In this study, a field depletion experiment with 45 transects was conducted to estimate the capture efficiency of a toothed dredge for seafloor litter in bottom-tow surveys in southeastern Osaka Bay, at a water depth of approximately 12 mover a muddy substrate. Seafloor litter catches generally decreased with increasing transect number, and plastics, particularly plastic fragments, were the dominant component. Using the patch model, capture efficiencies for total seafloor litter were estimated to be 19.2%, 32.6%, and 19.7% for tooth-bar, cod-end, and pooled samples, respectively, indicating differences in capture efficiency among collection parts of the gear. Capture efficiency also varied across litter categories and plastic subcategories, e.g., pooled sample estimates were 18.1% for plastics, 44.1% for natural fiber/leather, 29.3% for plastic bags, and 16.3% for plastic fragments, reflecting differences in physical properties and interactions with the gear. These results demonstrate that conventional approaches assuming 100% capture efficiency substantially underestimate true litter abundance. Analyses based on partial depletion datasets further showed that at least 25 transects are recommended to obtain statistically significant estimates, whereas approximately 20 transects are sufficient to produce stable point estimates. Overall, this study provides experimental estimates of capture efficiency for toothed dredges and empirical guidance on survey effort, contributing to improved abundance estimation and robust monitoring of marine litter.

Seafloor litter has become a growing environmental concern to marine ecosystems and coastal economies. To better understand the environmental behavior of seafloor litter, it is important to characterize its spatial patchiness at local scales. However, the local-scale patch size of seafloor litter has rarely been quantified. Thus, a patch size spectrum method was developed to estimate local-scale patch size from bottom-trawling survey data. The method utilized geostatistical simulations to relate observed area-normalized catches with simulated area-normalized catches under varying potential patch sizes. Cross-validation demonstrated predictive accuracy with a typical relative error of ∼20% and 31% in worst-performing cases. Application to experimental bottom-tow data collected from Tokyo Bay showed the local-scale patch size of 134 m, substantially smaller than the trawling transect length (∼2.8 km). Moreover, the local-scale patch size remained temporally stable between two survey periods. This method provides a novel framework for quantifying local-scale heterogeneity in seafloor litter distribution, and it is broadly applicable to other forms of integrated line transect sampling, such as benthic organism surveys and environmental monitoring where within-transect spatial resolution is limited.

Based on the developed biplane hyper-lift trawl door (HLTD), a structure optimization approach integrating a radial basis function neural network, a multi-objective genetic algorithm, and computational fluid dynamics (CFD) analysis was employed to design a biplane hyper-efficiency trawl door (HETD), with the objective of achieving reduced fuel consumption for the same swept sea area. The mechanism by which the biplane HETD outperformed the biplane HLTD in hydrodynamic performance at an attack angle of 20 degrees was clarified through variance analysis and CFD visualization. Hydrodynamic experiments confirmed that the maximum lift coefficient of the biplane HETD was 2.11, which was slightly smaller than that of the biplane HLTD (2.18), while the corresponding drag coefficient decreased by 27.6%, resulting in a significant increase of 30.0% in the lift-to-drag ratio. The stall angle of the biplane HETD was also reduced by 5 degrees compared to the biplane HLTD. Subsequently, a pair of sleds was attached to the bottom of the biplane HETD to reduce contact with the seabed and thereby reduce friction without affecting its hyper-efficiency characteristics.

In this study, we numerically investigated the drag force and deformation characteristics of a closed flexible membrane fish cage subjected to uniform flow. The target fish cage is circular, representing a real cage with dimensions of 25 m in diameter, a sheet height of 12.5 m, and a freeboard of 2.5 m. A 1/50 scale model was employed for numerical simulations. The numerical result was validated against experimental data from previous studies. Simulations were conducted at flow velocities ranging from 5 to 25 cm/s, with 5 cm/s intervals, under different filling rates to evaluate the drag force and deformation of the flexible cage. The results were also compared with those from traditional flexible net cages of the same dimensions. As a result, consistent with existing experimental findings, the flexible membrane fish cage did not exhibit large deformations like net cages. However, as the filling rate decreased, the deformation of the flexible structure became more pronounced. The deformation of the front membrane was more significant than that of the rear membrane. Additionally, drag force was observed to increase with flow velocity, consistent with prior research. Notably, the drag force on the flexible membrane fish cage was approximately several times greater than that on net cages. The study also provided a detailed tension distribution across the entire membrane fish cage. Overall, these results are expected to serve as valuable references for the design and operational optimization of closed flexible membrane fish cage systems, contributing to the advancement of sustainable aquaculture practices.

