Lateral Structure flows into Multiple Cross Sections

Written by Chris Goodell | January 26, 2010

Written by Chris Goodell, P.E., D. WRE | WEST Consultants
Copyright © 2010. All rights reserved.

Not sure if this will be changed for the soon-to-be-released version 4.1, but for now, the option to drop flow from a lateral structure into multiple cross sections is a little awkward.

First thing you do is specify that you want to drop your lateral flows into cross sections. Then set your upstream TW river station (the downstream one will be greyed out).


Then, go to the Weir/Embankment editor for your lateral structure. At the bottom left, you’ll see weir stationing reference. Select TW flow goes: “over multiple XS’s”. Then on the right hand side you’ll see a table for specifying weir stationing for your TW cross sections.



  1. SMIAH

    on February 1, 2010

    Storage area connected to a XS, an inline structure at least 2 cross sections away from the upstream end and then the boundary condition defined as a Lateral Inflow.

    Is this a way to simulate a dam with a reservoir upstream? Instead of simulating the reservoir with cross sections (dynamic routing).

  2. Chris G.

    on February 1, 2010

    YES! That's the way. The trick is making sure RAS recognizes the storage area as "connected" to the XS. You do that by turning on the "move object" mode in the geometry, and then clicking and dragging the end of the reach to a point inside of the storage area.

  3. SMIAH

    on February 1, 2010

    So there is nothing wrong (hydraulically speaking) to put the boundary condition as a lateral inflow at the XS located after this connection (SA to XS)?

  4. Chris G.

    on February 2, 2010

    Technically nothing wrong with that, although if you want to route the flow through the reservoir, to see the reservoir's affect on attenuation of the flow, then it would be better to add the lateral inflow to the Storage Area itself, rather than the cross section. That's how I do it. Just make sure your pilot flow at the dam is consistent with the base flow in your lateral inflow hydrograph. And also make sure you set an internal RS initial stage (unsteady flow data editor, under Options…) at the cross section just upstream of the dam that is consistent with the initial conditions storage area elevation.

  5. Jon Parker

    on April 3, 2014

    Can this method of lateral structure flows into multiple cross sections be used to put a storage area behind the lateral weir in order to do a level pool routing breach lateral to the reach? Didn't find yet where or whether a storage area can be connected to a lateral weir to input a flows laterally into a reach, not the opposite way of modeling flow out of a reach over a lateral structure and into a storage area. Thanks

  6. Chris G.

    on April 4, 2014

    Jon-Great question! And yes, you can do it. Set up the lateral structure in the reach and connect it to the storage area. When the storage area exceeds the threshold breach elevation, the lateral structure will breach. RAS monitors both sides of the lateral structure for a water surface that exceeds the breach threshold. Be aware, that discharges through the breach will be listed with negative values since flow is going from the SA to the reach. I put an example RAS project that demonstrates this on my Dropbox site. Feel free to download it.

  7. jp

    on April 13, 2014

    Thank you for the response Chris. Got everything running great thanks to your example.

  8. Chris G.

    on April 14, 2014

    You're welcome.

  9. MTR

    on November 7, 2014

    Hello Chris, do you mind putting the breachSAto reach zipfile back into dropbox? I am working on a similar project. Thank you!

  10. Chris Goodell

    on November 7, 2014

    Sorry about that. Everything got moved into a Google Drive Space (more free GBs!). Here is the new link:

  11. Chris Goodell

    on November 7, 2014

    The referenced HEC-RAS data set is now located at:

  12. SamihaT

    on July 10, 2015

    I am modeling a river where there is a power canal separated by a lateral structure in HEC-RAS.
    1. I have laid out the canal as a river tributary. Is this a correct approach? Is there any other approach to better represent the system?
    2. Modeling as a tributary creates a junction where the river and the canal meets. Now, I am not being able to draw a lateral weir across this junction. Whenever I put the Head water (HW) River station, no reach thalweg appears as supposed to. Can I go modeling without this structure?

  13. Chris Goodell

    on July 11, 2015

    1. Sounds like a good approach to me.
    2. You can leave the junction there if it makes sense physically. It sounds like you are hesitant about connecting with a junction. You could have 2 separate and parallel reaches, connected in multiple locations with lateral structures. Each reach would have to have its own upstream and downstream boundaries, but that's easy to do.

  14. SamihaT

    on July 11, 2015

    Thanks Chris. Just another related question, can I somehow merge two river reaches into one reach (by eliminating the junction) and somehow save all the cross section data? I can not imagine drawing the layout again!!

  15. Chris Goodell

    on July 11, 2015

    Yes, just left click on the junction and choose "Delete Junction".

