Fish attempt to optimize their swimming performance within their environments to minimize energy consumption. In this study, we conducted experiments using the Japanese dace (Tribolodon hakonensis), a simple flat plate, and a flume tank to simulate the swimming behavior in which fish exploit the stagnation area of the flat plate for maintaining their position. Two observed cases were: the fish localized by straightening the tail fin along the flow and by bending the tail fin like a cambered wing. The results of drag and lift forces suggest that fish adaptively flex tail fin based on the flow field, and selectively choose the optimal swimming posture for maintaining position.

Abstract Satellite telemetry devices can record movement data of animals along with the environmental data. Such data are relayed remotely via satellite systems, but are constrained by the limited bandwidth availability. A satellite relay data logger (SRDL) that can abstract dive profiles and compress the data for transmission using a broken stick model (BSM) has been widely used in studies on dive behavior and physiology of marine animals. However, there is still uncertainty in the abstracted dive profiles. Here, we aimed to evaluate the certainty of abstracted dive profiles (via satellite communication) in terms of dive performance (dive depth, duration, and dive type) by comparing it with the actual dive data (from the retrieved tag) in a loggerhead turtle deployed with the SRDL throughout a 1.4-year foraging period. There was no significant difference in the maximum dive depth between the retrieved and satellite transmission data; however, there was a slight but significant difference in the dive duration. The dives from both datasets were classified into five types. Inconsistent dive classifications occurred in 1.7% of the data. There was no significant difference in the proportion of time spent diving between the retrieved and satellite transmission data for each type during the common recording period. In monthly scale comparisons, however, a significant difference was detected when the amount of data via satellite transmission was the smallest. Our results demonstrated that the dive data abstracted using BSM almost reconstructed the actual dive profiles with certainty in a loggerhead turtle, although slight inconsistencies were observed.

The Mekong giant catfish (MGC), Pangasianodon gigas, is one of the world's largest catfish species. Endemic to the Mekong River system, the MGC is critically endangered. Six MGC were transported from Thailand to the Gifu World Freshwater Aquarium (GWFA) in Japan in May 2004. The MGC have not been measured since they were introduced to the GWFA due to the challenges associated with their large size. Traditional methods, such as anesthesia and net-capture, could harm the fish, potentially reducing the population further. However, understanding the ecology of the species is essential to conservation efforts. Seasonal feeding rhythms and long-term fasting were previously observed in captive MGCs. To investigate the effect of long-term fasting on MGC growth, total length measurements are necessary. In this study, we applied a noncontact method to estimate the total length via image analysis. We shot a stereo-video of free-swimming MGC in the aquarium tank using two digital video cameras and analyzed the 3D images using the direct linear transformation method. We successfully estimated the total length of each MGC individual without contacting the fish. This accurate estimation method is versatile, simple, and useful in aquarium breeding and is also recommended from the viewpoint of animal welfare because it is a noninvasive method of measurement.

In marine aquaculture, fish populations constantly decrease throughout the cultivation period because of mortality and escape. Current production management systems provide limited opportunities to count the cultured fish, making it difficult to estimate accurately the fish population in the cage. To overcome this problem, an automatic fish identifying method based on particle tracking velocimetry (PTV) flow visualization technology is proposed in this paper. The proposed method utilizes an image processing unit that extracts individual fish from the acquired image and a motion analysis unit that calculates the motion vector for each individual. Thus, the accuracy of the extraction results in the image processing unit affects the system's counting results. To validate the efficiency and robustness of the image extraction performed by the image processing unit, individuals were extracted from images using the open-source image deep learning semantic segmentation method (SegNet), which is able to distinguish between the background and foreground in the images via analysis at the pixel level. SegNet is able to improve the image discrimination performance by multiplying the learning paths, and the robustness of the detection results can be ensured by changing the layer structure according to the detection target. Accordingly, the use of SegNet was evaluated in terms of the number of layers and images in the training set. The results of this study indicate that the application of SegNet with PTV technology represents a promising method for the automatic identifying and behavioral tracking of fish in an aquaculture net cage.

Fish form schools because of many possible reasons. However, the hydrodynamic mechanism whereby the energy efficiency of fish schools is improved still remains unclear. There are limited examples of fish models based on actual swimming movements using simulation, and the movements in existing models are simple. Therefore, in this study, we analyzed the swimming behavior of Biwa salmon (Oncorhynchus sp., a salmonid fish) using image analyses and formulated its swimming motion. Moreover, computational fluid dynamics analysis was carried out using the formulated swimming motion to determine the fluid force acting on the fish body model with real fish swimming motion. The swimming efficiency of the fish model under parallel swimming was obtained from the calculated surrounding fluid force and compared for different neighboring distances. The flow field around the fish model was also examined. The swimming efficiency of two fish models swimming parallelly was improved by approximately 10% when they were separated by a distance of 0.4L, where L is the total length of the model. In addition, the flow field behind the fish body was examined under both inphase and antiphase conditions and at inter-individual distances of 0.8L and 1.2L. The apparent flow speed in the distance range of 0.5-2.0L from the midpoint of the snouts of the two individuals was lower than the swimming speed. The pressure distribution on the fish model showed an elevated pressure at the caudal fin. Interestingly, we obtained an isopleth map similar to that of a caudal peduncle. To avoid a negative thrust, the aft part of the body must be thin, as shown in the isopleth map obtained in this study.

