To assess the variability and mechanisms of nutrient transport and upper ocean primary production associated with fluctuations in the Kuroshio path off the coast of Enshu-nada Sea, numerical analyses were conducted on the nitrate flux transport processes maintaining primary production by using a three-dimensional regional ocean circulation model coupled with a nitrogen-based ecosystem model. The differences in nitrate transport were evident depending on the presence or absence of Kuroshio meandering, with upper ocean primary production being approximately 1.5 times higher during a non-meandering period compared to a meandering period. Nitrate flux budget analysis revealed that during the non-meandering period, the vertical nitrate transport within the mixed layer was amplified, primarily due to vigorous upwelling of nitrate on the northern side of the Kuroshio axis and around the Izu-Ogasawara Ridge. The nitrate supplied to the upper ocean was then transported horizontally by the counterclockwise rotating, topographically constrained standing mesoscale eddy formed between the coast and the Kuroshio, to feed the phytoplankton in Enshu-nada Sea.

 River plumes forming in coastal oceans significantlly impact on the marine environment because they contain large amounts of dissolved organic matter and nutrients. Since their dynamics vary greatly in space and time, high-resolution observations in the coastal oceans shouled be essential. This study investigated the spatio-temporal variation on colored dissolved organic matter (CDOM), a bio-refractory substance, as an indicator of terrestrial loads from the watershed of Kumaki River into the coastal sea located around the west bay of Nanao Bay, Ishikawa Prefecture, to capture the land-sea material dynamics. We conducted the integrated observations over the land and ocean using a CDOM logger and a automated vessel. As a result, the CDOM concentration was low in the upstream of Kumaki River mostly covered by the forest, and the concentration increased in the middle watershed retaining the large area of rice paddy fileds. Finally, the concentration in the downstream tidal areas decreased as salinity increased due to seawater dilution. Our marine observations using the automated vessel indicated that the water mass characterized by the low salinity and the low CDOM concentration was transported by the tidal flow and ocean currents from the estuary of Kumaki River to the oyster aquaculture and offshore areas, suggesting the the advection-diffusion process of river plume and its dilution.

 To mitigate greenhouse gas emissions, dredged sediments have been utilised as an under-ground carbon bunker in constructed tidal flats. However, it's important to note that the incorporation of dredged sediments may lead to methane production, a potent greenhouse gas. This study focuses on assessing the impact of carbon storage and methane production in the constructed tidal flat (Oshima tidal flat in Shunan City, Japan). The field survey was conducted in June 2023, with sampling undertaken at two sites within the subtidal and intertidal zones of the tidal flat. Vertical core samples, measuring 2.0 m in length and 10 cm in diameter, were collected using a vibration core sampler and subsequently divided into vertical layers. Analyses were performed to determine grain-size composition, water content, particle density, total organic carbon (TOC), ignition loss (IL), methane. The estimated residual organic carbon content was found to be 86.5 ± 2.5% in intertidal zones and 86.7 ± 2.5 % in subtidal zones, with methane concentrations ranging 0.068-0.602 µL/g-dry in intertidal zones and 0.037-0.087 µL/g-dry in subtidal zones. Additionally, based on the TOC content of the dredged sediments, the estimated carbon storage rate was calculated to be 9.02 × 10³ g-C/m³. Conversely, the estimated methane production from the dredged sediments was calculated to be 0.202 g-C/m³. These findings confirm minimal methane production and highlight the significant organic carbon storage ca-pacity of the constructed tidal flat.

 In August 2021, pumice from a volcanic eruption near the Ogasawara Islands drifted over 1200 km, reaching the Nansei and Okinawa islands in just 2 months. This pumice caused disruptions in port activities and environmental disturbances along the sandy beaches of Kagoshima and Okinawa, adversely affecting their economies. While several research institutes have developed models to predict pumice raft drift in the ocean, the complex forces at play have limited investigations into their behavior in shallow waters (<10 m). Therefore, we conducted quantitative experiments in the swash zone, employing a large water tank with a gentle slope and machine learning-based object tracking analysis. Our findings reveal distinct classifications of pumice behavior influenced by wave action in the swash area, leading to varied run-up distances and deposition patterns. Moreover, we propose a method to estimate the maximum run-up distance based on pumice particle size and wave characteristics.

