Precision farming in aquaculture: non-invasive monitoring of Atlantic salmon (Salmo salar) behaviour in response to environmental conditions in commercial sea cages for health and welfare assessment

Journal article

Frontiers in Robotics and AI

Sonia Rey Planellas

Research aim:

In the present study, underwater cameras were deployed, and the real-time videos were analysed using a machine-learning algorithm that converts the videos into a numerical format as a proxy for fish abundance and shoal cohesion. The aim of this study was to understand fish distribution patterns related to the environmental conditions within the sea cages. The hypothesis was that fish would follow vertical thermal gradients and change horizontal distribution based on hydrodynamics. Gaining a more profound insight into the distribution patterns of fish under typical conditions provides farmers with a valuable baseline to then discern whether deviations in these behaviours can be attributed to stressors of environmental (e.g., storm events or hypoxic conditions) or health origins (e.g., sea lice or gill health issues; Sadoul et al., 2014).

Background:

In sea cage aquaculture systems, the challenges of poor visibility (e.g., from water turbidity, fish density, variable lighting conditions), and restricted access, due to the remote location of many farms, further emphasise the need for arrays of sophisticated technologies to monitor fish effectively under farming conditions (Føre et al., 2018). There are many technologies that can be employed for farmers to monitor their fish such as biosensors, fish telemetry, hydroacoustic sensors (echosounders) and cameras, with the latter two being the most prominent in commercial sea cages (Føre et al., 2018; O’Donncha et al., 2021; Georgopoulou et al., 2021).

Approach:

This study used a computer vision algorithm at three marine farms to analyse fish group swimming behaviour termed “activity” (measured in percent), which includes fish abundance, speed, and shoal cohesion. The activity metric inferred the depth distribution of the main fish group and was analysed with respect to environmental conditions to explore potential behavioural drivers and used to assess changes in fish behaviour in response to a stressor, a storm event. During winter conditions,

Key finding:

Studies show that Atlantic salmon in captivity adjust their distribution in sea cages based on environmental gradients like temperature, waves, and photoperiod. This study used a computer vision algorithm at three marine farms to analyse fish group swimming behaviour termed “activity” (measured in percent), which includes fish abundance, speed, and shoal cohesion. The activity metric inferred the depth distribution of the main fish group and was analysed with respect to environmental conditions to explore potential behavioural drivers and used to assess changes in fish behaviour in response to a stressor, a storm event.

Industry or policy relevance:

Understanding fish behavioural responses to environmental conditions can inform management practices, while using cameras with associated algorithms offers a powerful, non-invasive tool for continuously monitoring and safeguarding fish health and welfare.

Route for practical application:

To enhance fish welfare, future research could consider designing cages that can be moved both vertically and horizontally to align with fish preferred temperatures and oceanographic conditions, while also considering management strategies that account for the behavioural differences between reactive and proactive fish.

Confidence in findings and next steps towards realising impact:

Next step is to scale up on more farms and cages whithin the farms to validate this first findings.