Animals

ObjectiveStall housing of pregnant sows raises welfare concerns, whereas conventional group housing systems often reduce space efficiency. This study evaluated the effects of periodic group housing (PG) on reproductive performance and welfare compared with continuous stall housing (CS). MethodsSows in the CS group (n = 15) were continuously housed in stalls. In the PG group (n = 15), sows were housed in groups of three and allocated 1 day of group housing and 6 days of stall housing per week over 10 weeks. During group housing sessions, the sows had access to a group housing area containing sawdust. Behavioral observations and salivary cortisol measurements were conducted on the first day of the stall housing session in weeks 1, 6, and 10. Behavioral indices were expressed as proportions based on 90 sampling points recorded at 1-min intervals. ResultsThe number of stillbirths was significantly lower in the PG group than in the CS group (0.63 vs. 1.49 piglets per litter). whereas other reproductive outcomes, including total litter size and average birth weight, did not differ. In older parity sows, the PG treatment markedly increased the proportion of time spent lying, suggesting reduced discomfort associated with restricted movement. Furthermore, the proportion of exploratory behavior decreased markedly, and drinking behavior showed a decreasing trend across parity levels in the PG group, suggesting partial satisfaction of motivations for environmental exploration and oral manipulation. The proportion of oral abnormal behavior showed a pronounced interaction between housing treatment and experimental week, increasing from week 1 to week 6 in the PG group. Salivary cortisol concentrations did not differ between the groups. ConclusionPG may improve reproductive performance and partially satisfy the behavioral motivations restricted under continuous stall housing. This system may represent a practical alternative for improving animal welfare while minimizing economic losses.

Zoological gardens strive to prioritize excellent animal care and adhere to the highest standards of animal welfare, ensuring that the conditions of animals` social and physical environment are as close as possible to those in the wild. This study investigates the effects of group split and relocation to the new enclosure on the behaviour of western lowland gorillas (Gorilla gorilla gorilla) (N = 7) living in Prague Zoo, Czech Republic. We conducted over 289 hours of behavioural observations focusing on daily activities, social interactions, and behaviours that serve as potential stress and welfare indicators. The group split led to establishment of two groups in two separate enclosures; the old enclosure consisted of a "bachelor group" (i.e. three males) and a new enclosure consisted of a mixed-sex group (i.e. three females and one juvenile male). The behavioural comparisons across different study periods were conducted using linear mixed models (LMMs). The changes led to an increase in time spent moving, feeding, being in social proximity, and higher rates of approaches among the gorillas, as well as to a decrease in rates of self-directed and "undesirable" behaviors. Our findings indicate that the gorillas effectively adapted to the changes, most likely by relying on social support, to navigate new conditions. This study contributes to our understanding of how socio-cognitively complex species cope with necessary alterations in captive animal care programs. Furthermore, these observations may inform strategies to enhance the welfare of zoo-housed animals and to improve their captive care.

The recent shift in Bangladesh from tradition backyard rearing system to modern commercial layer farming has made the birds immune to infectious diseases. Pediculosis, however, continues to pose a challenge in modern production system due to its invasive nature, often going unnoticed and neglected as it is typically non-lethal, yet capable of causing significant production losses. Lice infestation is a persistent threat in poultry production; however, its implications in battery-caged commercial layer hens in Bangladesh remain insufficiently characterized. The study aimed to identify the causative louse species and evaluate its associations with clinical pathology, hematological alteration and productive performance in 30 white-feathered (15 infested + 15 non-infested) and 30 brown-feathered (15 infested + 15 non-infested) laying birds from two commercial farms in Tangail. Morphological characterization confirmed the parasite as Menacanthus stamineus, distinguished by a dorsoventrally flattened body, parabolicallly rounded head wider than long, concealed club-shaped antennae, an oblong-oval abdomen with fine setae and three pairs of short legs each bearing paired claws. Infested birds exhibited consistent clinical pathology, including pale combs, petechial hemorrhages around the vent, severe feather damage with alopecic and exudative areas and incidence of irregular and broken-shelled eggs. Production performance analysis revealed significant reduction in hen-day egg production, egg weight, and feed intake, accompanied by significantly increased feed conversion ratios. Hematological evaluation demonstrated significantly reduced hemoglobin concentration, hematocrit and erythrocyte counts in infested hens, indicating mild anemia and compromised oxygen-carrying capacity. Collectively, pediculosis was strongly associated with lice-induced self-inflicted injury and cannibalism, systemic physiological stress, impaired erythropoiesis, reduced production efficiency and compromised welfare in caged laying hens. To best of our knowledge, it was the first integrative reports from Bangladesh documenting M. stramineus infestation in battery-caged commercial layer system with concurrent evidence of hematological disruption and measurable productivity losses, underscoring its epidemiological and economic significance and urgent need for targeted, evidence-based ectoparasite control strategies.