In this study, we conducted three experiments to determine the performance of a multi-stereovision technique to enhance the accuracy of fish body measurement for aquaculture management. We compared the performance of the stereo cameras using two, three, and four stereo cameras in each experiment. First, we compared the performance of camera calibration. Next, we calculated the difference between the estimated three-dimensional positions of the interpolated or extrapolated control points and the true values. Further, we confirmed the relationship between the observational error of the stereo cameras and the distance from the centroid of the analytical space to measure the interpolated or extrapolated control points. Subsequently, we estimated the fork length of the fish in an aquaculture net-cage. We investigated the relationship between the observational error of the stereo cameras and the distance from the centroid of the analytical space to measure the fish positions. The result of each experiment suggests that the multi-stereovision using four cameras has higher performance than that using two or three cameras. The accuracy of stereovision using four cameras was improved more than the conventional stereo measurement using two cameras. As the error in the depth direction is reduced by the multi-stereovision using four cameras, the multi-stereo measurement may have a wider measurement space than the conventional one.

The behavior of fish in the vicinity of fishing gear and in an aquaculture tank should be analyzed to understand their responses to materials such as nets, tank walls, and other physical objects. To monitor the 3D trajectory of target fish in recorded videos automatically and thereby reduce the workload of scientists, we used a particle filter as a state-space model, which is an image processing technique, to analyze video. The behavior of a single chub mackerel of fork length of 0.34 m was monitored using stereo vision cameras in an experimental fish tank. The fish body in the recorded video was binarized using threshold and background subtraction methods of image processing. Then, a segmented image of the fish body including color information was generated by obtaining the product of the raw image and binarized image. The segmented fish body was automatically tracked in the video recorded by the stereo camera using a particle filter. The 3D trajectory of the fish was calculated using direct linear transformation methods based on the 2D trajectories in the stereo image. The trajectory was smoothed using the Kalman filter to minimize measurement error in estimated fish position. Thus, using the proposed method based on image processing, we automatically obtained the 3D trajectory of the target fish. Although the wave effect in field observation remains an issue in applying this method to fisheries research, it can be put to immediate use in tank experiments.

Because escape from a net cage and mortality are constant problems in fish farming, health control and management of facilities are important in aquaculture. In particular, the development of an accurate fish counting system has been strongly desired for the Pacific Bluefin tuna farming industry owing to the high market value of these fish. The current fish counting method, which involves human counting, results in poor accuracy; moreover, the method is cumbersome because the aquaculture net cage is so large that fish can only be counted when they move to another net cage. Therefore, we have developed an automated fish counting system by applying particle tracking velocimetry (PTV) analysis to a shoal of swimming fish inside a net cage. In essence, we treated the swimming fish as tracer particles and estimated the number of fish by analyzing the corresponding motion vectors. The proposed fish counting system comprises two main components: image processing and motion analysis, where the image-processing component abstracts the foreground and the motion analysis component traces the individual's motion. In this study, we developed a Region Extraction and Centroid Computation (RECC) method and a Kalman filter and Chi-square (KC) test for the two main components. To evaluate the efficiency of our method, we constructed a closed system, placed an underwater video camera with a spherical curved lens at the bottom of the tank, and recorded a 360 degrees view of a swimming school of Japanese rice fish (Oryzias latipes). Our study showed that almost all fish could be abstracted by the RECC method and the motion vectors could be calculated by the KC test. The recognition rate was approximately 90% when more than 180 individuals were observed within the frame of the video camera. These results suggest that the presented method has potential application as a fish counting system for industrial aquaculture.

A constrained swimming test showed that at fastest flow speeds, the tail-beat frequencies (TBF) of followers in a school of Japanese chub mackerels were significantly lower than those of the leading fish. The gross cost of transport (GCOT) of schooling fish was 4.8-6.6 J·kg^-1·m^-1, whereas that of solitary swimming fish was significantly larger (9.2-11.5 J·kg^-1·m^-1). To examine the association between the configuration of individuals in the school and energy-saving, the three-dimensional positions of individuals in a school were measured in a large tank. Although the school was not arranged in the diamond pattern known to be most effective for energy conservation, the TBF of the followers remarkably decreased when within the range of 1.4 BL (Body Length) from the leading fish in the school. The energy gain of schooling fish was greater than that derived from the potential-theory basis, implying a potential hydrodynamic synergy effect.

“The tunny proper, the pelamys, and the bonito penetrate into the Euxine in summer and pass the summer there as do also the greater part of such fish as swim in shoals with the currents, or congregate in shoals together. And most fish congregate in shoals, and shoal-fishes in all cases have leaders”.

Vision is an important sensory system for most fish species. Fish eyes are similar to the eyes of terrestrial vertebrates such as birds and mammals, except that they have highly spherical lens. The retinas generally have both rod and cone cells (for scotopic and photopic vision, respectively), and most species have colour vision. Unlike humans, fish normally adjust their focus by moving the lens closer to, or further from, the retina.

