This study presents experimental findings of wave-vortex interactions observed in wave propagation over vertical vortices. A variery of wave responses had been observed; incident waves trapped within vortices or docomposed into multiple short-crest waves above vortices as well as transition to diffracted waves behind the vortices. Wave amplification in the vortical filed may be explained with wave-current interactions, while pressure depression in the vortices may need to consider to indentify the interactions.

 The δ-SPH method, one of the Lagrangian meshfree computational methods, demands appropriate setting on kernel radius size (or smoothing length) for obtaining best performance. Large influence domain setting would lead to increase of number of neighbor particles, which results in improved accuracy but increased computational cost, and vice versa. Due to this fact, users have been required to do empirical tuning of influence length with careful consideration of balance between accuracy and computational load.

 In this work, we develop δ-SPH-CG method, which can provide highly accurate results under small kernel radius. In the proposed model, the pressure gradient model in linear momentum conservation equation is enhanced through incorporation of corrective matrix in a methodical manner that would keep linear momentum preserved. For verification and validation of the method, numerical examples related to water waves are performed, where improved accuracy and conservation property of the method are well presented.

 The ultimate goal of this study is to accurately predict local scour around a cylinder in a wave-current coexisting field. Therefore, in this study, we investigated the non-reflective boundary condition for the coexisting field, including forward and reverse current in a three-dimensional (3D) analysis, and conducted reproduction analyses of the flow regime around a cylinder under the reverse current condition. First, we proposed a non-reflective boundary condition that takes into account the energy decay zone and water volume correction in that zone. Introducing the non-reflective boundary condition into the 3D model showed that a spatially and temporally stable coexisting field could be created. Next, the model was shown to reproduce well, albeit qualitatively, the experimental results on the surface flow around a cylinder as a function of the Keulegan–Carpenter (KC) number under reverse current condition. Finally, the model was used to study the internal flow field around the cylinder and showed that when the KC number is around 7 to 8, a complex and turbulent internal flow field may develop downstream of the cylinder regardless of the forward or reverse current conditions.

 Air density distribution above the bubbling water surface was visually measured using the Background Oriented Schlieren (BOS) method. It was confirmed that the BOS method is applicable to a density measurement of the air surrounding a water droplet that is slowly evaporating in a still air. The BOS method sucsessfully visualized the high-density air released from a bursting bubble is transported upward through diffusion. The air just above the bubbling surface, therefore, tends to have a higher average density than the bulk air, which generates the negative vertical gradient of air pressure. The pressure gradient near the bubbling surface is one of the important factor to determine the initial conditions of the sea spray droplets; further study about the air density and pressure field above the bubbling surface would be needed for its quantification.

 This study estimates the spatio-temporal changes in the whitecap coverage process from the development of whitecaps associated with wave breaking to the residual foam, and discusses the statistical characteristics of the whitecap coverage process by using sea surface image observations for storm surge events associated with the passage of a bomb cyclone and a typhoon. We propose a new method to estimate the crest lengths of whitecaps by using optical flow to estimate the local velocity of the whitecap image. This method can represent the characteristics of wave crest lengths that grow with the development of breaking waves and remain behind the wave breaking. The exponential scaling of the time-averaged wave crest length shows a -6 power law in the high wave speed region, consistent with previous reports for both bomb cyclones and typhoon events. Spatio-temporal spectrum of whitecap image enables us to separate active whitecaps from residual bubbles, which develop with the development of wave breaking.

 In this study, interactions between free-surface and a sub-surface vorttex pair have been numerically investigated with the aim to understand evolution of vortices under breaking waves. Three dimensional Large Eddy Simulation has been performed in numerical water tank with analytical vortex pairs as an initial condition. When the vortices vertically translated by self-induction approached a still water surface, the surface was lifted above the vortices owing to vortex-induced flow. Additional small scale vortices were produced on the resulting surface defomation with high curvature, resulting in small scale surface fluctuations.

 In Ibusuki port coast, a coastal improvement project is in progress applying the idea of integrated shore protection system combining several coastal works and beach nourishment. In the project, the local groundwater environment including the “natural sand-steaming hot spring” should be carefully conserved, because that is an important tourism resource in the region. The natural sand-steaming hot spring bathing in Ibusuki is a unique one that warms the body in a layer of sand heated by hot spring groundwater. This method is unprecedented in the world, and the conditions for its practice are not fully understood. We have been conducting field observations on the groundwater and the thermal environment since 2018. This study intends to clarify the conditions for the implementation of the natural sand-steaming hot spring by analyzing these results. The feasible conditions are found to be specified as that the ground table is T.P. +0.7 m to +1.4 m, the surface temperature (the temperature at 0.2 m below the sand surface) is 65 °C to 80 °C, and the distance from the sand surface to the groundwater level 0.3 m to 0.6 m. It is also clarified that the surface temperature decreases linearly with the distance from the sand surface to the groundwater table.

