS 5 roadway overtopping modeling capability. We need to be able to toggle off Villemeonte when not using a sharp crested weir.

Villemonte was developed for sharp crested weirs and the equation does not plot well on a logarithmic scale for broad crested weirs in the research... The HDS 5 provides roadway overtopping submergence curves specific to this modeling scenario.

We are halfway there now for modeling capability for culvert/roadway overtopping in SWMM with the implementation of the HDS 5 inlet control methodology awhile back. Culvert/roadway modeling is a very common need in the civil engineering industry and we need this to have properly integrated storm drainage models.

Here are some research articles:

http://stream.fs.fed.us/fishxing/fplibrary/Villemonte_1947_Submerged_weir_discharge_studies.pdf

http://digitalcommons.usu.edu/cgi/viewcontent.cgi?article=1079&context=water_rep

http://www.utexas.edu/research/ctr/pdf_reports/0_5492_1.pdf