It is necessary to understand the processes involved in sinking death in Pacific bluefin tuna Thunnus orientalis aquaculture in order to develop methods to prevent or minimize this problem. We observed the nighttime vertical distribution of Pacific bluefin tuna in the water column on 2-9 DAH and the morphological characteristics of the larvae, in order to clarify the processes involved in sinking death. A cuboid tank (height 300 cm) was used to measure vertical distribution. The number of larvae was counted in each of 4 regions in the observation tank: upper layer (water depth 0-100 cm), middle layer (100-200 cm), lower layer (200-300 cm), and bottom area. The distribution of larvae in these regions at 4 days after hatching was polarized to the upper layer and bottom area. Individuals with inflated swim bladders were observed in the upper layer 3 days after hatching. No larvae with inflated swim bladders were observed in the bottom area on any day after hatching. Total body length and caudal fin aspect ratio of larvae with both inflated and un-inflated swim bladders were greater in the upper layer than those of larvae in the bottom area. Larvae with un-inflated swim bladders that failed to develop sufficiently for swimming sank to the tank bottom and died. Swim bladder development and caudal fin swimming ability are strongly related to sinking death.

We numerically modeled the geometry of a purse seine net and evaluated the model's dynamic behavior during operation. The vessel position, wire payout and roll-up lengths, and current flow speed and direction in situ were collected as the dataset for computation. The mesh sizes of the upper, middle, and lower sections of the main net were changed to double and half their respective sizes, and the sinking speed and depth were compared among nets of different mesh sizes. The net with doubled mesh size in the lower section and constant mesh bar diameter strongly affected the sinking speed and depth. This implies that a large mesh size in the lower part of the main net, which eliminates the need for providing a thicker mesh bar, should be used for enhancing the efficiency of fishing operations.

Behavioral traits under light and dark conditions were evaluated in larval and juvenile stage Pacific bluefin tuna (PBF), Thunnus orientalis. Light/dark conditions affected behavior of PBF throughout the growth stages examined in the present study (from 20 to 55 days post hatching). PBF displayed schooling behavior under light conditions from 27 days post hatching, while they did not display well-coordinated schooling behavior under dark conditions at any point during the present study. Under dark conditions, PBF swam slower than they did under light conditions. From 20 to 29 days post hatching, PBF demonstrated particularly slower swimming behaviors under dark conditions. Results of the present study suggested that both larval and juvenile PBF swam actively under light conditions, and that they formed coordinated schools from 27 days post hatching when exposed to an environment with sufficient visibility.

Swimming trajectories of aquatic animals that are estimated using the dead-reckoning technique below the sea surface tend to have very large associated observational errors. Therefore, the aim of the present study was to develop a technique for removing accumulated errors from such trajectories for Pacific bluefin tuna Thunnus orientalis. Horizontal and vertical speeds and heading angle were measured in an aquaculture net cage using 2 types of data loggers, and current velocity was recorded at a depth of 12 m to measure the tidal current speed around the net cage. Fourier analysis indicated that the primary source of error in trajectory estimates was the effect of ocean currents, which resulted in drift, and further analysis revealed that the frequency contributing to drift was consistent with the low-frequency signal in a spectrum analysis of horizontal speed. Therefore, a high-pass filter was applied to horizontal speed data to remove any frequencies lower than the cut-off frequency (0.0015 Hz), following which these data were back-transformed into a time domain that no longer included the drift effect caused by the current. The reconstructed trajectories fit within the inner diameters of the net cage, indicating that they were realistic. To confirm the validity of the resultant swimming trajectories, a flume tank experiment was conducted, which demonstrated that the high-pass filter effectively removed current drift from the estimated trajectory. Furthermore, since the method was estimated to have a precision of approximately 0.20 m, it not only allows the 3-dimensional trajectories of circling tuna to be estimated but can also be applied to the behavior of fish in the wild.

A digital stereo-camera system was used for 3-dimensional monitoring of cultured Pacific blufin tuna Thunnus orientalis individuals, which were free-swimming in a net cage. The fork lengths and the length frequency distribution of tuna individuals in a net cage were estimated using the Direct Linear Transformation (DLT) method. Information obtained from stereo-images is useful for managing the growth of tuna during rearing. In this research, we are trying to develop the simple method involving a combination of DLT and a commercial image processing software to enable aquaculturists to obtain three-dimensional measurements of fish. Then, we used a digital stereo-video camera system to evaluate the precision and validity of fish size estimates determined from repeated measurements. Of the total assessed individuals swimming within a distance of <9.2m from the camera system, estimates for 100% (37/37) were found to be valid, with an error ratio (standard error/mean) of 5%. Therefore, we believed that our proposed simple non-destructive method for monitoring free-swimming fish could be very useful for aquaculture management.