 In Ibusuki Port Coast, a coastal improvement project is in progress applying the idea of integrated shore protection system combining jetties, offshore breakwaters, seawalls and beach nourishment. The preservation of the hot spring groundwater environment, especially the "Natural Sand Bath", which is an important tourism resource in this area, is an essential issue in the project. Incorporating understandings by the field observations focused on feasible conditions of the sand bath, this study analyzes the impact of the beach nourishment on the groundwater and sand temperature characteristics both for the present and after nourished beach forms based on a groundwater and heat transport model. The sand temperature after the beach nourishment work was estimated from the computed ground water level and by use of an empirical formula to relate the level to the sand temperature. As a result, although even after the nourishment the sandy beach meets the feasible conditions of the sand bath at the same ground level as the present situation, it is predicted that various risks may arise by the beach nourishment, such as narrowing of the feasible area for the sand bath and decreasing in the ground surface temperature at the optimum sand bath level by shifting the beach area offshore where the temperature is affected by a cooler water mass of the sea.

 The occurrence of drowning accidents on the beaches is mainly caused by rip currents. In this study, we created an AI model that can detect rip currents in Yuigahama beach. As a result of creating the AI model under various conditions, the accuracy, the precision and the recall rates of the rip current detection were 88 %, 17 % and 100 %, respectively at the final point. As a result of actually operating this AI model on the study beach during summer season, except for stormy wave conditions caused by typhoons, the number of detected rip currents tended in the center of the pocket beach to increase with the wave height and wave energy flux. Also, the number of detected rip currents was 10 % or less per hour in most cases. In these cases, because the wave height and the wave energy flux were relatively small, it was considered that flash rip currents occur on the study beach even under calm wave conditions.

 The Tanabe Nakashima Storm Surge Observation Tower, located at the mouth of Tanabe bay in the southwestern part of Wakayama Prefecture, has been continuously monitoring meteorological and oceanographic conditions.

 The wave characteristics at the observation tower varied greatly with the seasons, and wind waves were especially dominant in winter. It was confirmed that the distribution of waves before the closest approach of typhoons was concentrated in a region with a wave gradient of about 0.005 to 0.01, which are characteristics of swell waves. In order to investigate the characteristics of changes in wave heights during typhoon approaches in September 2018, we introduced a non-dimensional distance that takes into account the length of the long axis of the storm radius corresponding to the typhoon force as a parameter, and analyzed the correspondence with the ratio of individual wave heights exceeding the threshold value. As a result, a linear change in the ratio with an approximately equal slope was confirmed for the two typhoons that approached the tower in September 2018.

 Field experiments were conducted at the Hiratsuka coast in Kanagawa prefecture from September 8^th to November 7^th, 2022, to estimate wave height using wave sound observed by a sound level meter and an IC recorder. The wave sound levels were observed at 55 m landside from the average shoreline position and 11.7 m above the average sea level. Every 10-minute observation (sound level meter, IC recorder, wind speed, wind direction, and air pressure data) was used in the analysis. Wave height can be estimated using the IC recorder as accurately as using the sound level meter (R² = 0.51), although a slight estimation delay occurs when the observed wave height is high, i.e., approaching a typhoon. However, the observations by the IC recorder included wind noise. Thus, sound pressure corrected based on the windscreen effect made it more accurate to estimate wave height by the IC recorder (R² = 0.53). Even if the way to reduce wind noise is very simple, the wave height was also sufficiently estimated (R² = 0.54).

 The intensification of future typhoons and the increase in extreme weather events due to climate change are of concern, making it an urgent task to accurately estimate the extreme wave height for future climates. In this study, we improved the precision and resolution of the wind field using WRF dynamical downscaling simulation with d4PDF meteorological data, and examined a method to efficiently estimate the extreme wave height focusing on specific return periods. To calculate the typhoon central pressure on specific return periods, we performed probability distribution and bias correction of the central pressure based on typhoon track data of actual typhoons and d4PDF. As a bias correction method, we proposed a method of polynomial approximation by double-logarithm transformation of the probability distribution. We suggested estimating the extreme wave height from the mean and standard deviation of the maximum value data set, based on wave calculations for cases with the same typhoon central pressure in the same return period from each climate change scenario. The 50-year extreme wave height under the 2K warmer scenario is estimated to increase approximately 1.05 times compared to the present climate in the central part of Tokyo Bay.

 Wave forecasts are used in various fields, such as marine leisure, fishery, and marine construction, and the necessary wave prediction period differs depending on the purpose. Medium-term forecasts are also important for construction planning for marine construction. This study investigated the effect of the difference in spatial resolution of long-term weather forecast data of the meteorological model on predicted wave height using the statistical wave model. As a result, it was found that the difference in the spatial resolution of meteorological data does not greatly affect the monthly mean wave height prediction, and the wave height calculation using the GPV initial meteorological data has the same accuracy.

 To improve its accuracy, we developed a system to assimilate the drifting buoy observations into the spectral wave model, WAVEWATCH III. We implemented Optimal Interpolation as data assimilation method. We conducted typhoon wave simulations in the Western North Pacific, targeting the typhoon waves in the summer of 2022. The accuracy of wave simulations was estimated using the Japanese coastal wave observation network. At first, we conducted the initialized experiment in which the initial condition is assimilated with drifting buoy observations, and the model runs for three days without assimilation. As a result, the model improves immediately after data assimilation (initial condition), and the improvement continues for 71 hours. Next, we conducted a sequential data assimilation experiment in which the assimilation process was operated every 1 hour. The sequential data assimilation reduced the wave model error by up to 2.72 m when drifting buoys observed extreme typhoon waves near the center. Even when the assimilation process is operated every 3 hours, the wave model error was reduced by up to 2.36 m. Our results show that the assimilation of drifting buoys observation data can meaningfully improve the wave model for extreme typhoon waves.

