Novel use of deep neural networks on photographic identification of epaulette sharks (Hemiscyllium ocellatum) across life stages.

Photographic identification (photo ID) is an established method that is used to count animals and track individuals' movements. This method performs well with some species of elasmobranchs (i.e., sharks, skates, and rays) where individuals have distinctive skin patterns. However, the unique skin patterns used for ID must be stable through time to allow re-identification of individuals in future sampling events. More recently, artificial intelligence (AI) models have substantially decreased the labor-intensive process of matching photos in extensive photo ID libraries and increased the reliability of photo ID. Here, photo ID and AI are used for the first time to identify epaulette sharks (Hemiscyllium ocellatum) at different life stages for approximately 2 years. An AI model was developed to assess and compare the reliability of human-classified ID patterns in juvenile and neonate sharks. The model also tested the persistence of unique patterns in adult sharks. Results indicate that immature life stages are unreliable for pattern identification, using both human and AI approaches, due to the plasticity of these subadult growth forms. Mature sharks maintain their patterns through time and can be identified by AI models with approximately 86% accuracy. The approach outlined in this study has the potential of validating the stability of ID patterns through time; however, testing on wild populations and long-term datasets is needed. This study's novel deep neural network development strategy offers a streamlined and accessible framework for generating a reliable model from a small data set, without requiring high-performance computing. Since many photo ID studies commence with limited datasets and resources, this AI model presents practical solutions to such constraints. Overall, this approach has the potential to address challenges associated with long-term photo ID data sets and the application of AI for shark identification.

From little things big things grow: enhancement of an acoustic telemetry network to monitor broad-scale movements of marine species along Australia's east coast.

BACKGROUND: Acoustic telemetry has become a fundamental tool to monitor the movement of aquatic species. Advances in technology, in particular the development of batteries with lives of > 10 years, have increased our ability to track the long-term movement patterns of many species. However, logistics and financial constraints often dictate the locations and deployment duration of acoustic receivers. Consequently, there is often a compromise between optimal array design and affordability. Such constraints can hinder the ability to track marine animals over large spatial and temporal scales. Continental-scale receiver networks have increased the ability to study large-scale movements, but significant gaps in coverage often remain. METHODS: Since 2007, the Integrated Marine Observing System's Animal Tracking Facility (IMOS ATF) has maintained permanent receiver installations on the eastern Australian seaboard. In this study, we present the recent enhancement of the IMOS ATF acoustic tracking infrastructure in Queensland to collect data on large-scale movements of marine species in the northeast extent of the national array. Securing a relatively small initial investment for expanding receiver deployment and tagging activities in Queensland served as a catalyst, bringing together a diverse group of stakeholders (research institutes, universities, government departments, port corporations, industries, Indigenous ranger groups and tourism operators) to create an extensive collaborative network that could sustain the extended receiver coverage into the future. To fill gaps between existing installations and maximise the monitoring footprint, the new initiative has an atypical design, deploying many single receivers spread across 2,100 km of Queensland waters. RESULTS: The approach revealed previously unknown broad-scale movements for some species and highlights that clusters of receivers are not always required to enhance data collection. However, array designs using predominantly single receiver deployments are more vulnerable to data gaps when receivers are lost or fail, and therefore "redundancy" is a critical consideration when designing this type of array. CONCLUSION: Initial results suggest that our array enhancement, if sustained over many years, will uncover a range of previously unknown movements that will assist in addressing ecological, fisheries, and conservation questions for multiple species.

Directed conservation of the world's reef sharks and rays.

Many shark populations are in decline around the world, with severe ecological and economic consequences. Fisheries management and marine protected areas (MPAs) have both been heralded as solutions. However, the effectiveness of MPAs alone is questionable, particularly for globally threatened sharks and rays ('elasmobranchs'), with little known about how fisheries management and MPAs interact to conserve these species. Here we use a dedicated global survey of coral reef elasmobranchs to assess 66 fully protected areas embedded within a range of fisheries management regimes across 36 countries. We show that conservation benefits were primarily for reef-associated sharks, which were twice as abundant in fully protected areas compared with areas open to fishing. Conservation benefits were greatest in large protected areas that incorporate distinct reefs. However, the same benefits were not evident for rays or wide-ranging sharks that are both economically and ecologically important while also threatened with extinction. We show that conservation benefits from fully protected areas are close to doubled when embedded within areas of effective fisheries management, highlighting the importance of a mixed management approach of both effective fisheries management and well-designed fully protected areas to conserve tropical elasmobranch assemblages globally.

Fatty acid profiles of more than 470 marine species from the Southern Hemisphere.

Lipid and fatty acid datasets are commonly used to assess the nutritional composition of organisms, trophic ecology, and ecosystem dynamics. Lipids and their fatty acid constituents are essential nutrients to all forms of life because they contribute to biological processes such as energy flow and metabolism. Assessment of total lipids in tissues of organisms provides information on energy allocation and life-history strategies and can be an indicator of nutritional condition. The analysis of an organism's fatty acids is a widely used technique for assessing nutrient and energy transfer, and dietary interactions in food webs. Although there have been many published regional studies that assessed lipid and fatty acid compositions, many only report the mean values of the most abundant fatty acids. There are limited individual records available for wider use in intercomparison or macro-scale studies. This dataset consists of 4856 records of individual and pooled samples of at least 470 different marine consumer species sampled from tropical, temperate, and polar regions around Australia and in the Southern, Indian, and Pacific Oceans from 1989 to 2018. This includes data for a diverse range of taxa (zooplankton, fish, cephalopods, chondrichthyans, and marine mammals), size ranges (0.02 cm to ~13 m), and that cover a broad range of trophic positions (2.0-4.6). When known, we provide a record of species name, date of sampling, sampling location, body size, relative (%) measurements of tissue-specific total lipid content and abundant fatty acids, and absolute content (mg 100 g-1 tissue) of eicosapentaenoic acid (EPA, 20:5n3) and docosahexaenoic acid (DHA, 22:6n3) as important long-chain (≥C20 ) polyunsaturated omega-3 fatty acids. These records form a solid basis for comparative studies that will facilitate a broad understanding of the spatial and temporal distribution of marine lipids globally. The dataset also provides reference data for future dietary assessments of marine predators and model assessments of potential impacts of climate change on the availability of marine lipids and fatty acids. There are 480 data records within our data file for which the providers have requested that permission for reuse be granted, with the likely condition that they are included as a coauthor on the reporting of the dataset. Records with this condition are indicated by a "yes" under "Conditions_of_data_use" in Data S1: Marineconsumer_FAdata.csv (see Table 2 in Metadata S1 for more details). For all other data records marked as "No" under "Conditions_of_data_use," there are no copyright restrictions for research and/or teaching purposes. We request that users acknowledge use of the data in publications, research proposals, websites, and other outlets via formal citation of this work and original data sources as applicable.