The balance between capture fishery and aquaculture production is extremely important for maintaining and developing the fishery industry. In this study, we addressed the problem of optimizing aquaculture and capture fisheries production from the viewpoint of effectively using finite aquatic resources. Our goal was to maximize total fishery production by controlling the ratio of aquaculture to capture fishery production. A simple mathematical model was constructed to describe total fishery production as the sum of capture fishery production and aquaculture production. This model did not consider any economic aspects, but simply focused on the production flow. Aquaculture and capture fishery production interacted in the model via the parameter α, which was defined as the ratio of capture fishery production that turned into aquaculture bait. It was assumed that there was a bounded exponential curve relationship between the amount of bait used in aquaculture and aquaculture production. It was also assumed that aquaculture production was not used for bait in aquaculture or capture fishery. Within this framework, total production was critically dependent on the parameter α. We also analyzed the FAO data (nonfood uses, aquaculture, and capture production data were used). The results indicated that total fishery production was maximized if the amount of the bait used in aquaculture is limited to about 15% of capture fishery production. Sensitivity analysis also indicated that the upper bound of α should be maintained at less than 18.9% to maximize total fishery production.

In order to make quantitative statements regarding behavior patterns in animals, it is important to establish whether new observations are statistically consistent with the animal's equilibrium behavior. For example, traumatic stress from the presence of a telemetry transmitter may modify the baseline behavior of an animal, which in turn can lead to a bias in results. From the perspective of information theory such a bias can be interpreted as the amount of information gained from a new measurement, relative to an existing equilibrium distribution. One important concept in information theory is the relative entropy, from which we develop a framework for quantifying time-dependent differences between new observations and equilibrium. We demonstrate the utility of the relative entropy by analyzing observed speed distributions of Pacific bluefin tuna, recorded within a 48-hour time span after capture and release. When the observed and equilibrium distributions are Gaussian, we show that the tuna's behavior is modified by traumatic stress, and that the resulting modification is dominated by the difference in central tendencies of the two distributions. Within a 95% confidence level, we find that the tuna's behavior is significantly altered for approximately 5 hours after release. Our analysis reveals a periodic fluctuation in speed corresponding to the moment just before sunrise on each day, a phenomenon related to the tuna's daily diving pattern that occurs in response to changes in ambient light.

We explore how a stochastic model provides the most promising avenue towards predicting fish movement. To construct a stochastic model describing fish movement, trajectories of ten juveniles in a water tank were analyzed from a stochastic point of view. The heading angle was defined as a random variable. Our analysis found that the most probable forward heading angle was between 0 degrees and 22.5 degrees (probability similar to 78%), followed by angles between 22.5 degrees and 45 degrees (probability similar to 10%). We also found that the choice of future heading angle depends on the current heading angle. Therefore, we treated heading angle state as a first-order Markov process and constructed a correlated random walk model describing juvenile movement in a water tank. Our stochastic model simulated a trajectory similar to observed trajectories. We used the model as a tool for estimating the probability distribution of potential fish path outcomes. We derived the distribution of potential outcomes from a large number of simulations (N = 1000) and investigated these trajectories. We collected a set of juvenile trajectories that collided with the tank and estimated the probability of juvenile collisions with the tank.

The effects of vision development and light intensity on schooling behaviour during growth in juvenile Pacific bluefin tuna Thunnus orientalis were investigated using both behavioural and histological approaches. The schooling behaviour of three age groups [25, 40 and 55 days post hatching (dph)] of juvenile T. orinetalis were examined under various light intensities. Subsequently, schooling variables, such as the nearest neighbour distance (D(NN)) and the separation swimming index (I(SS)), were also measured under different light intensities. Furthermore, retinal indices of light adaptation in juvenile fish at each experimental light intensity and visual acuities in six stages (25-55 dph) of juveniles were examined histologically. During growth, the light intensity thresholds of I(SS) decreased from 5 to 0.05 lx, and D(NN) under light conditions (> 300 lx) also decreased from 9.2 times the standard length (L(S)) to 12 times L(S). The thresholds of light intensities for the light adaptation of retinas in juveniles (25-55 dph) similarly decreased from 5 to 0.05 lx with growth. In addition, the visual acuities of juveniles developed from 0.04 to 0.17 with decreasing D(NN). These data clearly indicate that the characteristics of schooling behaviour strongly correspond to the degree of vision development. Juvenile T. orinetalis also appear to be more dependent on cone rather than rod cells under low light intensity conditions, resulting in a relatively high light intensity threshold for schooling. These results suggest that juveniles can adapt to darker conditions during growth by developing improved visual capabilities. (C) 2011 The Authors Journal of Fish Biology (C) 2011 The Fisheries Society of the British Isles

We used a digital stereo-video camera system for three-dimensional monitoring of cultured Pacific bluefin tuna, Thunnus orientalis, swimming freely in a net cage. We estimated the fork length and length frequency distribution of individual fish using the direct linear transformation (DLT) method. Information obtained from stereo images is useful for managing the growth of tuna during rearing. Our aim was to develop a simple method involving a combination of DLT and commercial image-processing software to enable aquaculturists to obtain three-dimensional measurements of fish. In this study, we used a stereo-video camera system to evaluate the precision and validity of fish size estimates determined from repeated measurements. Of the total assessed individuals swimming within a distance of <5.5 m from the camera system, estimates for 99% (106/107) were found to be valid, with an error ratio (standard error/mean) of <5%. Therefore, we believe that our proposed simple method for monitoring free-swimming fish could be very useful for aquaculture management.