In this study, bottom-tow experiments using beam trawl and toothed dredge were conducted simultaneously in the same region to collect seafloor litter. Litter from both gear types, and from different collection parts (cod-end and gear-entangled samples), was characterized and compared. The results showed that dredges captured substantially greater amount of seafloor litter than beam trawls. For cod-end catches, dredge was 38.5 ± 12.7 times more efficient than beam trawl in terms of litter count, while for total catches (cod-end and gear-entangled samples), dredge collected 9.42 ± 2.40 times greater. Gear entanglement increased the total number of samples collected. In beam trawl, entangled litter has 28.2 ± 19.3 times larger number of samples than the cod-end, whereas in dredges, the ratio was 3.55 ± 1.45. These findings demonstrate that relying solely on cod-end samples can cause substantial underestimation of seafloor litter abundance, and that beam trawl surveys underestimate abundance compared with dredge surveys. Composition analysis revealed that dredges were more effective than beam trawls at capturing heavier and denser materials such as glass and metals. In addition, plastics dominated seafloor litter, and polyethylene (PE) was the dominant polymer type. On the other hand, large variability among sampling events indicates that estimates from a single tow are unreliable, highlighting the need for repeated sampling. This is the first study to directly compare the capture efficiency of different bottom-tow gears in the same area. The results provide a reference for future seafloor litter surveys, particularly in selecting gear type and developing sampling plans.

In this study, depletion experiments using a toothed fishing dredge were conducted to estimate the gear efficiency for collecting seafloor litter in a nearshore area on the north side of Tokyo Bay. Using the catch data of cod-end samples and gear-entangled samples, four models (Leslie-Davis model, DeLury model, normal patch model, and negative binomial patch model) were applied to fit the dataset. Analysis of the collected seafloor litter showed that plastics were the dominant category, accounting for 61.8-90.0 % by count (depending on whether the sample was from the cod-end or gear entanglement), with polyethylene (PE) as the predominant polymer type (50.5 %). Cross-validation identified the normal patch model as the most appropriate model for inference, yielding a gear efficiency estimate of 34.8 % and a corresponding seafloor litter abundance of 4000 pieces/km2 based on cod-end samples. Compared to conventional methods that assume 100 % gear efficiency, this suggests that seafloor litter abundance is underestimated by 38.6 % and 79.4 % given current experimental conditions. When all litter samples were included, i.e., both cod-end samples and gear-entangled samples, the estimated gear efficiency was 18.7 %, with a corresponding seafloor litter abundance of 45,200 pieces/km2. This suggests that excluding gear-entangled litter, mainly plastic sheet fragments from bags, can result in underestimation by as much as 91 %. These findings highlight the importance of accounting for gear efficiency and incorporating all gear-collected litter in bottom-towed surveys to more accurately assess seafloor litter pollution.

An innovative trawling strategy for transitioning from conventional bottom trawling to semi-pelagic trawling (elevating the otter board but allowing the trawl net to contact the seabed) was developed to mitigate the effects of seabed impact, and its dynamic performance was evaluated. Experiments were carried out using a 1/18 scale physical model of the otter trawl at a speed of 0.6 m/s, and the warp length was tuned in the range of 4-8 m under constant-speed control of the trawl winch for validating the numerical simulations. Numerical simulations were then conducted to investigate the effects of towing speed and winch speed on the dynamic loads of the otter trawl system and movement of the otter board, eventually develop a fuzzy or proportion-integrationdifferentiation (PID) controller for the trawl system. Numerical results showed the sharp increase of the warp tension and the overshoot of the otter board at the constant-speed control occurred, and the magnitudes of both raised as the towing speed and winch speed increased. The otter board spread was significantly reduced during the transition to semi-pelagic trawling at a low towing speed. The use of fuzzy and PID control successfully eliminated the overshoot and diminished the high warp tension.

