When waves go upstream against gradually-increasing opposing current, wave breaking will occur and/or be blocked inthe case of strong current. Since wave action density is conserved in a current field, the wave action balance equation issuitable for the numerical model of wave-current interaction. In this study, the numerical simulation model based on thewave action balance equation is developed by using QUICK scheme in order to improve the accuracy of wave transformationdue to the current. The usability of the present model is verified, through the computation of the wave heightchange in the fields where the caustic exists or the area around the mouth of a river.
This paper studies on interactions between morphological changes and wave and current fields around the river mouthduring severe storm. Installing six cameras, authors successfully captured the sand bar collapse around the Tenryu rivermouth when typhoon T0704 hit the site in 2007. Obtained images were analyzed with several image processing techniqueand, coupled with the other hydrodynamic data, showed clear evidence for interactive features of morphology changesand wave and current fields. Finally, numerical model based on depth-integrated non-linear shallow water equations wasapplied to the observed conditions and it was found that topography changes during the storm is one of the most essentialfactors that determine the characteristics of surrounding wave and current fields.
The characteristics of highly nonlinear waves reflecting from a vertical wall in intermediate water depths are investigatedboth experimentally and theoretically. The results of regular-wave experiments indicate that the changes in phasevelocity in the presence of reflected waves are significant. The phase speeds become smaller than those predicted fromlinear theory and can be relatively well described by the third-order interaction theory. The amplitude modulation in thereflecting wave field tends to be pronounced for incident waves having large wave steepness. In the modulation processthe sharp-pointed highest peak appears in every third wave. The nonlinear wave groups having bimodal spectrumundergo strong modulation and their maximum wave height in the group increases with time above the initial height.
The MPS (Moving Particle Semi-implicit) method has been proven useful in free-surface hydrodynamic flows. Despiteits applicability, the MPS method suffers from some shortcomings such as non-conservation of momentum and spuriouspressure fluctuation. By introducing new formulations for pressure gradient and new formulation of source term ofPoisson Pressure Equation, we have proposed modified MPS method, namely CMPS-HS (Corrected MPS with HigherorderSource-term) method, for the prediction of wave impact pressure on a seawall. The improved performance of theCMPS-HS method is shown through the simulation of wave impact problem in comparison with the experimental data.
This study provides dynamic analysis of fluid-elastic solids using MPS (Moving Particle Semi-implicit) method. A twodimensionalcode based on particle interaction model of MPS method is developed to simulate fluid-structure interaction.The government equations of elastic structures are characterized by material density, Young's modulus and Poisson'sratio in solids. Meshes are not needed in the MPS method, so that fracture as well as large deformation can be analyzedwithout mesh distortion. The applicability of the present model is demonstrated for dam breaking. The result is in goodagreement with experimental data. Furthermore, structure analysis in a breakwater is carried out with wave propagation.The results rose the possibility that the present model can simulate fluid-structures interaction.
This report describes large-scale simulations on impulsive wave pressures using CADMAS-SURF/3D (SUper Roller Flume for Computer Aided Design of MAritime Structure in 3D), which is a three-dimensional numerical wave tankunder development. This code is also adapted to the parallel computing system and multi grid size system. Prediction ofimpulsive wave pressure using empirical formula is one of difficult problems. So, the tool for the practical use is neededto solve this problem. In this report, the validity of this model for predictions of impulsive wave pressures is verified.Large scale simulations are conducted by using the field data and the occurrence of impulsive wave pressures is discussed. At the last, the calculation cost is reported for the practical use.
We have developed a new numerical scheme which combines a Eulerian scheme with Lagrangian particles to computefluid-structure interaction caused by impact pressure. The proposed numerical model was applied to several problemssuch as interaction between a solid structure with dam breaking, and hydroelastic due to impact pressure acting on a verticalwall. The Lagrangian particles are useful and available to capture an interface between different phases. The modelcan estimate impact pressure acting on a vertical wall, its deformation, vibration, stress and strain. We have investigatedthe three dimensional impact pressure field of structure in space and time and dynamic response of a structure caused bythe pressure.
