ABSTRACT A tsunami wave effect on the oil tank storage was numerically investigated with the Fluid-Structure Interaction (FSI) analysis. The calculation was conducted for the tsunami water tank experiment in which the scale ratio of modeled tank to real tank storage was 1/100. In order to assess the dependency of tank stiffness on fluid and tank motion, the Young's modulus was parametrically changed and the stress on tank side was calculated. When the tsunami arrived at the tank and the hydrodynamic force took a maximum value, it was found that the fixed tank with high stiffness had large risk for side wall buckling that led to oil spill. INTRODUCTION When the Great East Japan Earthquake occurred in 2011, a massive tsunami arrived at Miyagi prefecture, and 22 oil storage tanks at Kesennuma bay area were broken and collapsed by the tsunami and a large amount of oil were flowed into the sea. The spilled oil was fired due to the sparks from washed automobiles and broken electric wires, and the second fire disaster caused serious damage (Zama, 2012). This tsunami fire disaster due to the oil spill from tank storage was firstly observed at the 2011 Earthquake although Japan has experienced many earthquakes. However, the ocean countries that have many industrial parks at bay area take a risk of tsunami fire disaster and we have to develop the risk management system for such accidents. Especially, in Japan, the Nankai Trough Earthquake assumes to be occurred in a few decades, we also have to find a new innovative technology to prevent the accidents related to tsunami. Although tsunami simulation has been carried out by many researchers with the standard grid-based Computational Fluid Dynamics (CFD) or particle method, the advanced CFD coupling oil spill with multiphase flow calculation was not frequently conducted. Kyaw (2017a) carried out the hybrid simulation coupling a generic tsunami calculation in a wide area with oil parcel tracking. In their study, the initial condition of the oil parcel tracing was given based on the estimation by the Osaka Prefecture Petrochemical Disaster Prevention Cabinet Headquarters (2014) and the setting of the initial amount of spilled oil were arbitrary. As discussed in Kyaw (2017b), the motion of oil tank storage in tsunami flow can be classified into some types such as sliding, floating, collapse, slide buckling, and bottom drop. Kyaw (2017b) investigated the dynamic motion of tank storage under tsunami overflow with multiphase CFD but they did not considered the structure change of tank body by hydrodynamic force. The Fluid-Structure Interaction (FSI) analysis is a useful simulation method for evaluating the relationship between hydrodynamic force of tsunami and structure change of tank but it requires large numerical cost compared with a pure CFD for solving only fluid motion, and the study on dynamic tank motion with FSI analysis is not so much reported. Sugatsuki (2013) carried out the simplified FSI analysis to investigate the tank damage by tsunami but the method of the FSI was one-way coupling and the effect of structure deformation was not considered in the calculation of fluid region. In order to investigate the relation between hydrodynamic property of tsunami and structural property of tank precisely, we have developed the simulation method based FSI analysis for the multiphase flow around single tank storage and estimated the risk of tank buckling due to tsunami.

ABSTRACT In the whole Pacific Ocean, the probability, which a super-great earthquake of Magnitude 9 will occur, is by no means low, like the 1960 Chile earthquake, the 1964 Alaska earthquake, the 2004 Sumatra earthquake, and the 20" off Pacific Ocean of Tohoku earthquake. Moreover, since the scale of the topographical change by the supergreat earthquake will become large, the prediction of the topographical change is important. Therefore, in this research, by proposing a method for determining a bed load coefficient rationally, the working efficiency of topographical change prediction by a tsunami is raised. INTRODUCTION As numerical simulation models for predicting topographical change by a tsunami, there are models of Fujii et al. (1998), Takahashi et al. (2000), Nishihata et al. (2006), Kihara and Matsuyama (2007), Nakamura and Mizutani (2008), and Ca et al. (2010). The calculation process of these models is as follows: The distribution of fluid velocity is gotten using nonlinear long wave theory. Then, the bottom distribution of non-dimensional shear stress (i.e. Shields parameter) is gotten using the fluid velocity. The distribution of bed load rates is gotten using an empirical equation with the non-dimensional shear stress. Moreover, the distribution of deposition rates of a suspended load and entrainment rates from the bottom are gotten using the kind of a diffusion equation or empirical equations. Ground surface elevation is gotten by substituting bed load rates, deposition rates and entrainment rates to a continuity equation of sediment. (4) The calculations (1)∼(3) are repeated. In the above-mentioned process, a suitable coefficient of the empirical equation for the bed load must be selected in order to obtain correct bed load rates. For the purpose, we need some verification simulations for every coast. However, supposing we can produce diagrams which can determine the suitable coefficient by setting basic parameters instead of the verification simulations, laborsaving of the topographical change prediction can be attained. Therefore, in this research, to a numerical simulation model using Ribberink's formula (1998) as the empirical equation for getting a bed load rate in an oscillating flow, the diagrams were produced by inverse analysis using scour data from many hydraulic experiments in our lab with a scale of 1/20. Especially, since we could not find existing experiments conducted by changing uniformity coefficients and dry densities, our experiments were implemented by changing median grain sizes, uniformity coefficients, and dry densities. Moreover, we confirmed that numerical simulations using bed load coefficients taken from the diagrams can reproduce topographical changes by the huge tsunami due to the 2011 off Pacific Ocean of Tohoku earthquake.

