Flow Change Locations and Incremental Analysis: Multiple Dam(s) break

[tpcolson]

All, I'm a GIS-guy supporting some H&H folks with HEC-GEORas for several large innundation projects, and I have a question pertaining to my assumption that I "junction" a river so that different flow volumes can be assigned along the river reach(es)...e.g. incremental analysis:

I assume that performing an incremental analysis requires setting up the RAS river geometry in such a manner that allows the specification of different flow rates (cfs) and boundary conditions along Big Creek to account for the influence of “stand-alone” contributing areas. The definition of stand-alone implies that each sub-drainage area that we had specified will be analyzed in HEC-1 to determine the additional volume of water that it contributes to Big Creek during each of the three events we have to model for. E.g, the volume of water coming out of Reevoir 1 is 1000 cfs during a 100 yr event, at the point in Big Creek where Small Branch confluences with Big Creek, we’re going to add another 1000 CFS.

In order to “tell” HEC-RAS that we’re changing the hydrograph at some random point along the 20-mile Big Creek study area, I assume that “junctions” needed to be added to the river in the form of small, several hundred foot long side tributaries (at each location we identified on the topo map). Standard RAS geometry was generated for each of these (x-section, banks, flowlines) so that they would import into RAS with no geometry errors. The idea is to use these “reaches” in order to assign the incremental-additional flows to them as to add to the initial volume of water coming out of each resevoir.

The RAS geometry file now contains a complex, dendritic river reach file. In addition to Big Creek now having junctions, which break the river up into several different reaches to which different flows can be assigned, there are now additional side tributaries to which flow must also be assigned. E.g, if HEC-1 determines that the watershed labeled “Small Branch” generates 1000 cfs, we have to a) assign 1000 cfs to reach labeled “Big Creek Small Branch” IN ADDITION to the 1000 cfs that HEC-1 determines is coming from the reevoir, for a total of 2000 cfs for this reach. Furthermore, as we have a river junction at the bottom of “Big Creek Small Branch”, labeled “Small Branch Reach 1”, we need to add a flow to this river reach as 1000 cfs. So now our flow is 3000 cfs.

Now, bear in mind..I'm a GIS guy. Not a H&H guy. So my knowledge of such matters is quite limited. But I can wrap my head around what HEC-GEORas is outputting, and I took it upon myself to read the RAS manual to make sure the data I'm generating meets the assumptions put forth in the manual so the H&H guys can do their job.

So I created a bogus (assigned 1000 CFS to everything) plan, and ran the model (actually, hit the compute button, read the pages of errors caused by my GIS errors, and fixed those 'till I finally got something to run in terms of being able to read the geometry and produce a profile). At least it works in steady...

Now the manual states (on page 4-10) that all you need to do to change flow along a river/reach is specify a station and hit the "Add A Flow Change Location" button, and I can add my incremental flows in this manner...assuming that I had characterized the contributing sub-watersheds appropriately to determine the additional flow they are dumping into the river (Arc Hydro, HEC-GeoHMS, HEC-1).


So my question is...am I introducing complexity and error with this "Use side tributaries in order to trick RAS into incremental analysis mode", and I should be using the "Add A Flow Change Location" button to accomplish the same thing? I'm only interested in the WS profile created by the dam(s) breaking, not the contribution of the dozens of downstream tributaries.

 

[RWF7437]

Yes

 

[SMIAH]

hmm!

I would probably let the H&H guys set up those flow change location when the model is ready ("geometric model").

You say that you're only interested by the dam break flow profile. It might be useful in a dam break analysis to compare the inundation area created by the dambreak to the X-year flood inundation area.

Do you have to specify a flow change in a junction - is it necessary in HEC RAS?
Because you have to do it with the other option (add a flow change location one) so if the H&H guys don't think it's necessary to add a flow change to a specific junction then they will be able to do so.

 

[cvg]

I agree with SMIAH, the engineer should be specifying all this to you. Regardless of if you are comparing to existing 100-year floodplains or not, extending the model some nominal distance along each of the tributaries may not be necessary. However, it might be desired to study the effect of the Big Creek flood on the backwater going up the tributaries of Small Branch. If so, then your H&H guy should tell you how far to go up the Small Branch channel and any other channels.

 

[froude]

It seems that this is a question of hydrology versus hydraulics. From what I understand, you are concerned about the hydrologic contribution to the main channel, from the tributary, but that the hydraulics along the tributary are not really in your scope. If this is true, then amending your HEC-RAS flow file to add flow at a cross-section near the confluence of the main channel and tributary may be the way to go.

I think the location of these flow changes can be a critical factor in the resulting flood elevations and should be evaluated carefully; so if you aren't comfortable with that, definitely get a second opinion.