The iFarm project, featuring AI and machine learning, expands to a full sea site
The knowledge gained from the initial phase was utilized to make modifications in the second phase. The iFarm equipment was initially installed in individual net pens at the Martnesvika site in Steigen. The primary objective at Martnesvika was to observe the behavior of the fish and familiarize with the new equipment in the pens. The insights gathered thus far have been instrumental in tweaking the equipment and setup for the project's second phase. Consequently, the iFarm equipment deployed in the second phase at Langøyhovden differs slightly from the equipment used in the first phase at Martnesvika. Karl Fredrik Ottem, the iFarm project manager, noted that the design of the sensor housing and the openings through which the fish swim to reach the surface influence their swimming pattern. The fish need to choose to swim through the sensor house, so in this phase, six different sensor houses with varying geometric designs are being tested to determine which ones the salmon prefer. This will also inform the choice of sensor houses for the third phase when the fish will be stocked for the third time in the project. The six houses that have been deployed were assembled on land at Myre in Vesterålen before being transported to the sea site and installed in the net pens at a depth of eight meters. Ottem stated that they have numerous high-quality cameras to continuously monitor the fish's swimming patterns and ensure their well-being. So far, the fish seem to be adapting well to the new equipment in the net pen at both Martnesvika and Langøyhovden. Despite the fish appearing to thrive and grow, not everything has been smooth sailing in this innovative development project. With a significant amount of equipment in the net pen, finding effective solutions for camera cleaning and equipment maintenance has been challenging. Ottem noted that these large and complex installations either need to be lifted for maintenance or require underwater access, which has made it difficult to establish efficient routines while ensuring the fish's welfare. However, these experiences will be valuable in designing and adjusting the equipment and setup for the subsequent phase. The project has shown promising results in terms of sea lice. As the fish are kept ten meters down in the net pen with iFarm, fewer lice have been observed on the salmon. Ottem estimates that at least one de-licing operation was saved during the first phase at Martnesvika, even though iFarm was only set up in single net pens. With iFarm set up in all the net pens at Langøyhovden, it will be intriguing to see its impact on lice levels. About iFarm: iFarm is a concept that leverages artificial intelligence and machine learning to transition from population-based to individual-based care in aquaculture, which is crucial for fish health and welfare. It is a joint venture between Cermaq and BioSort, with ScaleAQ being the primary supplier of farming equipment for the project. Cermaq has been granted four development licenses with iFarm.