ABSTRACT Submerged wall and trench systems are investigated in this study to identify their usability as a defense measure against tsunamis. This paper focuses on several arrangements of submerged wall and trench systems via a numerical model, which is calibrated by physical experiments of tsunami like wave transformation. A dam break event is used to model a tsunami like wave interaction with structures. Resulted wave properties were investigated to identify an optimum solution from the viewpoint of structure configuration. Our results clearly show that the submerged wall-trench systems at near-shore can suppress the impact of tsunami like waves on the shore. INTRODUCTION A tsunami is a water waveform (or series of waves) that occurs in the ocean (in most cases) that sends a surge of water, sometimes reaching heights of over 30 meters on to land. This kinetic energy within the wave front can cause widespread destruction when they land onshore. The most common tsunami generation mechanism is an abrupt vertical displacement of a large area within the epi-central region (or its vicinity) associated with a strong submarine earthquake. However, massive submarine landslides and the fall of large soil masses from steep slopes cannot be excluded in generating tsunamis. The amplitude of a typical tsunami wave in the open ocean is rather small (normally from 1 to 30 centimeters) (Goring, 1992). Its length can reach hundreds of kilometers, which is much greater than the depth of the ocean. As this wave approaches the shore, with decreasing speed, its amplitude increases dramatically, which can lead to the destruction of various facilities in the coastal zone leading to flooding. Coastal dikes and sea walls are the most popular defense structures against tsunamis, which are commonly used across the world. Countries like Japan have strongly invested in building such structures along its tsunami vulnerable areas, especially since 2011 great east Japan tsunami, which has been identified as one of the largest natural disasters in recent history. Some of these structures can reach a height of 17 meters with a length of several kilometers along the East coast of Japan (Raby, 2015). There are however increasing concerns of cons to their pros with these massive walls that are built within socially and environmentally sensitive areas.

ABSTRACT Offshore tsunami records recorded by deep ocean bottom pressure gauges are very helpful in developing a real-time tsunami forecast system and substantially improving theoretical understanding. In these studies and systems, a simplified proportional relationship in which the ocean bottom pressure changes are linearly proportional to tsunami wave height is usually used under the long-wave approximation. However, the ocean bottom pressure changes should be attenuated by the product of the wavenumber of the tsunami and the water depth at the location of observation. In order to evaluate this effect of this attenuation, we investigated a short-wavelength tsunami in real-time tsunami forecasting. INTRODUCTION Widely distributed offshore tsunami observation networks, such as the Deep-ocean Assessment and Reporting of Tsunamis (DART; Bernard and Meinig, 2011; Table 1; Fig. 1), have tremendously improved the theoretical understanding of tsunami propagation, such as confirming the existence of dispersive waves (e.g., Saito et al., 2010; Saito et al., 2011; Miyoshi et al., 2015; Baba et al., 2017). DART stations, which has 51 stations in the world ocean as of March 22, 2019 (Fig. 1), are located at sites in regions generating historical destructive tsunamis to perform real-time tsunami forecast. Each station is located at the range of water depth from approximately 2,000 m to 6,000 m (Table 1). In addition, recent dense offshore observation networks will possibly help us to develop a real-time tsunami forecast system for reducing damage (Tsushima et al., 2009; Baba et al., --- Yamamoto et al., 2016a; Takahashi et al., 2017). These observation networks consist of a number of ocean bottom pressure gauges connected by a satellite network or optic fiber cables to transfer the data in real time. For example, the Dense Oceanfloor Network system for Earthquakes and Tsunamis (DONET; Kaneda et al., 2015; Kawaguchi et al., 2015; Fig. 2) has been constructed by the Japan Agency for Marine-Earth Science and Technology (JAMSTEC) and started observations in August, 2011. DONET is consisted of 51 stations in Kumano-nada and off Kii Channel, to monitor earthquakes and tsunamis in the Nankai Trouph region southwest Japan. These stations are located at the depth of approximately from 1,000 m to 4,500 m. The data is transmitted to research institutes and universities in real-time, and improves precision and warning times of earthquake early warning and tsunami warnings and/or advisories by Japan Meteorological Agency (JMA). DONET is currently operated by the National Research Institute for Earth Science and Disaster Resilience (NIED).

ABSTRACT Tsunami-induced bottom shear stress has commonly been evaluated using a friction coefficient for steady flow, such as Manning's roughness coefficient, Chézy formula and Darcy-Weisbach equation, assuming development of bottom boundary layer up to the water surface under tsunami. However, applicability of steady friction law to tsunami wave has never been scientifically investigated, and it has simply been assumed that long-period wave motion satisfies quasi-steady flow condition. In the present study, the k-ω model is used to calculate velocity profile and resultant sea bottom shear stress under shoaling hypothetical tsunami propagating from tsunami source area to nearshore region. It is found that steady friction is not valid in the entire computational region from the source area to shallow region due to extremely thin boundary layer thickness beneath tsunami. INTRODUCTION In numerical modeling of tsunamis, bottom shear stress is usually evaluated by Manning roughness coefficient n or friction coefficient f which is valid under steady flow motion. Tsunami energy loss due to bottom friction is an important physical process causing attenuation of tsunami wave height during wave shoaling process. In addition, when we deal with sediment movement and subsequent sea bottom morphology change due to both bed load and suspended sediment movement, it is highly important to estimate accurate bottom friction exerting on the sea bottom, since both modes of the sediment transport rate have generally been formulated as a function of acting bottom shear stress (e.g., Takahashi et al., 2000). It is well known that in a steady open channel flow, bottom boundary layer develops up to the water surface, whereas the thickness of bottom boundary layer under wind-generated waves is extremely thin as compared with water depth (Jonsson, 1966), resulting in a fact that the water depth is not dominating representative length in turbulent sea bottom boundary layer. In addition, it is noted under wave motion that due to steep velocity gradient near the sea bed, bottom shear stress acting on the sea bottom surface is much larger than that estimated from steady friction coefficient. Amid these two limitations of "steady flow friction law" and "wave friction law", a question arises which is valid under tsunami wave as depicted in Fig. 1. This is a research motivation for the present investigation.

