Verification of morphodynamic models on channels, trenches and pits

Abstract

This study is part of the SAND PIT project, which objective is to develop reliable prediction techniques and guidelines to better understand, simulate, and predict the morphological behaviour of large-scale sand mining pits and the associated sand processes at the middle and lower shoreface as well as in the surrounding coastal zone. In this study the morphological development of pits/trenches are investigated with two process based morphological models, namely the SUTRENCH 2DV-model and the Delft3D (2DH/3D) modelling system. The aim of this study is to verify the performance of both models on channels, trenches and pits. Both models are applied on four benchmark tests with identical model settings as far as possible. The benchmark tests consist of two laboratory experiments (test 1: waves propagating perpendicular to the current direction and test 2: waves propagating parallel to the current direction) and two field experiments (test 3: trench Scheveningen and test 4: dumping near Hoek van Holland). Recently new updates of SUTRENCH and Delft3D are composed, in which the updated transport formulas of van Rijns TRANSPOR2000 (Van Rijn, 2000) are integrated besides the TRANSPOR1993 transport formulations (Van Rijn, 1993). In case of Delft3D the sediment transport can also be computed with relative simple approximation formulae for depth-integrated sediment transport based on TRANSPOR2000 (Van Rijn, 2001). In order to make good comparison between the different simulations and models, the initial (suspended) sediment transports at the boundaries are equalized. This is achieved by adapting the correction coefficients for the suspended and bed-load transport to the measured (suspended) sediment transport in both models. Furthermore, in both models (SUTRENCH and Delft3D) the algebraic turbulence model has been used in the simulations. Simulations are made with the SUTRENCH-1993 and SUTRENCH-2000 model for both the laboratory experiments as well as the field experiments. The results of the simulations of both models are compared with the measurements. Both models showed reasonable well morphological results for both laboratory experiments and both field experiments, whereby in general the results of SUTRENCH-2000 are an improvement compared with SUTRENCH-1993. With the Delft3D model, simulations are made with the TRANSPOR2000 formulations (2-dimensional horizontal and 3-dimensional) and the approximation formulae (2-dimensional horizontal) for both the laboratory experiments and the trench at Scheveningen. Simulation of the dumping near Hoek van Holland experiment demands too much simulation time in Delft3D. The results of the simulations were not fully satisfactory. In case of laboratory experiment test 1, deviations occur between the measured and computed bottom profile. The sedimentation and erosion were both under-estimated. Also the simulation results of the trench Scheveningen model are not satisfactory. The sedimentation in all simulations (2DH and 3D simulations with the TRANSPOR2000 formulations and 2DH simulation with the approximation formulae) are largely under-estimated. In contrast with the simulation results of laboratory experiment test 2 that were satisfactory. Besides a basic simulation a sensitivity analysis per parameter is made with the SUTRENCH model and the Delft3D-TRANSPOR2000 (3-dimensional) model for both laboratory experiments. With SUTRENCH also random simulations are made in order to study the combined influence of the variation of the most important physical parameters on the computed bed levels. In general, the performance of SUTRENCH-2000 is at this moment more satisfactory than Delft3D for the used model settings. SUTRENCH-2000 shows more agreement with the measurements than Delft3D-TRANSPOR2000. However, the Delft3D results of test 2 are hopeful.