Methods to evaluate flow modelling results quantitatively have been standardised on the basis of practical cases. This helps to improve the validation of complicated 3D flow models and makes the evaluation more objective, easier to reproduce and explain to the parties commissioning the studies. Traditionally, flow modelling results are often evaluated purely in graphic terms. In this case, the salinity distribution by tidal motion and vertical and horizontal mixing modelled by Delft3D-FLOW (red and magenta curves) is compared to the salinity measured at positions along the physical scale model in Delft Hydraulic's Tidal Flume (green and blue curves).
In cases where salinity profiles are measured by vertically moving salinity probes, the visual evaluation of modelled and the measured salinity profile can be made intuitively. The salinity probe systematically measured the vertical salinity distribution by moving periodically in five vertical steps from the top to the bottom measurement level (indicated by dots).
Quantitative evaluation of modelling results requires that suitable acceptance criteria to be formulated, given the intended use of the model. In the present case the parameter is salinity intrusion, and the acceptance criterion is that the maximum RMS salinity difference at given locations does not exceed an agreed percentage of the local measured dynamic salinity range. |
