BREACH output hydrograph

[c.campbell]

1) For an overtopping failure, is there a way to extend the falling limb of the outflow hydrograph since a normal termination of the program abruptly cuts the hydrograph off way before the dam dries out? The breach develops to the dam bottom elevation.

2) For a piping failure, can it be assumed that the when there is a weir flow transition followed by termination of the simulation that the material above slumped and created a new crest elevation above the water surface elevation?

[Chris Maeder]

The DAMBRK simulation is controlled by the length of the inflow hydrograph. If you want to continue the routing simulation, then you need to lengthen the inflow hydrograph defined at the upstream end of the model.

Sincerely,

Chris E. Maeder, M.S.
Senior Technical Engineer
chris.maeder@bossintl.com

[carlo]

Assuming that you are talking about the BREACH program and not DAMBRK...

I think that the reason that the BREACH model stops the outflow hydrograph abruptly is that the reservoir runs out of volume. You can check the volume that has been released from the dam by measuring the area under the outflow hydrograph. The documentation seems to be pretty unclear about halting the breach runs, but this seems to make sense...

I totally agree with your assumption about the material above the piping failure falling in and blocking the breach. Anyway, by plotting breach elevation vs water surface elevation you'll see this. Again, the documentation seems to be pretty poor in explaining all this.

Though when you have 8000cms flowing out through a breach, it doesnt seem intuitively right that a slump of that dam material into the breach would halt the flow...

Hope this is useful...
carlo.

[c.campbell]

Yes, I'm talking about BREACH!

As all ready stated, the model halts before the dam has a chance to dry out. To elaborate, I've done a volume check in which the area under the hydrograph equals the change in dam volume. However, I still have 9 ft of water in the dam, the breach is fully developed to the bottom of the dam, and the program terminates itself normally for whatever reason. Is there a way to overcome this premature stop to the simulation?

[carlo]

Yeah, I seem to recall something like that happening to me...models that would halt when there a large outflow (compared to the peak outflow, what is the q when the model halts?) and seemingly still head in the reservoir.

Sorry if I am suggesting things you may have already done, but try:
(1) Check the duration of the simulation...
(2) There is an error tolerance you can play with...I never played with this, so I am not sure if it will have an effect, or if it is merely for each time step...Perhaps Chris could confirm this???
(3) Depending on the shape of your storage-elevation curve and the volume in the lower part of it (ie, what volume is left in your reservoir when there is 9ft of water in it), the next time step may be enough to drain more from the dam than is available and it seems to halt with no explanation.
(4) Throw your hands in the air in exasperation and go home for the day. :)

I actually use the NWS dos version with my own front end, so I am not sure if there is anything else in the BOSS version to play with.

I am curious to know how you get on...When the first three didn't yield any clues, I generally opted for #4.

cya
carlo.

[Support Staff]

Note that BREACH has a finite number of time steps that it will model the breach. Perhaps in your simulation you have reached the maximum number?

From my recollection, I think that the maximum number of time steps is 4000.

Sincerely,

Chris E. Maeder, M.S.
Senior Technical Engineer
chris.maeder@bossintl.com