DEVELOPMENT OF THE LONGSHORE CURRENTS MODEL USING NON ORTHOGONAL BOUNDARY FITTED TECHNIQUE: A CASE STUDY OF SANUR BEACH, BALI

• Main Author: Bahari Muslim, Arung
• Format: Theses
• Language: Indonesia
• Subjects: Teknik (Rekayasa, enjinering dan kegiatan berkaitan)
• Online Access: https://digilib.itb.ac.id/gdl/view/61312
• Institution: Institut Teknologi Bandung

Ocean currents have become an important factor in everyday life, such as sediment transport, generating energy, ship navigation, and others. Ocean currents can be formed due to tides (tidal current), wind (wind-induced current), large currents (sub stream current), density differences (density induced current), and waves (waveinduced current). Wave-induced currents are ocean currents formed due to ocean waves that come towards the shoreline and then bring changes in momentum and mass so that currents occur. The occurred current is moves parallel to the shoreline and it called longshore current Longshore currents occur due to breaking waves that propagate from offshore at a certain angle to the onshore, and when the waves break, the momentum of the wave decreases so that to balance the momentum, the water level elevation must increase, or wave set up and current. The waves that carry momentum and mass are called radiation stress. The radiation stress equation is still being developed, for example, based on the depth-dependent by Mellor and Xia (Mellor, 2008) and (Xia et al., 2004), and then there is also being developed based on the mild slope equation by Zeng (Tang et al., 2016). Another research is based on mild slope equation and transformed to orthogonal curvilinear coordinate (Cui et al., 2011). With the lack of research so far, the developed equations are still in Cartesian coordinates and are directly transformed to orthogonal curvilinear coordinates. These two things have disadvantage which cannot accommodate complex coastlines. Muin and Spaulding (Muin & Spaulding, 1997) has developed hydrodynamic equations in spherical coordinates and transformed them to curvilinear coordinates in a non-orthogonal technique. The advantage of this research is that it can accommodate complex shapes such as rivers and can also be applied to complex coastlines. The drawback of Muin and Spaulding's research is that it has not included an additional force, namely radiation stress. Therefore, in this study, development of longshore current with non orthogonal boundary fitted technique was occurred and the model was applied to Sanur Beach, Bali The advantages of this model are that it can accommodate complex coastlines, and the calculation of longshore currents does not become a blowup and quick computation. To test whether the model is feasible to use, a model test called steadystate wave set up is made by making a grid with bathymetry with a particular slope so that the wave set down and wave set up values are generated. In the model test, 5 scenarios were made in which scenario-1,2 and 3 were time step variation to determine whether the hydrodynamic model developed was stable or not. Furthermore, the sensitivity test is to test whether the smaller resolution used, the closer numerical results are to the analytical results. There are 2 scenarios in the resolution variation test, namely scenario-4 with a resolution of 50 m x 50 m while scenario-5 with a resolution of 30 m x 30 m The model test results of time step variation have the same trendline as the analytic where there is a wave set down and break at a distance of 2500 m and experiencing a wave set up with a value of 1.37 m. For the results of variation time step, the numerical value has the same trendline as the analytic. Where the numerical error is same for each scenario which is 21.9 %. For the sensitivity test, the model states that the smaller the resolution, the better the numerical result which closer to the analytical value. Resolution 30 m x 30 m (scenario-5) has the lowest numerical error among the other resolutions which the set up error on onshore is 20.6%. For other resolution which 50 m x 50 m (scenario-4) creates error set up on onshore is 21.09% while 100 m x 100 m resolution (scenario-1) creates an error 21.9 %. So, it can be concluded that the developed model has passed the model test and can be applied to a case study in Sanur Beach, Bali. The model was compared to an experiment that was studied by Hamm (Hamm, 1993) . The model result shows that at 1500 m there is no rip current occurred, but the model shows that the current moves toward to the shore. This indicate that the model is not optimized enough to predict current’s pattern, but the model shows a good agreement to predict set up and set down. So, the model can be applied to a study case in Sanur Beach, Bali. When the model is applied to Sanur Beach, Bali, there are longshore currents and rip currents. There is a rip current because there are currents against each other and meet each other so to order balance the momentum the currents must moves outward from the shore. There is no current pattern difference between scenario- 6,7 and 8 while the scenario-9 shows a different pattern in top area which there is no whirlpool. The elevation results show that in bottom area a set down occurs with value -0.003 m while the set up 0.009 for waves with period 100 seconds (scenario- 6) and period of 50 seconds (scenario-7). Waves with period 25 seconds (scenario- 8) shows a different set up on shore with value 0.008 whereas period 5 seconds (scenario-9) shows the set up that was obtained is 0.006 m The model shows that there is set down and set up where the value are -0.003 m and 0.009 m. The model vector shows that the value of the current is not uniform where the value is from 0.1 m/s to 3 m/s. Although the value is reasonable, the model needs to be validated with field data. The validation is needed because to know if the developed model can be represented as a real case.