ABSTRACT In this paper, we apply a coupled numerical method that employs Smoothed Particle Hydrodynamics (SPH) and Distinct Element Method (DEM) to solve failure problem of a breakwater caused by long-lasting tsunami. We conduct case study simulations to confirm effects of countermeasures, namely embankment, foot-protection block, inclined parapet and trapezoidal caisson, on stability of composite type breakwaters. Moreover, we discuss how to deal with simulation results for evaluating stability of breakwaters on large deformation problems such as caisson collapse. INTRODUCTION In the Japan's 2011 Tohoku Earthquake, caisson breakwaters in several ports were seriously damaged by tsunami. Damage investigations after the tsunami showed that scouring and seepage affected the destruction of breakwaters as well as horizontal tsunami forces, and a long-lasting effect was characteristic. These complex mechanics had not yet been considered in Japanese design standard before the tsunami hazard. Thus, the analysis on the complex failure modes is necessary to improve the design criteria against tsunami. Hence we have developed a coupled analysis method (Iwamoto, Nisihura et al., 2015–2019) that employs Smoothed Particle Hydrodynamics (SPH) and Distinct Element Method (DEM) to solve such complex problems. Fluid is modeled by SPH, solid objects such as the foundation rubble mound and the caisson are modeled by DEM, and the both scouring and seepage effects are considered to use locally averaged Navier-Stokes momentum equations (Anderson and Jackson, 1967) and Gidaspow drag model (Gidaspow, 1994) which consists of the Ergun (1952) and Wen-Yu (1966) Equations. Numerical parameters were determined from permeability tests and triaxial tests, moreover, simulation results were validated by comparing with scouring failure tests of a breakwater using a drum centrifuge apparatus. As a result, our method can simulate following physical phenomena, overflow and scouring, loss of embankment and foundation, decrease of foundation stability, and caisson collapse that were caused by long-lasting tsunami. Here we try to apply our method to the stability evaluation of additional countermeasures against long-lasting tsunami for composite type breakwaters. Furthermore, we discuss how to deal with the numerical simulation results, and what are important points to evaluate structural stability on large deformation problems such as scouring failure caused by loss of foundation mounds.

ABSTRACT The 2011 Great East Japan Tsunami caused enormous damage to coastal urban areas in the Tohoku region of northeast Japan, despite numerous engineered breakwaters. Conventional two-dimensional shallow-water models of urban inundation, and tsunami wave-pressure models that approximate hydrostatic pressures on a single building, do not accurately predict actual flooding processes or tsunami wave-pressures. This study investigates urban flooding scenarios in which the tsunami overtop breakwaters, explicitly evaluating tsunami waveforms, typical breakwater shapes, and typical building placements based on model and numerical experiments. We propose use of specific energy rather than hydrostatic-pressure approximations to model wave pressures on multiple buildings. INTRODUCTION The Japanese government has classified the tsunami initiated by the 201 1 Great East Japan Earthquake as a Level 2- and Level 1 tsunami (L2 ts unami and L1 tsunami, respectively) and has recommended implementa tion of additional disaster-prevention activities in response to each type of tsunami. The basis of this disaster-prevention activity is a tsunami ha zard map, which depicts the supposed inundation caused by a tsunami b ased on predictions. Such a map has been created and released in almost all coastal municipalities in Japan (Shuto et al.,2007;Arikawa et al.,2013). However, these inundation simulations are based on a two-dimensio nal shallow-water model in which the effects of coastal structures such as breakwaters cannot be evaluated. Previous studies have pointed out t hat the actual range and level of inundation during the 2011 Great East Japan Earthquake was vastly different from the expectations recorded o n previous tsunami hazard maps and have suggested that these differenc es are caused by factors such as the breakwater shape (Yanagawa et al., 2016). Simply put, to improve the accuracy of conventional tsunami haz ard maps, it is necessary to use a three-dimensional numerical model ca pable of taking into account the three-dimensional arrangement and sha pe of coastal structures, rather than rely on planar two-dimensional mod els(Nagayama,2016;Nagayama,2017).

