Using the “Wormhole Method” to introduce internal boundary conditions into HEC-RAS 2D models
Below is an example that shows how to use a wormhole culvert (essentially a SA/2D Area Connection with upstream and downstream culvert centerline stations located some distance apart) to introduce an internal boundary condition. In the schematic below, I’ve assumed the tributary on the left is outside of my area of interest and that I want to bring some flow into the tributary on the right without moving the external boundary around that tributary. I first use the Pac-Man method (slicing the external boundary) to add a boundary condition for inflows into the left tributary. (This is just to show what I mean by the Pac-Man method; I could just as well choose any area along the existing external boundary that doesn’t affect the hydraulic conditions in my area of interest.)
I then add an SA/2D Area Connection that crosses each of the tributaries as shown here:
I then edit the SA/2D Area Connection and click on the Weir/Embankment editor and copy/paste the terrain profile data into the weir/embankment profile. I add a culvert with the upstream station in the left channel and the downstream station in the right channel. Because there is no other flow crossing the connection line, I could just as well raise the elevation of the entire weir embankment, but in this example I have only raised the weir elevations within the left tributary to function as a dam and block the flow (raising the head sufficiently to push water through the culvert).
When I run the model, flow that is stored in my artificial reservoir in the left tributary is then introduced into the tributary on the right. Essentially we’ve added an internal boundary condition (similar to SA boundary conditions in TUFLOW). As with the dam outlet works example, it requires a few iterations to match the desired hydrograph (which would have been provided from HMS, RORB, or other rainfall-runoff model) but it gets the job done in the end. Here’s what the results look like in this example:
As shown above, where the connection alignment crosses the tributary on the right, the weir embankment elevation is set at the existing ground elevation. Because the tributary is relatively flat along its longitudinal slope, flow ends up going both directions (upstream and downstream). If I change the connection to act as a dam or levee by raising the embankment to a constant elevation of 316 metres along its entire alignment, the weir prevents backflow and flow only travels downstream as shown in the results below:
It is definitely a bit cumbersome, and hopefully this workaround won’t be required in future versions of HEC-RAS, but it seems to do the trick for now! Click here to go back to the introduction of the Wormhole Method or feel free to contribute to the discussion on the blog here.
May 2017 Update
- Introducing wormhole culverts
- Putting wormhole culverts to the test
- Wormhole culvert sensitivity analysis
- What’s the best shape for a wormhole culvert?