The power of national acoustic tracking networks to assess the impacts of human activity on marine organisms during the COVID-19 pandemic.

COVID-19 restrictions have led to an unprecedented global hiatus in anthropogenic activities, providing a unique opportunity to assess human impact on biological systems. Here, we describe how a national network of acoustic tracking receivers can be leveraged to assess the effects of human activity on animal movement and space use during such global disruptions. We outline variation in restrictions on human activity across Australian states and describe four mechanisms affecting human interactions with the marine environment: 1) reduction in economy and trade changing shipping traffic; 2) changes in export markets affecting commercial fisheries; 3) alterations in recreational activities; and 4) decline in tourism. We develop a roadmap for the analysis of acoustic tracking data across various scales using Australia's national Integrated Marine Observing System (IMOS) Animal Tracking Facility as a case study. We illustrate the benefit of sustained observing systems and monitoring programs by assessing how a 51-day break in white shark (Carcharodon carcharias) cage-diving tourism due to COVID-19 restrictions affected the behaviour and space use of two resident species. This cessation of tourism activities represents the longest break since cage-diving vessels started day trips in this area in 2007. Long-term monitoring of the local environment reveals that the activity space of yellowtail kingfish (Seriola lalandi) was reduced when cage-diving boats were absent compared to periods following standard tourism operations. However, white shark residency and movements were not affected. Our roadmap is globally applicable and will assist researchers in designing studies to assess how anthropogenic activities can impact animal movement and distributions during regional, short-term through to major, unexpected disruptions like the COVID-19 pandemic.

Using tri-axial accelerometer loggers to identify spawning behaviours of large pelagic fish.

BACKGROUND: Tri-axial accelerometers have been used to remotely describe and identify in situ behaviours of a range of animals without requiring direct observations. Datasets collected from these accelerometers (i.e. acceleration, body position) are often large, requiring development of semi-automated analyses to classify behaviours. Marine fishes exhibit many "burst" behaviours with high amplitude accelerations that are difficult to interpret and differentiate. This has constrained the development of accurate automated techniques to identify different "burst" behaviours occurring naturally, where direct observations are not possible. METHODS: We trained a random forest machine learning algorithm based on 624 h of accelerometer data from six captive yellowtail kingfish during spawning periods. We identified five distinct behaviours (swim, feed, chafe, escape, and courtship), which were used to train the model based on 58 predictive variables. RESULTS: Overall accuracy of the model was 94%. Classification of each behavioural class was variable; F1 scores ranged from 0.48 (chafe) - 0.99 (swim). The model was subsequently applied to accelerometer data from eight free-ranging kingfish, and all behaviour classes described from captive fish were predicted by the model to occur, including 19 events of courtship behaviours ranging from 3 s to 108 min in duration. CONCLUSION: Our findings provide a novel approach of applying a supervised machine learning model on free-ranging animals, which has previously been predominantly constrained to direct observations of behaviours and not predicted from an unseen dataset. Additionally, our findings identify typically ambiguous spawning and courtship behaviours of a large pelagic fish as they naturally occur.