| one publication added to basket [96729] | Modelleren van ritmische morfologie in de brandingszone = Modelling rhythmic morphology in the surf zone
Klein, M. (2005). Modelleren van ritmische morfologie in de brandingszone = Modelling rhythmic morphology in the surf zone. Communications on Hydraulic and Geotechnical Engineering, 05-5. Delft University of Technology. Faculty of Civil Engineering and Geosciences: Delft. XII, 165 pp.
Part of: Communications on Hydraulic and Geotechnical Engineering. Delft University of Technology. Department of Civil Engineering: Delft. ISSN 0169-6548, more
|
| Available in | Author |
|
Document type: Dissertation
|
| Keywords |
Earth sciences > Geology > Geomorphology > Coastal morphology > Beach morphology
Modelling
Topographic features > Beach features > Surf zone
Marine/Coastal |
| Abstract |
Many of the world’s sandy coasts, such as parts of the Dutch coast, are erosive. Since a largepart of the Netherlands is lying below mean sea level a ‘healthy’ dune and beach systemis of utmost importance for the very existence of the Netherlands. The Dutch governmenthas therefore decided to dynamically maintain the coastline of 1990 by means of beach andshoreface nourishments. The latter ones are increasingly being applied, since they are cheaperand more flexible in execution. The positive contribution of a shoreface nourishment to thecoastline position is based on transport towards the coast of the supplied sand by naturalprocesses. Observations of a 2000 m long shoreface nourishment executed near Egmond, theNetherlands, in the Spring of 1999, however, show that shoreface nourishments do not alwayssimply diffuse. The Egmond nourishment remained in tact for a long time and in fact becamepart of the bar system, locally creating an additional third breaker bar. Besides, the rhythmicbar crest topography that was already present before the execution of the nourishment (i.e.the autonomous or free-behaviour) was affected by this nourishment. Despite its unexpectedbehaviour, the Egmond shoreface nourishment in combination with a beach nourishment isconsidered a success, since in the four years after the nourishment no other nourishments hadto be executed.The hypothesis forming the basis of this thesis is: the response of the rhythmic topography of the surf zone of Egmond to the shoreface nourishment has been so strong because the length of the shoreface nourishment was of the same order as the spacing of the rhythmic features on the outer bar. To validate this hypothesis, the occurrence of rhythmic bed features in the surf zone must be investigated systematically in order to be able to discern the autonomous behaviour from the behaviour caused by the nourishment. The major part of this thesis concerns the investigation of this autonomous behaviour. The first question that needs to be answered is ‘What is the linear free-behaviour of coastal systems in absence of shoreface nourishments?’ This step in the research into the occurrence of rhythmic bed forms in the surf zone is an exploration of the initial growth of bed forms on otherwise longshore-uniform coasts by means of linear stability analyses (LSA’s). These LSA’s describe the linear autonomous behaviour in terms of a bed perturbation with a certain preferred longshore wavelength, cross-shore amplitude function and associated growth and migration rates. Chapter 3 and a part of Chapter 4 deal with this question. The second research question reads: ‘What is the non-linear free-behaviour of coastal systems in absence of shoreface modelling rhythmic morphology in the surf zone nourishments?’ In this second step, the non-linear autonomous behaviour is investigated by means of morphodynamic experiments in which the bed forms can freely evolve in time. This question is addressed in Chapter 4. The third and final question that needs to be answered is ‘What is the non-linear behaviour of coastal systems in the presence of shoreface nourishments?’ In this final step morphodynamic experiments including shoreface nourishments are performed. Differences in the morphodynamic evolution with and without nourishments are indicative of the impact of shoreface nourishments on the non-linear evolution of coastal systems. Chapter 5 deals with this question. Both the LSA’s and the morphodynamic experiments are performed using a numerical model computing the water motion, sediment transport rates and bed changes. Due to the use of a numerical model, another method than the usually applied, analytical one has to be conducted to perform the LSA’s. An advantage of this method is that the linear stability of an arbitrary coastal profile can easily be assessed. Besides, the physical processes do not necessarily have to be simplified, contrary to the analytical method. First of all, LSA’s of planar sloping beaches are performed, despite the fact that the central part of the Dutch coast is characterized by the presence of two or three breaker bars. Many studies concerning LSA’s of planar beaches have appeared in the literature, which are used in this thesis for validation of the presently applied method. Besides, these LSA’s of planar beaches are performed to gain insight into the dependence of the linear stability characteristics on process formulations and parameter values. From these LSA’s it is apparent that the shape and the growth and migration rates of bed perturbations on planar sloping beaches are highly sensitive to process formulations, especially to the sediment transport formulation, and to parameter values. This explains the large variation in the linear stability characteristics of planar beaches obtained in this and previous studies. Nonetheless, under current-dominated conditions similar results as presented in the literature have been found. Secondly, LSA’s of double-barred beaches have been performed to describe the linear autonomous behaviour of the central part of the Dutch coast under different hydrodynamic and morphological conditions. In these LSA’s the Bailard [1981] sediment transport formulation is applied. The bed forms resulting from these LSA’s can be characterized as rip channel systems with alternating channels and shoals along the crests of the breaker bars. Whether bed perturbations emerge on top of the outer breaker bar or not, depends