ABSTRACT Experiments are conducted to investigate the solitary wave-induced forces and moments on a submerged horizontal plate in the threedimensional situation for the first time. The hydrodynamic loads are measured by an underwater load-measuring system. The measured wave loads acting the plate at four submerged depths are presented to understand the characteristics of the hydrodynamic loads. The horizontal force has a maximum uplifting force and a maximum downward force in the test cases. The variation of the vertical force and pitch moment depends on the different submergence depth. The loads keep fluctuating after the solitary wave leaves away from the plate in the threedimensional situation. INTRODUCTION The flat-form structure works as a typical coastal construction, including the wave energy absorbing or converting devices, floating structures, docks and the bridge-deck (Yu, 2002). When the waves impact on the flat-form structure, the vertical and horizontal forces dominate the hydrodynamic loads. Since the length of the flat-form structure is larger than the thickness, the vertical force is usually blamed for the structure destruction. Apart from the exceeded vertical force, the pitch moment caused by the vertical force may lead to the overturn of the whole structure. Most constraints on the flat-form structure are fabricated to overcome the vertical force resulted from the severe wave conditions. While the vertical force defeats and destroys the opposite constraint, the structure will get out of control and float away due to the not massive horizontal forces (Hayatdavoodi and Ertekin, 2016). Fully understanding the hydrodynamic loads on the flat-form structure brings great benefits for the coastal structures and bridges. When it comes to the severe wave conditions, the tsunami deserves the most destructive wave in coastal waters. The impact of the tsunami and the consequent bores on the flat-form structure have received extensive concerns in the recent years (Iemura et al., 2007; Kosa, 2011; Istrati et al., 2016), especially after the 2004 Indian Ocean tsunami and the 2011 Tohoku tsunami. The offshore ocean platforms and the coastal bridges usually sustain the great damage under the attack of tsunami (Mori et al., 2012). Maruyama (2013) indicated that it is the earthquake-induced tsunami ruins the bridge-deck, instead of the shake from earthquake. Although the real waveforms of tsunami wave is of inconvenient to be simulated in laboratory, the tsunami-like wave, such as the solitary wave, has been widely used as the simplification to study the tsunami evolution (Hsiao and Lin, 2010), run-up and the maximum impact on structures. As a consequence, the research of the solitary wave loads on the flat-form structure is deemed to understand the performance of the coastal structures under the action of tsunamis.

ABSTRACT In the present paper, the liquefaction potential and the change in effective stress associated with the gap between the water pressure on the ground surface and the pore water pressure in the ground induced by tsunami inundation have been numerically studied. From the numerical analysis, the following results have been found: The effective stress increases as the inundation depth increases. In contrast, degradation of the ground occurs as the effective stress decreases during the lowering of the water level amidst a tsunami. The degradation of the ground extends to a maximum depth of 5.0 m. INTRODUCTION The 2011 Great East Japan Earthquake and the associated tsunami (the 2011 Tsunami) were massive events which left approximately 20,000 people dead or missing (The Cabinet Office of Japan, 2015). The tsunami was the largest in over 1,000 years of tsunami history in Japan. Pile-supported reinforced concrete (RC) buildings were uplifted and overturned by the tsunami, as shown in Fig. 1 (NILIM and BRI, 2011a, 2011b; see also, Chock et al., 2013). This kind of building damage has been reported only once before, i.e., a lighthouse was overturned by the 1946 Aleutian Tsunami. It is expected that the overturning may be caused when the overturning moment of the tsunami hydrodynamic force and the buoyancy force exceeds the resisting moment of the self-weight of the building and the pile-resisting force. It is known that large ocean waves cause the liquefaction of the sea bed (Oka et al., 1994). Sassa et al. (2001) studied the liquefaction of the sea bed and its propagation considering fully liquefied soil as an inviscid dense fluid. Liu et al. (2009) extended their theory by considering a two-layer viscid fluid model for the liquefied soil and the surrounding region. However, these studies deal with short-period waves, such as wind waves and swells. For a tsunami, it is necessary to study the effect of waves with longer periods, i.e., 10 to 60 minutes. The pile-resisting force can be affected by the liquefaction of the ground induced by the seismic ground motion and the successive tsunami.

ABSTRACT Targeting huge earthquakes and subsequent tsunamis, coastal structures require more improvements to enhance their durability and resilience. Thanks to recent studies, the accuracy of estimation of the wave forces by the overflowing tsunamis have been improved, in particular, for design of breakwaters. While, quay walls have not been studied sufficiently about such a situation. Some surveys shows the terrible damage of quay walls by the backwash of the tsunami. In this study, the wave force by backwashes of tsunamis acting on a quay wall are examined through hydraulic experiments, and influence of the permeability around the caisson is focused on. INTRODUCTION In Japan, there is increasingly concern over occurrence of huge earthquakes and subsequent tsunamis, namely the Nankai or Tonankai earthquakes or a huge earthquake directly hitting Tokyo area. From the circumstance, the coastal structures such as quay s and breakwaters have been widely improved as an urgent subject for enhancement of the durability against earthquake and tsunami disasters. However, the costs for the improvement of structures or grounds come to be significantly huge. In addition, the improvement work should be done with the targets being in service. Therefore, some economical and simple approaches are newly required to achieve the improvement. As for quays, they generally can keep the relatively good durability against surging tsunamis thanks to the soil pressure of the grounds behind the quays. On the other hand, the ground's assist does not work against backwashes of the tsunamis, and to make matters worse, it may deteriorate the damage due to push of the quay s by the soil pressure. As a result, the quays possibly have significant damages by the backwashes even if they come through the surging waves. This study implements hydraulic experiments with targeting a quay under backwashes of tsunamis. And, an effective and simple approach is proposed for enhancement of the durability of the quay against the tsunamis by fixing rubble mounds behind the target quay with injection of soil solidifier. The injection of soil solidifier does not include any complicated or large-scale process so that it is applicable to the practical construction of quay s being in service. The effectiveness to reduce the tsunami force is shown by comparing the values of tsunami pressure measured in the experiments. And, the trend of variation of the acting pressure depending on the hydraulic condition is shown.