 The ensemble wave climate projection dataset for the present and ~4K future warmer climate around Japan clarified the impact of climate change on the cargo handling operation rate in Hokkaido ports. The average rate of Tomakomai Port could increase with the decrease in wave height, and the Mombet-su Port’s rate could decrease because the fetch becomes longer with the decrease of the sea ice in the Okhotsk Sea. Furthermore, the ensemble climate dataset with a much longer data period than observa-tion one makes the probability analysis of the operation rate averrable. It is studied that the impact of climate change could vary by the target of the operation rate.

 Highly accurate weather and wave predictions are necessary for marine activities. Although a dynamical model with many ensemble members can consider long-term forecast uncertainty and provide probabilistic information, it is computationally expensive and cannot cancel out the model biases. In this study, we propose a wind and wave forecasting system combining dynamical ensemble models and observations by long short-term memory (LSTM). The proposed system provides probabilistic forecasts that are more accurate than the base model, which has biases caused by the coarse numerical grid. The Brier score for the short-term forecast up to 10 hours ahead was lower than that of the base model, indicating the effectiveness of the proposed system in planning and deciding on marine construction.

 Although tsunami inundation calculations for urban areas have been performed using high-resolution topography and taking buildings directly into account, it is important to use a calculation method that uses a resolution coarser than the width of the building from the viewpoint of computational load.

 In this study, the model is improved by feeding back the building volume to the mass conservation law, which has not been considered in the subgrid-scale model iDFM, the averaged individual building drag model by Fukui et al. Replication of the idealized numerical and tsunami inundation experiments confirmed that the improved iDFM causes water exclusion in the grid due to the building volume, resulting in higher water levels and inundation depths than before the improvement. As a result, the accuracy of the inundation depth on the buildings near the shoreline, which tended to be underestimated before the improvement, was improved. In contrast, the overestimation of the inundation depth in the inland area (especially where the wet fraction in the grid is small) was confirmed.

 In the 2011 Great East Japan Earthquake, the so-called “black tsunami”, which involved bottom mud and sludge, were observed, and their effect on tsunami loads is concerned. Although it is required to quantitatively evaluate the effect of changes in density and viscosity due to the inclusion of bottom mud, there are few previous studies on the bottom mud transport for unsteady and large pressure gradient flows such as tsunamis. This study aimed to investigate the bottom mud transport and wave force characteristics, and the suspended mud experiment using a moving bed and the tsunami impact force experiment with different density fluids were conducted. The results indicated that the bottom mud movement patterns are different depending on the relationship between the yield shear strength of the bottom mud and the pressure gradient of the flow, and that the suspension characteristics of the bottom mud differed depending on the pressure gradient even at the same flow velocity. In addition, it was suggested that the relationship of the tsunami wave force to the density of muddy water depends on the incident waveform of tsunami.

 Aiming for the advancement of evaluation of wave force due to tsunami inundation flow, a theory for evaluating water surface profile in the tip region of tsunami inundation flow with sediment over a uniformly sloping movable bed, which is based on a simple tsunami inundation flow model reported previously, is proposed including the cases using only mass conservation law and using only momentum conservation law. It is indicated that the water surface profile in the case using both mass and momentum conservation laws is expressed by the complementary error function and as the specific gravity of inundation water becomes greater, the water surface gradient and therefore wave force increases. A theory for evaluating volume of bed material movement and average depth of bed scour in the tip region of tsunami inundation flow, which is based on the same tsunami inundation flow model, is also proposed. It is indicated that both the volume of bed material movement and the average depth of bed scour monotonically increase with time. Except for the early and the final stages in the inundation process, validity of the proposed theories is also confirmed through a comparison between existing experimental values obtaind under a condition of horizontal bed composed of sand and theoretical values on the water surface profile and the average depth of bed scour in the tip region of inundation flow.

 Volcanic activity, such as the collapse of Krakatau volcano in 2018, requires calculations that consider various possibilities, as it is difficult to estimate the collapse slip surface and the resulting sediment as part of a predictive study. In this context, a numerical model that accurately represents volcanic mountain collapse needs to accurately portray landslides and tsunami propagation, while also introducing a simple method with low computational cost. This study aimed to investigate the feasibility of using two methods to reproduce the mountain collapse of Krakatau volcano: a two-layer flow model based on non-linear long-wave theory and a simplified method of modeling. The results of the reproduction calculations showed that the analysis using the two-layer flow model demonstrated high reproducibility in the vicinity of the volcano, while the analysis using the simplified method was sufficiently accurate in the coastal area of the Sunda Strait. These results suggest that the two-layer flow model and the simplified method can be used to predict and reproduce mountain collapse phenomena.