This paper is intended to newly construct a two-dimensional multiphase flow numerical model “DOLPHIN-2D” withsolid-gas-liquid phase interaction based on a CIP method and an extended SMAC method, taking into consideration a Lagrange particle-based method for the dynamic motion analysis of multiple rigid bodies. The constitutive law of Bingham fluid is, moreover, introduced into the DOLPHIN-2D in order to compute non-Newtonian fluid dynamics. Theutility and validity of the model is verified by applying it to bore-floating body collision/dam-break problems and comparing with the laboratory experiment. The results reveal that the model can appropriately simulate not only thedynamic behaviors of complex shaped multiple bodies but also non-Newtonian fluid motion.
Ocean-wave dependence of sea surface roughness is investigated using a wave boundary layer model, which is based ona recently obtained knowledge of air-sea interaction. The momentum flux is estimated by solving the model with meteorologicaland ocean wave information, which is obtained by carrying out the WRF and WWIII simulations. When windwaveis predominant, the results show clear relationships between Charnock coefficient and the wave age, between theroughness normalized by significant wave height z0/Hs and the wave age, and between the z0/Hs and the wave steepness.On the other hand, when swell is predominant, the results do not show clear dependence of Charnock coefficient on thewave age, and that of z0/Hs on the wave steepness, although that of z0/Hs on the wave age is observed.
Field measurements of wind stress over the ocean surface were made by the combined use of theeddy correlation andinertial dissipation methods, which are referred to as “ECM” and “IDM”, respectively. The stresses obtained from bothmethods should agree well under ideal surface layer conditions. Thus, confirming the degrees of agreement between thewind stresses from ECM and IDM makes it possible to reveal how air-sea momentum transfer varies depending on wavefieldconditions. The present data show that the degrees of agreement depend significantly on the wave age and the deflectionof the wind stress vector from the wind direction. The deflection of the vector is caused by the presence of swell.
Behavior of vortices near the wind wave surface is measured by using three high-speed video cameras and the Moving Least Square (MLS) method for vorticity estimations. The high-speed video cameras can increase the sampling rate andthe MLS method can estimate accurate vorticity profile from randomly distributed velocity profiles. The Particle Tracking Velocimetry, PTV, is applied to measure two dimensional flow fields close to the interface in detail. The vorticesare generated at the surfaces by the interaction between the wind shear and wind waves. The generated vorticesseparate from the surface and fall down to the bulk water. The generated vortex increases their spatial scale by unitingthe other vortex.
Sinusoidal wave generation in shallow water produces higher order free waves. These high frequency free waves are responsiblefor wave height fluctuation in a wave flume. The wave height fluctuations are measured on a sloping bottom, and shown to be explained by KdV models. These high frequency waves also affects wave breaking and wave heightdecay after breaking, since phase difference between primary waves and second order free waves causes significantchange in local wave profile. Role of phase difference is confirmed by a series of experiment with varied slope locationrelative to wave maker
Splashing water jets produces numbers of sprays and finger-shaped secondary jet. In this study, they were visualizedusing a fluorescent imaging technique. It is found two mechanisms for forming sprays determine fundamental featuresof spray distributions and transport velocity:(A) ejection of sprays due to moving contacts between the splashing jet andreceiving water surface, and (B) fragmentation of finger jets into sprays through a surface tension instability. The spraysformed in (A), transported at high speeds, precede those formed in (B). Size distributions of the finger jets and sprays are found to be approximated by log-normal distributions.
The two-phase flow measurements using dual-tip void probe for bubble size and its moving speed measurements wereconducted for the surf zone breaking waves. The characteristics of air bubbles in the surf zone as void fraction, bubblenumber distributions, bubble size spectra, and turbulent characteristics of fluid velocity were measured with highly temporaland spatial resolutions in the wave flume. The horizontal and vertical distributions of void fraction and characteristicbubble size were measured and modeled empirically. The relationship between the void fraction and wave energydissipation are proposed for the surf zone breaking waves.