ABSTRACT A tsunami scenario determination methodology based on probabilistic tsunami hazard assessment is demonstrated. First, contributions of potential tsunami sources to the hazard curve, which is the relationship between tsunami height and the annual frequency of exceedance, are analyzed to identify scenarios that would generate a tsunami at the target coast in a target height range. By considering similarities in tsunami wave form and current direction, similar tsunami scenarios are grouped, and major tsunami scenarios can be determined. INTRODUCTION Assessing the tsunami hazard of coastal industrial facilities requires the consideration of multiple scenarios with regard to tsunami arrivals and their potentially severe impacts on society and business continuity planning. Identifying the potential spatial and temporal distributions of inundation depth and velocity for such events can aid in these determinations. Probable maximum tsunamis (PMT) are sometimes used for identifications of tsunami inundation depths (e.g., Grilli et al., 2015; JSCE, 2012; U.S.NRC, 2016). Based on historical tsunami data and field surveys, potential tsunami source mechanisms are identified. By considering uncertainties regarding future tsunami generations, many tsunami source models are prepared. Among the source models, that of PMT is selected such that the highest tsunami height in front of the site were predicted (JSCE, 2012). However, the frequency of PMT is unknown. The probabilistic tsunami hazard assessment (PTHA) methodology is used to evaluate the scale and frequency of future tsunamis (e.g., Mori et al., 2017). The logic-tree approach often used for PTHA systematically considers both epistemic (lack of knowledge) and aleatory (random variability) uncertainties regarding future tsunami prediction (e.g., Geist and Parsons, 2006; Annaka et al., 2007; JSCE, 2016). The former are expressed by tree branches while the latter are expressed by probabilistic density functions for predicted tsunami heights. JSCE (2016) develops a PTHA methodology and shows examples of logic trees for various sea areas around Japan. The PTHA methodology proposed by JSCE (2016) was based on the probabilistic seismic hazard assessment (PSHA) approach proposed by Cornell (1968).

ABSTRACT Since the frequency which the tsunami by the super-great earthquake of magnitude 9 generates in the Pacific Ocean has increased to once for several decades recently, examination of measures to the huge tsunami is becoming very important. Therefore, this paper reports noteworthy facts which are obtained from investigation on hazard map creation by numerical simulations and tsunami measure examination. Furthermore, since a refuge mound made of soil is the one of long-life facilities, the development of a method which can evaluate the durability over scour of the mounds easily is also important. So, the examination result of this subject is also reported. INTRODUCTION In the whole Pacific Ocean, the probability that super-great earthquakes of Magnitude 9 level happen is by no means low, like the 1960 Chile earthquake, the 1964 Alaska earthquake, the 2004 Sumatra earthquake, and the 2011 off the Pacific ocean of Tohoku Earthquake. Therefore, the examination of tsunami disaster prevention and mitigation is very important. In this research, the following problems are assessed using numerical simulations and experimental data. When the local government of a province level performs tsunami flood predictions, the calculation range is wide, a grid interval of finite difference calculation is set to 10 m or more in many cases, therefore, water gates and land gates are neglected because they are shorter than the grid interval. When a super-great earthquake occurs, many of these gates will be broken and cannot be closed, consequently, a lot of sea water will flow inside. However, this phenomenon cannot be considered in their predictions. Moreover, although a banking road serves as a flood prevention wall, in the case that another road crosses it in a tunnel, a huge tsunami can enter through the tunnel of which this inundation also cannot be considered in their predictions. Although earthquake-proof high dikes are being provided on many coastlines in Japan, but if a huge tsunami ascends a river, a lot of water is flooded into the land area from a low river bank, the inundation water will pile up in inland because the strong dike will obstruct the drainage of the flooded water. Since the probability that a super-great earthquake generates at the same area is about once in 500 years, long-life protection and refuge facilities are in demand. Although the life of reinforced concrete buildings for refuges are about 100 years, the life of soil refuge mounds are about 500 years. However, in the case of soil mounds, countermeasures to destruction by scour are required.

ABSTRACT On the 2011 off the pacific coast of Tohoku earthquake, many of road embankments along the coast were less damaged. Following this, a lot of municipalities along the coast adopt a road embankment as a countermeasure against tsunami in Japan. This paper aims to make clear the performance requirements for road embankments in coastal areas. First, we reviewed damage examples of road embankment in the earthquakes, and we focus on two topics. The first concerns the settlement of road embankment. We introduce our field survey to measure the settlement of road embankment behind bridges in Mashiki town, where is damaged by 2016 Kumamoto earthquake. And we propose a performance index of settlement of the road. The next concerns the performance of the reinforced soil wall against tsunami. Here we introduce our water tank experiment to verify tsunami-resistant performance of the reinforced soil wall. Our experimental results showed that reinforced soil wall has high stability against tsunami if there no gap of front panels. From these results, we confirmed that it is important to determine the tolerance level to settlement of embankment even when we perform the capacity design of road embankment along the coast. INTRODUCTION The 2011 off the pacific coast of Tohoku earthquake (Mw=9.0) caused great damage of geotechnical structures in the vicinity of the eastern coasts. In contrast, many of road embankments along the coast, which is less damaged by the earthquake, prevented the tsunami and protected a lot of people against tsunami. Following this, a lot of municipalities along the coast adopt a road embankment as a countermeasure against tsunami in Japan. In recent years analytical technique of the road embankment structure was improved, and it became easier to estimate the deformation of the embankment in the earthquake. For these backgrounds, a performance design was adopted for the road embankment in Japan, and checking for the deformation of the embankment came to be conducted. However, there is no defined criterion about the performance of the road embankment required in earthquake, particularly the allowable deformation.