 Although the landslide scale and behavior significantly affect the tsunami water level, various analysis methods with different theories and parameters have been proposed. Since there are few examples of comprehensively verifying the characteristics of methods and their impact on tsunami evaluation, this study focuses on understanding the impact on tsunami water level evaluation and compares typical software and analysis methods. Verification using benchmark problems and comparison with hydraulic experiments are performed, and three-dimensional analysis using a spherical slip surface is considered appropriate for slope stability analysis. In addition, regarding landslide behavior, it is concluded that the appropriate setting of general parameters, such as bottom friction, is more important than the differences between software.

 In the Great East Japan Earthquake, the ground behind the coastal levees was scoured by the tsunami, causing the levees to breach and increasing the damage from flooding. Elucidating the mechanism of damage caused by scour will help to strengthen the tenacity of coastal levees. The open source fluid analysis code DualSPHysics implements a non-Newtonian fluid model and has shown applicability to scour phenomena such as dam failure. However, there are few examples of analyses of scour caused by embankment overtopping, and the parameters of non-Newtonian fluid flow are not well understood in many cases. In this study, the hydraulic experiments of Mitobe et al. were reproduced using DualSPHysics, and the issues and possible improvements of the non-Newtonian fluid model for scouring phenomena were discussed. The results show that the reproducibility of scouring phenomena can be improved by dividing the ground behind the beach ridge into three layers, setting the density and kinematic viscosity of the non-Newtonian fluid in each layer, and setting the layer thickness for all tsunami inflow rates.

 On January 15, 2022, at around 13:00 JST, a massive eruption occurred at Hunga Tonga-Hunga Ha’apai volcano in the Tonga Islands, South Pacific Ocean. The eruption was accompanied by rapid atmospheric pressure changes and tsunamis along the Pacific coast. Along the Japanese coast, tsunami waves were observed several hours before the expected arrival time due to the pressure change of about 2 hPa, and the subsequent tsunami heights exceeded 1 m at some stations. These are caused by the pressure waves that excited the sea surface waves, but the external force factor that can fully explain the maximum amplitude has not been clarified. This study performed tsunami simulations for the area around Japan and estimated the factors that could explain the maximum amplitude of the tsunami along the coast of Japan by comparing the results of the simulations under several patterns of external force conditions with the observed records. The results showed that a component corresponding to the resonant period caused continuous amplification in bays along the coast of the Tohoku region. On the other hand, in Amami and Shikoku, the amplification mechanism generated by the propagation on the water surface may have contributed to the amplification and to the excitation of the water surface by the external barotropic force.

 The Great East Japan Earthquake tsunami swept away and washed ashore many trees, vessels, and vehicles. In order to prevent such damage, it is necessary to understand the behavior of drifting debris and tsunami in ports and harbors. This study conducted drifting simulations of a floating body using DualSPHysics targeting the experiment conducted in a wave basin by Fujii et al.(2005), and the applicability of the model was evaluated. The coupling simulation with SWASH, DualSPHysics could simulate tsunami propagation by reducing the computational cost. The use of the mDBC well reproduced the drifting behavior of the floating body, and the 3D simulation was able to simulate the phenomenon of the floating body runup from the water to the land, although the reproducibility of the change in the moving speed due to the difference in the installation angle of the floating body remained as an issue.

 Since 2011, offshore ocean floor observation networks such as DONET and S-net have been installed off the Pacific coast of the Japanese archipelago, and real-time observations by seismographs and ocean bottom pressure gauges installed offshore have been implemented. Although research and development of tsunami forecasting technology using the offshore observation data has been conducted, an examination of tsunami prediction by coupling real-time observation network and tsunami interferometry has not been sufficient. In this study, the possibility of real-time tsunami prediction applied for tsunami interferometry using offshore observation network was investigated from the DONET area to the Wakayama Prefecture coast. The synthetic waveforms constructed using the Green's function by the tsunami interferometry showed good phase agreement for the long-period component, but it was found to be difficult to reproduce the secondary oscillation component excited in the bay. Although further validation with other event waveforms is required, the possibility of constructing synthetic waveforms that contribute to tsunami prediction using the tsunami interferometry Green's function and offshore observational data was discussed.

 In the Shinmaiko beach area of Iwaki City, Fukushima Prefecture, expectations for coastal forests went beyond academic findings, as tsunami attenuation was visually recognizable during the 2011 off the Pacific coast of Tohoku earthquake tsunami. However, it is highly likely that the attenuation process included considerable topographical characteristics such as sand dunes. Although several academic studies have already been conducted on the relationship between tsunami and coastal forests in this area, the tsunami attenuation process has not been summarized. In this study, numerical simulations were conducted to clarify the tsunami attenuation process in the Shinmaiko beach area, especially the effect of different tsunami periods. The main result is that the apparent attenuation of the dunes is large due to the drop of the water level after the coughing event, but the actual attenuation is about 25%. The attenuation due to coastal forests was about 10%, which was attributed to the attenuation effect and the capping effect by micro-elevation.

