Abstract:
The uncertainty associated with simulations of process-based coastal area morphodynamic models is assessed through numerical experimentation. Appropriate metrics of uncertainty are defined based on the
standard deviation of the model results at each location and each time step. Uncertainty is examined using a set of realistic one year morphodynamic simulations of the evolution of a highly dynamic tidal inlet. Results indicate that uncertainty increases linearly with time, and suggest that its rate grows with increasing
sediment fluxes. Hence, the limits of predictability of morphodynamic model applications are higher for slowly varying systems. Attempts to reduce uncertainty by aggregating model results at larger spatial scales met with limited success. Ensemble simulations are suggested as a possible avenue to investigate the longterm evolution of tidal inlets using process-based models.