If the modeler uses the "User Entered Data" option for the breaching method, it is up to the user to enter the key parameters that must be estimated for the breach dimensions, such as the breach formation time and the maximum size of the breach opening. Several researchers have developed regression equations to estimate breach sizes and times from historical dam breach information. Additionally, a few researchers have tried to develop computer models to simulate the physical breach process. The bulk of the research in this area has been summarized in a 1998 publication entitled "Prediction of Embankment Dam Breach Parameters", by Tony L. Wahl of the U.S. Bureau of Reclamation. Wahl documents the data from most of the historical dam breaches that have occurred in the world, as well as describing the equations and modeling approaches developed for predicting the dam breach parameters.
For the HEC-RAS software, the modeler must estimate the maximum breach dimensions and breach formation time either outside of the HEC-RAS program, or with the Breach "Parameter Calculator" provided on the Breach Data editor. Because the breaching process is complex, it is suggested that the modeler try to come up with several estimates of the breach parameters, and then put together a matrix of potential breach sizes and times. One example would be to use two different sets of regression equations and one of the breach simulation models to estimate the breach parameters. In several studies performed at HEC we have used both the Froelich (1995), MacDonald\Langridge-Monopolis (1984), and the Van Thun and Gillete (1990) regression equations, as well as the BREACH model by Dr. Danny Fread (Fread, 1988). All four methods give different answers for the breach dimensions, as well as the time for the breach to form. In general, a range of breach parameter estimates should be run as separate trials within HEC-RAS in order to test the sensitivity of the results to the breach dimensions and times. It is always good to test the sensitivity of the breaching parameters, since they are the most unknown factor in this process.

Breach Parameter Calculator

To assist users in estimating the Breach dimensions and development time, HEC has added a "Parameter Calculator" to the Breach Data editor. To use this calculator select the Parameter Calculator Tab from the breach editor, and the editor will look like the following screen:

Figure 14 11. Breach Parameter Calculator from Regression Equations.

As shown in Figure 14-11 above, the Breach Parameter Calculator contains five regression equations (MacDonald\Langridge-Monopolis; Froehlich 1995; Froehlich 2008; Von Thun and Gillete; and Xu & Zhang 2009). The user is required to enter several parameters that describe the Dam and the volume of water behind the structure at the time of failure. These parameters include: Top of Dam elevation; Breach Bottom Width; Pool elevation at Failure; Pool Volume at Failure; Failure Mode; Dam Crest width, upstream and downstream embankment slopes; Earth Fill Type; Dam Type; and Dam Erodibility factor. Not all values are used for all regression equations as noted on the editor. Some of the variables are specific to the MacDonald equation, and the last two are specific to the Xu & Zhang equation. Once the values are entered the calculator computes Breach Bottom Width; Side Slopes; and Breach Development Times from each of the regression equations. The user can then select the answers from one of the equations to by pressing the Select button next to the equation results that they would like to use.
Each of the breach parameter estimates will yield a different outflow hydrograph from HEC-RAS. However, once these hydrographs are routed downstream, they will tend to converge towards a common result. How close they get to each other will depend on the distance they are routed, the steepness of the stream, the roughness of the river and floodplain, and the amount of floodplain storage available for attenuating the hydrograph. If the populated areas below the dam are quite a distance away (say 20 miles or more), then the resulting hydrographs from the various dam breaches may be very similar in magnitude by the time they reach the area of interest. However, if the areas of interest are closer to the dam, then the resulting breach hydrographs could produce a significant range in results. In this situation, the selection of the breach parameters is even more crucial.