  16. Anonymous

    on September 9, 2015

    Good Afternoon Chris, I have a question regarding the combined use of inline and lateral structures. I am working on a model where the reach "Zs" across a roadway. That is, the reach runs parallel to the roadway upstream of the inline structure, crosses the roadway at a skew through the inline structure, and then resumes parallel to the roadway downstream of the inline structure. Assume the roadway grade is flat. What is the best way to model this – a lateral structure upstream of the inline structure, ending just upstream of the inline structure with flow connected to downstream cross sections? How would cross sections be drawn for this – cross sections ending at the CL of roadway (assume to by high point)? If the roadway is flat do I just propagate the inline structure roadway profile to the bounding lateral structure cross sections (thus capturing all over-the-road flow by either lateral structure weir flow or inline roadway weir flow)? I realize this will likely result in the inline structure cross sections being bound by frictionless walls at the edges of the cross section. Please advise if a more detailed description/sketch would be helpful. Thank you!

  17. Chris Goodell

    on September 9, 2015

    I think you are on the right track. If it's strictly a low flow model, no overtopping of the roadway, then just draw your cross sections perpendicular to the flow lines, terminating at the CL of the roadway. You may have to dogleg them to make the turn through the bridge opening.

    If you expect overtopping, and that is what you are most interested, then yes, I think a lateral structure would work well. But there are any number of ways to set this up, depending on the terrain around the bridge, and where you expect the water to go during high flow. For example, will some of it continue to flow parallel along the roadway, while some overtops? Will it all go over upstream of the opening? If you want to link me to a sketch, or even an aerial photo, I'd be happy to give you my recommendation for cross section/structure layout.


  18. Anonymous

    on September 9, 2015

    Thanks for the timely response Chris!

    That would be fantastic if you could look at an aerial, I would greatly appreciate any recommendations you might have. Here is a public link to a Google Drive stored pdf:

    I appreciate your time.

  19. Chris Goodell

    on September 10, 2015

    Here's my suggestion. Keep in mind this is from me sitting at my desk, never having been to the site. Hope this helps.

  20. Darrell

    on November 16, 2016


    I am working on a project where there are two parallel stream channel that have the potentional to influence one another when larger storms occur. I have two questions:

    1. I setup geometry to use lateral structures across multiple cross-sections to convey the flow from one reach to the other. Does a lateral structures need to be input to allow flow conveyance in either direction or can one be used.

    2. When running steady flow analysis the following statement results…"In reach 2 the lateral weir extends past the tailwater reach." Why does this occur and how can this be resolved.



  21. Chris Goodell

    on November 17, 2016

    1. You can have just one lateral structure between the streams. Flow can go both directions over a lateral structure. If it's a long reach, you may want to break it up into multiple lateral structures though.
    2. Sounds like your tailwater connection is not consistent with the reach on the tailwater side of the lateral structure. You have to go into the tailwater connection table and make sure that the tailwater reach's reach lengths are consistent with the stationing of the lateral structure weir. It's very confusing, so it's helpful to sketch out your reaches, cross sections, and weir stationing to find the problem.

  22. Unknown

    on March 30, 2017

    I am having trouble with the same issue–specifically what you discuss in your second point above. What do you mean by consistent lengths between tailwater connection and the weir length? Do they need to be the same length? I have been working on this for several hours now and cannot seem to get this error message to go away…

    any help would be appreciated.

  23. Scott Shipley

    on March 30, 2017

    I am having the same problem with lateral structures being too long. Can you describe what the relationship needs to be between weir length and tailwater connection? Thanks!


  24. Chris Goodell

    on April 3, 2017

    Each tailwater river station that receives flow needs to be tied to a stationing on the lateral structure. That's what the TW Connection table does. If the stationing on the lateral structure is longer than what you allow in the TW Connection table, the weir can extend beyond the end of the tailwater reach. That's the genesis of that error message. As I said, it can be very confusing, especially the first time you do this. I've found that sketching out the two reaches and the lateral structure and where it is all connected can really help to understand how to set this up correctly.

  25. Anonymous

    on November 29, 2017

    Hello Chris, I have two roadside ditches on either side of a road. I am using a lateral structure along the centerline of the road to simulate flow from one ditch overtopping the road and flowing into the other ditch on the other side of the road. I don't quite understand what the "upstream TW river station" is referring to. Is it the most upstream cross section in the receiving ditch that the flow will spill into?

  26. arkads1

    on December 12, 2017

    I have model with fish passage where water should flows into the fish passage directly by lateral structure. The lateral structure is connected with main river (headwater side) and with the fish passage (tailwater side) as next to left bank station (unfortunately HEC-RAS can't connect lateral structure tailwater side directly to the last XS of the river reach). Model works properly but I have no flow in fish passage beside initial flow 0.01 m3/s although detailed results table shows gate flow in lateral structure as 0.77 m3/s. Should I use the junction rather than the lateral structure in this case? Why I have no flow in fish passage?

    The model is here

    I'll be gratefull for your help.

