The automatic classification of fish species appearing in images and videos from underwater cameras is a challenging task, albeit one with a large potential impact in environment conservation, marine fauna health assessment, and fishing policy. Deep neural network models, such as convolutional neural networks, are a popular solution to image recognition problems. However, such models typically require very large datasets to train millions of model parameters. Because underwater fish image and video datasets are scarce, non-uniform, and often extremely unbalanced, deep neural networks may be inadequately trained, and undergo a much larger risk of overfitting. In this paper, we propose small convolutional neural networks as a practical engineering solution that helps tackle fish image classification. The concept of “small” refers to the number of parameters of the resulting models: smaller models are lighter to run on low-power devices, and drain fewer resources per execution. This is especially relevant for fish recognition systems that run unattended on offshore platforms, often on embedded hardware. Here, established deep neural network models would require too many computational resources. We show that even networks with little more than 12,000 parameters provide an acceptable working degree of accuracy in the classification task (almost 42% for six fish species), even when trained on small and unbalanced datasets. If the fish images come from videos, we augment the data via a low-complexity object tracking algorithm, increasing the accuracy to almost 49% for six fish species. We tested the networks with images obtained from the deployments of an experimental system in the Mediterranean sea, showing a good level of accuracy given the low quality of the dataset.

Streptococcus agalactiae is one of the most important fish pathogenic bacteria as it is responsible for epizootic mortalities in both wild and farmed species. S. agalactiae is also known as a zoonotic agent. In July 2018, a stranded wild sandbar shark (Carcharhinus plumbeus), one of the most common shark species in the Mediterranean Sea, was found moribund on the seashore next to Netanya, Israel, and died a few hours later. A post-mortem examination, histopathology, classical bacteriology and advanced molecular techniques revealed a bacterial infection caused by S. agalactiae, type Ia-ST7. Available sequences publicly accessible databases and phylogenetic analysis suggest that the S. agalactiae isolated in this case is closely related to fish and human isolates. To the best of our knowledge, this is the first description of a fatal streptococcosis in sandbar sharks.

Requiem sharks (genus Carcharhinus) have previously been reported to form large aggregations around marine infrastructures in the eastern Mediterranean Sea. While this behaviour may offer fitness advantages at the individual level, the implications of extended residency at human-altered habitats for population persistence have yet to be assessed. In this work, we investigated the phylogeographic and demographic composition of sharks near a coal-fired power and desalination station in Israel. Our aim was to infer habitat use and the mechanisms underlying the aggregation behaviour, and to highlight potential conservation impacts. We sampled, measured, and released 70 individuals between 2016 and 2022 to assess genetic variability within the cytochrome C oxidase I (COI) region and to analyse the aggregation’s structure based on the sharks’ size and sex distribution. In addition, we performed meristic counts on a reference specimen collected dead at another power station in Israel to supplement species identification using the abovementioned techniques. Our findings indicate size-based sex segregation of adult female dusky and male sandbar sharks (Carcharhinus obscurus and Carcharhinus plumbeus, respectively), with each species comprising two COI haplotypes. In the dusky shark, one haplotype corresponded to an Indo-Pacific lineage, and the other matched an Atlantic lineage. In the sandbar shark, we observed a haplotype previously sampled in the Mediterranean Sea, the Red Sea, the Northwest Indian Ocean, and South Africa, and another haplotype that was unique to our study site and genetically closer to the former than to sequences sampled in other ocean basins. This study provides the first indication of sympatric aggregation amongst phylogeographically distinct dusky and sandbar shark lineages, suggesting that human-altered habitats in the eastern Mediterranean Sea may influence the distribution of these species. Based on the observed segregation pattern, we conclude that the site does not function as a nursery, parturition, or mating area, and discuss other plausible explanations that warrant further research. Finally, we highlight important directions for future research and the implications of our findings for management and conservation.

SUMMARY: Identifying and preserving biological diversity is fundamental for the management of wild populations. The Atlantic bluefin tuna ( Thunnus thynnus ; ABT) is a vital species to the North Atlantic Ocean ecosystem that is now rebounding from decades of overfishing. This teleost fish is notable for its large body size, unique form of endothermy, and trans-oceanic migrations that enable individuals to move rapidly between spawning and foraging locations. Here we combine high-resolution whole genome sequencing data with spatial and environmental data from electronic tagging to improve our understanding of population structure in ABT. We analyzed whole genome sequences (n=82) from both larvae and adult fish representing the two recognized stocks (western and eastern) of ABT, which originate from geographically distinct spawning grounds. Analyzing these data, we identified 11,181,223 single nucleotide polymorphisms (SNPs), a dataset of unprecedented size and resolution. Coverage across the entire genome resulted in increased power in analyses of population structure, patterns of selection, and demographic history between the two recognized stocks. Notably, we report that both neutral and putatively adaptive SNPs are differentiated between populations, and through analyses of F _ST outlier SNPs, we discovered candidate genes with potentially adaptive roles. We suggest that both demographic history and oceanographic variation of the spawning grounds have contributed to shaping bluefin tuna genomic diversity. Our results characterize adaptive variation that will be consequential for management decisions and critical for preserving locally adapted populations.

The eastern Mediterranean Sea is among the warmest and most oligotrophic marine regions globally. Shark aggregations at coastal infrastructure in this basi