Applicability of the 3rd order theoretical solutions for internal waves in a two-layer system is investigated by using a strongly nonlinear and dispersive internal wave model. The 3rd order solution is derived using the 9th order internal wave equations. The 3rd order solution is found to give larger wavelength scale compared to KdV theory. The applicability of the 3rd order solution is confirmed to be high when the amplitude of internal solitary wave is 5% of the lower layer thickness. A fully nonlinear strongly dispersive internal wave model reveals that high frequency internal waves are induced behind an internal solitary wave when the initial shape of internal solitary wave is larger than critical level.
The purpose of the present study is to newly propose a numerical wave tank by introducing a wave generation source into a CADMAS-SURF/3D and verify the validity of the numerical wave tank. The numerical experiments with the change of the coefficient α, which is related to the strength and extensity of wave generation source, revealed that the most appropriate waves are able to be generated in the case of α=2. The applicability of the proposed numerical wave tank was confirmed by applying it to oblique wave generation problems and comparing the numerical results with the results of the laboratory experiment on wave deformation under Tsunami action.
Generally, wave transformation in shallow water should be estimated for a representative offshore wave spectrum. However it may be difficult for a Boussinesq wave model which requires a relative narrow computational region for calculating wave fields that an incident wave boundary is expanded to offshore. In this paper, multi-directional random waves are generated by using Boussinesq wave model on the linear incident boundary with several wave spectra transformed in shallow water. Wave height distributions behind a spherical shoal can be reproduced with spatial distribution of multi-directional spectra estimated by an energy balance equation. In an actual harbor, the significant wave height in front of a revetment can be estimated larger in generation with several wave spectra.
The third-order interactions between the normal incident and reflected wave groups due to a vertical wall are investigated by numerical simulations based on Zakharov's equations with kernel functions valid for the cases of finite depth. The amplitude modulation occurs due to the unstable disturbance of class I for water depths except for near kh=1.36. The results of the temporal surface elevations computed on the basis of Zakharov's equation indicate that in intermediate water depths nonlinear waves both having side-band disturbance and bimodal spectrum undergo strong modulation and their maximum wave height in the group increases with time above the initial height. The changes in phase velocity of the reflected wave groups are also significant.
Laboratory experiments were conducted for the concentration of long period waves due to sudden alongshore change in nearshore bathymetry. Waves and currents were measured around a boundary between mild and steep sloping beds. Alongshore variation in wave run-up showed significant peaks near the boundary, which were developed by the interaction of waves with nearshore currents and the excitation of long wave components by breaking group waves. Introduction of shore protection structures on the mild slope side was found to be effective in decreasing run-up heights on the steep slope side.
A set of nonlinear internal-wave equations, which have been derived on the basis of the variational principle without any assumptions concerning wave nonlinearity and dispersion, is applied to compare numerical results with experimental data of internal waves propagating in a deep-water region. Internal waves propagating over a uniformly sloping beach are also simulated. Internal progressive waves show remarkable shoaling after the interface reaches the critical level, after which the physical variables change discontinuously near the wave-breaking point. In the wave-breaking case of internal waves reflecting at a vertical wall, the vertical velocity of the particles in the vicinity of the interface is different from that of the moving interface on the wall, which means that the kinematic boundary condition on the interface is not satisfied.
To calculate wave breaking, a numerical simulation method for fluid composed of air, water and intermediate bubbly flow is presented. Bubbly flow is assumed as a fluid with density continuously varying between air and water and calculated as an incompressible uni-phase flow. To satisfy momentum conservation over the computational domain with large density variation, Navier-Stokes equation described in terms of momentum advection is employed as the governing equation. The advection terms of the momentum equation and advection equation for Density Value Method are solved by R-CIP method to avoid numerical diffusion. The results are compared with experimental results for a dam-break, bore and wave breaking on a slope. It is found that the model can reproduce bubbly areas due to wave breaking fairy well.
