Tuesday, March 31, 2009

Stage-discharge curve for Dam Breach modeling

Written by Chris Goodell, P.E., D. WRE | WEST Consultants
Copyright © RASModel.com. 2009. All rights reserved.

In the NWS DAMBRK model, a single stage-discharge curve could be constructed external to the program and used as a method for defining the reservoir. Though this is an easy and stable way to define a reservoir, it lacks the ability to account for submergence effects, which could significantly affect the results once the breach has fully developed. My guess is this is why HEC-RAS does not allow for a stage-discharge curve to be used to define a reservoir.

If you use cross sections to define your reservoir (full dynamic hydraulic drawdown routing), then you can use an inflow hydrograph at the upstream-most cross section. If you use a storage area to define your reservoir (level pool drawdown routing), then a simple stage-storage curve is required. A storage area can also have a lateral inflow hydrograph attached to it. You also have the option to determine your breach hydrograph external to HEC-RAS and just enter it as the upstream boundary to a cross section. That way you could avoid modeling the reservoir in HEC-RAS altogether.

A family of stage-discharge curves (rating curves) CAN be used at the inline structure to define flow through the structure prior to the breach. This can be used to describe a complicated gate strcture or spillway-one that is not available explicitly in HEC-RAS. However, the weir equation will be used to define flow through the breach-in addition to any other outflow you may have at the dam. And this is not the same as using a stage-discharge curve to define the reservoir.


  1. Hi Chris, I am simulating a simple 2D dam breach and leaning towards using a generated breach hydrograph as the upstream boundary condition, instead of using storage area for the reservoir route, because the reservoir data (Elev. v Storage curve) will not be always available to me currently.

    My question on here is, is attaching upstream BC to a cross-section the only available option even in the newly released 5.0? If so, I figured it will give me just as inaccurate/rough results as I would use the estimated Elev v. Storage curve from my terrain data, because I will have to roughly sketch the cross-section too, as this is not always available to me as well... Thanks for your help...

    1. No, you can establish an upstream boundary condition right on the perimeter of a 2D area using BC Lines.