Currently, global conservation efforts for wildlife focus on a limited number of cell types and species. Although protocols from domestic and non-threatened related species have been applied to endangered species, cryopreservation techniques are species-specific and are constrained by a lack of understanding of reproductive biology in these species. Based on a review of 126 original studies from 27 countries, encompassing 160 species, we assess the current state of cryopreservation in wildlife, including gametes, embryos, somatic cells, and various tissues. Furthermore, we focused on the most homogeneous and frequently studied cell type in wildlife cryobanking: mammals sperm.. A meta-analysis of 27 studies was conducted to examine species-specific and protocol-dependent factors that affect post-thaw sperm quality. Our findings provide quantitative estimates of cryopreservation for various cell types and tissues in wildlife taxa. Furthermore, they serve as a crucial research roadmap, identifying major challenges in cryopreservation and proposing solutions.

Non-human primates are crucial animal models in biomedical research including neurosciences. Legal frameworks ensure that participation of monkeys in biomedical research follows an ethical assessment, driven by specific guidelines for ensuring animal well-being. Those guidelines promote continued refinement of practices to improve animal wellbeing, yet quantitative data to inform such refinements is limited. In this context, we investigated the impact of modifications to the housing that created either navigational complexity or visual access to neighbouring animals on the behaviour of rhesus macaques. Then we quantified the value of different enrichment programmes based on manipulable objects. We showed that making navigation within a modular housing more complex, or adding transparent separations between housing units is associated with significant reduction of severe aggression, stereotypical behaviours and a significant increase of appeasing behaviours. We also observed a positive effect of manipulable objects with an increase in the expression of appeasing behaviours and a decrease in abnormal or stereotypical behaviours. Finally, our study revealed significant inter-individual variability that was not explained by sex, age, social network size or social status. Overall, our results point to the importance and benefits of environmental changes and enrichment protocols, and of the critical need for personalised enrichment programmes.

Development and use of assisted reproductive technologies (ARTs) in non-domestic species provides novel tools for species conservation. As a first step towards in vitro embryo production, we developed an OPU technique for two antelope species, scimitar horned oryx (Oryx dammah) and roan antelope (Hippotragus equinus) utilizing a custom-made needle guide and existing OPU equipment utilized by livestock and human practitioners. Females were anesthetized and placed in sternal recumbency for transvaginal OPUs. Prior to OPUs (36 - 45 hours), SHO and roan were either hormonally stimulated with follicle stimulating hormone (FSH, 140 or 250IU) as a single injection or not. A total of 32 and 26 OPUs were completed in SHO (n=10) and roan (n=7), respectively, representing one to four OPUs per animal at monthly intervals. A total of 141 oocytes were recovered from 215 follicles in SHO and 31 oocytes from 58 follicles in roan. FSH dose (250IU) increased (P<0.05) the number of follicles aspirated and the number of oocytes recovered in SHO. No effects of FSH were observed in roan (P>0.05). Good quality oocytes were recovered from all females and procedures were conducted in four consecutive months with no evidence of scar tissue buildup or reduced capacity to recover quality oocytes. These ARTs can be used to develop in vitro embryo production tools for population management and the preservation of female genetics; bolstering genetic diversity and guarding against extinction.

