We are in the process of modeling two dams in series. The upper dam is being analyzed using Level Pool Routing and the lower dam is being modeled using Dynamic Routing. The upper dam discharges directly into the pool of the lower dam.

The upper dam is considered full to the top of the dam with a minimal inflow hydrograph and the breach is triggered almost immediately.

The downstream dam is at normal pool with a rating curve entered for the principal and auxiliary spillways. The overtopping is modeled using the top of dam elevation and crest/length coefficient. We have input an uncontrolled spillway crest and based the discharge rating curve heads off of this elevation. The spillway crest and normal pool elevations are the same.

The model runs correctly up to a point. The upper dam begins to breach correctly and the lower lake starts to fill up. The lower lake is large enough that the breach from the upper lake is contained and the lower lake doesn't overtop so a failure is not modeled. The problem occurs as the lower lake continues to recede. It continues below the original starting elevation and large swings in the outflows begin to occur along with non-convergence. Prior to the point when the lower lake drop below the starting elevation the discharges appear to match the flows input in the rating curve.

We believe it is tied to the rating curve for the lower lake not being interpolated correctly near the starting pool elevation.

Any insight would be greatly appreciated,

Thanks

Jeffery G. Powers, P.E.
Director of Water Resources
Ross Consulting Engineers, P.C.
770-924-2955 ext. 206