 When a tsunami with the wave breaking approaches a coastal area, it is necessary to consider the effect of the wave breaking in order to evaluate the impact of tsunami-borne debris on port facilities. However, there have been few studies that investigated the relationship between the motion of the floating tsunami-borne debris and the wave breaking. Addirionally, the mechanism of the moving behavior of the debris caused by the tsunami with the wave breaking has not been investigated in detail. This study aims to examine the mechanism of the tsunami-borne debris behavior affected by the wave breaking through a series of hydraulic experiments. By conducting spatially dense hydraulic measurements and trajectory analysis of the debris using image measurements, the results indicate that the presence or absence of wave breaking, as well as the relationship between the location of wave breaking and the initial position of debris, affect the characteristics of moving behavior of floating debris. This study also reveals that the debris drifts at a speed equivalent to the wave celerity, driven by the rapid water surface gradient caused by wave breaking.

 Storm surge disaster caused by explosive cyclones frequently occurred around Hokkaido in the winter season. Future changes in winter explosive cyclones and storm surge due to climate change needs to be considered. In this study, the explosive cyclones classified by genesis and maximum development location into three trajectory types using d4PDF. Statistics compared the frequency of passage, migration speed, central pressure, development rate, and maximum wind velocity both in the past and in the future. In addition, water levels and flow fields calculated using advanced circulation model for oceanic. The calculation results evaluate the impact of changes in the characteristics of the meteorological field on storm surge levels. The comparison results show that in the future climate, the passing frequency increased from the Sea of Japan to the Sea of Okhotsk. The maximum magnitude low-pressure intensity could intensify. The storm surge calculation results show areas that could be dangerous in the future.

 The study clarified the generation status and characteristics of micro-pressure waves around Osaka Bay in 2021 through the use of several micro-pressure gauges installed in the area. It was found that typhoons and fronts passing near the target area were related to the generation of micro-pressure waves. The WRF analysis depicted the process of micro-pressure wave passage around Osaka Bay and indicated its potential to cause significant fluctuations in water level. While no significant meteorological tsunami was observed in this study, water level fluctuations possibly caused by the micro-pressure wave were observed from November 30 to December 1, 2011. An ocean long wave analysis, based on a hypothetical sinusoidal wave model using a nonlinear long wave equation, demonstrated generally consistent amplitudes and water level fluctuations at the observation points. However, there are still some issues to address regarding reproducibility.

 This study aims to verify the impact of typhoon characteristics and changes in levee conditions on the start time of inundation by focusing on the presence or absence of typhoons earlier than the start time of inundation assumed by the Tokyo Metropolitan Government and the spatial distribution characteristics of the start time of inundation. The results showed that the distribution characteristics of the inundation start time were more influenced by topographical characteristics than typhoon characteristics, and that the inundation start time tended to be earlier on the western side of the inner part of Tokyo Bay and in areas where the bends and branches of the Sumida River overlap. Regarding the inundation start time under the conditions of this study, no typhoon was more dangerous than the path assumed by the Tokyo Metropolitan Government. In studies where only certain parameters were varied, there is a tendency for the start time of inundation to be delayed by more than 30 minutes in many areas when the typhoon speed is slow. On the other hand, when the levee is raised by 1 m, the start time of inundation is delayed by more than 5 min when the typhoon speed is slow, but in many areas the difference is within 5 min when the typhoon speed is high.

 We have investigated a probabilistic storm surge risk assessment methodology that simultaneously alters multiple typhoon parameters. This study focused on Tokyo Bay, implementing storm surge computations for a diverse set of typhoons generated via the Monte Carlo method. We confirmed that the hazardous typhoon tracks extracted through this method generally cover instances mentioned in previous research. Within the scope of this study, the exceedance probability distribution of storm surges essentially converged with the execution of 200 numerical simulations.

 Subsequently, we conducted a preliminary investigation on the substitution of numerical simulations with machine learning to efficiently evaluate storm surge return periods. Specifically, a machine learning model was trained on the input-output relationships of numerical simulations. It was used to carry out pseudo storm surge simulations for approximately 5, 000 years’ worth of hypothetical typhoons, thereby demonstrating the feasibility of estimating the return periods of maximum class storm surges.

 The future challenge lies in improving the variety of the storm surge database and the estimation accuracy for extreme storm surges where the data is relatively insufficient.

 The present and future compound coastal flooding risk by storm surges and river floods at the river mouth of Japan’s three major bays was diagnostically evaluated based on the d4PDF climate projection. The number of extreme cases in which large storm surges and large river discharges occur by the same typhoon tends to increase slightly in the future. On the other hand, the time difference between the peak storm surge and the peak river discharge in the future climate tended to shorten by about 0.7 hours. In addition, such shortening of storm surges and river flood peaks will be significant for the rivers with the larger time difference under the present climate. The future changes in peak-time shortening are caused by the shortening of the start of the flow to the maximum flow. Finally, the relationship between drainage area and the peak time difference was examined for the top 10 cases of high river discharge in each river. The results showed that the variability of each case tended to decrease in the future. This result suggests that storm surge and flood peaks will likely overlap even when the typhoon track is not the worst.