Research gillnets use combinations of different gillnet mesh sizes to avoid the effect of mesh selectivity on the body length distribution of fish caught. In this study, pooled relative sampling efficiencies (PRSEs) for body length were obtained by assuming various gillnet selectivity curve parameters when the common ratio or common differences between mesh sizes were held constant in the available gillnet mesh sizes for research gillnets, and the conditions under which the PRSEs were kept nearly constant over the length range of target fish species for survey were examined. As a result, the PRSEs for body length were observed to be constant to some extent only when the common ratio between mesh sizes remained constant in the cases of gillnet selectivity parameters with large optimum relative length (6.0 to 7.0) and a medium 50%-selection relative length range (2.5 to 4.5). In the cases of narrow 50%-selection relative length range, the PRSEs for body length often showed a multimodal shape. Possible countermeasures in such cases include using a smaller common ratio between mesh sizes, adjusting the number of gillnets of a given mesh size, or correcting the length composition of the fish caught by mesh selectivity.

Longnose lancetfish ( Alepisaurus ferox ) may has been studied as an indicator of marine pollution caused by marine litter. The objectives of this study were to determine the difference in frequency of occurrence of plastics ingested by longnose lancetfish in different ocean area. In this study, we compared the incidence and characteristics of anthropogenic debris in the stomachs of longnose lancetfish. We examined 91 longnose lancetfish caught by pelagic longline fishing in Sagami Bay, the North Pacific Ocean, approximately 200 km south of Shikoku, and in the Indian Ocean. Broken down by ocean area, the incidence of anthropogenic debris ingestion was highest in Sagami Bay (23 of 34 specimens, 68%), followed by the North Pacific Ocean (1 of 9, 11%), and the Indian Ocean (8 of 48, 17%). The frequency of occurrence increased in area close to the sphere of human habitation. The anthropogenic debris collected in this study were more than 70% classified as plastic sheeting. Stomach content analysis revealed that more than 90% of the plastic fragments were composed of PP and PE, which have specific gravities that are less than that of seawater. The results of this study show that some of the plastics flowing from the land into the sea are spreading through under the water surface of the ocean.

The oscillation cycle and intensity in each part of the pelagic longline were investigated in Sagami Bay, Japan, to evaluate the effect on bait loss. The results of three longline operations were analyzed. The longline consisted of 25 ordinary and 15 midwater float baskets, with six branch lines per basket. Acceleration loggers were attached to the upper end (main float) and lower end of the float line, and the hook of the first and third branch lines. A depth logger was attached to the upper and lower ends of the branch line (hook). A video logger was attached, facing downward at a position 0.6 m above the hook. The periodicity of the oscillation was clarified by autocorrelation analysis, and the sum of the absolute values of the dynamic acceleration of the three axes (ODBA) was compared. From the video footage and autocorrelation analyses, the oscillation cycle of the hooks was approximately 3 s, which is the same as that of the main float. The ODBA of the first hook was 1.9–2.3 m/s2, which was significantly larger than the main float (1.1–1.4 m/s2) and the third hook (0.2–1.8 m/s2, p < 0.001). Therefore, the oscillation of the hook was derived from the vertical movement of the main float due to the wave, and it decreased as the distance from the main float increased. The third hook in the midwater float basket had an unclear cycle, and the ODBA was significantly smaller than that of the ordinary baskets.

To improve the lift performance of the biplane-type otter board with a high aspect ratio, a monoplane-type hyper-lift trawl door (HLTD) with a spanwise slit h was designed by vertically assembling a pair of square HLTD models. Flume experiments were conducted using four models with slit ratio s (= h/c, c: chord) changed to 0, 0.1, 0.2, 0.3. Experiment results coupled with a CFD analysis revealed that the wingtip flow generated from the 0.2c slit is beneficial for suppressing the flow separation at moderate to high angles of attack, associated with increased lift and drag coefficients. Based on these results, a monoplane HLTD with a slit ratio of 0.2 was adopted for the fore wing, rear wing, and both wings of the biplane HLTD (composed of two monoplane HLTDs), denoted as model A, model B, model C. Flume experiments showed that model B was superior with a maximum lift coefficient of 2.18 (stall angle: 32°) in the free stream and 2.22 (28°) near the bottom. A bottom-trawling sea trial for the new biplane HLTD design (model B) was conducted. The new biplane HLTD with 80% of the wing area spread about 1.3 times more than the comparative biplane-type otter board.