 Many studies on the long-term trend of wave characteristics along the Japanese coasts have utilized NOWPHAS data and few of them have delved into the factors in detail. In this research, we investigated the long-term trend of significant wave height along the Japanese coast annually and seasonally, using data from 1979 to 2020 collected by Water and Disaster Management Bureau of the MLIT and RSMC best track data by JMA, and analyzed their relationship with typhoons. The results show that the significant wave height does not tend to increase nor decrease nationwide, but particularly on Tokai region, the maximum significant wave height in autumn tends to increase. In addition, the results demonstrate similarities between the changes in the annual and seasonal maximum significant wave height at some observatories and an increase trend in typhoons passing south of Honshu Island. From the above, we conclude that changes in typhoon’s path are related to the trends of the maximum significant wave height.

 Wave conditions are crucial parameters for projecting shoreline changes under future climate change scenarios. The high-resolution dataset of future wave climate, available around Japan, is a valuable resource. However, the model biases inherent in this dataset remain insufficiently evaluated. In this study, we conducted a comparative analysis between historical wave climate experiments and hindcasted wave modeling off Omaezaki. Our findings reveal that the monthly mean wave heights in the wave climate dataset are lower than those calculated through the hindcast. To address these biases, we proposed a novel correction method and applied it to the frequency spectrum of future wave climate projections. The corrected wave climate projection indicated a decrease in wave height for the future. Notably, the monthly mean wave heights were higher when compared to estimates obtained without bias correction. These results suggests the critical importance of bias correction to prevent underestimation of future shoreline changes.

 We collected open satellite data and read the shoreline positions for the ten years from 2013 to 2022 in the Enshunada (East: Benzaiten River, West: Imagiriguchi, Extension: about 35 km, with the Tenryu River in the center of the target area). We examined the reliability of shoreline reading by survey results. As a result, we confirmed that the trend of shoreline variation (advance, retreat) can be captured within the resolution of satellite observation. Next, we calculated the shoreline change rate for each position and confirmed that it did not contradict existing reports and decreased in sections where structures were installed. We investigated whether shoreline fluctuations respond to the flooding of the Tenryu River (sediment supply) and high waves (coastal sediment movement). As a result, a tendency for the shoreline near the mouth of the Tenryu River to advance after flooding was observed. Examining the relationship between wave offshore energy flux and shoreline fluctuation, shoreline retreat was observed in sections without structures after high waves. The shoreline remained stable in sections with structures.

 Tugboats push the ferry against the quay to maintain cargo handling functions in case of agitation in the port of Tomakomai. The annual number of tugboats pushing ferries showed a positive correlation with the PDO index, and this paper discusses the relation between the wave powers and so on at the port for the period 1980 to 2021 and climatic indices which are PDO index, NPGO index and AO index. The major conclusions are as follows.

 (1) The annual maximum wave power shows a positive correlation with the AO index before 1999 and the PDO index after 2000. The reason why is that the correlation between the AO index and the PDO index are getting weaker as strong in 1980's.

 (2) Because typhoons are coming closer than before, the ratio of the annual maximum wave power more than 100 kW/m after 2000 is twice than before 1999.

 (3) The annual frequency of waves that interfere with ferry cargo handling operations at the Tomakomai Port shows a positive correlation with the annual number of tugboats pushing ferries after 2000.

 Border islands are politically and economically important, but sufficient monitoring and maintenance systems have not been established. In this study, we created a database concerning the geographical and physical characteristics of border islands and developed indicators assessing conservation priorities. In the database creation process, we compiled quantitative data, such as the "amount of territorial sea loss due to the disappearance of border islands (groups), " based on geographic information. In evaluating conservation priorities, we conducted more detailed analysis and ranking by adding vulnerability indicators such as elevation and volume to border remote islands deemed to have high conservation priority based on territorial sea loss and island group area. Finally, we identified some islands requiring urgent conservation efforts, such as Okinotori Island.