on the height of the outer breaker bar and the wave height. If the conditions are such that no significant bed perturbation develops on the outer bar, then the preferred longshore length-scale of the rip channel system is 700 m. If, on the contrary, significant bed perturbations do emerge on the outer bar, then the preferred length-scale increases to 2200 m. The length scales of these bed forms and their associated migration rates correspond well with observations of 3D morphological features in the surf zones of Egmond and Noordwijk.Thirdly, the non-linear autonomous behaviour of bed forms emerging on the bars of a double-barred beach is explored with various initial bed perturbations, demonstrating that the modelling rhythmic morphology in the surf zone iii non-linear evolution can be subdivided in two phases. The first phase is a phase of exponential growth lasting 50 to 80 hrs. The second phase is a non-linear phase in which the morphodynamic evolution is highly dynamic. No morphodynamic equilibrium is reached, since the aggregated amplitude of the bed forms in the area of interest keeps varying in time. Furthermore, it is apparent that the preferred mode, as found with the LSA, has a significant contribution to the total bed forms. However, it is certainly not the only important mode, since 4 to 7 modes are significantly contributing to the bed perturbation. Besides the differences in (the number of) length scales, large differences in the shapes of the bed perturbations obtained in the linear and the non-linear regime exist. The bed perturbations in the linear regime are symmetrical whereas in the non-linear regime they are asymmetrical, ranging from crescentic to undulating features. Although highly dynamic, the shapes of the bed perturbations found in the non-linear phase are similar to bed forms observed in nature. Moreover, also the simulated migration rate of the order of 100 md−1 and the overall wavelength of 1000 m on the inner bar and 2000 m on the outer bar correspond well with observations. An important difference with previous studies is the fact that in this study no morphodynamic equilibrium is reached. This is attributed to the sediment transport formulation. Bed slope related sediment transport in the formulation of Bailard [1981] does not essentially influence the non-linear evolution. Furthermore, the initial bed perturbation is of importance to the non-linear evolution, mainly on the amplitude of the bed perturbations and in some exceptional cases also on the length scales of the bed forms, if the hydrodynamic conditions are sufficiently energetic and/or the outer bar is sufficiently high. If the conditions are, however, such that no bed forms grow on the outer bar the initial bed perturbation is not of importance to the non-linear evolution. Fourthly, the impact of shoreface nourishments on the non-linear morphodynamic evolution is explored by performing a number of non-linear experiments including shoreface nourishments. Whether shoreface nourishments have an effect on the morphodynamic evolution depends on the wave height and the height of the outer bar as well. If the conditions are such that no significant morphodynamic activity takes place on the outer bar, then a shoreface nourishment executed at the seaward side of the outer breaker bar hardly has an influence on the morphodynamic evolution. No significant morphodynamic development is triggered on the outer bar and the evolution of bed forms on the inner bar is hardly affected, although the amplitudes of the bed forms on the inner bar slightly decrease. If the conditions are such that bed forms do grow on the outer bar, then the influence of a shoreface nourishment on the morphodynamic evolution depends on the length of the nourishment. An 800 m long nourishment initially has a damping influence on the bed perturbations, but in the course of time the root-mean-square amplitude ARMS of the bed perturbations in the area of interest obtains a value similar to the one obtained without a nourishment. Both the shape and the length scales of the bed forms obtained in the experiments including the nourishment are similar to the ones obtained without the nourishments. iv Modelling rhythmic morphology in the surf zone in the case of a 2000 m long nourishment, however, ARMS significantly increases after an initial period of decay. In the dynamic phase, the value of ARMS becomes much larger than without a shoreface nourishment. Besides, the length scales on both the inner and outer bar become larger when a 2000 m long nourishment is included. The difference in response of the coastal system to the presence of an 800 m long nourishment and to a 2000 m long nourishment confirm that the length of the nourishment is of great importance to the reaction of the coastal system. An additional experiment with a 3000 m long nourishment results in the dynamic phase in ARMS values smaller than ARMS of the experiment with the 2000 m long nourishment. Although the results of this additional experiment should be used with care, they suggest that the fierce and unexpected response of the rhythmic patterns in the surf of Egmond to the execution of the shoreface nourishment is indeed caused by resonance. Hence, the reaction of the Egmond coastal system was so fierce since the length of the nourishment corresponded with the length scale of the bed forms on the outer breaker bar. The evolution of the coastal system including the 2000 m long nourishment has a number of characteristics in common with observations. First of all, the onshore movement of that part of the outer bar covered by the shoreface nourishment, thereby locally creating a third breaker bar. Secondly, the part of the outer bar moving onshore becomes somewhat higher. And thirdly, a region of erosion develops between the shoreface nourishment and the outer breaker bar. These characteristics also result from an experiment in which only the influence of a 2000 m long nourishment on the morphodynamic evolution is accounted for by omitting the initial bed perturbation at the start of the experiment. This demonstrates that, at least initially, the morphodynamic evolution due to the nourishment is to a certain extent decoupled from the morphodynamic evolution induced by the initial bed perturbation. |
|
