Pseudo-Timestepping Method

The Pseudo-Timestepping Method requires initial estimates for flow and stage at each model node specified in the Initial Conditions. These initial conditions are used for the steady state run, with the boundary conditions held constant for the time at which the solution is required.

The initial timestep is input at the start of the run, but after each five pseudo-timesteps the user is asked whether it should be changed. If the initial conditions are very approximate, the flows may be highly unsteady or unstable and a very small timestep (may be as low as 10 seconds) should be used. The timestep can be increased as the flow becomes less unsteady (measured by the flow ratio).

The model is run until all the irregularities and inaccuracies in the guessed initial conditions have propagated or been dissipated out of the system. This can be a time consuming process, especially when the model contains large reservoirs which have long response times. The monitoring parameters are the flow and head ratios, which are automatically printed on the screen. When these values become very small (usually less than 0.5 x 10^-3) and the timestep is relatively large, then a steady solution should have been attained. This can sometimes be difficult to achieve since some models may be sensitive to sudden increases in timestep. It should be noted that when the timestep is changed the solution may appear to diverge before starting to converge again.