ABSTRACT A numerical model was developed for tsunami generation by fully submerged landslide. The model is based on a finite-volume Godunov-type scheme for nonlinear shallow water equations with a source term arising from prescribed bottom motion. The obtained numerical results on the influence of geometric variation in landslide mass revealed that the evolution of depressive wave became significant when the slope of the rear side of the sliding model was steep. The influence of geometric deformation was shown to be generally weak and the distribution of wave height was mostly governed by the shape of the slide in shallow area. INTRODUCTION In the past, destructive tsunamis have been induced by submarine landslides over the world (Lovholt et al. , 2017). In the offshore area of Norway, for example, submarine landslides have occurred repeatedly and resulting tsunamis are recognized as one of the important natural hazard to coastal areas (Bugge et al., 1988). It is also considered that giant waves were generated in the past around Hawaiian island by underwater landslides (Moore et al. , 1994). Recently, the evolution of Papua New Guinea tsunami in 1998 is considered to be influenced by a submarine slide (Tappin et al. , 2001). Nevertheless, the existing field observation (e.g., Bondevik et al. , 1997; Moore et al. , 1994; Tappin et al. , 2001) and experimental investigation (e.g., Enet et al. , 2002; Grilli and Watts, 2005) on tsunamis induced by submarine landslides are very limited compared with studies on co-seismic tsunamis induced by earthquakes. Accordingly, the characteristics of landslide-generated tsunami evolution over wider range of conditions awaits further investigations. In addition, it is necessary to gain deeper physical understanding on related mechanisms as well as to develop reliable prediction methods in order to design effective counter-measures against landslide tsunamis. For these purposes, numerical approaches are expected to be effective. In the numerical studies on tsunami generation by submarine landslide, various numerical models have been used such as non-linear shallow water equations, Boussinesq equations, and two-layer flow model (Yavari-Ramshe and Ataie-Ashitiani, 2016). On the basis of these models, the numerical investigations have been successfully applied to various cases recently. In some cases, however, the obtained numerical results are found to be model-dependent (The Tsunami Evaluation Subcommittee, The Nuclear Civil Engineering Committee, Japan Society of Civil Engineers, 2016); Different models might predict different characteristics. Furthermore, the characteristics of landslide tsunamis are related to various features of submarine slide such as geometry, submergence, and kinematic characteristics. The relation among these parameters and tsunami evolution seems to be quite complicated. Previous studies on the relation between tsunami evolution and the characteristics of submarine landslide are, however, relatively scarce. It is therefore important to conduct systematic numerical examination and to enhance physical understanding on the basic characteristics of tsunamis generated by submerged landslides.

ABSTRACT Debris generated during extreme events like the tsunami can impose substantial impact loading on structures closer in the coastal zone. Majority of design codes do not quantify the impact forces close to reality owing to uncertainties in defining the wave characteristics and a lack of knowledge in understanding the underlying physical processes. The present study focuses on the measurement of forces which a coastal structure would encounter due to the impact of debris in motion during such an extreme event. The study herein focuses the motion of the debris due to undular bores. 1:20 scaled model of structure and the debris were used. Experiments were conducted in a wave flume 72.5 m long and 2 m wide. A beach slope of 1:30 is laid to replicate the coastal zone. By varying the wave heights and time periods, different types of waves such as elongated single pulse waves, symmetrical N waves and unsymmetrical N waves were generated replicating the characteristics of a tsunami as close as possible. The impact tests were conducted using box shaped smart devices as debris of weight 4.62 kg, 5.82 kg and 7.02 kg. The debris is attached with an accelerometer for measuring the impact acceleration. In order to have a better understanding of the behavior of debris during the impact, a camera at a speed of 120 fps is operated. The force acting on the structure is measured with a load cell. The forces due to the velocity of the debris and the mass is compared with the force measured using load cell. The details of the testing facility, model parameters, test set-up, test procedure, analysis of results and discussion are presented in the paper. INTRODUCTION Debris generated during a tsunami or any coastal floods can strike the residential buildings, commercial or any other structures in the coastal zone. The debris generated during such process on striking the structure induces large impact loading on the structure. The magnitude of the force may be large enough causing local or global failure of the structure. The maximum force acting on the structure due to this tsunami driven debris is difficult to estimate as the impact of the debris on the structure is influenced by the mass, velocity, draft and the orientation of the debris with respect to structure. The study on the impact of debris on structure has received consideration, after the studies on impact of debris on the structure (Haehnel and Daly, 2004, Nouri et al., 2010, Riggs et al., 2013). Their study reported that debris driven during tsunami induces large impact loading on the structure compared to that due to the impact of tsunami alone. Further, the contact duration of the debris with the structure is less. ASCE/SEI 7–05 suggests contact duration of 0.03s between the structure and debris for estimating the impact force. Few guidelines for the design of structure around the globe have included the impact of debris driven during floods and tsunami on the structure (ASCE/SEI 7-05(2005), FEMA55 (2011), FEMA P-646(2012) etc.). These guidelines provide rational formulae for estimating the loads acting on the structure during a tsunami.