A multiphase model for water waves was developed consisting of the fifth order space accurate WENO (weighted essen-tially non-oscillatory) method for advection terms and a level set method for capturing the gas-liquid interface. The model was applied to a solitary wave propagation over a rectangular structure, 2nd-order Stokes waves over a trapezoidal struc-ture, and a solitary wave over a semicircular structure. Through comparisons with existing experiments, the model was found to be capable of reproducing the experimental results with relatively coarse grids and without an explicit turbulence model because of the adoptation of high-order and high resolution scheme with the concept of Implicit LES.
We develop three-dimensional coupling model of fluid-rigid body interaction by MPS method. The model takes into consideration a force of fluid-solid interaction which used the pressure difference between particles. When the force of fluid particle is calculated using the gradient model of MPS method, the values of pressure of rigid particles are used. We verify that the fluid-solid interaction model makes the analysis of coupling simulation more accurate by comparing numerical results to theoretical values. Further we run experiments that measure pressure on a structure on an apron by generated solitary wave. Our numerical results are in good agreement with the experimental data. The coupling model will be possible to estimate the fluid force due to tsunami and storm surge for the structure considered as rigid body.
A highly precise numerical calculation method of the gradient as a differential operator in a computational method with the MPS has been proposed in this study. The method is not dependent on the particle arrangement. Gradient calculations of a linear and nonlinear function have been introduced in the proposed method to verify the numerical accuracy. The results show high accuracy on the boundary in regular grid arrangement case but on both the boundary and inside in random arrangement case. High accuracy and computational stability are also obtained when applied to the calculations of the hydrostatic pressure and free surface of water.
A higher order Laplacian model is proposed for enhancement and stabilization of pressure calculation by the MPS method. The higher order Laplacian model is derived by taking the divergence of a particle-based gradient model. The proposed higher order Laplacian is then applied for discretization of both Laplacian of Poisson pressure equation and that of the viscous forces in the CMPS-HS method. A few numerical tests, namely, exponentially excited sinusoidal pressure oscillations and a violent sloshing flow are carried out to demonstrate the enhancing and stabilizing effect of the proposed higher order Laplacian model.
The MPS method has been proven useful in simulating free-surface hydrodynamic flows. Despite its applicability, the MPS method suffers a high computational load. The main objective of this study is to develop a GPU-accelerated MPS code with using CUDA language. Several techniques have been briefly shown to optimize calculations. Some specific three dimensional calculations including a numerical wave flume have been carried out by the GPUaccelerated MPS method. The developed GPU-based code distinctly improves computational efficiency and shows comparable reliability to CPU-based codes.
We develop a hybrid 2D-3D simulation of Lattice Boltzmann Method (LBM) that increases the capability of simulating offshore/nearshore tsunami propagation and coastal inundation. For that purpose, the following schemes are newly developed to reproduce the free-surface movement of inundation flow including water splash in front of the structure, and nesting schemes of 2D/3D model for more efficient simulation of wave propagation and inundation in wider area. The model is validated by laboratory experiments of dam break and bore propagation.
In the past years, the Lattice Boltzmann Method (LBM) has been developed and applied to simulate single phase flow with a free surface. In this paper, the free-surface tracking algorithms are newly implemented in a multi-phase LBM model to expand a capability of LBM to simulate a complex behavior of fluid flow. The model was verified by standard dam-break (single-phase) and oil-slick (two-phase) experiments and were found to be in good agreement with the results.
In this paper, flashing responses of bioluminescent dinoflagellates in oscillatory turbulent boundary layer flows over rippled bed are studied with the aim with future development of imaging technique to aquire planar distributions of turbulence in a shear layer. Foundamental statistical properties of the bioluminescence involved in strong turbulence caused by separation of the boundary layer from the ripple crest are discussed, and it is found that there is a correlation of time variation of the mean flash intensity with the turbulent energy.
Lateral instability of overtopping jets is manifested when a breaking wave hits a vertical wall, resulting in formation of so-called finger jets. This study presents that the instability is caused by changes in orientation of capillary waves on the stretched surface rising at high acceleration, and that the evolution of the resulting fingers depends on relative distance between the breaking point and location of the wall. Population of the sprays with very high rise velocity over 9 m/s significantly increases when incident waves break in front of the wall due to expansion of air pocket squeezed between the jet and wall.