The swimming path of a reared Pacific bluefin tuna, Thunnus orientalis, was measured in a submerged aquaculture net cage to understand how reared fish use the space in such a cage. A bluefin tuna (fork length, FL, 0.51 m) was captured by angling in the cage, and two micro data loggers (PD3GT, Little Leonardo; DST Comp-Tilt, Star-Oddi) were attached to its body. The fish was then released back into the net cage. The PD3GT measured its swimming speed and depth at 1-s intervals and recorded these in flash memory. The DST Comp-Tilt measured the magnetic field strength at 1-s intervals and recorded the heading estimated from the magnetic field strength in flash memory. The fish moved through the water in the cage at speeds of 0.7-0.8 ms(-1) and attained a maximum speed of 3.6 ms(-1). Burst swims exceeding 2 ms(-1) were confirmed only after dark and a significant difference was found between the daytime and night-time swimming speeds (p < 0.001). The fish moved at depths between 2 and 22 m, swimming near the bottom during the day and at 10-15 m at night, with a significant difference in swimming depth between day and night (p < 0.001). The swimming path reconstructed by dead reckoning was visualised using night-time data. For this period, the absolute speed was corrected from 0.75 +/- 0.09 ms(-1) to 0.71 +/- 0.15 ms(-1) by removing the accumulated error from the reconstruction vector. This study allowed us to examine the behaviour of a tagged tuna in three dimensions and is the first to monitor the behaviour of a bluefin tuna in a submerged net cage. Although only one fish was analysed, this study provides useful information on the space use of reared fish in aquaculture net cages. Future studies must obtain sufficient data to understand the underlying generalities of tuna behaviour.

We investigated schooling behaviour and the rheotactic reaction in the dwarf sweetfish Plecoglossus altivelis. Under 400lx illumination, the fish formed schools with low polarity; these aggregations were highly compact. There was no schooling behaviour in darkness (0.01lx). A positive rheotactic reaction occurred at current velocities >3.3cm s^<-1>. At current speeds ≥5.2cm s^<-1>, fish ground speeds declined to zero. There was no clear rheotaxis at current speeds <3.3cm s^<-1>. These results suggested that the stimuli by these slow currents might be too weak to increase the rheotaxis of P. altivelis. Current speed was thus an important factor affecting swimming effort and direction in P. altivelis. Current effects may be particularly strong in dwarf sweetfish because the schools have low polarity.

Schooling behavior traits during the process of retinomotor response from scotopic to photopic vision were examined in cultivated juvenile Pacific bluefin tuna (PBT) at 3 different ages. After a sudden change in illumination from darkness to 300 lx, retinal adaptations changed from scotopic to photopic vision. Retinomotor and schooling indices showed strong agreement, with juvenile PBTs forming polarized schools upon complete retinal adaptation to photopic vision. The behavioral and retinal adaptation to sudden illumination took 20, 15, and 10 min after illumination in PBT 25, 40, and 55 days after hatching (dah). At 40 dah, PBT took a longer time to adapt than fish aged 55 dab and showed the highest swimming speed, including momentary bursts of swimming immediately after illumination. This suggested that these fish were swimming at high speed under poor visibility conditions. In contrast, PBT at 55 dah showed a gradual increase in swimming speed that correlated with their retinal adaptation. Therefore, behavioral and retinal adaptation traits changed during growth, suggesting that the high mortality in PBT around 40 dah, due to collisions with the tank and net walls at dawn, may be because these adapt more slowly than fish at 55 dah and were swimming at a relatively high speed under conditions of poor visibility. (C) 2010 Elsevier B.V. All rights reserved.

Schooling was first observed at 25-27 days after hatching (26.2-33.8 mm, total length) in the Pacific bluefin tuna Thunnus orientalis. At this time, the mode of swimming changed from intermittent sprinting to continuous cruising, and this allowed the fish to adjust to an inertial hydrodynamic environment. (C) 2010 The Authors Journal compilation (C) 2010 The Fisheries Society of the British Isles

In this study, we determined the color vision and spectral sensitivity of pointhead flounder, slime flounder, and red halibut by recording electrophysiological S-potentials from isolated retinas; this basic data was collected to aid the selection of appropriate colors and effective wavelengths for selective fishing gear for sustainable fisheries. Four kinds of L-response and one C-response were recorded in the pointhead flounder and red halibut. Two kinds of L-response were recorded in the slime flounder. The occurrence of a C-response indicates the possession of color vision. The C-responses show that the pointhead flounder has superior distinction from blue to yellow, and red halibut has superior distinction from blue and green to red. The most frequently recorded L-responses indicate high spectral sensitivity at 544 nm in pointhead flounder and 518 nm in slime flounder and red halibut. All recorded L- and C-responses indicate sensitivity to ultraviolet wavelengths.

