Preview of the new Finite-Volume Approach for 1D Reaches

One of the most anticipated new features soon to come in the next major version of HEC-RAS (Version 5.1) is the option of running unsteady 1D reaches with a finite volume solution scheme.  This will be a fantastic addition to HEC-RAS.  Gary Brunner recently gave me a brief overview of the new finite volume feature we can expect.  But before you ask, there is no set release date for Version 5.1 yet.  But I'm hoping we'll see it within the next year or two.

1D Finite Volume Solution Algorithm

By Gary W. Brunner, P.E., D.WRE

Senior Technical Hydraulic Engineer

Hydrologic Engineering Center

A brand new solution algorithm has been developed for 1D modeling.  A Finite-Volume solution approach, similar to what was added for 2D modeling will be available for 1D modeling in HEC-RAS version 5.1.

The current 1D Finite Difference solution scheme has the following deficiencies:

1. Cannot handle starting or going dry in a cross section
2. Low flow model stability issues with irregular cross section data
3. Extremely rapidly rising hydrographs can be difficult to get stable
4. Mixed flow regime (i.e. flow transitions) approach is approximate
5. Stream junctions do not transfer momentum

The new 1D Finite Volume approach has the following positive attributes:

1. Can start with cross sections completely dry, or they can go dry during a simulation (wetting/drying)
2. Very stable for low flow modeling
3. Can handle extremely rapidly rising hydrographs without going unstable
4. Handles subcritical to supercritical flow, and hydraulic jumps better.
5. Junction analysis is performed as a single 2D cell when connecting 1D reaches (continuity and momentum is conserved through the junction).

Additionally, the new 1D Finite Volume approached is solved in the same matrix as the 2D equations.  Solving in the same matrix allows for faster 1D/2D model solutions and more accurate flow transfers between 1D and 2D elements.  The equations are solved together and all hydraulic connections are updated together on an iteration by iteration approach, rather than separately, as in previous versions of HEC-RAS.

Using a HEC-RAS Storage Area and Lateral Structures to Replace Standard Reach Junctions

Written by Lonnie Anderson, P.E., CFM  |  Pape-Dawson Engineers, Inc.
Copyright © The RAS Solution 2018.  All rights reserved.

Situation:

Advances in HEC-RAS now allow for improved simulation of overbanks using 2D Flow Areas. Coupled 1D/2D models where 1D cross-sections represent the bank-to-bank cross-section data and 2D flow areas represent overbank areas has greatly improved the accuracy and robustness of HEC-RAS models. This is particularly the case in flat, urban areas with significant overbank flow paths.

These coupled 1D/2D models have highlighted several simplifications and shortcomings of the traditional HEC-RAS Junction methodology. Note that the standard method is still required in 1D steady flow modeling.  The following points suggest an alternative Junction method is worth considering when building a coupled 1D/2D model.

 Figure 1 - Traditional HEC-RAS Junction Methodology with a Coupled 1D/2D Unsteady Model - Simple Junction

• 1D to 2D offline flow transfer over the junction length is not possible in HEC-RAS.  In other words, a lateral structure cannot span across a junction. In complex confluences, this transfer region may be critical. Simply reducing the distance between bounding cross-sections to minimize this region may not be an option depending on the channel and bank alignments (see Figure 1). 

• The volume of water within the bounding cross-sections of junctions is not accounted for, whether forcing the water surface elevation to match the downstream bounding cross-section or using the Standard Step one dimensional Energy equation. Both solution techniques simplify the hydraulics of the region, and in particular, the main reach which conveys the greatest volume (see Figure 1).

Simplification of Junction hydraulics has been “accepted” as reasonable, as                          demonstrated by the following guidance from the USACE-HEC: 

1D? 2D? or 1D/2D? How Should I Build my Model?

Written by Christopher Goodell, P.E., D.WRE  |  WEST Consultants 
Copyright © The RAS Solution 2016.  All rights reserved. 