The characteristics of bubble entrainment process owing to a vertical circular jet at the water surface arediscussed. The two-phase flow measurements using imaging technique for two-dimensional bubble measurements, so called Bubble Tracking Velocimetry (BTV) were conducted in the laboratory. The dependence of the air bubble statistics onthe Froude number and salinity beneath the water surface is discussed both downward and rising bubbles. The clear dependence of bubble size distributions on the Froude number is obtained through the experiments.
Sizes and numbers of sprays vertically ejected by wave overtopping through a flip-through process were measuredusing a back-light imaging technique in this experimental study. It has been found from the measured spraysize spectra that there are three primary mechanisms to form the sprays in the flip-through overtipping;(a) fine sprays are formed at the initial stage of overtopping and are vertically transported at very high speed.(b) numbers sprays with wide-ranging sizes are formed via disruption of an air-pocket compressed under breaking waves.(c) sprays are deintegrated fromthe vertically enteded overtopping jets.
Spatial variation of time varying rip current and surface displacements around a rip channel were measured by arrayed five wave gauges and current meters in the breaker zone. It is found that small change in incident wave direction caused a significant shift of rip current pattern and grouping incident waves caused low frequency fluctuation in mean current velocity and mean water surface displacement. Although the effect of incident wave grouping on the low frequency motion was reproduced by the numerical simulation, the shift of flow pattern caused by the change in incidentwave direction was not reproduced perfectly.
The formation of rip currents affected by regularly arranged beach cusps has been investigated quantitatively using 2-dimensional Boussinesq type numerical wave simulation. The numerical model has nonlinearity, dispersion, wave runup on the beach and wave breaking effects. The rip currents are generated regularly at the edge of the each cusp. Shape of the beach cusp enhances the formation of rip currents. The rip current takes its maximum velocity at H/Lo=0.013 and corresponding surf zone similarity parameter 0.7-1.0. We found the constantly low water level point at the edge of the beach cusp.
X-band radar measurements have been applied for rip current observation. Radar images were collected at research pier HORS on a straight sandy coast. Rader images capture wave propagation, breaking and run-up motions. Hourly averaged images have been processed, which frequently display characteristic cross-shore patterns that resemble to neck and head of a rip current. Floater experiments and radar measurements were done simultaneously to confirm rip current pattern and flow speed. Wave propagation speeds on rip channel were smaller compared to that of outside with same magnitude of the offshore-wards drift speed of floaters. Temporal variation of wave celerities at inside and outside of a rip current was measured for several tide cycles. The difference was large for low tides suggesting growth of rip current.
An X-band nautical radar system was employed to examine wave run-up during a typhoon around the research pier HORS in Hasaki, Japan. Analyses on radar echo images were done to estimate temporal and spatial variations of water fronts by manual digitizing of cross-shore time stack images. Digitized instantaneous water fronts were validated with wave gauge measurements with an acceptable agreement. Long-shore distribution of mean shoreline positions and intertidal foreshore slopes were then estimated using time averaged images. Run-up, the height of discrete waterlevel maxima, was estimated from the digitized water fronts with the help of foreshore profile. Run-up variationswere parameterized with surf similarity parameter. Low frequency variances in run-up motion were observed, whichwere traveling in the longshore.
The ratio of water surface velocity to wave celerity is applied as the wave-breaking criteria, and the hydraulic experiment is carried out to clarify the relationship between Froude number and the wave-breaking criteria. The estimation method of wave-breaking criteria including flow effect is proposed and its validity is confirmed by comparing to the experimental result. In addition, we study the governing equation, space grid size and scheme of advective term in order to express the transformation of flow-ascending tsunami adequately in numerical simulation.