ABSTRACT The fate of the energy, momentum, and volume (or mass) of the leading-elevation N-type tsunami considering geophysical scale is studied. From the epicenter to the land the physical properties of the positive wave of tsunami are calculated using a second order depth integrated model. Not only the incident wave characteristics but also the scale and shape of bathymetry are very important for the propagation of tsunami physics. INTRODUCTION Tsunami is caused by undersea earthquake and usually propagates across the entire ocean. Tsunami is not detected well over deep and intermediate ocean area due to its very long length and small crest amplitude. As tsunami enters shallow coastal region, its wavelength reduces and the amplitude increases, which is believed to lead low altitude inland around coast to catastrophic inundation damage. For decades, various aspects of tsunami propagation have been studied experimentally, numerically and theoretically. Because the inundation is directly related with water surface elevation, interests have been focused on water surface elevation of tsunami. For example, Synolakis (1987) presented an analytical model for non-breaking solitary wave runup on plane beach. Briggs et al. (1995) presented intensively measured data of solitary wave runup heights around a circular island in laboratory scale. Matsuyama et al. (2007) carried out relatively undistorted experiment on the shoaling and fission of tsunami. Considering the complexity in real fluid motion and bathymetry, numerous numerical models based on shallow water equation (Titov and Synolakis, 1995; Li and Raichlen, 2002) and Boussinesq equation (Goring, 1987; Fuhrman and Madsen, 2008; Lynett, 2007; Kim and Lynett, 2012; Kalisch and Senthilkumar, 2013) have been developed and applied. Based on many analytical, experimental and numerical studies, it was proposed that the runup height and amplification of nonbreaking wave generally increased as the bottom slope was milder (Synolakis, 1987; Suh et al., 1997; Li and Raichlen, 2002). In field experience, however, tsunami damage has not been frequently reported where a continental shelf with very gentle slope existed, for example, the sea of north Australia, east China and west Korea. On the contrary, east coast of Japan, India and Sri lanka where the under sea bathymetries are relatively steep have experienced severe tsunami attacks. Recently, Madsen et al. (2008) and Kim and Son (submitted) studied on the contradictory results: They examined the importance of geophysical scale for the tsunami wave study and proposed that unrealistic wave could be developed under improper scale and geometry considerations. For the damping by friction Horsburgh et al. (2008) found that the frictional dissipation was not primarily responsible for tsunami attenuation using a numerical model on tsunami crossing ocean and continental shelf. This implies the damping could be resulted by the geometry of geophysical scale, not by frictional effect.

ABSTRACT To analyze tsunami risk in detail, drift damage should be considered using numerical tsunami simulations. However, no validation method for drift simulation has been established. This study proposes a validation method for drift simulation based on analysis of the actual drift of ships during the 2011 Tohoku Tsunami for three coastal areas in Japan: the Otsuchi and Kirikiri coasts in Iwate Prefecture and the Namie coast in Fukushima Prefecture. The analysis included a simulation that reconstructed tsunami-induced drift in these three areas. INTRODUCTION Tsunamis cause inundation and drift damage, such as destruction of structures and obstruction of road and port traffic. In the 2011 Tohoku Earthquake Tsunami off the Pacific Coast of Japan (denoted hereafter as the 2011 Tohoku Tsunami), drift damage, such as houses destroyed, were reported in various studies (e.g., Kawasaki et al., 2012). When tsunami risk is assessed, inundation height and area are mainly used as indicators of damage. However, to analyze tsunami risk in detail, drift damage should also be considered. In risk assessment of tsunami drifting, numerical simulation is generally used because we cannot determine specific drift tracks in actual tsunamis. Validation of a simulation model requires a reproduction simulation using drift position data from a tsunami. Among the things that drift during a tsunami, such as cars, ships, containers, and debris, ships are the best choice for verification because their initial position before the tsunami can be reasonably determined, that is, they are usually moored at a port. The initial positions of most other drift objects are generally unknown. However, data on the drift position of ships following tsunamis are rare, and few simulations have attempted to reproduce tsunami drift at actual coasts, except for studies like Suga et al.,2012 and Hashimoto et al, 2009. For validation of a drift simulation, one could use the sensitivity analysis considered by Suga et al.,2012 and Shigihara et al., 2016., but this method requires tremendous calculation effort. In other words, a standard method has not been established.

ABSTRACT Tsunamis cause tremendous damages and loss of life at many coastal areas around the world. The main purpose of this study is to investigate propagation of tsunami in order to validate tsunami run-up and inundation and assess ocean environment at shallow water region. We used Smoothed Particle Hydrodynamics based on Shallow Water Equation (SWE-SPH) to reproduce the previous tsunami event. The results were compared with water elevations at the survey locations. Moreover, we applied to compute wave propagation and velocity filed around offshore structures such as a wind farm. INTRODUCTION Tsunamis cause tremendous damages and loss of life at many coastal areas around the world. Tsunamis with destruction at spreading areas should be accurately predicted to establish evacuation routes and to find out safety locations at inundation areas. Tsunami inundation process at flooding area and tsunami behaviors become a key factor to protect coastal areas and to reduce number of victims. In particular, it is difficult to estimate wave deformation and its propagation at shallow water region caused by shoring due to bottom topography and coastline. In general, wave propagation at shallow water region can be represented by Sallow Water Equations (SWE) and its computation is lower cost comparing with that of full-3D model. In Grid Based Method, to obtain reliable results dynamically, adaptive structured (Liang, 2009; George, 2010) or unstructured grid systems (LeVeque, 2007) were employed. However, the Grid Based Method needs to generate grids at complicated domains, and then it is difficult to compute water elevation and wave propagation at focused areas. On the other hand, in Particle Based Method, Rodriguez-Paz and Bonet (2005) introduced a shallow water formulation based on SPH method (Monaghan (1994)) with variable smoothing length, which treats the continuum as a Hamiltonian system of particles. And also, de Leffe et al. (2010) employed Riemann approach proposed by Vila (1999) to realize more robustness for computations. Moreover, R. Vacondio et al. (2012a) applied open boundaries conditions using SWE-SPH for shallow water flow to simulate flood inundations due to tsunami attacking.