 Coastal areas near cities with the potential for large tsunamis require improvements in facilities such as breakwaters. As an approach, the mechanism of the tsunami hazard in coastal structures has been widely studied, especially since 2011. However, most of the existing studies do not consider the 3D characteristics of the target hazards. In recent numerical/physical experiments on upright composite breakwaters, one such hazard has been newly revealed as the scouring of the mounds around the breakwater heads. This 3D mechanism should also be considered for the effective design of other types of breakwaters. In large wave conditions, breakwaters covered with dissipating blocks are often used instead of upright composite breakwaters. This type tends to have relatively large mounds. In order to maintain sufficient width of shipping lanes, the design of this type of breakwater often attempts to reduce the number of blocks around the head of the breakwater as much as possible. Given the recent understanding of the 3D mechanism of tsunami disasters, such a strategy may involve a fatal design. To avoid this, 3D experiments targeting the head of this type of breakwater are required to investigate the details of its failure processes. In this study, large-scale 3D experiments are carried out by targeting breakwaters covered with dissipating blocks, modeling a real port, under large tsunamis. The damage processes, including scouring around the head of the breakwater, is investigated.

 In order to use d4PDF for future predictions in storm surges and high waves due to the climate change, it is required to develop a practical method to utilize a huge amount of data. In this study, we analyzed wave climate in the Tosa Bay by using d4PDF typhoon track data in future +4K environment. The wind speed due to the future typhoons was found to be underestimated by the existing relationship between the atmospheric pressure and wind fields. Corrections for the wind speed were therefore necessary to estimate wave climate appropriately. We also confirmed that waves and surges in the Tosa Bay were highly sensitive to the pattern of SST. Therefore, it is important to keep a close eye on SST to facilitate climate change adaptation.

 Estimating the risk of fire spreading in tsunami-affected areas is crucial for mitigating the impact of a tsunami disaster, as fires pose a significant threat to tsunami evacuation buildings and hinder rescue and recovery operations. During a tsunami event, combustible debris resulting from earthquakes and tsunamis is not only displaced and accumulated within inundated areas but also carried onto the sea surface by the force of the tsunamis. Consequently, this debris significantly contributes to the spread of fires. However, existing risk assessment models for tsunami-related fires have primarily relied on data from individual target cells to estimate the risk, without taking into consideration the influence of tsunami debris from other inundated cells and the sea surface. To address this limitation, we have developed a new risk assessment model for the spread of tsunami-related fires, considering the effect of debris and large amounts of oil spills in inundated cells other than a target cell and in the adjacent sea surface. The results of our regression analysis underscore the importance of considering the effect of such debris and oil spills in the risk assessment of fire spreading.

 This study aims to quantify the sensitivity of fault depth and rake to the coastal tsunami height. The target earthquake source is the subduction zone along the Nankai Trough, conditioned on the southwest of Japan. First, possible ranges of the fault parameters are investigated. The spatial distribution of fault depth is compared using existing two major fault models, Slab2 and the Cabinet Office of Japan. Regarding the rake angle, a statistical analysis is conducted using historical earthquake catalogs. Second, the numerical tsunami simulation for many hypothetical source models is conducted. Different eight patterns of fault depth and rake angle with the same slip distributions are considered. Each pattern contains 100 source models with a moment magnitude of 9.1, thus, 800 tsunami simulations in total are performed. At coasts facing the open sea, the median tsunami heights between 100 scenarios with 5 km deeper faults were 1 m larger than those with the original depth of the existing model. And at same coasts, the median tsunami heights between 100 scenarios with rake rotated -14.5 degrees were 0.4 m larger than those with the original rake of the existing model.

 At the time of the 2011 Tohoku Earthquake and Tsunami, rescue activities were delayed due to obstruction by debris. In Japan, rapid search and rescue after a disaster is very important to mitigate human suffering and to reduce the number of unidentified victims. In this study, we focus on the number of days elapsed until the victims found in the rubble of the 2011 Tohku Earthquake and Tsunami, and analyze the factors affecting the number of days elapsed in each municipality and zip code district. We compared the number of days elapsed between outdoor victims and victims found in the rubble, and found that victims found in the rubble took more days to be found. The comparison of the number of days elapsed for victims found in rubble by municipality suggests that topography and the rate of total building destruction may have an effect on the number of days elapsed for victims found in rubble. Furthermore, the analysis of the characteristics of the number of days elapsed for each zip code area using quartile deviations suggests that the characteristics of the number of days elapsed differ depending on the topography.

 In urban tsunami inundation calculations, the influence of buildings is often indirectly considered by assigning roughness values. However, in order to accurately assess the tsunami arrival time and temporal variations in flow velocity at any given location, explicit consideration of buildings is necessary. In this study, we conducted high-resolution tsunami inundation calculations for Osaka Prefecture, explicitly considering buildings using the 3D urban model PLATEAU developed by the Ministry of Land, Infrastructure, Transport and Tourism. We investigated the changes in maximum inundation depth and tsunami arrival time due to the inclusion of buildings. The results showed that in calculations that considered buildings, an increase in maximum inundation depth was observed at 45.5 % of the grid points within the entire inundation area. Furthermore, the tsunami arrival time was earlier at 53.1 % of the grid points within the calculation domain due to the consideration of buildings. These trends were particularly significant in coastal areas and near rivers, which can be attributed to the concentration of overflow flow at grid points without buildings.