An obliquely descending eddy caused by wave breaking is a strong three-dimensional phenomenon. However, there have been a limited number of studies using hydraulic model tests that examined the distribution of obliquely descending eddies along coasts and the distribution of the three components of velocity. In the present research, the structure for the formation of obliquely descending eddies was examined. As a result of the study, it was found that obliquely descending eddies formed along coasts can be categorized into different types: "A structure wherein two vortexes come into close proximity," and "a structure wherein the distance between the two vortexes remains constant."
The novel imaging technique for measuring planar free-surface shapes using a PC projector and digital camera is examined through model tests of planar wave field with coastal structures. The measured results reasonably represent local reflected and diffraction waves behind a breakwater and near a port mouth, which are consistent with computed ones. This technique is also applicable to the measurement of complex surface shapes of breaking waves. Typical finger-shaped jets formed after plunging overturning jets and local depression stretched behind a breaking-wave front are able to be obtained using the current technique.
Solitary wave is considered to resemble a tsunami wave. Thus solitary wave study is very important for coastal area. The shallow water equation (SWE) model is commonly used in tsunami modeling. The Manning method is generally used to assess the bed stress term. Nevertheless, boundary layer approach in assessing the bed stress would provide a more accurate prediction. In this study, a new method for solitary wave run up modeling has been developed by simultaneous coupling between SWE and k-ω model. Bed stress in the SWE is assessed directly from the boundary layer equation using k-ω model. The new method was used to simulate solitary wave run up (Synolakis, 1987). Water surface and run up height were compared. Further analysis to the momentum balance was conducted.
We examined field observation data of the friction velocity on the sea surface to describe the roughness and the drag coefficient in terms of windsea parameters. Estimations of the friction velocity were made by the combined use of the eddy correlation and inertial dissipation methods. Based on directional wave spectra, we extracted only pure windsea data to carry out the parameterizations for ideal wave-fields. The sea-surface roughness was parameterized in terms of two dimensionless windsea parameters, i.e., the wave-wind coefficient and the windsea Keulegan number. The drag coefficient was formulated empirically on the basis of the result of the roughness parameterization. The validity of the empirical expression for the drag coefficient was confirmed in comparison with the field observation data.
A new linear relation between sea surface drag coefficient and the saturation ratio of wind wave spectrum was obtained. The theory is based on the 3/2 power law of wave heights and periods, which was proposed by Toba (1972). The saturation ratio of wind wave spectrum is calculated by using spectrum analysis in principle, however a new parameter derived by the 3/2 power law gives us a saturation ratio from significant wave height and period instead of spectrum analysis. It is found that the negative correlation between sea surface drag and saturation ratio of wind wave spectrum exists. The relation can be explained that the saturated wind wave conditions reduce the momentum exchange from wind to wave, so the sea surface drag reduce in proportion to the momentum exchange.
This paper develops a quasi-three-dimensional neasrshore current model that accounts for volume fluxes due to waves and surface rollers. While the present model follows the concept of depth-integrated 2DH mass and momentum conservation equations, the model splits the water column at the wave trough level and has 2DH equations respectively in two layers. Decomposing velocity fields into multiple components due to different phenomena, such as waves, surface rollers and slowly varying mean current, and time-averaging over the wave period, the model expresses the influence of additional volume fluxes and momentum forces in explicit formulae. The model showed excellent predictive skills both in undertow and longshore currents on the long straight beach and more complex circulation currents behind the submerged breakwater.
Field measurements of bottom topography together with the wave height and water particle velocity in the shallow water region were carried out in September, 2008 and 2009 to investigate the relation between rip channel and rip current. On the relict rip channel, rip current is usually generated provided the wave incident angle does not change significantly or the incident wave height increases largely. The rip channel easily decays when the incident wave angle undergoes great change from normal to oblique. Even in the case where there is not any typical bottom configuration, rip current occurs from the hot spot where longshore currents converge that has a potential for generating rip channel through the interaction between wave and wave-induced current.