ABSTRACT Design standards of vertical evacuation buildings against tsunami loads have been developed in Japan and the USA. In this paper, we compare the overall tsunami loads and structural design requirements of the two standards. Hydrodynamic forces from each standard are further compared for example cases, and it is found that for a given inundation depth, the standard in Japan generally requires higher capacity for vertical evacuation buildings than in the USA. The USA standard explicitly requires a 2,500-year Maximum Considered Tsunami design hazard. In Japan, it is expected that the local governments will determine a worst case deterministic scenario for their region, and the design return period for the maximum considered tsunami is not specified. INTRODUCTION The purpose of this paper is to provide an overview of the technical methodology utilized in Japan and the United States of America for the tsunami-resilient design of vertical evacuation buildings. Taller structures in a community can provide effective secondary alternative refuge when evacuation out of the inundation zone is not possible or practically achievable for the entire population. During the 2011 Tohoku Tsunami in Japan, many taller buildings were successfully used as evacuation buildings saving tens of thousands of lives (Fraser et al., 2012). To design and construct buildings resistant to tsunami loads, quantitative evaluation of tsunami inundation and loads applicable to structural design is essential. The tsunami evacuation building also needs to reach sufficient elevation such that the refuge areas are located well above the tsunami water elevation considering possible splash-up during tsunami inundation and the inherent uncertainty in estimating tsunami run-up elevations. Design guidelines for tsunami evacuation buildings in Japan were developed by a task committee under the Japanese Cabinet Office in 2005 (JCO, 2005) referring to "Structural Design Method of Building to Tsunami" (Okada et al., 2004) which introduced a formula to compute tsunami loads expected to act on buildings constructed at the coastlines. In November 2011, The Housing Bureau of the Ministry of Land, Infrastructure and Transport issued updated Interim Guidelines on the Structural Design of Tsunami Evacuation Buildings (MLIT, 2011), after considering new findings of the Great East Japan Earthquake of March 11, 2011. In February 2015, a chapter on tsunami loads was established in "AIJ (Architectural Institute of Japan) Recommendations for Loads on Buildings" as the outcome of the Tsunami Loads Subcommittee under the Committee for Loads on Buildings in AIJ (AIJ, 2015).

ABSTRACT A flap-gate breakwater is a coastal defense structure that ordinarily lies down on the seabed, and rises due to its buoyancy to form a continuous seawall before a tsunami attack. Because the 2011 Tohoku earthquake tsunami seriously damaged many coastal defense structures, the resilience of these structures must be considered so they do not immediately collapse against tsunamis exceeding design levels. In this study, we examined resistance properties of the flap-gate breakwater against tsunamis through a series of hydraulic experiments using a flap-gate model that could detect reaction forces from a rubble mound. In addition to this study, we conducted a field experiment to prove the flap-gate breakwater's basic performance. The paper also describes the results of this experiment. INTRODUCTION A flap-gate breakwater (hereafter called a flap-gate) for tsunami and storm surge protection was previously developed, and is expected to be used as a movable flap-type tsunami barrier. Figure l shows the flap-gate's mechanism. The 2011 Tohoku earthquake tsunami severely damaged many coastal defense structures, which are required to be resilient enough to not immediately collapse when subjected to tsunamis that exceed design levels. The flap-gate was initially developed through hydraulic model experiments (Kimura et al., 2009; Kiyomiya et al., 2006; Kiyomiya et al., 2007), and a field experiment was later conducted to consider its function, applicability, and effectiveness (Kimura et al., 2012). The purpose of this study is to determine the flap-gate's resistance properties against tsunamis exceeding over the design level through an experiment that subjects a hydraulic model of the flap-gate to excessive external force. Moreover, we considered how a seepage flow formed in a rubble mound due to sea level differences between a port inside the flap-gate and the ocean outside would affect the bearing force of soil. Figure 1 illustrates the flap-gate behavior during a tsunami or storm surge. The flap-gate, which usually lies on the seabed, rises above the sea surface due to its buoyancy, and then closes of the port or canal entrance. The flap-gate stands up to its prescribed angle (90° in Fig. 1 (c)) due to the water elevation outside of the flap-gate. Upper and lower tension rods support the forces of water pressure acting on the gates. Resistance plates between the lower tension rods, shown in Fig. 1 (c), reduce the speed at which the gate rises and reduce impacts on the tension rods when the gate is upright. A substructure, shown in Fig. 2 (a), holds gates, tension rods, and other movable equipment, and the weight of the substructure supports all forces caused by water elevation.