The concept of "wild" is poorly defined when applied to non-human animals, despite its central role in conservation, wildlife management and environmental policy. Despite numerous attempts to define "wildness", the distinction between "wild" and "non-wild" remains inconsistently applied. A shared, empirically grounded understanding of wildness would strengthen communication, support policy and legislation, and clarify human attitudes towards wildlife and nature. We assessed whether a shared understanding of wildness exists within the research community. A total of 358 professionals from different fields completed an online survey comprising 44 wildness-related statements along established measures of attitude and acceptability toward wildlife (AATW) and nature-relatedness (NRS). . Exploratory Factor Analysis revealed six components (19 statements): Human-mediated animal availability, Urbanisation, Independence from humans, Human-wildlife perceived conflict, Individual history of wildness, and Feralisation. These factors explained 43.91% of the variance. Mediation analysis showed that urbanisation and human-wildlife perceived conflict shape nature-relatedness by influencing attitudes and acceptability towards wildlife. Respondents demographic and professional backgrounds influenced their conceptualisation of wildness. While there is a shared latent structure of wildness across fields, it is multidimensional. This has implications for developing working definitions applicable to research, policy and practice.

The white shark Carcharodon carcharias (Linnaeus, 1758) is one of the most iconic and threatened marine predators globally and is listed as critically endangered by the IUCN in the Mediterranean Sea, where populations are primarily affected by fishing pressure and the degradation of critical habitats. Available information on Mediterranean white sharks mostly relies on fishing interactions and opportunistic surface records, thereby limiting evaluations on the species ecology. Here, we report an exceptionally rare underwater encounter with this species, documented by divers involved in an expedition aimed at removing ghost fishing gear from a World War II shipwreck in the Strait of Sicily (SoS), central Mediterranean. Given the exceptional nature of this observation, we analysed the video footage in detail with the aim of providing morphological information and novel insights into the species ecology. The footage analysis also allowed close observations on behavioural aspects and on interspecific and parasitic associations, thus providing information that is rarely achievable through surface sightings or from captured individuals, which historically represent the majority of records for this species in the region. To contextualize this observation within the SoS--an area suggested to function as a nursery and reproductive ground for the species--we also provide an updated compilation of records reported in the scientific literature over the last decade, adding 18 new records to existing datasets. This study highlights the value of shipwrecks as observation sites for rare and elusive species, and underscores the importance of non-invasive, video-based approaches in advancing knowledge of a species whose ecology in the Mediterranean remains poorly understood.

Classical swine fever (CSF) is a highly contagious disease caused by Classical swine fever virus (CSFV), posing a serious threat to the global swine industry. This study aimed to investigate the effect of CSFV on differential genes of histone lactylation at the H3K18 site in the PI3K-AKT signaling pathway. The site with the most significant change in histone lactylation antibody level was screened by Western blot. Omics analysis was performed using CUT&Tag technology to identify differential genes in the PI3K-AKT pathway between the CSFV-infected group and the mock group, followed by validation using RT-qPCR. Functional analysis of significantly differential proteins was conducted, and the protein expression level of THBS4 was detected by Western blot. The results showed that after CSFV infection of 3D4/21 cells, the H3K18la site exhibited the most significant difference in antibody level. A total of 8,859 differential genes at the H3K18la site were identified by CUT&Tag analysis, including 6,349 up-regulated genes and 2,510 down-regulated genes. Further focusing on the PI3K-AKT signaling pathway, 10 differential genes were identified, comprising 6 up-regulated genes and 4 down-regulated genes. Compared with the control group, the mRNA expression levels of CD19, LAMA1, PDGFRA, BDNF, ANGPT4, and THBS4 were up-regulated in the CSFV-infected group, while FOXO3 and NRTN were down-regulated. Western blot results showed that the protein expression level of THBS4 increased after CSFV infection. These findings lay an important foundation for understanding the molecular mechanisms regulating viral replication and immune evasion, and have significant scientific implications and potential application value.