 This study presents river plume mixing induced by internal tide breaking over a shallow slope using a nonhydrostatic model SUNTANS. When a low salinity layer exists in the surface rive plume layer, internal tide disturb the surface layer resulting in internal soliton wave packets propagating toward the coast. Two types of internal soliton wave packets are generated in the surface laeyr during run-up and receding phases of internal tides. The internal soliton wave packets enhance the surface kinetic energy in shallow coastal area. In addition, the internal soliton wave packets increase the onshore ward internal wave energy flux, possibly causing energy dissipation in shallow coastal areas. The surface kinetic energy and internal wave energy flux are intensified as the surface low salinity layer decreases, which implies that the rate of river discahrges modulated by precipipations would influence internal tide energy toward the coast.

 In this study, we simulated a three-dimensional flow for the vertical profile of flow velocity observed at Amitori Bay, Iriomote Island, Okinawa Prefecture, during the arrival of Typhoon No. 1515. In order to improve the reproducibility of the vertical velocity distribution of the observed results, we conducted a sensitivity analysis for the vertical viscosity coefficient and the bottom roughness coefficient of the sea area. As a result, the flow propagating from the surface to the bottom layer was suppressed by reducing the vertical viscosity coefficient. The reproducibility of the observation results was improved by increasing the bottom roughness coefficient to a value larger than that used in normal tsunami and storm surge analysis because of the coral reefs that develop in the sea area. The flow distribution of the simulation results revealed that different flow fields were formed in the surface and bottom layers of Amitori Bay during Typhoon 1515.

 On the water surface under a wind stress, many surface streaks are formed even in a case of almost no wind waves. In this study, field observations and numerical simulations were conducted in order to investigate features and the formation mechanism of such streaks. Movie data analyses obtained at Yata River revealed that such streaks were stable in time and space. Previous studies by authors showed yet that, beneath similar surface streaks, horizontally-periodic velocity profiles of wind-driven currents were formed, and it revealed the existence of secondary circulations. Numerical results by using LES with the rigidsurface model also indicated the formation of surface streaks and secondary circulations. However, they were more unstable in time and space than observed streaks and circulations. Movie data analyses also indicated low surface elevations at high-speed jets at surface streaks. According to these results, spatially-periodic high-speed jets and low surface elevations in the lateral direction were given as the surface boundary conditions. Numerical results revealed surface streaks and secondary circulations were formed stable in time and space. Its suggested that these surface streaks and circulations with almost no wind waves were formed due to the interaction between wind-driven currents and surface boundary conditions.

 Turbidity currents in growth process to equilibrium are analyzed using OpenFOAM. The Smagorinsky model of LES is used as the turbulence model. The variations of the travel speed, the maximum thickness of turbidity fronts and the maximum concentration of turbidity currents are obtained from numerical results. Numerical results show that the accelerating turbidity fronts eventually reaches an equilibrium state as it flows over a long propagation. After reaching an equilibrium state, turbidity fronts repeatedly increases and decreases in travel speed and maximum thickness, while its average value remains constant.

 Coastal development in the Seto Inland Sea(SIS) has been underway since before 1900, and the reclaimed area will continue to expand. Coastal land reclamation affects the flow and transport of water and nutrients; therefore, understanding how reclamation affects the surrounding marine environment is necessary. When coastal development is undertaken, environmental assessments are conducted, but they analyzed the impact of land reclamation at local scales to assess its influence on coastal environment; however, don’t analyzed at large spatial scales. However, in SIS, which is composed of multiple bays and nadas, there is a possibility that the effect of landform changes extends to adjacent area. In this study, we analyzed the effects of coastal landform changes on the flow in SIS using a 3D hydrodynamic model that could resolve the topography in detail with unstructured grids. The calculation results showed that the differences in temperature and salinity distribution were localized in many areas, while in Osaka Bay, the spread of river water changed and the differences were significant. Analysis of the effect on tides showed that the resonance condition with the M2 tide was strengthened by the decrease in the natural oscillation frequency period. In additions, the decrease in the amplitude in Osaka Bay reduced the water exchange between Harima-nada, and changed the scale of counterclockwise circulating current in the central part of Harima-nada. The results of this study show that tidal currents have changed over a wide area due to past and present landform changes, and that the influence of human activities on the sea area is significant. We suggest that future coastal development should be carefully considered, paying attention to the impact of human activities not only on local scales but also large scalese.

 The Tsushima Straits have important port facilities such as the Port of Hakata and population centers of Fukuoka City and Kitakyushu City, and offshore wind power generation facilities are being planned in the Hibiki-Nada Sea facing the straits. Therefore, it is important to understand the characteristics of the sea-level anomalies that occur along the coast of the straits from the viewpoint of coastal disaster prevention and marine infrastructure development. In this study, the characteristics of sea-level anomalies along the coast of the Tsushima Straits were investigated in detail by analyzing observed data and conducting numerical simulations. As a result, it was found that the tidal level anomalies along the coast of the straits are strongly influenced by the Ekman transport by the wind blowing parallel to the straits rather than by the wind blowing into the straits, that the spatial characteristics of the Ekman transport are completely different from those of the wind blowing into the straits, and that the wind direction on which the sea-level anomalies most depends differs depending on inside the bay, outside the bay or angle between wind direction and coastline.