The kurtosis of the surface elevation, Benjamin-Feir Index (BFI) and directional spread are measures of nonlinear four-wave interactions and freak waves. The dependence of kurtosis, BFI and directional spread under typhoon conditions are examined by numerical simulations and field data. The Value BFI is significantly large in the fourth quadrant of the typhoon while the directional spread is small. The potentially possible area of freak wave occurrence is the fourth quadrant of the typhoon rather than the other quadrants. These parameters of freak waves under the typhoon condition are verified by the field data.
Annual maximum series data of the storm surge heights above the astronomical tide and the highest tide levels around Japan were analyzed in search of regional frequency distributions using the L-moments method. Storm surge height data at 34 stations were divided into five homogeneous regions, while tide level data at 36 stations were divided into different five regions. Both data sets had one station each that was excluded from homogeneous regions. Storm surge heights of the five regions were fitted to the Weibull distribution with the shape parameter of k = 1.01, 1.10, 1.57, 1.68, and 2.20, respectively, while annual highest tide levels of the five regions were fitted to the General Extreme Value distribution with k = -0.08, -0.10, and -0.18, and the Weibull with k = 1.19 and 1.92, respectively.
Shallow water wave hindcasting for each of many intense storms in the past 45 years was conducted on 3 sea areas of the Seto Inland Sea using input wind distributions of the wind data acquired at on-land meteorological stations around the Inland Sea. A high quality of wave hindcasting is ensured by a reasonable agreement between the measurements and hindcasts for many cases of the peak wave height during any given storm. Extreme value analysis yields an estimate of 50-year return wave height in the areas (H50) and the H50 value ranges from 10 to 12 m in the outer area connecting to the open ocean and from 3 to 8 m in the inner area depending greatly on the topological situations. Also the effects of AM or POT-based data extraction method data period length and other factors on the estimate of H50 are discussed.
Trend analysis and least-squares-based extreme value analyses are made for the data sets of annual maximum (AM) values and peaks-over-threshold (POT) values which are constructed from the long-year measurement data of sea wind speed wave height and storm surge height acquired in inner sea areas from Tokyo Bay to the western Seto Inland Sea. The analysis detects hardly any substantial trend with statistical significance in any of the AM data sets. The POT data analyses indicate the following findings: 50-year return wind speed is around 30 m/s depending on the sea area. 50-year return wave height ranges from 3.2 to 5 m in the inner sea areas and from 5 to 7 m in the areas partly exposed to open sea. 200-year return storm surge height is widely distributed from 76 to 372 cm in the inner sea areas.
This paper deals with statistics of wave climate. NOWPHAS data from 7 points along the sea of Japan are analyzed. Spectrum of a time series of significant wave heights from these points can be commonly divided into three frequency regions. Those regions are roughly (a) f<0.01 d-1, (b) 0.01<f<0.2 d-1 and (c) f>0.2d-1 in which d-1=1/day. Very few components in the (a) region give affect to the seasonal wave height change. Amplitudes of the Fourier components in the (b) region are almost uniform, but proportional to f-1 in the (c) region. Probability distributions of the data from (b) and (c) regions agree with the Gram-Charlier distribution. Statistical models which give probabilities of wave height that exceed any threshold value are shown in the last.
Sea extremes (annual maximum sea levels significant wave heights over a certain threshold etc) will be modelled with a temporal trend and they may be also governed by the climate factors e.g. Southern Oscillation Index (SOI). The fitting becomes better in general when any explanatory variable is added in the regression model. The sensitivity for the residuals should be examined to avoid the over-fitting. The outliers detection for extreme values can be firstly discussed by the degree of experience which we proposed in the previous study. It will conduct to the robustness of estimation. The judgements for the removal of outliers are demonstrated in a diagram of leverage and residual of extremes.