Large offshore net cages have been rather successful in reducing coastal contamination and developing aquacultural technology for raising large body-size species such as bluefin tuna. Deformation and shrinkage of net cages due to severe current and waves is one of the main causes of mortality of cultivated fish, and is thus of great concern for marine cage aquaculture. Even though the cage depth can be determined by pressure sensors at several locations on the cage, this is generally insufficient to measure the deformation and shrinkage. In this study, the dynamic shape and volume of a net cage under the influence of current and waves was analyzed using a numerical net geometry simulator previously validated by tank tests. The dynamic behavior of finite mass points distributed on the net cage under various wave and current conditions was simulated in detail, and reduction coefficients of the volume were calculated by the positions of these mass points on the cage. A drastic reduction of the cage volume occurred for a current velocity of 0.28-0.39 m/s. As the current velocity increased, the deepest point of the cage changed position, moving toward the downstream direction. The effective reduction of cage volume for a wave height of 3 m and a current velocity of 0.26 m/s was the same as that for a wave height of 5 m without any current. These results suggest that the volume deformation must be based on the actual measured depth of each part of the net cage, that the combination of wave levels and current velocities may have synergistic effects on the reduction of cage volume, and that our computational method is valid for estimating the volume reduction of a net cage under the influence of a concurrent wave and current field.

The visual acuity, visual axis and visual accommodation of pointhead flounder, slime flounder, and red halibut were determined to obtain basic knowledge for developing appropriate fishing gear and fishing methods for sustainable fisheries. Each of these species has a different ecotype in terms of habitat, depth and prey species. Thus, it was hypothesized that they may differ in terms of visual acuity, visual axis and visual accommodation. Few studies have compared these characters in flatfishes from different ecotypes. We used histological methods to determine visual acuity (i.e. cone cell density) and visual axis (i.e. cone cell distribution) in each of these species. The maximum visual acuity was 0.127 in pointhead flounder (total length, TL 344 mm), 0.092 in slime flounder (TL 372 mm) and 0.109 in red halibut (TL 336 mm). Based on the cone cell distribution in the retina, the visual axis was upward and forward in pointhead flounder, forward and downward in slime flounder, and downward in red halibut. Finally, the mean angle of lens movement was -2 degrees in pointhead flounder, -13 degrees in slime flounder and -32 degrees in red halibut. This measurement of lens movement indicated that the average near-point distance was 0.87 x TL in pointhead flounder, 0.65 x TL in slime flounder and 1.02 x TL in red halibut. At similar TL (336-355 mm), the visual acuity of these species differs depending on the direction in which they are looking.

Schooling behaviour and histological retinal light adaptation in juvenile Pacific bluefin tuna Thunnus orientalis were examined under various light intensities to determine the effect of light intensity on behaviour. After monitoring the schooling behaviour of juveniles 35-36 and 45-46 days post hatching, schooling variables such as nearest neighbour distance and separation swimming index were measured under different light intensities. Furthermore, retinal indices of light adaptation were investigated histologically for each experimental light intensity. Under intensities >5 lx, schooling variables in the two juvenile growth stages were nearly constant, allowing schooling. In contrast, the schooling variables indicated that the fish gradually swam more widely and randomly with decreasing light intensities <5 lx. The retinal indices also showed a shift from light adaptation to dark adaptation at light levels <5 lx. From 5 to 0-01 lx, retinal adaptation and fish schooling behaviour changed with light intensity. These data suggest that the schooling behaviour of juvenile Pacific bluefin tuna is greatly affected by retinal adaptation. (c) 2007 The Authors Journal compilation (c) 2007 The Fisheries Society of the British Isles.

Population density and space limitation have proven important considerations for both fisheries management and aquaculture, resulting in intense interest in the development of new techniques and technologies for management and hatchery applications. To investigate the effects of space limitation and population density on the schooling behavior of fish, we examined the schooling behavior of captive juvenile chub mackerel. Three groups of fish were collected; collections were made at 18, 42, and 73 days post-hatch (dph) at which mean body lengths were 2.28, 8.83, and 11.4 cm, respectively. The two-dimensional movement of individuals during 200-s observation periods was digitized and processed. A mathematical model based on Newton's second law of motion was used to quantify the forces dominating schooling behavior. The forces of swimming motion were quantified for each fish in a school as the propulsive force due to the swimming ability of each fish, the interactive force used to keep the proper distance and maintain similar velocity with neighboring fish, and the repulsive force used to keep a proper distance from the wall. The magnitude of the repulsive force was minimized in the 18-dph school, maximized in the 42-dph school, and decreased in the 73-dph school. The magnitudes of the propulsive and interactive forces increased with growth. Thus, the interactive force, which was critical for school formation and maintenance appears to reduce the importance of the repulsive force and causes the decline in the repulsive force between 18 and 73 dph.

Although visual information can help to monitor fish behavior in situ, quantitative analysis from images is difficult. There is little quantitative information from images on fish behavior in situ. In this study, we used digital still-picture loggers (DSLs) to monitor the bearing and tilt angles of walleye pollock, Theragra chalcogramma, in a trawl in situ. We attached two DSLs to a trawl net and recorded still pictures of fish near the mouth of the trawl. Eleven pictures of walleye pollock were obtained during 79 trawl surveys. From the still images, we calculated the bearing and tilt angles of the fish using the ratio of body length and body height. We confirmed the accuracy of the calculated angles by comparing the computed angles using image-processing software. The calculated angles agreed with the angles computed using the software. Therefore, our simple method can be used to monitor fish bearing and tilt angles in situ. Based on our results, bearing angles of walleye pollock were not shown in the same direction to a trawl. Furthermore, the fish tilt angles differed significantly between adults and juveniles. (C) 2005 Elsevier B.V. All rights reserved.