Wave braking is one of the most complex and important phenomena in the coastal engineering. For this reason, various empirical formulas based on the linear statistical approach have been developed to estimate breaking height and water depth, called to breaker index. This study presents an artificial neural network (ANN) model as an alternative method for predicting breaker index on a gravel and sandy beach. The published available experimental data for sandy beach is used as input system. Further, hydraulic model experiments are performed to obtain the experimental data for gravel beach. A fundamental three-layered feed forward type of network trained using the usual back-propagation training is developed to obtain breaker index from the input of the deep water height, wave period and sea bed conditions. The predicted breaking height and water depth confirmed usefulness of the proposed ANN model for the application of breaking wave.
Quasi-resonant four-wave interactions may influence the statistical properties of deep-water surface gravity waves such as a freak wave. The freak wave prediction method was developed by Mori and Janssen (2006) based on the quasi-resonant wave theory assuming unidirectional wave condition. In this study, the directional effect is considered for kurtosis estimation from directional wave spectra assuming a semi-empirical relation. The validation of the kurtosisestimation including directional information is performed by operational wave forecasts and observed data.
We present a corrected form of a universal frequency spectrum of wind waves in finite depth water under ideal generation condition made by Yamaguchi (1988). It is a generalized form (G-TMA) of the TMA spectrum proposed by Bouws et al.(1985) as an extension of the JONSWAP spectrum in deep water. The G-TMA spectrum covers other universal deep-water spectra such as the Donelan spectrum and shallow water extensions such as the FRF spectrum. We indicate that the supposed peak frequency fp in the G-TMA spectrum is actually not the peak frequency. This is remedied in our generalized form, with a non-trivial term. In addition, we present rather accurate approximations of spectral moments in terms of total wave energy and various average frequencies.
“Long tail” has been well known for the studies on the internet business. A long tail of statistical distribution shows peculiar properties. Long tail has the moments of lower orders diverge and does not allow us to calcurate the mean nor variance of the focusing variables. Though we belive that there are no such properties in random sea wave fields, almost all of the distributions of wave periods derived by previous theories have long tails. This isa big obstacle for the practical uses of statistical properties of random sea waves. This study shows a path to escape the difficulties.
Based on a 3 to 23-year comparison between hindcast and measured wave data at 20 widely separated stations in the Seto Inland Sea, the applicability of a long term wave hindcast system is investigated. The system consists of data sets of onehourly sea surface wind distribution made from ECMWF analysis/reanalysis wind data in the open sea and measurements at many stations in the Inland Sea, and a backward ray tracing model on a grid with high spatial resolution. One major finding is that the system yields reasonable estimates of the time variations of waves and the resulting climatic properties both at stations exposed to open sea waves and inland sea waves, although a consistent discrepancy appears in the comparison of the inland sea wave periods because of the poor quality of the measurements.
Shallow water wave hindcasting for each of many intense storms in the past 23 to 29 years is conducted separately in 3 sea areas of the Seto Inland Sea. The wind distributions are made on the basis of a spatial interpolation of wind data measured at many sea and coastal stations. An extreme value analysis yields estimates of 50-year return wave height and its confidence interval in the areas. A hindcast-based return wave height at each of 10 measurement stations is shown to give reasonable estimates. The return wave height is 9-11 m in the sea areas connected to the open sea, 7m in the strip areas of the inner sea where open sea waves propagate directly and a maximum of 5 m in the area dominated by local wave generation. A reanalysis of data in the public domain produces a rough agreement with the present estimate.
Least-squares-based extreme value analyses are conducted for annual maximum (AM) and peaks-over-threshold (POT) wave height data, extracted from measurements carried out over a period of 16 to 36 years at 37 coastal stations around Japan. The main findings are as follows:(1) a statistically significant trend is not detectedat almost all stations, (2) a POT analysis yields a more proper estimate of the 50-year return wave height (RWH) than an AM analysis, (3) the inclusion of unusual wave height data in 2004 and 2005 into the POT data increases the estimatesof both RWH and its confidence interval, (4) neither separating storm types nor censoring the data improves the efficiency of the RWH estimated in the POT analysis.
