|LTV O&M thema veiligheid: deelproject 1. Verbetering hydrodynamisch NEVLA model ten behoeve van scenario-analyse|
Maximova, T.; Ides, S.; De Mulder, T.; Mostaert, F. (2009). LTV O&M thema veiligheid: deelproject 1. Verbetering hydrodynamisch NEVLA model ten behoeve van scenario-analyse. versie 2.0. WL Rapporten, 756_05. Deltares/Waterbouwkundig Laboratorium: Antwerpen. VI, 32 + 7 p. tables, 69 p. figures pp.
Part of: WL Rapporten. Waterbouwkundig Laboratorium: Antwerpen, more
In the framework of different projects, among which “LTV O&M thema Veiligheid - Deelproject 1” and “LTV O&M thema Toegankelijkheid - Ontwikkeling van een slibtransportmodel”, a calibration of the existing hydrodynamic NEVLA model of the Scheldt estuary is executed. This report gives a summary of all the steps which are undertaken to improve the model.
The objective of the study “LTV O&M thema Veiligheid -- Deelproject 1” is to analyze the effect of different changes and their influence on the hydrodynamics of the estuary. Based on a literature review and data analysis some hypotheses concerning the change in tidal penetration in the estuary will be made. Afterwards these hypotheses will be verified with numerical models. In order to obtain reliable results, it is necessary that the numerical model performs well. Therefore a sensitivity analysis, a calibration and a validation of the NEVLA model, which will be used for 2D-3D hydrodynamic simulations, were carried out.
The objective of the sensitivity analysis was to understand the impact of different model parameters on the tidal penetration. The results of this analysis are described in (Ides et al., 2008) and (Vanlede et al., 2008a). In (Vanlede et al., 2008b) the calibration was performed for the calculated water levels and discharges, based on the phase and magnitude of the most important harmonic tidal components of these parameters. In (Maximova et al., 2009) the calibration of the NEVLA model was extended. The resulting model parameters from (Vanlede et al., 2008b) were used as the reference simulation. The methodology was based on the comparison of phase and magnitude of the calculated and measured high and low water levels. This extended calibration was mainly focused on the Upper Sea Scheldt and the tributaries Zenne and Dijle because in these regions the differences between calculated and measured water levels were the largest. As a result of the calibration, the accuracy of the model for high water levels and low water levels was improved for most stations along the Scheldt estuary.
The calibrated model was validated for a period with normal tide and for a period with an extreme high water. The calibrated model performs rather well for a period with normal tide. However, it does not simulate accurately a period with an extreme high water level. The reason for this and the possibilities to improve the model accuracy for such circumstances should be studied more in the future.
The roughness field used for the calibrated model is strongly related to the bathymetry that was used during calibration of the model. If we use a different bathymetry for the scenario analysis the roughness field might change as well. Therefore, a simple roughness field without variation in the transversal direction was found, which gives rather similar results as the calibrated model with the space varying roughness field.