In this study, the visual acuity, visual accommodation, and spectral sensitivity of the elkhorn sculpin were determined. The elkhorn sculpin Alcichthys alcicornis is a typical groundfish that has conspicuously large eyes. They are caught by bottom gillnets, trawl nets, bottom line angling, and other fishing methods. Although fish behavior arising from visual stimuli is important in these fishing methods, there is little information available regarding the visual sensitivity of the elkhorn sculpin. Three specimens, with total lengths of 300-310 mm, were studied. Visual acuity was calculated, from the maximum cone density in the retina, to be between 0.10 and 0.14. The minimum separable angles were calculated to be 7.14-10.37 min. Measurement of lens movement in five specimens indicated that the near-point distance was between 0.87 and 1.53 times the total length. Lens direction was estimated to be 11-22 degrees in the downward visual axis. Spectral sensitivity was also determined electrophysiologically by recording S-potentials. Three types of response were obtained from 21 specimens, including L responses that showed hyperpolarization at all wavelengths of stimulating light, and two C responses that showed wavelength-dependent hyperpolarization and depolarization. Based on these results, the elkhorn sculpin has color vision and its sensitivity is most acute in light with a wavelength of 554 nm.

Understanding the behavior of Jacopever Sebastes schlegeli to visual stimulation is important and requires determination of visual sensitivity through the use of effective stimulation. In the present study, we Measured the visual acuity, visual accommodation and spectral sensitivity of Jacopever. Visual acuity was calculated from the maximum cone density in the retina. In five specimens, the total length ranged from 17.6 to 24.5 cm and the calculated minimum resolving angles and visual acuities were 9.5-10.7 min and 0.09-0.11, respectively. Measurement of lens movement for visual accommodation indicated a mean distance and direction of movement of 0.31 mm and 16.0degrees in an upward and forward direction, respectively. The direction of accommodation and visual axis measured from the retinal area of maximum cone density were almost identical. The calculated focal distance was 0.74-fold the total length in the nearest point to infinity. We also determined the spectral sensitivity electrophysiologically by recording S-potentials from 45 specimens of total length of 15.4-47.5 cm. The S-potentials were recorded from 300 horizontal cells in the retina. All responses showed a similar spectral pattern, consisting of hyperpolarization in response to various wavelengths of stimulating light; the maximum amplitude of the response was recorded at 522 nm.

賢く獲る漁業技術の開発が近年求められているが，生産手段の最適化は対象生物(魚類)の行動特性の理解なしには実現できない。したがって，魚類の群行動についての基本的理解とそのための研究は不可欠である。 本研究では，魚類が群れを形成する意味を物理的側面からアプローチすることにより，その本質的意義の解明を目指す。背景となる力学的論理基盤から群を形成したことによる移動エネルギーコストの利得が単一個体よりどの程度のアドバンテージとなり，魚群の規模によりどのような差がでるのか明らかにすることを試みている。 PTV（流体可視化計測）実験では，マアジを対象種として3尾で形成される魚群を回流水槽で巡航遊泳させるケースと実際のマアジと翼形状の魚体模型を用いて魚群のように遊泳させるケースについて，魚体後流域の流況可視化実験を2種類のレーザー光源と高速カメラを用いて実施した。当該実験で撮影した画像から流況を明かにするために，魚体後流域の流速ベクトルや渦度分布の解析および圧力分布の算定を現在実施中である。 CFD（数値流体力学）解析実験では，ビワマスを対象種としてデジタルビデオカメラで撮影した遊泳運動から魚体の遊泳運動関数を算定した後，魚群を形成する3個体の魚体モデルに遊泳運動を与えて群泳を再現し，数値流体力学解析により魚体モデルに作用する圧力とその分布を算出することで流体力学的効果を求めた。個体間の距離を全長の0.4倍から2.0倍まで変化させながら様々な配置で3個体の魚群を形成させた結果，単体で遊泳した場合と推進効率が変わらない配置と推進効率が10%程度向上する配置があることが分かった。