1. Video-tracking programs have now become an essential tool for researchers measuring animal behavior across biological fields. The panel of available programs is growing rapidly, providing researchers with numerous specific tools that will match their precise needs. However, their proliferation may complicate post-tracking data processing, and some programs do not even provide tools for correcting tracking errors or analysing tracking data. In the case of commercial software, the loss of access to a program due to budget limitations or researchers' mobility from one institution to another could prevent them from accessing and visualizing their tracking data. 2. There is therefore a growing need for an accessible and flexible tool to handle post-tracking processes such as the correction and analysis of tracking data obtained across different video-tracking programs. 3. We present here the latest update of the video tracking and analysis program AnimalTA. With this new release, we propose to solve the above-mentioned problems by providing the scientific community with a program that will allow for data importation from other video-tracking programs. Like in its previous versions, AnimalTA remains a free, open-source, and highly user-friendly program, ensuring that it will always be accessible without restriction. Now, with this new importation option, users who performed their tracking with other programs can benefit from AnimalTA's complete toolset of data visualization, correction, and analysis. 4. Finally, this article gives an overview of the other main improvements associated with this new release. The program is now faster in both video importation and tracking, proposes an amplified toolset for data visualisation and correction, and features new options for data analysis.

Dog (Canis lupus familiaris) vocal communication occurs across a wide range of social contexts, including dog-human social play, a common and dynamic interaction in which diverse vocal signals are expressed, particularly in young dogs. However, most available open datasets of dog vocalizations focus primarily on barks, leaving other sound types underrepresented. Here we present a bioacoustic dataset of dog-human social play sessions recorded under standardized laboratory conditions, comprising 30 play sessions involving 17 young dogs (6-24 months old) of different breeds playing with familiar humans. Raw audio recordings are accompanied by two layers of annotations covering the dog sounds produced during sessions. Video recordings of the sessions are provided for contextual reference, along with metadata describing each dog and the experimental sessions. Additionally, permutation-based classification analysis showed that annotated sound categories exhibit above-chance and generalizable acoustic differences across individuals. The dataset may support research on dog vocal communication and expand the range of sounds documented during positive dog-human social interactions.

The advancement in deep learning techniques has made Convolutional Neural Networks (CNNs) a powerful tool for the fully automated classification of zooplankton images. In this study, we systematically investigate how network selection, colour information and differences in imaging instruments affect the classification of zooplankton images by comparing multiple state-of-the-art CNNs on images of zooplankton and marine snow from the in situ Continuous Particle Imaging and Classification Sensor (CPICS), Video Plankton Recorder (VPR), In Situ Ichtyoplankton Imaging System (ISIIS), and the on-board Plankton Imager (Pi-10). With differences between models of 7.8 to 19% in F1-score, we find that model selection strongly affects the classification performance, with EfficientNetV2S showing the most reliable overall performance. Moreover, differences between model architectures are largest for the least abundant classes (<100 labeled images), which implies that when these are present, careful model selection is most beneficial. The high image quality of the Pi-10 strongly increases the performance for the least abundant classes compared to the other instruments. In addition, we find a significant correlation (r = 0.597) between ImageNet the performance and F1-score on zooplankton images, which implies that more generally, a model that performs well on ImageNet will perform well for zooplankton classification. Colour information increases the F1-score of the best performing classifier with 2.8%, but provides a stronger benefit (25% F1-score) for classes with <100 images. The overall performance increase of colour information is less than expected and questions the advantage of recording colour information for zooplankton.