Sea turtles that have entered the submerged bag net of setnets repeatedly push their heads up against the ceiling netting (referred to here as push-ups) to ascend to breathe. Consecutive push-ups are essential to the success of escaping in a turtle releasing system developed to reduce the incidental death in setnets. Existence of differences in behavioral characteristics were suggested among turtle species, but no detailed evaluation has been performed. The objectives of this study were to establish extraction methods of the consecutive push-ups with a triaxial acceleration logger, and to clarify the differences in the behavioral characteristics between green Chleonia mydas and loggerhead Caretta caretta turtles under the simulated condition of bycatch in setnets for appropriate adaptation and modification of this system. The behavior of each turtle mounted with an acceleration logger in an experimental bag net was recorded using a video camera. The consecutive push-ups were extracted by filtering the time-series body angle converted from the static surge acceleration and the overall dynamic body acceleration (ODBA) from the triaxial dynamic acceleration. A threshold of 20 degrees in body angle and 1.0 m/s2 in ODBA was detected among 89 % of all consecutive push-ups in green turtles. On the other hand, the ODBA was not adopted as the extract condition in loggerhead turtles, although a threshold of 20 degrees in the body angle detected 93 % correctly. Mode of ODBA (1.0-1.5 m/s2) in the green turtles was larger than that in the loggerhead turtles (0.5-1.0 m/s2) during consecutive push-ups. In contrast, the number and duration of the consecutive push-ups in the green turtles were less than those of loggerhead turtles. Consequently, modifying the gear with a larger inclination angle of the ceiling netting should be considered for the setnets where the bycatch of green turtles dominates.

Hydrodynamic experiments in a flume tank in the current range of 40-80 cm/s and computational fluid dynamic (CFD) analysis at the current of 60 cm/s of trawl plane nettings with solidity ratios of 0.06-0.32 at small angles of attack (0-25 degrees) were conducted to refine the understanding of the relationship between resistance performance and flow behavior. The rigid model representing the plane netting in a taut configuration was adopted. A finite volume method was used for solving Reynolds-Averaged Navier-Stokes equations associated with the shear-stress transport k-omega model in CFD analysis. Referred to experimental results, the parallel drag coefficient decreases from 0.4 to 0.25 with a reduction of 38% but, the drag coefficient at the angle of attack of 25 degrees rises from 0.62 to 0.79 with an increment of 27% as the solidity ratio increases from 0.06 to 0.32. Such dual effect of the solidity ratio on the drag coefficient of plane nettings with respect to the inclination angle is also observed from CFD analysis. A transition of flow behavior from flowing above to passing through the netting occurs. The changing tendency of the local flow velocity is consistent with the resistance performance relative to the solidity ratio for the tested nettings.

The hydrodynamic characteristics of cambered otter boards with a camber ratio of 15% and aspect ratios of 0.5, 1.0, 1.5, and 2.0 were investigated through flume tank experiments and CFD analysis for the cases in which the wingtips were blocked or unblocked. In the blocked wingtip case, the maximum lift coefficient, stall angle, and lift slope (α: 0° to 15°) were approximately 1.75, 15°, and 0.115 (deg−1), respectively, regardless of the aspect ratio. Comparatively, in the unblocked wingtip case, maximum lift coefficients were 1.74 (stall angle: 40°), 1.65 (35°), 1.54 (25°), and 1.50 (20°), and the lift slope increased from 0.063 to 0.097 (deg−1) as the aspect ratio increased from 0.5 to 2.0. The drag coefficients for all unblocked wingtip models were larger for all considered angles of attack. The effect of the wingtip flow formed around the otter board on the hydrodynamic characteristics was visualized using CFD analysis. A small-scale wingtip plate with a width of 0.1c was designed. Particularly, for boards with aspect ratios of 1.0–1.5 and a 0.1c wingtip plate, lift slopes and maximum lift coefficients approached those for the blocked wingtip cases, and stall angles were close to those for the unblocked wingtip cases.