"Long swell" which attacked Toyama Bay from February 23 to 24, 2008, serious damaged breakwaters at Fushiki-Toyama Port. We used the wave prediction model which considered data assimilation to predict offshore wave. And we calculated wave transformation in a shallow water using Boussinesq-type wave transformation model (NOWT-PARI for short). We combined these both of them and built the prediction model of "Long Swell" at Fushiki-Toyama Port. To improve the precision of the made prediction model of "Long Swell", we built the wave watch system. We showed that a prediction of "Long-Swell" is possible to use wave prediction model and a wave watch system.
A 45-year wave hindcast system on a nested grid with high space resolution is applied to investigate the long-term wave conditions in the inner bay areas and inland sea of Japan from Tokyo Bay to the western Seto Inland Sea. The system is driven station by station by hourly sea wind distributions evaluated using both the typhoon-embedded analysis wind data archive in the open ocean and the land-based measurement(SDP) wind data in the bay areas. Reasonable agreement between the hindcasts and measurements for wave statistics in each bay area shows that the system is sufficient to closely reproduce the long-term wave conditions in the concerned sea areas but that use of other types of wind data archives with temporal homogeneity may be desirable to more properly evaluate the long-term variation of wave conditions.
A long-tied tombolo with the length of 800m at Chiringa-shima island emerges only during ebb-tide in spring tide condition from March to October. In this study, the physical processes of the emerging are investigated. The field surveys have revealed that the base underneath the surface sands consists of rocks, and a shallow area with huge amount of sand extends around the tombolo, and beyond the area, the sea bottom suddenly deepens. A series of numerical simulations for the wave fields, current fields, sediment motion and bathymetric changes were conducted. The result suggests that the emerging process of the tombolo can be explained by the seasonal change of thewind direction and the resultant changes in the wave fields.
The present study validates real-time wave prediction by using wind forcing from hourly analysis GPV during the typhoon season of Japan. Large variations in weather fields occurring in typhoon events provide an opportunity to assess the quality of forecasts. Observed data at 16 NOWPHAS wave gauge stations were used to evaluate predictions during the typhoon season. Deterministic statistic error indices showed that real-time predictions give good results. In addition, probabilistic statistic error indices that can evaluate errors in detail also showed that the correspondence between predictions and observations is good.
Accuracy of numerical simulations by Boussinesq model for long waves in harbors was investigated through a comparison between numerical and observed long wave spectra at two stations in Shibushi harbor, Kagoshima prefecture, Japan. To reduce the computational load, which is formidably large in two-dimensional computations for long waves accompanied by wind waves, one-dimensional Boussinesq model computation was conducted first, and incident wave spectra at a point closer to the harbor were estimated from the observed spectra. It was found that the long waves in harbors can be well predicted by using the model at least in terms of long wave spectral power and overall frequency distributions.
The generation of long waves due to air-pressure change was numerically simulated in model basins of uniform water depth, as well as East Chine Sea, where the location of air-pressure change moved in the eastern direction, using a nonlinear shallow-water model. Long waves of large wave height require larger change rate of air pressure. If the moving velocity of the location of air-pressure change is larger than the celerity of shallow-water waves, the wave height of generated long waves is not so large, while the wavelength becomes longer. The long-wave train increases the wave height near Kamikoshiki Island. The air-pressure change was simply assumed according to the air-pressure data actually observed on an island, resulting in a long-wave period close to that of the oscillation mode in Urauchi Bay, Kamikoshiki Island.
The propagation of infragravity waves in intermediate water depths near and in a port is investigated with field observations. A Green function method which is used for seismic interferometry is applied to estimate wave profiles of infragravity waves. The equations used in the method include Boussinesq equations with external forcing terms by fluctuating components of short gravity waves, atmospheric pressure and surface wind stresses. It is shown that when external forces are uncorrelated noise source, Green functions between two points can be retrieved by taking cross-correlation of records. The wave profiles of infragravity waves are estimated with the Green function and compared with the measured data. The agreement between them shows fairly good.