 We developed a new system to evaluate the amounts of COD (Chemical Oxygen Demand), TN (Total Nitrogen), and TP (Total Phosphorus) generated in watersheds and their inflows into rivers and coastal ocean. This system calculates the generated amounts from point and non-point sources, classified by municipalities into categories such as residential, industrial, livestock, and others, and estimates the inflow amounts by considering the delivery ratio. Additionally, it predicts changes in generated and inflow amounts based on variations in population, industry, land use, and rainfall. The system was applied to the Osaka Bay and Harima-Nada watersheds. For system validation, we compared the observed and calculated values of COD, TN, and TP time series in the Yamato River, and the calculated values generally reproduced the observed values. Moreover, predictions of generated and inflow amounts for changes in population, industry, land use, and rainfall in Osaka Prefecture indicated that population decline and industrial structural changes from FY 2004 to FY 2045 significantly reduced the generated amounts. Additionally, changes in rainfall conditions greatly affected the extreme values and periods of inflow amounts.

 There are only a limited number of locations where the sand used to make foundry sand, which can be used as molds for casting, an ancient metalworking method, can be collected, and Fukiagehama on Awaji Island is known as one of the oldest locations. In order to utilize and preserve the beach, it is necessary to develop areas where sand of the same quality can be obtained, but the method for doing so has not been established. Image analysis, elemental analysis, and ignition loss tests were conducted on sand collected from 22 locations on the beach, and particle size characteristics, mineral composition, and amount of attached organic matter were investigated.

 We proposed a screening method to select candidate sites for alternative sampling analysis by performing cluster analysis based on these. The selected candidate site was confirmed to have maintained a stable sandy beach area based on spatial information on land use changes since 1976, and was found to be suitable as an alternative site.

 The GPS buoy for wave observation such as NOWPHAS generally adopts a cylindrical shape with a diameter of about 7 meters. In order to improve the economy and maintainability of buoys and mooring devices, a new catamaran-type GNSS (GPS + QZSS Quasi-Zenith Satellite System) buoy has been developed. In this study, a prototype (L12m × B7m, W63.9t) was installed in an area with a water depth of 238 meters off Shirahama, Wakayama Prefecture, and a demonstration test was conducted.

 In the demonstration test, we examined wave observation accuracy, buoy motion characteristics, economy of buoy fabrication and installation, improvement effects of observation accuracy due to GNSS, and standard design methods. By comparing the observation values of GPS wave gauges in the surrounding waters with the wave estimates at the installation site, it was found that the wave observation accuracy was equivalent to that of conventional wave observation buoys. Moreover, since fluid forces are reduced, it was found that an economically designed mooring chain is possible.

 There are from 2,000 to 3,000 rescues every year on the beaches of Japan. Also, the occurrence of drowning accidents is mainly caused by rip currents, it accounts for 46 % of accidents. It is difficult for beach-goers to recognize rip currents including flash rip currents under the momentary changes in natural phenomenon. In this study, we investigated differences in the subjects' recognition of rip currents using Virtual reality and Eye tracking for six different situations in which rip current locations and shooting locations. In the eye tracking, rip current areas were set as AOI, and the number of visits, the total duration of visit, and the time to first fixation entered the AOI were calculated. As a result, it was considered that it was relatively easy to recognize rip currents from the hilltop where an entire coast can be seen and the location where rip currents could be seen in front of the swimming area, whereas it was difficult to recognize a flash rip current which was generated 30 seconds after the start. In addition, it was thought that a knowledge of rip currents was effective for rip current recognitions.

 Weather routing (WR) using meteorological and wave forecasts is effective for safe and efficient ship operations. This study assumes an increase in maritime weather observation points due to future maritime demand and automatic observation efficiency. We analyzed the impact of assimilating data from a hypothetical maritime weather observation network on typhoon and wave forecast accuracy and its influence on route selection. Results showed that assimilating atmospheric pressure observations significantly improved typhoon track, intensity, and wave forecasts. This improvement in forecasts also affected optimal ship routing, indicating that future ship observation networks could enhance operational efficiency. Further case studies under diverse meteorological and observation conditions, considering observation costs and more realistic scenarios, are necessary.

 This study aims to provide more accurate forecasts of future sea ice thickness in response to climate change, and to track past variations in sea ice thickness for statistical evaluations of these trends. Therefore, we proposed a method using neural networks (RNN, LSTM) for predicting the temporal changes in sea ice thickness and the heat balance on ice, which serve as indicators of coldness, from meteorological data including solar radiation and temperature, and investigated its accuracy. We found that even a simple RNN could make good predictions for thermal balance with daily periodic variations, and that LSTM, with its high capability for long-term memory in sequential data, could also predict relatively well for sea ice thickness without periodic variations. Furthermore, we confirmed a further improvement in accuracy by adding the cumulative value of atmospheric temperature data (Freezing degree days) as input data, by normalizing the input data, and by evaluating with ensemble averaging using multiple initial weights.