Now that the official release of HEC-RAS 5.0 is out with 2D capabilities, I'm getting a lot of questions about whether 1D or 2D (or 1D/2D combined) is the best way to set up a specific model.  The answer is very simple.  Like everything else...It depends!  Fortunately, there are some guidelines.  

1.  The general rule of thumb is that if the length-to-width ratio is larger than 3:1, a 1D model can possibly be used; otherwise, a 2D model is needed (source:  Desktop Review of 2D Hydraulic Modelling Packages, UK Environment Agency, 2009).  For example, if a river reach is 10,000 m long and has a 100 m wide floodplain, the ratio is 100 to 1, so a 1D model is likely okay.  For a river reach that is 10,000 m long but has a 5,000 m wide floodplain, the ratio is 2 to 1, so a 2D model will probably be needed.

2.  Features such as a narrow bridge crossing causes significant expansion/contraction are best modeled using 2D capabilities.

3.  If knowing the flood patterns around buildings and other discrete features is important, a 2D model will be necessary.

4.  Detailed animations showing floodwave progression in multiple directions at a local scale is best represented using a 2D model.  If simple water surface elevation graphics are needed, both 1D and 2D models can be used to produce these results.

When will a 1D model be suitable?

1.  Locations where flow isn’t required to ‘spread’ significantly (flow maintains primarily uni-directional flow patterns).

2.  Well-defined channel/overbank systems (channel is bounded by steep slopes, constricting the lateral expansion of flows).

3.  Simply-connected floodplains where flow in main channel is well connected to flow in the overbank and that flow in both is primarily uni-directional in nature.

4.  When elevation data of only limited quality/quantity are available.

When is a 2D model usually preferable?

1.  Anywhere flow is expected to spread

2.  Urbanized Areas

3.  Wide Floodplains

4.  Downstream of Levee Breaks

5.  Wetland Studies

6.  Lake or Estuary Studies

7.  Alluvial Fans

Other Considerations:

Like anything else, there is rarely a definitive answer to the subject question, rather a lot of gray area.  Frequently, a model could be constructed in 1D or 2D and provide excellent answers either way.  In this case, the experience of the modeler with 1D modeling or 2D modeling becomes very important.  Someone who is very skilled at setting up a 1D model to represent 1- and 2-D conditions (a quasi-2D model) may end up with a much better model than if that same person tried to build a 2D model without much experience in 2D modeling.  And vice-versa. 

There are pluses and minuses to going purely 2D.  First of all, if you can justify using Diffusion wave, a purely 2D model will most definitely be more stable and robust than a 1D or 1D/2D unsteady flow model.  You’ll be surprised how easy it is to set up and run.  Even if you do have to use Full Momentum, typically if your Courant Condition is well satisfied, the model will be more stable.  With multiple streams arranged with complicated junctions and loops, the 2D version will do a much better job – especially around junctions and flow transfers from one stream to another.  And you get to remove subjective modeling techniques like ineffective flow areas, levee markers, cross section orientation, etc.  Some downsides to a fully 2D model are: 

1.  Run times.  If your 2D area is very large and you have relatively small cells (i.e. a lot of cells), then run times can be long.  By a lot of cells, I’m talking about 100,000 to 1 million or more.  Making your model 2D in areas where you need detail and 1D everywhere else can help solve this problem. 

2.  Output.  Getting output from 2D areas is a bit more cumbersome and limited. Still, you can get quite a bit of stuff out of your 2D areas, it just might take more time. 

3.  In version 5.0, there is no direct way to model pressure flow at bridges in a 2D area. Hopefully this will change for the next version.

4.  Learning curve.  Being new to 2D modeling, there will be some overhead just learning how to do it. 

5.  Your client may not be okay with it.  Make sure your client is aware of the benefits of 2D modeling.  There is generally a perception that 2D modeling is more expensive.  This is not (should not) always be the case. 

And remember…

Make everything as simple as possible, but not simpler.

            -Albert Einstein (paraphrased)

For every complex problem there is an answer that is clear, simple, and wrong.

            -H.L. Mencken

For more information, make sure to give Chapter 6 of the new HEC-RAS 2D ModelingManual.