Aug 16 2011

The Water Goes One-Way

I have talked about one of the biggest assumptions you have to make when setting up a HEC-RAS model, and how it is difficult to accurately assume cross section locations in an area of complex flow.  Another thing to be aware of when using a 1-D model is how it calculates flow at each cross section, and a 1-D models’ inability to handle transverse flow.


From the perspective of a 1-D model, the direction of flow is exactly as you tell it, perpendicular to the cross section.  When a model like HEC-RAS calculates the hydraulic parameters at a cross section, the available conveyance at that section is the main source of how things like water surface and velocity get determined.  Obviously downstream controls, and upstream diversions, splits flows, structures, etc. will have some impact, but the conveyance at that cross section is key.

In a primarily 1-D environment, this is ok, but in an area of complex flow, the lateral movement of water between cross sections and between the channel and floodplain are not 1-D in reality and this lateral movement can have a big impact on the results that a model will give you.

Let’s use the Lynches River model I have been talking about to show what I mean.  We ran a model condition that represented the same section of river without the RR or Highway 1 embankments.  I will use that model and results to demonstrate the importance of properly accounting for the transverse flow in a floodplain.  Below is the elevation contours of the 2-D mesh representing this “natural” condition.


Let’s say, for examples sake, that we are modeling this reach of river with HEC-RAS and two of our cross section are laid out as shown below.


Now, before I point out what a model like HEC-RAS can not account for, let’s look at the velocity contours and vectors for this reach and look at what happens at the sharp 180° bend in the meander.


In the upper section, most of the flow is being conveyed within the main channel.  This is also the case just upstream of the lower section.  As the 2-D model shows, all of this water exits the main channel and enters the floodplain.   So in reality at the lower section, most of the flow is now being conveyed in the river right floodplain.

HEC-RAS would calculate the hydraulic parameters in the lower section based on conveyance only, and cannot even take into account that most of the flow can’t even physically get to that part of the channel.  HEC-RAS (being a 1-D model) is not concerned with the more complex movement of water from one side of a floodplain to another.  It is just calculating conveyance.

I hope I do not sound like I am bashing HEC-RAS!  I am not anti 1-D models!  I just want to make sure that they are used where appropriate, and that the correct tool be used when a reach of river to be modeled is more complex.

1 ping

  1. » The Water Goes That Way Hydraulically Inclined

    […] Part 3 of 4 […]

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>