  27. anouk bomers

    on January 10, 2018

    Dear Chris, I have a question regarding breaching a lateral structure. Is it correct that only one combination of water level + duration can be set which determines the moment of breaching? Is it perhaps also possible to give multiple combinations of water levels and overtopping durations (e.g. with the use of scripts outside the Hec-ras interface?). Thank you in advance. Regards, Anouk

  28. Chris Goodell

    on January 10, 2018

    Interesting question! You are correct, RAS only allows one combination of water level and time duration. You could probably accomplish this by trying each water level + duration combination in a different plan. See which one breaches first, then just use that plan. The only way I can think of to do this with external code is to wrote a controlling routine that will run RAS for short durations (e.g. 1 minute). You would write a restart file after each short duration so you can start up a new simulation where you left off previously. In between simulations, check the water level and keep track of it. Then in your code, you can write some conditional statements to check the different combinations of water level and duration. Once one of them is exceeded, then have RAS set to breach. Sounds complicated though. I prefer my first suggestion.

  29. anouk bomers

    on February 21, 2018

    Dear Chris, thank you for your reply. I also have another question related to breaching a lateral structure. For my study I would like to breach a lateral structure at multiple locations and identify resulted flow patterns. However, is it correct that it is not possible to breach a single lateral structure at multiple locations? Do I have to split the structure into multiple lateral structures and then add the breaching criteria? Thank you in advance. Regards, Anouk

  30. Chris Goodell

    on February 21, 2018

    Correct. One breach per lateral structure. You can try breaking it into multiple lateral structures or set up multiple plans, each with a different location.

  31. Rebekah Perkins

    on August 20, 2018


    I have a situation where there is a watershed overtopping point lateral to a bridge and I want to simulate that the water will overtop that point prior to overtopping my bridge but I am unable to put a lateral structure using the US bridge face as the US XS for the lateral structure. Do you have any suggestions?

  32. Christina Hughes

    on July 19, 2019

    Hey Chris,

    What is the purpose of specifying the TW Position location (LOB, ROB, or next to L/R channel bank)? If you are already specifying the weir geometry, what does changing the TW Position location affect?


  33. Chris G.

    on July 30, 2019

    If you're delivering the flow to multiple XSs then the TW connections are a function of reach lengths. And reach lengths can be different between the overbanks and the main channel.

  34. IP

    on November 19, 2020

    Hi Chris,

    I need to model a lateral structure in the confluence of the mainstream and tributary. But I was not able to create the one just above the confluence, can you please give me some idea on creating lateral structure at that point?

  35. Chris Goodell

    on November 20, 2020

    I believe you need at least one cross section between the lateral structure and the confluence.

  36. Mo

    on July 15, 2022

    Hi Chris,
    I have a problem with my lateral structure. I have two reaches that run parallel and connected by a lateral structure, and there’s no error with structure and the model runs perfectly. However, after the model runs, there’s no flow contribution from lateral structure to the second reach. I should mention that flow is going over the weir and the water surface elevation is also increasing in second reach but no flow contribution
    Thanks in advance for responding to this comment

  37. Chris Goodell

    on August 12, 2022

    If you are running this in steady flow, make sure to turn on the split flow optimization for that lateral structure. If it’s an unsteady flow model, that would be strange. Not sure what would cause that.

  38. Collin

    on May 5, 2023


    I have selected to do a 1D steady flow model for a situation where I have two streams converging to a junction, but there is a flat floodplain area between the two streams where in high flow situations, becomes one continuous floodplain.

    I’ve modeled the streams as two reaches. The right extent of cross sections in reach 1 abut exactly the left extent of cross sections in reach 2 at a defined ridge line inside the flat floodplain area. I have modeled multiple lateral structures between individual cross sections and applied a weir coefficient of 0.1. I have selected headwater connections as cross sections along reach 2 (higher flow), and tailwater connections along reach 1 (lower flow).
    I have selected to have each lateral structure optimized and increased the maximum number of iterations to 40.

    The model has no trouble running and there are no warnings about not obtaining a solution within the maximum number of iterations. When I go to analyze the detailed output results table for the lateral structures the only warning is that “the cross-section end points had to be extended vertically for the computed water surface”. I’m able to see weir flow leaving and the math checks out for the flow at the next downstream cross section of reach 2, but the “flow leaving” is not being added to the flow in reach 1. I would think that if 100 cfs “leaves” reach 2, then that same 100 cfs should enter into reach 1 at the equal corresponding cross section.

    Additionally, the water surface elevations throughout the area with the lateral structures differ by up to 0.6′.

    The fact that I can’t trace the lost weir flow to the other reach, the warning of end points being extended vertically for the computed water surface and the water surface elevations not matching is giving me caution to the validity of the model.

    Is there anything you can recommend?


  39. Chris Goodell

    on May 18, 2023

    Yes, I would not try this in steady flow. Use unsteady flow instead. You can still run a steady state solution (constant inflow discharges, long enough simulation to reach an equilibrium condition), just use the unsteady flow editor and unsteady flow analysis windows. Then you don’t have to worry about optimizing the lateral structures and you’ll be able to track how flow moves across the lateral structures. OR…turn the junction area into one big 2D area and then you don’t have to use lateral structures at all. That’s probably the best option.

Add Your Comment

Leave a Reply

nineteen − five =

Related Posts