The confidence interval for R-year return wave height is examined for the design of coastal structures. It is, however, a relative discussion whether it is wide or narrow. Even when we have just a short record of extreme wave height, there is no restriction to estimate a wave height of a longer return period, though it is accompanied with a very large interval. Engineers have been bothered with a problem: how much wide interval is no worth being considered. As a solution we introduce newly the degree of experience for the estimation of occurrence intensity of extremewave height.
On February 24, 2008, a twin low pressure system caused very high wave condition on the whole coast of the Japan Sea. The Yorimawari-Nami, peculiar long period swells to the sea, entered Toyama Bay and gave unexpected damages in the coastal and port areas. This paper, therefore, describes the significant waves and spectrum profiles of the swells observed by the coastal wave observation network NOWPHAS. The significant wave height and period on thewave record reached 9.92m and 16.2s at Toyama Station on a submarine peninsula in the bay. The shallow water wave simulation with the Boussinesq model led the wave energy focusing in some areas due to the peculiar bathymetry with submarine peninsulas and steep slopes in the bay, called Aigame in the district.
High wave with long wave period, which is known by local residents as “Yorimawari-Nami”, hit the east coast of Toyama Prefecture on February 24, 2008. The high wave, in particular, caused huge human and property damages in Ashisaki district of Nyuzen town. The purpose of this study is, at first, to examine coastal damages induced by the swelllike wave throughout filed surveys. The characteristics of high wave deformation, wave overtopping of a gentle-slop type revetment and inundation flow in the residential area are, furthermore, discussed by performing wave hindcasting with a GFS-WRF-SWAN coupled model and numerical simulation on wave overtopping of the revetment in Ashisaki district using a two-dimensional numerical wave flume “CADMAS-SURF”.
Severe coastal disasters were caused by high waves due to winter depression along the Toyama Coasts on February 2008. Hindcast simulations of wind and wave were carried out for the Hokuriku area by using the GFS-WRF-SWAN Wave Prediction System, and these results were found to be enough accurate compared with observed data. It was also confirmed that a real time wave prediction system using hourly analyzed atmospheric GPV is usable for the prediction of ‘Yori-Mawari Wave’ along the Toyama Coasts. The wave concentration along the breakwater of Fushiki Toyama Port due to refraction and diffraction due to unique topography called ‘Aigame’ was simulated, and the concentration location corresponded well to the damaged location estimated from the damage analysis of the breakwater
Wave hindcasting study of so-called ‘Yorimawari-waves’, huge swell caused by the low pressures in February 2008 was conducted by using the meso-scale meteorological model (MM5) and the third generation wave spectrum models (WW III and SWAN). It is considered that both the reproduction of the wind fields of wide region precisely and the estimation of the swell propagations from north are the most important factors of such abnormal wave simulation. Numerical simulation results indicate that swell from NE and wind waves from NW play a key role of the generation of Yorimawari-waves and it is clarified that the wave model need to improve its diffractive effects more for swell-dominant wave field simulation.
As a part of the intensive research on storm surges and high waves in Japan Sea, we conducted both tide observation data analysis and numerical simulation of weather and surge fields by the atmosphere-ocean coupled model for the severe storm in the central Japan Sea in February, 2008. It was made clear by numerical simulation of storm surge and data analysis that the range of surge heights generated by the storm that caused severe damage on the coastal structure, was 20 to 40cm. However, the actual tide record includes the long-period sea level changes, such as continental shelf seish, bay oscillation and surf beat, except wind waves.
Recently, the disaster that high waves attack any coastal area increases. For example, the caissons were moved by extreme forces of long period waves in Kuji port, Iwate Prefecture in 2006 or the boat was capsized by the high waves suddenly appeared in a calm weather in the Kurihama Bay, Kanagawa Prefecture in 2004. Such waves are caused by the swell propagated into the shallow water. In this study, the characteristics of the swell propagations along Japanese coast are classified by using ‘TIME LAG’ defined by cross-correlation analysis on swell data observed at several stations in NOWPHAS
Historically, many disasters at Shimoniikawa Coast has occurred by high waves called Yorimawarinami from east. The direction of longshore sand transport of the Shimoniikawa Coast is eastward. In Nyuuzen-machi and Asahi-machi located in the east from the Kurobe River mouth which becomes the source of supply of the sand, many houses moved because of beach erosion. In such circumstances, intense high waves by the low pressure hit these areas on February 24, 2008, so that coastal protection facilities in Kurobe-shi, Nyuuzen-machi and Asahi-cho got damaged and, furthermore houses were flooded and destroyed by overtopping waves. In this paper, various observation data in Shimoniikawa Coast of Yorimawarinami was analyzed by and the cause of the disaster was considered.
