HEC-RAS Flood Study

[Lexwater]

Hello,
I am fairly new to this analysis, but in one of my models I am getting a couple floods crossing one another.

For instance. Upstream of one culvert it is telling me that my 50 - year and 100 - year storm events have a higher elevation than my 500 -year.

I know these models are extremely complicated, but has anyone every ran into this situation where floods seemed to be jumping one another?

Thanks

 

[bltseattle]

One idea -
It's possible that the 50- and 100-year results come from subcritical flow conditions but the 500-year is likely to be critical or supercritical conditions. Sometimes HEC-RAS assumes critical depth when the energy balance cannot be achieved from section to section, meaning that the change in cross-section geometry is too abrupt for the model to handle.

Check your expansion/contraction upstream of the culvert and make sure you have enough incremental x-sections such that the model doesn't make assumptions on it's own, e.g. critical depth. If you are getting warnings for the culvert inlet and upstream section, you may need to insert sections.

 

[LeoSC]

I agree with bltseattle. Most of the time the problem is due to the program not being able to converge to a solution on the section downstream of the crossing.
In addition to more sections, you should check the elevations you input for the ineffective flow areas on each side of the culvert, upstream and downstream. It could be that the failure to converge is the result of one trial energy level being high enough for the flow to use the entire section and the next trial being low enough for the flow to be fully inside the channel. This back and forth is enough to prevent convergence.
And if after you try these suggestions you can't make the program work, you may have to use the good old calculator and input the upstream water levels, or use a culvert program.

 

[Bayou]

Usually when a less frequent event water surface elevation is calculated below more frequent event wsel, it is a matter of ineffective flow areas. The program probably calculated that the flow of the 500-yr overtopped the roadway while the 50 & 100-year did not (based on the inputted information). Check the EG of the profiles and look at the velocities. Remember that HEC-RAS calculates WSEL by first calculating the EG then subtracting the velocity gradient (V^2/2g). Therefore the presented WSEL sometimes does not correctly represent the 'true' wsel of the upstream side of a bridge.
You might have to change the ineffective flow area elevations or the Manning's values within the overbank area to get better answers.