HEC RAS split flow lateral weir with perpendicular ditch

[hmv519]

I have an interesting situation and am not sure how to approach it.

I have two parallel streams that share flow in a few areas so I am using lateral weirs to account for the flows. No problem. However in one instance I have this ditch. It is located between the two streams. It is perpendicular and connected to the left reach. Generally, it is a dry ditch however for the 100 year event the right reach shares flow at this point. Most if not all of the right reach right bank Q is going to end up in that ditch. What would be the appropriate way to model this? The next cross section downstream in the right reach is also sharing flow to the left reach. I am thinking if I have a cross section both just before and after the ditch I could run the weir up to the ditch, pull the right Q from the right reach and add it to the left reach then pick up the weir again beyond the ditch.

Any brilliant ideas out there??

 

[gbam]

Hmm interesting. Could it be modeled as a junction, a third reach or??? I presume the Rt OB and Lt OB are connected by this ditch, correct?
Have you evaluated the capacity of the ditch? Could the OB flow skip over it, thus ignore it?

 

[hmv519]

Yes, the Rt OB and Lt OB are connected. Since the Rt. OB flow would end up in the left reach I did try to ignore it. I ran the lateral weir right through the ditch but my WS elevs were not balancing. I was getting a vertical wall of water on the right end of my cross section for the right reach.

 

[gbam]

Sounds like you didnt have the iteration option triggered (I cant remember witch button or name). Are your flows changing in the downstream direction?

I have modelled flows that would exit one channel fill up the parallel and then flow back into the main channel further down stream. I had to trigger something and I cannot for the life of me remember what it was.

Can this reach be modelled as a single reach with two thalwegs? How important is this ditch between the overbanks? Some thing you should consider when modeling.

 

[hmv519]

I have the flow optimization toggled on as well as "initial split flow from output" to minimize iterations if that is what your are thinking of. Is there another iterations option that I have missed? I get convergence after a few itirations but yet a vertical wall on the right.

The ditch really isn't important at all. It is dry unless there is more than normal flow in the two streams. My next attempt is to delete the dip representing the ditch in the weir station and elevation chart to see if I can get the WS elevs to balance.

 

[gbam]

Optimization is what I brain farted on, earlier. I am surprised that the WSEs did not balance with optimization triggered. I do not recall ever getting a vert wall with the trigger on.

You may want to step back and think about the overall result and what you are trying to find in terms of the big picture. Is this a floodplain study or what?

 

[hmv519]

Yes, floodplain study and it is my first one. My colleagues took on the project and then "ditched" me. I have only worked with HEC RAS for determinations on a single home (two or three cross sections). I have painstakingly taught myself everything I know by trial and error along the way. This is really the last piece of the puzzle and boy would I love to get it out of my hair.

Thanks for your help to date.