High waves attacked fishery harbours and towns faced the Japan Sea both in Niigata and Toyama prefectures on February 24, 2008 and brought sever damages in the breakwaters and houses. The generation mechanism of the high waves has been investigated using the weather charts, wind velocity data, wave and tide gage data. The significant wave heights reached about 6m or higher at many ports along the Japan Sea, especially, the waves with a wave height of 9.92m and period of 16s were observed at Toyama port and harrbour. Long period waves have also been observed under the condition of high wind speed and high waves and exited continental edge waves of 0-th to second modes. The weather conditions for generating abnormaly high waves called the Yorimawari Nami has also been investigated.
This study developed a real-time wave prediction system by using SWAN (Simulating WAves Nearshore) with the data of hourly analyzed atmospheric GPV (Grid Point Value). The validity of this system was carried out by comparing the predictions with observations from NOWPHAS (Nationwide Ocean Wave information network for Ports and HArbourS) during one month of December, 2007. The predicted wave heights and periods well corresponded to the observed ones. By using this system, wave situations at present and short-term future can be estimated for locations where there are no wave observation equipments.
Long period waves are sometimes significantly large and cannot be neglected even for engineering purposes in the nearshore zone. The growth of amplitude of long waves and phase shift between short wave envelope and incoming long waves on sloping bottom have been studied but not clearly understood. We present a theoretical model for the long wave evolution on the basis of the linear long wave equation with forced term and the model of breakpoint forced long waves. The present model shows good agreement with experiment results of random wave field. Then this model was applied to field observation data, and we show that the amplitude of long waves inner surf zone could express with squared of significant wave height.
The characteristics of infragravity waves are different depending on the frequencies and the sea areas. We analyzed wave height data of four components (30-60s, 60-300s, 300-600s and 600s-) of infragravity waves compiled by the NOWPHAS system. The spectral characteristics of the changes of the longest component (600s-) with the time lapse are very different from those of shorter components. The wave height longevities of the longer components are smaller than those of the shorter components. The wave height longevities in the southern areas are larger than those in the northern areas. When only the longest component develops and the shorter components do not, the significant wave height does not increase.
Field observation have been carried out to investigate spectral characteristics of natural wind velocity. Laboratory experiments on response of wind waves under periodically fluctuating winds have also been done. Spectral density of natural fluctuating wind velocity indicates a f-1 decay low. The measurements of the wind velocity and air pressure above wind waves show that the fluctuating components of wind velocity change in phase to the water surface elevations, while those of the air pressure change almost out of phase to wind waves. In the experiments, long-period waves appear to be generated by the periodically fluctuating components of wind velocity, not by those of air pressure through the spectrum analysis of the fluctuating components of the water surface elevation, wind velocity and air pressure.
This study investigates a coupling field of grouped wind waves and the resultant long waves on a bar type beach. The numerical results based on the Boussinesq equation have been compared with the analytical ones for the long waves generated by the time varying breaking point mechanism. The comparison aims at elucidating the mutual interaction process between the wind waves and long waves. The influence of the long waves on short waves is found to be significant for after-breaking zone through the alteration of eddy viscosity coefficient.
In the Segami River flowing into Hitachi Port facing the Pacific Ocean, swell waves with solitons in front often propagate from the river mouth, causing the inundation damages. These damages are closely related to the wave run-up of the long-period waves. Field observation of these long-period waves was carried out to investigate the mechanism of the propagation of long-period waves. Waves with an extraordinary long period of 1, 200s, which was generated as seiche on continental shelf, were observed as well as the long period waves with the period of 100s.