生産物が水面下にある魚類養殖では，生産管理や成育状態を日常的に把握することが困難となっている。そのため，きめ細かい生産管理や効果的な養殖技術開発のためにICT導入による高度化が強く求められている。本研究では魚類養殖の成育・生産を効率化するシステムの構築を目指して，非接触で生簀内養殖魚の収容尾数や，成育状態を計測する基盤技術開発に取り組む。 【個体数計数システムの開発】養殖クロマグロを対象として遊泳する養殖魚魚群をワイドコンバージョンレンズを装着した小型民生品カメラを用いて20m直径の大型生簀内のほぼ全景を捉えることができた。セマンティックセグメンテーション（画像内の全画素にラベルやカテゴリを関連付ける深層学習アルゴリズム）を適用して個体画像を識別したところ，およそ5000枚の教師画像データデータセットを個体画像と背景画像の比率7:3で学習させることにより80%の検出性能を出すことができたが，システムのロバスト性を向上させるためには，個体を移動追跡するアルゴリズムを見直し，実海域での個体同士の重なりや懸濁物等のノイズ等の識別を確実化することが求められことがわかった。 【行動計測による養殖魚状態評価技術開発】個体計数システムの開発では，連続する画像間での同一個体の移動ベクトルをKC法を適用することにより推定することが可能となる。このベクトルの状態を分析することにより，養殖環境や適切な給餌量を評価することが実現できる可能性がある。そこで，連続する移動ベクトルのノルム差D，交差角θ，渦度に類似した指標ωの評価指標を算出し，養成環境の違いがこれらの指標に差をもたらすか分析した。給餌時と非給餌時では，θの値で有意な差が確認され，台風接近による風浪の激しい環境では，ノルム差やωの値に有意な差が現れた。養殖魚の飽食度合いや，波浪によるストレス度合をこれらの指標により評価できることが示唆された。

The stocks of tuna species such as Pacific bluefin tuna, bigeye tuna and yellowfin tuna are recently decreasing. In order to maintain the abundance of tuna species, it is required to reveal the behaviour of tuna species. Therefore, we monitored the three-dimensional swimming behaviour of tuna species in the net cages and the purse seine nets using a stereo-video camera system and bio-logging technics. The environmental stimulations (light intensity, net shape, ambient depth, ambient temperature and water current) were also observed. Furthermore, we histologically determined the visual capabilities, which is predominant in tuna species. Based on these results, the integrated mathematical models for induced behaviour in response to the environmental stimulation in tuna species were established.

To develop the fundamental technique of counting number of fish individuals in an aquaculture cage, the basic algorithm for counting individuals was constructed by applying PTV (Particle Tracking Velocimetry) technique. The region above a certain area and brightness was defined as the candidate region of an individual. The movement vectors of the extracted regions were estimated by using Kalman filter, and the number of the vectors was possible to give that of individuals. When 200 individuals were put in an experimental tank, applying the algorithm showed that the averaged error was 5.9% and its availability was confirmed.

To assess the capture process during purse seine fishing of tuna species, the dynamic geometries of the purse seine fishing net under some operations were investigated by using a numerical simulation technique as well as by obtaining three-dimensional swimming paths of three bigeye tuna individuals during the purse seine fishing operation through dead-reckoning algorithms. The vectors of the swimming paths showed great variations in the vertical and horizontal directions. The hem of the purse seine net reached up to 200 m in depth, whereas the tuna individuals swam at depths shallower than 100 m and their swimming speeds decreased by 30% after pursing. The simulation results revealed that by 1500 s after casting the net, immature bigeye tunas could not pass through the body net because of the meshes of the net stretching in the longitudinal direction.

In order to make quantitative statements regarding behavior patterns in animals, it is important to establish whether new observations are statistically consistent with the animal's equilibrium behavior. From the perspective of information theory such a bias can be interpreted as the amount of information gained from a new measurement, relative to an existing equilibrium distribution. We demonstrate the utility of the relative entropy by analyzing observed speed distributions of Pacific bluefin tuna. We show that the tuna's behavior is modified by traumatic stress, and that the resulting modification is dominated by the difference in central tendencies of the two distributions. We find that the tuna's behavior is significantly altered for ca. 5 hours after release. Our analysis reveals a periodic fluctuation in speed corresponding to the moment just before sunrise on each day, a phenomenon related to the tuna's daily diving pattern that occurs in response to changes in ambient light

In the present study, we tried to create the basis for an optimum design for an offshore aquaculture facility for Pacific bluefin tuna (PBT) from not only mechanical aspects but also biological ones after considering the behavior of PBT. We used a numerical simulation system of the netting geometry developed by us to evaluate net deformation and the loads to the facility. The simulation results showed that the net cage was deformed and reduced to 50% of the original volume by current flow when the speed was 1 kt. The observation data obtained using a bio-telemetry system showed that the strong current blew up the net cage and caused a collision between the netting and the cultured tuna. In addition, the cold water mass acted as a strong constraint factor on the vertical swimming of PBT.

During growth of juvenils tuna (25-55 dph), the light intensity thresholds of schooling decreased from 5 to 0.05 lx. The thresholds of light intensities for the light adaptation of retinas in juveniles (25-55 dph) similarly decreased from 5 to 0.05 lx with growth. These data clearly indicate that the characteristics of schooling behaviour strongly correspond to the degree of vision development. These results suggest that juveniles can adapt to darker deep and off shore sea conditions during growth by developing improved visual capabilities

To examine responses to environment changes, we monitored the behavior of cultivated Pacific bluefin tuna in a submerged aquaculture net cage. We confirmed notable changes in the swimming depth of fish when cold water was added to the aquaculture net cage. Additionally, we successfully measured the three dimensional swimming path of the fish. This will enable us to evaluate which factors (i.e., a change in the living space of the submerged net cage by upward or downward movement or a decreased volume caused by ocean currents) affect the behavior of the fish by reconstructing their swimming paths.