Large island

[Bruce Smith]

Dear HEC-RAS experts:
I am new at HEC-RAS. I am examining a section of river using an older set of HEC-2 "FEMA" data. In addition, we have measured several additional transects in between the old ones. It just so happens that the study area includes a large island. I have treated the entire river as one reach, just as the original flood study did.

I have noticed that HEC-RAS apparently calculates a single water surface elevation regardless of the "main" channel versus the "secondary" channel. The output is indicating "divided flow" in the island area. In the study area, the "main channel" (I assume you must make a stab at this if flow seems equal between channels) is a longer channel and has a much gentler slope. The secondary channel is steeper.

The initial new transects were projected across the island with an effort to be perpendicular to flow at the channel locations. However, in so doing, the approximate water surface elevations as indicated on a 2-foot-contour topographic map were ignored. My question is this, because of the large difference in slope of the two channels and their divergent trend directions, the resulting transect locations cross the channels at quite different water surface elevations at each channel. That is, one channel water surface elevation may be 2 to 4 feet different from the other channel along the same transect.

In an effort to attempt some calibration of the model, I am not getting good results. Should ALL transects be constructed such that in the case of a large island, they cross the channels at roughly the same channel or water surface elevation? I noticed that this is not the case for some of the old FEMA transects. Do I need to split the reach into two reaches, if so, why?

B. Smith

[Matthias Fritz]

Hi Bruce,

what about a sketch??

if the two channel have different length I would create different reaches, because in the xsec you define the reach lenght, that means distance to the next xsec. Doing so with only one reach you make always a mistake. For hydrauloc calcs the distance is important.

If using only long transects crossing the island, HEC will treat it as one section and calculate only one water level. Thats normal.

Using two reaches you may end up with two different independant water levels. Thats even better for calibrating

[Bruce Smith]

Thanks for your response. I'm ignorant here. If I create a "loop" reach to model both channels, I assume all cross-sections are independent of one another. That is, I will not be using cross-sections that include both channels and the island? How far should the cross-sections extend for each channel?

[Matthias Fritz]

exactly, you model each channel separetly. You should extend the xsecs up to a dry part of the island.

[Bruce Smith]

Matthias,
I appreciate your suggestions. I will forge ahead and hopefully get back to you with confirmation of good results!

-Bruce

[Matthias Fritz]

good luck,

use the manual. take a chanche on the loop example provided by hec.

it is not that difficult...n