In Urauchi Bay of Koshiki Islands, long-period oscillations are observed in early spring. These oscillations are called “Abiki” as is the case with Nagasaki Bay. In this study, numerical calculation of long-period oscillation has been performed by applying the nonlinear shallow-water equations to simulate the oscillation characteristics of Urauchi Bay, which branches into two bays. Numerical results are compared among various types of bays, including I-type, L-type, and T-type bays, the last of which resembles to Urauchi Bay in bay shape. Urauchi Bay shows the second-mode oscillation corresponding to that of T-type bays. The first-mode oscillation can be large especially in Kojima Fishing Port.
The Lattice Boltzmann Method (LBM) has been developed as a new and promising numerical model to solve the fluiddynamics. In the present study, the applicability of LBM for the Shallow Water Equations is discussed in terms of tsunamirun--up problem and its accuracy that relates spatial resolution. As a result, we found that LBM can be applied tothe tsunami run--up problems, and the LBM Solution has the accuracy as same as that of Finite Difference Method inthe same spatial resolution.
The effect of coastal and river forests on tsunami run-up in a river has been investigated by numerical simulations basedon two-dimensional non-linear long wave equations. A simple coastal and river topography is considered where theriver course is straight and perpendicular to the shore line. The coastal forest zone consists of Rhizophora apiculata woods of 200m wide in the offshore side from the shore line at high tide and Pandanus odoratissimus woods of 100mwide on the backshore. In the river, Rhizophora apiculata woods of 1000m long are placed on the high water channelnear the river mouth. The results of numerical simulations with and without forests suggest the possibility that the coastaland river forests can reduce the run-up not only on the coast but also in the river.
The numerical simulator (STOC) has been developed to estimate tsunami damage on coastal areas in which there are lotsof structures. STOC consists of two sub-models: three-dimensional and non-hydrostatic model (STOC-IC) and multilayerhydrostatic model (STOC-ML). The present model, especially STOC-IC is validated by experiments on flowaround a rectangular weir and tsunamis passing a tsunami breakwater. In addition, applying the combination model of STOC-IC and STOC-ML to tsunami simulation in actual topography of harbor, it confirms that non-hydrostatic phenomenaappear around seawalls and bending section of channels.
Mesh division of geographic model effects on results of run-up tsunami simulation. Recently, approximate accuracy ofgeographic models increase with using LIDAR (airborne laser). And run-up tsunami simulation using mesh of severalmeters came to be done. In this study, we study effects of mesh division on inundation area of run-up tsunami simulation.To exclude the effects of various conditions of the tsunami simulation, we designed the simplified probability calculationmodel. The calculation model determines the reaching probability of tsunami inundation area. As a result, the tendencythat inundation area of run-up tsunami simulation decreases as mesh size decrease was theoretically confirmed.
It is important to establish a method to analyze the behavior of tsunami runup in river, because the effect of tsunamishould be considered in the examination on earthquake-resistant performance of river structure. Thus, the numericalsimulations of 1983 event in the Yoneshiro River were conducted using both dispersive and non-dispersive nonlinearlong wave equations. The simulation with previous one reproduced an undular bore which was observed in 1983.However, both of them provided the similar maximum tsunami heights in the river, and the computed heights agreedwith the measured heights well.
To predict the tsunami profiles along coasts precisely and quickly based on the tsunami profiles observed at offshore locations, the present study improved the inversion method by using the epicenter as a priori information. Numerical experimentswere conducted around Nankai Trough. Based on the tsunami profiles observed at 5 offshore locations for 15minutes, the initial water surface distribution and the tsunami profiles along coasts could be predicted with good accuracy.Sensitivity analysis was also conducted to investigate how the number of the offshore observation points and the length ofobservation time affect the accuracy of the prediction. Condition number and Dirichlet spread function were confirmed tobe effective for estimating the accuracy of the prediction even if the true initial water surface distribution was unknown.