A large amplified meteotsunami (seiche) occurred at Urauchi bay of Kamikoshiki Island on 24-26 February 2009. This study investigated the meteorological conditions related to the event using surface observation and non-hydrostatic meteorological model. The convective systems around the cold front, changed from the stationary front, evolved as the pressure wave band in the middle of the East China Sea while the extratropical low occurred in the north of the front system. The propagation of the pressure wave bands was 20-30m/s corresponding to the low troposphere wind (about 1.5~4 km ASL). Continuous wavelet analysis showed that the peak frequency of the pressure fluctuation was ranged between 10 and 30 minutes, corresponding to the eigenoscillation period of the Urauchi bay of Kamikoshiki Island.
In Kumaishi Fishing Port, which locates at the coast of Sea of Japan in Hokkaido, troubles on cargo handling / mooring of fishing boats caused by harbor disturbance with rush of long-period waves became a problem in the harbor. Therefore, maintenance of long-period wave measure facilities by rubble-mound structures was planned. On this account, in this study, local wave observations were performed at the time of temporal construction of long-period wave measure facilities. Reflection factors of long-period waves in an actual scale were estimated in comparison with past local wave observation results and observed values to demonstrate long-period wave measure effects.
Using Graphics Processing Unit (GPU) the authors conduct a development of real-time simulation model of tsunami propagation and coastal inundation. As a preliminary result we terminated the estimation of tsunami travel time and the maximum tsunami height within 8 minutes for 60 minutes of tsunami propagation. The future perspective to develop a tsunami warning system based on the realtime simulation is also discussed.
A storm surge and high wave-induced inundation model was developed in this study, which consists of the Myers-based typhoon model, a depth-averaged flow model with an one-way nesting method, the wave model SWAN and an inundation flow model using a CIP method. The influence of high wave on storm surge and inundation flow was taken into account by adding the half value of the siginificant wave height obtained from the SWAN to storm surge fields at each time step. The validity of the model was verified by comparing with the experimental result of flood inudation due to dam break and the observation data of storm surge and inundation induced by Ise bay typhoon.
The aim of this study is to develop the practical numerical analysis model of tsunami to reproduce run-up soliton fission waves in the river. We verify accuracy and the computing time about the developed model by using the large-scale flume experiment. It has been understood that scheme of advective term and evaluation of space grid size have an effect on numerical result. We study the evaluation of space grid size and the computing time necessary for the calculation of the run-up tsunami with a soliton fission waves in the river. In this study, numerical algorithm of the soliton fission waves proposed by Shigihara and Fujima(2007) is very useful to calculate practically.
This study calculated the numerical simulation of Tsunami on Tokachi river with a soliton fission waves and examined the spatial resolution whichi is necessary for it. The soliton fission waves were recorded on video in Tokachi river at 2003 Tokachi-oki earthquake. We analyzed this video and obtained the continuous location information of soliton fission waves. The numerical simulation of Tsunami run-up using nonlinear dispersive wave equations model was able to reproduce the soliton fission waves. It is necessary to use the space grid size of 1/20 or less of the wavelength of soliton fission waves to calculate the amplification of the wave height by soliton fission.
In order to evaluate the decay process of far-field tsunamis, we have revised the definition of tsunami coda and related measures - moving root mean squared amplitude (MRMS amplitude), and non-dimensional tsunami amplitude (NDA) - in this paper. We applied these measures to the tsunami waveform data of 2006 and 2007 Kuril Island earthquake tsunamis observed by tidal stations located in Japan. Analytical results of tsunami coda of these events indicate that decay time constants obtained at tidal stations along the Pacific coast of Japan have only small differences. Further we confirmed that the NDA seldom exceed three times of MRMS amplitude. Then we show the possible application of utilizing these characteristics of tsunami decay process to realize appropriate cancelation of tsunami warnings.
A surge-wave coupling model was developed for numerical analysis of the disaster phenomena in coastal areas. The basic equations of storm surge model are the continuity and momentum equations integrated. The shear stresses the pressure gradients and the radiation stresses are taken into consideration in the momentum equations. The wave fields of a coupling model are calculated by MRI shallow water wave hindcasting model. The developed surge wave coupling model is applied to the estimation of the wave and water level generated by T7010 in the inundation area in the Kochi city.