The non-consumptive use of sea turtles has become a rapidly growing sector of the global tourism industry and is increasingly recognised as an important source of economic benefit for coastal communities. In the Maldives, however, the socio-economic value of sea turtles remains poorly quantified. Building on a preliminary survey conducted in 2019, this study provides a comprehensive assessment of the socio-economic value of sea turtles to the tourism industry, post-pandemic, while identifying sea turtle viewing hotspots with high tourism pressure that may require stricter regulation. Questionnaires collected information on operations related to sea turtle excursions in 2022, including the estimated direct spend on sea turtle excursions and the perceived value of sea turtles to tourists and operators. Results include responses from 117 tour operators across 20 atolls in the Maldives, with 68% from resort operators, 27% from local island operators, and 5% from liveaboards. Maldivian nationals made up 55% of respondents and 78.8% of the people working, directly or indirectly, on sea turtle excursions in 2022. The direct revenue generated by sea turtle tourism in the Maldives is estimated at US$10.9-14.4 million in 2022, with individual turtles at high-use sites generating up to US$29,709 per year. These findings underscore the significant economic and social importance of sea turtles to the tourism industry in the Maldives, matching that of manta rays and sharks, and highlights the need for improved conservation efforts to safeguard local sea turtle populations and their associated benefits for Maldivian communities.

BioBlitzes are widely used citizen science events that combine biodiversity monitoring, public participation, and environmental awareness through short and intensive observation campaigns. However, applying this model to marine environments presents additional challenges related to safety, access, weather dependency, specialised equipment, species identification, and sustained participation. This paper presents the BioMARathon model as a case study of how BioBlitz-inspired events can be adapted to marine citizen science contexts. The BioMARathon extends the conventional BioBlitz format into a longer, seasonal, and distributed engagement model designed specifically for marine and coastal environments. The paper describes the conceptual foundations of the model in the Janus Engagement Framework, which informed both the design of the BioMARathon and the adaptation of the MINKA citizen science observatory to better support participation, validation, feedback, and continuity over time. BioMARato Catalunya, launched in 2021, is presented as the founding implementation of the model and as the basis for later replication in Portugal. Between 2021 and 2025, BioMARato Catalunya showed continued growth in participation, observations, and taxonomic coverage, while also contributing to the detection of non-indigenous species, first regional records, and climate-related ecological impacts. Beyond biodiversity outcomes, the case suggests that extending participation across a season, distributing activities through local mobilising organisations, and combining expert validation with visible feedback mechanisms can support recurrent participation, retention, and community reactivation in marine citizen science. Rather than offering a formal causal evaluation, this article contributes practical lessons for the design of citizen science initiatives in challenging environments.

Global efforts to prevent and mitigate the impacts of high pathogenicity avian influenza (HPAI) H5 on domestic animals, humans, and wildlife rely on timely and transparent information that is both accurate and interpretable across countries and sectors. International epidemiological and genomic databases, such as the World Animal Health Information System (WAHIS), the Global Animal Disease Information System (EMPRES-i+), the Global Initiative on Sharing All Influenza Data (GISAID), and the National Center for Technological Bioinformation Virus Portal (NCBI) provide essential information for surveillance, research, and decision-making. To evaluate how well these resources capture recent wildlife impacts, we consolidated information from these databases and complementary public sources including government reports, scientific literature, and news articles, on wildlife mortality associated with HPAI H5 in the Americas from November 2021 to July 2024. The consolidated dataset comprised 615,883 wild birds (287 spp.) and 63,409 wild mammals (39 spp.). In comparison, WAHIS represented 16,902 wild birds (261 spp.) and 6,323 wild mammals (31 spp.) while EMPRES-i+ captured a substantially smaller portion of affected host diversity for both wild birds (105 spp.) and wild mammals (27 spp.). Genomic databases (GISAID and NCBI) represented 7,027 whole genome equivalents of H5 viruses from wild birds (175 spp.) and 371 from wild mammals (26 spp.). These discrepancies indicate that international databases, while essential, provide an incomplete picture of HPAI impacts on wildlife, with significant geographic and taxonomic asymmetries attributable to differences in surveillance capacity, reporting practices, sequencing effort, and data-sharing pathways. Studies and management strategies relying on these resources without complementary validation may therefore mistake data gaps for real-world epidemiological patterns. Strengthening data reporting standards, improving validation procedures, and integrating international databases with national reports, scientific publications, and other sources will enhance the reliability of epidemiological analyses and support more effective One Health surveillance, risk assessment, and conservation action. Author summaryHigh pathogenicity avian influenza (HPAI) H5 viruses, often called bird flu viruses, can cause severe disease in birds and mammals, including humans. Because of their relevance for human health, livestock production, and wildlife conservation, international databases were established to share information on when and where these viruses are detected, which species are affected, and what virus strains are found. These databases are essential tools for governments, scientists, and conservation practitioners working to track outbreaks, understand how these viruses spread and evolve, and guide surveillance and response. In this study, we compiled and compared information on recent HPAI H5 events in wildlife in the Americas available in international databases with information from other public sources, including reports from governments, scientific literature, and news articles. We found important discrepancies in how countries and species affected were represented across sources. As a result, international databases might not fully capture the actual distribution or conservation impact of HPAI H5 on wildlife. Our findings also show why decision-makers and scientists should interpret database-derived patterns carefully. We provide recommendations to improve international databases to address these gaps and better inform One Health risk assessment and wildlife conservation actions.

