Data Driven Insight Into Fish Behaviour and Their Use for Precision Aquaculture

Abstract

Aquaculture, or the farmed production of fish and shellfish, has grown rapidly, from supplying just 7% of fish for human consumption in 1974 to more than half in 2016. This rapid expansion has led to the growth of Precision Aquaculture concept that aims to exploit data-driven management of fish production, thereby improving the farmer's ability to monitor, control, and document biological processes in farms. Fundamental to this paradigm is monitoring of environmental and animal processes within a cage, and processing those data toward farm insight using models and analytics. This paper presents an analysis of environmental and fish behaviour datasets collected at three salmon farms in Norway, Scotland, and Canada. Information on fish behaviour were collected using hydroacoustic sensors that sampled the vertical distribution of fish in a cage at high spatial and temporal resolution, while a network of environmental sensors characterised local site conditions. We present an analysis of the hydroacoustic datasets using AutoML (or automatic machine learning) tools that enables developers with limited data science expertise to train high-quality models specific to the data at hand. We demonstrate how AutoML pipelines can be readily applied to aquaculture datasets to interrogate the data and quantify the primary features that explains data variance. Results demonstrate that variables such as temperature, wind conditions, and hour-of-day were important drivers of fish motion at all sites. Further, there were distinct differences in factors that influenced in-cage variations driven by local variables such as water depth and ambient environmental conditions (particularly dissolved oxygen). The framework offers a transferable approach to interrogate fish behaviour within farm systems, and quantify differences between sites.

Welfare in Farmed Decapod Crustaceans, With Particular Reference to Penaeus vannamei

“…They can be automatic systems for behavioral observations or water quality sensors (Barreto et al, 2022). With the use of Artificial Intelligence (AI) and machine learning methods the monitoring of environmental parameters and behaviour can be done automatically, and the decision-making process becomes more accurate (O'Donncha et al, 2021) and autonomous. These methods of precision farming (PF) are being developed for finfish and molluscs with proven success.…”

A Review of Unmanned System Technologies with Its Application to Aquaculture Farm Monitoring and Management

“…With the vast and varied amount of data collection from the aquaculture site, data-driven management of fish production is now possible. This scheme improves the ability of farmers to monitor, control, and document biological processes on their farms so that they can understand the environmental conditions that affect the welfare and growth of fish [93].…”

Offshore aquaculture of finfish: Big expectations at sea

“…257 Acoustic technology and underwater cameras can be used to monitor feeding behaviour and ensure that feed is delivered at optimal times. 258,259 Real-time sensors can also be used to monitor water quality and can provide alerts, for example if oxygen levels are too low and may affect fish health and welfare. 109 The cage infrastructure can even be monitored using sensors and models to provide advance warning of cage deformation.…”

“…Remote monitoring and Precision Fish Farming (PFF) can be used to control and automate important tasks from a land base 257 . Acoustic technology and underwater cameras can be used to monitor feeding behaviour and ensure that feed is delivered at optimal times 258,259 . Real‐time sensors can also be used to monitor water quality and can provide alerts, for example if oxygen levels are too low and may affect fish health and welfare 109 .…”