 A 3D LiDAR (Light Detection and Ranging) has the potential to become an effective measurement device for detailed nearshore processes. To demonstrate its applicability, we tested a 3D LiDAR for observation of nearshore processes at the tip of a sandbar on the Tenryu River mouth in Shizuoka prefecture. As a result, time-varying water surface and beach topography in the swash zone were captured with high spatiotemporal resolutions, and it revealed that both aeolian sediment transport and runup waves contributed to the topography changes during the observation period. A LiDAR measured topography changes of several centimeters induced by aeolian sediment transport. It also captured migrations of a berm and step induced by wave runups and tidal water level change. Obtained data revealed that the characteristic topography change at the sandbar tip proceeded in the alongshore direction in accordance with a change in incident wave directions.

 When inundation occurs due to wave overtopping, it is necessary to measure wave run-up heights at the seawall and levee for the evacuation of coastal residents. In this study, the authors conducted wave generating experiment in order to develop a method to measure wave run-up heights using LiDAR. The characteristics of measuring noise and accuracy of LiDAR were examined by compareing the LiDAR data with wave gauge data in the experiment employing 1/25 scale seawall models in which regular waves were generated. As a result of the experiment, point clouds were obtained at the water surface containing bubbles from a surf zone to the seawall under breaking wave conditions. Furthermore, the wave run-up heights could be measured by extracting shoreline locations from point clouds. In addition, point cloud acquisition conditions were examained in comparison with modeled water level data by CADMAS-SURF/3D.

 Construction of a classification model, which can identify the coastal sediments, by 5 convolutional neural network(CNN) models, ResNet50v2, SqueezeNet, VGG16, VGG19, and the model constructed in the previous study was conducted in this study. In the learning process, the effectiveness of transfer learning using the ImageNet was investigated additionally. The learning results showed that VGG16 and VGG19 had high accuracy. The learning result of ResNet50V2 indicated the overfitting. The increase of the intermediate layer of CNN model was not related to the high accuracy of learning in the case of this study. The transfer learning did not show improvement in learning accuracy, in some models, the accuracy decreased. In the classification of an aerial photograph including a sandy beach, the models except ResNet50V2 provided desirable identification results, however, some areas such as the area covered by water that was not included in learning factors should be treated properly in the learning process. VGG16 had the best classification result of coral gravels which did not show a good result in the past study.

 Measuring the CO₂ absorption of algal beds is one of the key issues for achieving carbon neutrality, but identifying the area of algal beds from UAV images requires a great deal of labor and experience. In this study, we propose a method for automatic recognition of algal beds using UAV images. The proposed method uses a model that enables semantic domain segmentation at the pixel level, and employs ViT-Adapter, one of the latest models. The advantage of this technique is that it effectively utilizes the knowledge of a trained large-scale model to recognize algal beds, and it can identify algal beds at the pixel level by adjusting the parameters of the model. In this study, we conducted learning using mask images of visually identified algal beds from aerial photographs, and further examined data expansion and other processing to adapt the learning to UAV images. The effectiveness of this method was verified through a demonstration using UAV images of the Erimo coast of Hokkaido.

 Atmospheric CO₂ removal by macroalgal beds has attracted attention, and the macroalgal biomass is an important metric for calculating CO₂ removal. Conventional dive surveys are highly accurate but have limited areal coverage. On the other hand, remote sensing methods such as aerial photography have high areal coverage but do not provide detailed information on macroalgal beds. In this study, we conducted a survey of macroalgal beds using a UAV equipped with a green laser scanner to establish a survey method with high accuracy and areal coverage. Noise processing and annotation were applied to the point cloud data, and the results were compared with the measured vegetation height and wet weight. As a result, vegetation height (0.0–3.7 m measured, 0.0–4.0 m analyzed, R² = 0.97 for the test plot) was estimated with high accuracy. In addition, biomass (wet weight) could be estimated from vegetation height (P < 0.01), indicating the possibility of measuring with high area coverage.