 This study presents effects of barotropic/baroclinic, tidal/non-tidal motions and their frequency characteristics in the inner part of Tokyo Bay using long-term monitoring data. Observed flows are divided into tidal and non-tidal currents, and these components are decomposed into barotropic and baroclinic currents. Results showed that the non-tidal baroclinic component dominate in the total kinetic energy and is enhanced in summer. The relationship between the stratification and the kinetic energy of the non-tidal baroclinic component shows a clear positive correlation. This indicates that the non-tidal baroclinic component in the shallow inner bay is caused by enhanced estuarine circulations due to summer stratification. In addition, the four components are decomposed into long-period and short-period components (low and high frequencies, respectively) with a cut-off period of 36 hours. The short-period component dominates in the nontidal baroclinic component, and the long-period component dominates in the non-tidal barotropic component in the inner bay. The long-period non-tidal barotropic component is explained by low frequency winds.

 In recent years, it has become crucial to take into account changes in external forces due to climate change when formulating basic policy for coastal conservation.High waves and storm surges mainly induced by low atmospheric pressure occur on the west coast of Japan, specifically the Ishikawa coast, which is the focus of this study.While numerous studies evaluate the impacts of climate change on typhoons on the Pacific side of Japan, a standardized method for evaluating the impacts of climate change on low atmospheric pressure remains elusive.

 In this study, we analyzed observed tide and wave data to validate the long-term alterations in external forces.Additionally, this study proposes a method for predicting the external forces associated with low pressure.

 The database for policy decision-making for future climate change (d4PDF) global model GCM was used for future projections.Our method estimates future external forces through multiple regression analysis and does not necessitate high-performance computational resources like supercomputers. As a result, this approach can be readily employed in various locations.

 The existence of sand ripples on tidal flats is closely related to sediment movement and topographic changes. In this study, a UAV photograph of a tidal flat called Higashi-hazu tidal flat in Nishio City, Aichi Prefecture in Mikawa Bay was conducted to investigate the sand ripple characteristics by using image analysis based on luminance information of the obtained images. Based on the results, the sand ripple formation area was revealed and the relationship between the sand ripple direction and topography was analyzed. As a result, it was confirmed that the ripple direction can be detected with high accuracy by using the luminance information. Furthermore, by visualizing the ripple directions in the entire tidal flat, it was clarified that the normal direction of the contour line of the tidal flat and the sand ripple direction coincided to some extent. It was reconfirmed that investigation and analysis focused on sand ripples led to the estimation of topographic changes on tidal flat.

 On Ajigaura Beach, sand back pass has been carried out to recover a sandy beach for bathing during summer season. The beach changes after the nourishment were monitored by the survey using an UAV. The beach topography was measured immediately after the beach nourishment as well as the measurements after 3 and 6 months from the beginning. Because these surveys were carried out at long intervals, it was difficult to analyze the temporal changes of the beach. In this study, the measurement using an UAV was recurrently carried out at short time intervals together with the photographing from a fixed point. It was found that sand was transported even under the normal wave condition, although nourishment sand was in active during storm wave condition. Thus, beach produced by beach nourishment quickly disappeared, implying the ineffectiveness of sand back pass at Ajigaura Beach.

 Beach changes in an event in which rapid beach erosion due to offshore sand transport under the storm wave conditions superimposed with the beach changes by longshore sand transport, and the recovery of the beach after the storm were investigated, taking the coast with a stretch of 16 km between Point Omaezaki and the Katsumata River in Suruga Bay as an example. It was found that although the beach was severely eroded by storm waves associated with Typhoon No. 19 that hit the area on 12 October 2019, sand returned to the shoreline by shoreward sand transport during the calm wave condition after the storm in Susuki area south of Sagara Port, where longshore sand transport is blocked by the Sagara Port breakwater. On the other hand, in the Sagara-Katahama area in northern part, both seaward sand transport and northward longshore sand transport occurred during the storm wave condition, and then the beach recovered due to shoreward sand transport under the calm wave condition and southward longshore sand transport.

 The Hojo River floodway was constructed to improve flood control safty in the Hojo River basin flowing through the central part of Tottori Prefecture. In the mouth of the floodway, blockages by the for-mation of sandbars occur frequently because of little flow rate during normal condition. Currently, artificial excavation of sandbars is also frequently car-ried out using heavy machinery for a measure against the blockages. This study intends to examine the feasibility of a sandbar flushing method by deflating the diversion weir, which is a inflatable rubber dam, as a more efficient sandbar excavation method, through field surveys and numerical models. From the numerical analysis, we revealed that even with a normal flow rate of 0.3 m³/s during the irrigation season, it is possible to flush a sandbar in only a few hours by deflating the diversion weir, as long as the difference between the crown height of the sandbar and the water level at the mouth is small.