The precision aquaculture initiative at the Freshwater Institute is developing advanced technologies to increase fish production, improve product quality and optimize efficiency, all while ensuring fish welfare in recirculating aquaculture systems (RAS).

With growing interest in land-based RAS technology, both domestically and globally, research is needed to optimize operations. Current management practices in RAS rely on manual observation and fish handling. Manual approaches, however, can be subjective and labor-intensive, adversely affecting fish health, welfare and yield. Adopting advanced sensing technologies and data analytics can assist farm operators with knowledge-based decision-support tools to monitor, control and document biological processes. Such approaches also reduce subjectivity and dependency on manual labor through automation while improving sustainability, fish health and welfare, and productivity.

The Conservation Fund’s Freshwater Institute precision aquaculture initiative is part of efforts by the U.S. Department of Agriculture’s Agricultural Research Service to use advanced technologies for precision farming. The institute’s initiative uses sensors, computer vision and artificial intelligence (AI) to identify and mitigate bottlenecks in current land-based RAS management practices. The precision aquaculture team is developing solutions that can be integrated within RAS for labor- and cost-effective management.

Our Solution

Fish mortality is a vital welfare indicator — and avoiding excess mortalities is critical to maintaining normal RAS operations. Using AI and Internet of Things technologies, the team developed MortCam to provide round-the-clock mortality monitoring and alerting. The tool can automatically detect and log daily and cumulative mortality counts in RAS and alert farm managers to unusual mortality. Early recognition of unusual patterns of fish mortality may help prevent catastrophic losses. Moreover, such smart devices enable farm managers to create a mortality benchmark for the various stages of fish production. While the MortCam prototype is trained to monitor Atlantic salmon mortality, the AI model can be easily customized for other species — such as kingfish, rainbow trout and sea bass — grown in RAS.

Why This Project Matters

Mass mortality events in large-scale RAS incur substantial economic losses. MortCam could help farm operators make informed decisions on mortality management. Furthermore, the timely removal of fish remains helps maintain desired water quality and minimizes potential pathogen and disease outbreaks to ensure surviving fish health. In the future, real-time mortality monitoring data may be integrated with systems to automate the removal of fish remains.