The moonfish Mene maculata (Bloch and Schneider, 1801), the sole extant species of the family Menidae, is widely distributed in the Indo-West Pacific. Although its spawning season has previously been inferred from seasonal variation in the gonadosomatic index (GSI), the histological basis for interpreting GSI as an indicator of spawning-capable status remains limited. Here we describe the gonadal structure of male and female M. maculata and evaluate how GSI and standard length relate to spawning-capable status based on germ cell development. The testis was lobular and exhibited an unrestricted spermatogonial distribution, a structure widely observed among neoteleosts, indicating that the phylogenetic distinctiveness of M. maculata was not associated with a distinctive testicular structure. In the spawning-capable female, oocytes at multiple developmental stages, from primary growth to oocyte maturation, co-occurred within the ovary, indicating asynchronous ovarian development. This finding suggests that M. maculata may be a batch spawner rather than a total spawner as previously inferred. The spawning-capable female had a GSI consistent with previously inferred spawning season estimates. By contrast, histologically examined males were spawning capable at GSI values lower than those previously associated with the inferred spawning season, suggesting that male spawning-capable status may persist beyond the period inferred from elevated GSI alone. This study provides the first histological description of reproductive biology in the phylogenetically distinctive M. maculata and establishes a histological basis for interpreting GSI as a reproductive indicator.

The increasing impact of humans on natural habitats leads to an increase in human-wildlife conflict (HWC), specifically when there is competition for shared resources. In freshwater systems such as Lake Nasser, Egypt, co-occurrence of local fishermen communities with Nile crocodiles (Crocodylus niloticus) poses critical challenges for both livelihoods and biodiversity conservation. Understanding local perception of crocodiles by local fishermen is therefore essential for developing effective and socially accepted management responses. We used a structured questionnaire to assess how fishermen perceive and respond to crocodiles across three attitudinal domains: (1) perceived threats, (2) perceived economic benefits, and (3) conservation or co-existence values. Forty-two fishermen were interviewed across multiple khors (side arms of the lake). The responses showed a multifaceted picture: while many local fishermen associated crocodiles with gear damage, reduced catches, and livelihood risks, support for crocodile protection and recognition of their ecological role were also widespread. Interest in crocodile-based livelihood opportunities, such as harvesting or collecting the hatchlings, was generally low, suggesting social, cultural, or legal barriers to such approaches. Fishing experience influenced perceptions, with fishermen encountering crocodiles more frequently reporting decreased catches and greater concern. Cluster analysis further revealed three different respondent groups with different attitudes: conflict-oriented, moderate, and coexistence-oriented. Support for crocodile protection was the strongest predictor of belonging to the pro-co-existence group. Our findings underscore the complexity of human- crocodile co-occurrence in Lake Nasser and, on a wider scale, add to the existing cautions against simplistic mitigations of local HWC. Effective conservation and livelihood interventions will require participatory, context-sensitive approaches that integrate the different perceptions and attitudes of local people.

