On The Physical Mechanisms Of Wave-current Coupling In Nearshore Zone And Their Applications

Nearshore waves and currents are highly interacted. The currents could be driven by waves, andthe waves could be influenced by varing current and water depth. Hence the nearshore processessimulated with coupled wave-current models are more accurate than that computed with uncoupledmodels. However, the physical mechanisms of wave-current interaction are still controversial, and itsapplications on simulating wave-induced currents and storm surge need further study.Radiation stress, the excess momentum flux which is caused by the presence of waves andexerting on the current, is an important mechanism of wave-current interaction. The verticalintegrated radiation stress is widely used in computing wave set-up and rip currents. During the pastdecade many theoris concerning the vertical dependence of the radiation stress have been advanced.Although validated in different works, these theories still lack objective comparisions andevaluations. In this study a numerical wave tank is implemented to study the vertical distribution ofradiation stress. The simulated water level corresponds to airy wave theory well, and the verticalintegrated radiation stress computed by the water particle velocity and pressure corresponds toLonguet-Higgins and Stewart (1964)’s theory well, both of which indicated the validity of thenumerical wave tank. The vertical profile of radiation stress induced by the wave motion is alsocomputed with particle velocity and pressure, and it shows a maximum value between the wave crestand wave trough. If the radiation stress in this area is integrated as a surface stress, it would not becontinuent with the raidition stress under the surface. Both the radiation stress computed with Xia etal.(2004)’s and Zhang (2004)’s theories show a continuous transition from surface to bottom, whichare not consistent with the wave tank solution. Mellor (2008) expressed the radiation stress on thesurface with a Dirac function, which corresponds to the wave tank solution well.To compare the influnce of radiation stress, wave roller and vortex force on wave-inducedcurrent simulation, two historical experiments are simulated with coupled SWAN-POM model. Threeradiation stress expressions, one vortex force expression and one wave roller expression areimplemented in the coupled model. It is found that the wave setup simulated with radiation stressapproaches and vortex approach corresponds to observations well. The current velocity computedwith Longuet-Higgins and Stewart (1964)’s approach is very small, but gets larger and correspondsto observations better when the effect of roller is added. The circulations computed with Xia et al.(2004)’s approach correspond to the observations well in the barred beach run, but rotate in a reversedirection of the observations in the wave tank run. Both the results computed with Mellor (2008)’sapproach and vortex approach show the best correspondence with observations when roller effectsare included.In the nearshore region, the wave condition varies a lot in space. It induces large radiation stressgradient, which drives complex nearshore current. Rip currents could be formed when topographyvaries alongshore. It is dangerous for swimmers and plays an important role in sedimentaltransportation. However, rip current is hard to predict according to wave conditions. A series of numerical experiments are performed to investigate the relation of incident waves and rip currentvelocity. It is found that the rip current velocity increases with increasing wave height, wave periodand decreasing water depth, and it is also sensitive to the variation of wave breaking location. It isalso found that the dimensionless rip current velocity increases linearly with the dimensionless waveheight computed with the wave height at the seaward egde of the bars, and the relationshipcorresponds to observations well. Model results also indicate that secondary circulation is mainlydriven by the waves which propagate through the rip channel and finally break near the shoreline.The intensity of the secondary circulation is in direct proportion to the alongshore variation of waveheight gradient between the bars and the shoreline.Besides the radiation stress, model domain is also an important factor which would influencethe model result in the storm surge simulation. How to find a small domain with reasonablecomputation cost while limiting the model errors to within an acceptable range is an unsolvedquestion. In this study a set of numerical experiments are performed with the coupled model toinvestigate the influence of model domains on simulated storm surge under different typhoon andcoastal bathymetric conditions. Model results indicate that the peak surge increases with increasingdomain size and then approaches a constant value. A “threshold domain size” is defined accordingly,so that differences among model results with different domain sizes larger than the threshold domainsize are negligible. The threshold domain size is shown to be insensitive to hurricane intensities, butincreases linearly with increasing hurricane RMWs. It also increases with increasing translationspeeds and decreasing bottom slopes, when hurricanes approach land perpendicularly. The thresholddomain size can also be affected by other factors such as the direction of the storm track of alandfalling hurricane and the extent of continental shelf. Considering the complex dependence of thethreshold domain size on various hurricane parameters, a polar graphic chart is created for estimatingthe threshold domain size for possible hurricane landfall directions and RMWs. A real case study ofHurricane Charley (2004) is conducted with different domain sizes. Model results indicate that thethreshold domain size estimated from the idealized experiments is reasonable for practicalapplications.