Model Timestep Selection

The sensitivity of the model to timestep selection is dependent upon the slope of the discharge hydrograph ( ) and the grid spacing. Increasing the grid spacing size introduces additional water storage to a corresponding increase in nodal point flood depth values. Similarly, a decrease in timestep size allows a refined calculation of inflow and outflow values and a smoother variation in nodal point flood depths with respect to time. The computer algorithm may self-select a timestep by increments of halving (or doubling) the initial user-chosen timestep size so that a proper balance of inflow-outflow to control volume storage variation is achieved. In order to avoid a matrix solution for flood depths, an explicit timestepping algorithm is used to solve for the time derivative term. For large timesteps or a rapid variation in the dam-break hydrograph (such as is large), a large accumulation of flow volume will occur at the most upstream nodal point. That is, at the dam-break reservoir nodal point, the lag in outflow from the control volume can cause unacceptable error in the computation of the flood depth. One method that offsets this error is the program to self-select the timestep until the difference in the rate of volume accumulation is within a specified tolerance.

Due to the form of the DHM in equation 22, the model can be extended into an implicit technique. However, this extension would require a matrix solution process which may become unmanageable for two dimensional models which utilize hundreds of nodal points.


[Back to DHM Home] [Back to Research] [Cover] [Table of Contents 1] [Table of Contents 2] [Table of Contents 3]
[<-- Previous Page]  8   9   10   11   12   13   14   15   16   17   18   19   20   21   22   23   24   25   26   27   28   [Next Page -->]