The automated classification of animals from photos is important in ecology and conservation biology for organizing and understanding the immense diversity of species, as well as facilitating effective conservation and management practices. It is equally important for disease surveillance systems, allowing prompt detection of anomalies in species distributions and boosting citizen-scientist platforms by making user-reported data more accurate and convenient. Image classification uses photos and can also rely on the geographical locations of animals to improve performance. While image classification models have difficulties in classifying low-quality images, unbalanced datasets, and with a small number of images, species distribution models have difficulty in classifying species that coexist in a given region. We propose here strategies for combining image classification models based on deep neural networks with species distribution models using genetic algorithms. The proposal is applied to a real-world dataset comprising fifteen classes of animals from the Brazilian fauna obtained from Fiocruzs citizen-scientist Wildlife Health Information System (SISS-Geo). The SISS-Geo photos portray the reality of animals in their environments, with varying quality, and pose numerous difficulties for classification. Experimental results demonstrate that the proposed integration consistently outperforms standalone models. While individual SDMs achieve Top-1 accuracies of 27.79% (MaxEnt) and 31.76% (Bioclim), and CNN-based classifiers reach 58.17% with ResNet50 and 64.13% with ResNet-152, the hybrid strategies yield substantial improvements. The genetic algorithm-based integration with a single global weight achieves up to 67.96% Top-1 accuracy, whereas the class-specific integration using fifteen parameters attains the best overall performance, reaching 69.03%.

Intraspecific morphological variability presents a complex challenge for biological systematics and biomonitoring, particularly for organisms with high phenotypic plasticity, such as zooplankton. Morphological differences between individuals of the water flea species Bosmina longirostris (Crustacea: Cladocera) are difficult to distinguish visually, parthenogenetic females look morphologically uniform within the species; nevertheless, they demonstrate differences attributable to their geographic origin and developmental stage. A reference dataset of microscopic images was created for the study, including populations from two geographically separated regions (seven ones from European Russia and seven ones from Sakhalin Island in the Pacific Ocean (Far East of Russia) and two age groups, demonstrating the ability of a neural network classify to successfully the intraspecific morphological variation. This study demonstrates that deep learning methods are prospective for the detection and understanding of fine morphological intraspecific differences in the cladocerans.

Fixed larval specimens often shrink and curve, making length measurement labor-intensive. Although recent studies have demonstrated efficient fish-length estimation from images using deep learning, methods for estimating curved length remain limited. Furthermore, although deep learning is a powerful method for object detection in images, an essential step for length measurement, it requires preparing large amounts of training data, which can hinder practical implementation. In this study, we used a zero-shot model that requires no training to detect fish in an image. The curved length was then estimated using an image-processing approach that combines image thinning with Bezier curve approximation, and its accuracy was evaluated. We analyzed 1,040 larvae from five tuna species captured in stereomicroscope images. Manual measurements (notochord length or standard length; 1.5-8.5 mm) were conducted by two measurers and served as reference values. Fish regions were detected using GroundedSAM, and curved body centerlines were extracted through image thinning and approximated with Bezier curves. The curve length was used as the estimated body length. Estimation accuracy was assessed using bias and standard deviation between estimated and manual measurements. GroundedSAM detected all 1,040 fish, although there were 49 overdetections. Overdetection was caused by the double-detection of a single fish or by the misidentification of debris and light reflections as fish. Although the standard deviation of the differences between manual measurements and the image analysis-based (IAB) method was larger than the inter-measurer differences, the bias for [&le;]5 mm was comparable to or smaller than the inter-measurer bias. According to the strength-of-agreement criteria for the concordance correlation coefficient, the IAB method demonstrated substantial agreement in the [&le;]5.0-mm range. The IAB method accurately measured most curved tuna larvae without prior training, particularly in the [&le;]5.0-mm range. Combining the IAB method with manual remeasuring can improve the efficiency of curved-length measurement tasks.