The use of Tech in Aquaculture

Commercial aquaculture has been practised for a few decades and during that time there have been considerable advancements in the technology used to produce fish and optimise aquaculture business success.  As we progress through the 21st century, AI has become an important component in global technology, begging the question ‘to what degree can AI technology be used in aquaculture?’  In this brief I will present some of the technologies currently in use in aquaculture to provide a broad overview of what is available in the industry.  It should be stressed that each solution has a particular area where it is best suited, and areas where it is a poor fit; do not feel the need to fit any or all of these to your situation if the benefit does not appear to be convincing.
 
Technology falls into various categories and provides solutions from company management through accounting to human resources, maintenance scheduling and cold chain management, and includes an impressive array of biometric and software solutions.  Whilst important, these are not the focus of the present investigation which will concentrate primarily on technologies being used to optimise the farming side of an aquaculture business.
 
Feeding Systems
A range of fish feeding systems have been in use for some time, whereby the feed is distributed to the tanks by means of pneumatic air tracks (pipes), water flow and other methods.  These can be pre-set to deliver a specific pellet size and quantity to each tank or cage on a farm at selected times throughout the day and night and can operate to distances of up to 1.4km!
 
The software that runs these feeding systems is operator programable so that adjustments can be made as the biomass of fish increases.  A web-based app can be downloaded onto the managers’ smart phone to allow them to monitor the system remotely, to be sure that the feeding is happening as required.
 
Automatic feeding systems offer the advantages of reduced labour requirements along with more frequent, smaller meals.  The disadvantage is that the systems feed according to settings not fish hunger, and under or over feeding is likely if not closely managed.
 
Stock Management
The system utilised on most fish farms to monitor stock numbers and growth rates is to periodically harvest the container of fish, size sort them into batches, weigh the batches and a sample per batch, and then return the fish to tanks.  From this data the average and total mass are determined, and, when compared to the data from the previous sorting event, the growth rate of the fish, as well as survival and feed usage efficiency, can be established.
 
This process involves starving the fish prior to handling, and then subjecting them to a lengthy process involving much stress.  Typically, the fish are not fed on the day of handling, thereby loosing a day or two of feeding, and the associated growth.  If the handling is rough this can lead to external damage, infections and mortality during the days following handling.  Because the information gathered during size sorting and counting is so important, fish farmers execute this work knowing that it can result in losses.
 
Monitoring Fish Behaviour – tying it all together
Several companies have developed AI based solutions including AKVA (AKVA Observe), Alphabet (Tidal Moonshot), InnovaSea (BiomassPro) and AquaByte (AquaByte Solutions).  The essence of each system is a camera that monitors the fish underwater and in real time, feeding the image stream to smart technology.  From the behaviour of the fish the AI ascertains hunger, health and growth, and can be set to control activities such as feeding and treatments.  This means that fish are no longer subjected to the stress of being harvested, sorted, weighed and the consequent reduction in stress means the fish grow faster, overfeeding is avoided and the fish utilise their feed more efficiently, and they potentially remain healthier.  More advanced systems are now able to provide an estimate of individual fish mass, from which growth rate and FCR are determined.

Genetic Improvement
The final area of AI utilisation in fish farming is in the realm of genetic improvement.  Software is utilised to monitor the breeders and predict the expected outcome of specific spawning crosses.  The results are then fed back into the model to improve future predicting ability, and thereby aid the farmer in achieving the most desirable traits from spawning events.  Providers of these systems claim an improvement of at least 10% in the targetted parameter in each subsequent generation!
 
In summary, AI is taking the aquaculture industry forward at a rapid rate; offering solutions that reduce stress, improve growth and optimise feed usage efficiency. Embracing this exciting progress will keep your fish farm on the aquaculture frontier.