Model towing experiments of a bottom trawl net with hyper-lift trawl door were conducted to investigate the effect of the bottom sediment (concrete, sand, gravel, and rock) on the warp tension of the overall trawl system. The towing speed was from 50 cm/s to 70 cm/s and the ratio of warp length relative to the water depth was within the range of 4-6. Through the signal analysis of time-series warp tension, results reveal that there is a significant dependence of the warp tension on the type of bottom sediment, and the oscillation of warp tension in a frequency range of 1-10 Hz increases in the order of concrete, sand, gravel, and rock. Based on these characterizations, the time-series warp tension is thus represented by the feature vector for the input data of the selforganizing map (SOM) and learning vector quantization (LVQ) neural networks. A clustering method with an unsupervised SOM neural network acting as an updating tool for the bottom sediment database was successfully built using the validation of the prepared sediments. In combination with the output vector of labeled bottom sediment, the supervised LVQ neural network for sediment recognition performed excellently with a high classification accuracy of over 80%.

This study investigated a newly developed soft-type turtle releasing device (Soft-TRD) for setnets. Some setnet fish chambers (bag nets) are open at the sea surface, while others are closed (with ceiling netting) and submerged. Sea turtles that stray into submerged bag nets often drown because the ceiling netting prevents them from swimming up to the surface to breathe. The Soft-TRD has a slit that functions as an escape outlet at the center of a 2 m × 2 m netting. The escape slit remains closed, due to the buoyancy of a float attached at each end of the slit, and prevents fish from getting out. Turtles can push the escape slit upward, which opens and allows the turtles to escape. To decide the optimum design of the Soft-TRD, a turtle releasing trial was conducted in an outdoor water tank. Four loggerhead turtles (Caretta caretta) (standard carapace length (SCL): 67.1–72.8 cm) and five green turtles (Chelonia mydas) (SCL: 42.4–63.4 cm) were used for the trial. A single turtle was placed into the experimental bag net where the Soft-TRD (six models in total) was mounted at the center part of the ceiling, and its behavior was observed. All turtles successfully escaped the model with a 150 cm length escape slit, 10 cm overlaying width, and a 4 kgf buoyancy float at each end. The escape slit immediately closed after the turtles escaped. A sea trial (same model) was also performed using a submerged bag net (30 × 10 × 10 m) for two months. Approximately 90% (seven of the eight) of the entrapped loggerhead turtles successfully escaped out through the Soft-TRD. Fishing was not interrupted by the introduction of this device. The use of the Soft-TRD would be helpful in releasing sea turtles from the setnets and contribute to sea turtle conservation efforts.

In chelonids, oxygen is primarily stored in the lungs during a dive. Therefore, management of blood oxygen transportation to peripheral tissues by cardiovascular adjustments during submergence is crucial to maximize their dive duration, and consequently, the time spent for ecological activities such as foraging. However, the cardiac response to exercise, has rarely been examined in sea turtles. In this study, heart rate and its relationship with exercise during voluntary dives were determined in six captive green turtles (19.4±1.5 kg) by simultaneously recording depth, acceleration, and electrocardiogram. Our results demonstrated that the heart rate of green turtles was generally low (11.1±0.4 bpm) during resting dives, but they often exhibited instantaneously extreme tachycardia (up to 78.4 bpm). Green turtles elevated their heart rate up to 39.8±1.5 bpm during ventilation after resting dives, while up to 33.1±1.4 bpm after active dives. The heart rate immediately elevated with onset of exercise, and increased linearly with exercise. This result may indicate that turtles immediately need to transport oxygen from the lungs to peripheral tissues by pulmonary and systemic circulations to meet the metabolic demands of exercise because they mainly store oxygen in their lungs.

Blue shark (Prionace glauca) and shortfin mako shark (Isurus oxyrinchus) are recognized as pelagic sharks and highly migratory species, but these sharks appear in coastal areas. It is suggested that sharks migrate in the growth stage, but we do not know the details and how sharks use coastal areas. A fishing survey of these species in coastal areas like Sagami Bay is rare, and thus we carried out two types of longline operations to survey catch trends about the shark in Sagami Bay from August 2011 to July 2017: vertical and horizontal longline operations. Both species were caught throughout the year, with peaks in July and December for both blue sharks and shortfin mako sharks. Many of the male juvenile blue sharks caught had not reached sexual maturity, and pregnant blue sharks were also present among the females. The birthing period of blue sharks is reported to be from April to July, and the young are believed to be born and grow up in the open ocean of the North Pacific. However, the fact that we caught pregnant blue sharks close to giving birth suggests that they may give birth in coastal areas such as Sagami Bay. We were unable to catch young shark larvae because of the size selectivity of the longline fishing gear. Future studies, using net sampling, will be necessary to search for new-born blue sharks in this coastal area.