Intelligent Aquaculture
Traditionally, fishers went out to sea with fishing nets or long lines on boats to catch fish.
They took advantage of their experience gained through trial and error, through observation, and through knowledge passed down from previous generations.
With wild fish stocks gradually declining due to overfishing and the changing climate, more fishers are turning to farming instead of casting nets or lines.
In fact, according to a UN report, about 49% of the fish consumed worldwide in 2020 came from fish farms, not from the wild.
This figure is expected to continue to grow, increasing to 62% by 2030.
With the arrival of smart fish farming technology driven by artificial intelligence (AI), fish farmers are able to provide healthy, affordable seafood to families everywhere.
An essential element in providing healthy fish to consumers is the fish farmers’ ability to detect and prevent sea lice, a significant threat to fish health and profitability.
One of the most modern methods to do this is through the power of computer vision and precision sensors, allowing fish farmers to identify and track diseases before they spread.
An aquaculture system equipped with high-precision underwater tools has already been developed in Norway.
It can hold up to 150,000 salmon in a sea cage, regularly inducing each fish into a tube where high-resolution cameras and sensors record everything about each individual salmon, including their overall shapes and dot patterns, their weight and growth rate, as well as any signs of sea lice, physical wounds, or abnormalities.
Every detail is registered in the fish’s health record, along with millions of images and figures.
Such huge amounts of data and other information are taken care of by the AI system.
A computer system with deep learning capabilities is able to track and record fish behavior better than humans can, especially when the system is trained with enough relevant data.
It can distinguish and identify one fish from another and diagnose whether a small speck is dirt or sea lice, and this information is entered into the fish’s health record.
In a sense, the fish farm is monitored by an AI doctor on duty 24/7.
Farmers can then determine when they should take action, such as when to separate an abnormal group from the rest or spread out the fish population to reduce density, without resorting to harsh chemical treatments, which can harm the fish, the environment, and even the consumer.
In addition, all this data from cameras and sensors can be utilized for various applications and purposes.
One of the most noticeable and effective applications is real-time monitoring of fish behavior and other environmental changes, which serves multiple purposes, including diagnosing diseases in individual fish.
By processing large sets of data, an AI system can detect and recognize unusual group behaviors of an entire school of fish, possibly caused by changes in water quality, temperature, or oxygen levels surrounding the cage.
It can also ensure the quality of cage nets to prevent fish from escaping or predators from invading and protect the surrounding habitat.
The system can also tell farmers exactly when to feed the fish by identifying patterns of behavior that an individual fish or the school exhibits when they are hungry.
This is how farmers can accurately determine optimal feeding times and amounts.
The right amount of food at the right time minimizes cost, reduces waste and pollution, and maximizes fish growth rates.
Given the fact that feeding constitutes more than 50% of a fish farm’s operational budget, this precise feeding approach will ultimately contribute to a more affordable supply of fish for consumers as well as efficient and sustainable aquaculture operations in general.
AI continues to advance in aquaculture to extend beyond caged fish.
In particular, smart fish farming technologies can be applied to seaweed farming.
Seaweed is as valuable a source of food as fish.
It’s also a source of fuel and an important ingredient for cosmetics, fertilizers, and other medicinal and industrial products.
As such, ongoing research is exploring various ways AI can help in the seaweed industry, including the use of autonomous underwater vehicles (AUVs) and swarm robotics.
Using cameras and sensors, AUVs can do many of the analytical and diagnostic tasks on seaweed farms.
Swarm robots, or many small robots working together, can swim in and around seaweed farms to collect data and then make decisions on how to deal with problems that arise.
Single larger robots or even humans may be available to take care of the physical labor from removing harmful weeds to harvesting;
however, swarm robots may be more efficient and cost-effective.
The potential of AI in aquaculture is vast, and it brings a more sustainable and technologically driven future for fish and seaweed farming.
Families that gather together, eagerly anticipating a healthy seafood dinner, may not be aware of the carefully monitored fish and seaweed farms or the cutting-edge AI technology involved in getting food to their table.
However, the process of doing so is evidence of the considerable amount of investment in sustainable, high-quality aquaculture with AI.
With AI as an ally, the future of the oceans and the nourishment of humanity is just a little bit more secure.