The present study discusses the mechanism of abnormal storm surges occurred at Sakai-Minato located west coasts of the Sea of Japan due to Typhoon Songda. The storm surge simulations by SuWAT (Surge-Wave-Tide coupled model) are conducted with and without Coriolis forces and river discharges as well as using different calculation regions. The analysis indicated that the Coriolis force is a predominant factor for the abnormal storm surge with time lags of 15 hours at the coastal regions in the Sea of Japan if the meteorological forcing is correct. The maximum contribution of the Coriolis force to the peak storm surge is 70 %. The influence of the river discharge is insignificant. For the storm surges generated in the Pacific Ocean, the effect of the Coriolis force is insignificant as about 1 % for the Typhoon Anita's storm surge.
This study examines the sensitivity of tidal simulation to the model-domain size, and validates the accuracy of tidal simulation with broad model domain as the storm surge simulation. The tidal simulation models which adopt the driving force given by the global ocean tide model are developed. The models cover the Seto Inland Sea and coastal region of Kyushu-Shikoku with three different model-domain sizes. The size of the broadest domain is just as the storm surge simulation. Validation with the tide-gauge data shows that the accuracy is high in the narrower domain cases, however, in the broadest domain case the calculational error of tide becomes large in Bungo Channel and eastern coastal region of Kyushu.
It had been reported that many bridges in Sumatra island were suffered serious damages due to the 2004 Indian Ocean Tsunami event. We conducted the propagation and runup simulation of tsunami to clarify the external force acted on the bridges of the neighborhood of Banda Aceh. Numerical results showed that the horizontal tsunami force at Cut River Bridge and Lueng Ie Bridge were much larger than that of Kr. Ritting Bridge. A Safety Factor (S.F.), which could predict whether the girder slide occurred or not, was evaluated for the three bridges. The S.F. value was able to identify a clear difference in correspondence with the damage situation of each bridge. A present method is very simple, but it is applicable for the tsunami damage prediction of a bridge.
Not only hydrodynamic forces of tsunami current cause serious damage, but also driftage by tsunami may cause additional damage in a port and harbor area. Especially, in the 2004 Indian Ocean Tsunami, many stranded ships were observed. Therefore, when considering disaster prevention, the identification of expansion mechanisms of tsunami damage caused by drifted ships is very important. This study aims to develop a practical model to analyze the ship drifting and grounding motion by tsunami current. The simulated ship locus was consistent with the evidenced ship locus. And, as a hazard map, the potential damage due to ship drifting was evaluated by incorporating the possibility of collision with drifted ship.
Water surface profile and velocity distribution of 2010 Chilean earthquake tsunami was observed by using the Ultrasonic-type wave gage and ADCP placed on the sea bottom 11m deep and 3 km offshore of Choshi city, Chiba prefecture, Japan. The following findings were made. 1) Measured data can be linearly separated into tidal and tsunami component and wave component by adapting numerical filter with the threshold period of 60seconds. 2) Existence of tsunami does not have large effects on the wind waves and swells component distribution. 3) Observed tsunami current velocity profile was affected by the tsunami deformation due to the sea bottom topography. 4) The tsunami current velocity vertical distribution follows power low with 1/n=1/5 to 1/7. Based on these findings tsunami wave force acting on the small diameter ocean structure was proposed.
There have been many studies on tsunami forces acting on structures, but few studies on tsunami-induced water flows that move a lot of sands or soils, resulting in damages to such structures as road embankments and seawalls. In the present study, the damage of soil embankments by tsunami overflow is discussed. Hydraulic experiments on movable beds reveal that the erosion of the downstream slope and the scouring at the rear toe are important factors in the erosion of soil embankments. An erosion rate law is experimentally established. A numerical method based on the CADMAS-SURF to simulate the erosion of soil embankments is developed using these data. It is applied to gain insight into the Shuto diagram (2001) about the damages of embankment obtained from field data in the past.