Unlocks genetic value of feed intake
By genetically selecting fish that grow well but require less feed, feed conversion ratio may be improved by 10 percent in Norwegian salmon farming.
That means saving 1,5-2 billion NOK per annum, and a decrease of carbon footprint of 7 percent.
Feed is the single largest operational cost in a salmon farming, counting for more than half the cost. It also counts for 85 percent of the carbon footprint. The benefit of reducing these numbers is great, and it can be done by selecting feed efficient fish. However, a bottleneck is that feed intake for individual fish in water, is very difficult to measure in practice, unlike livestock on land.
Can measure individual feed intake
But with the explosion in machine learning and deep learning algorithms, it’s now trivial to segment out objects in an image. For example using x-ray to identify radio opaque markers. Nofima has given a technological facelift to the old x-ray method imaging beads in the feed to measure intake.
“Colleagues have successfully produced extruded feed with radio opaque markers, and they have live-x-rayed thousands of salmon from our partner Mowi’s breeding nucleus, says scientist in breeding and genetics”, Gareth Difford at Nofima.
First breeding value for feed intake
“I’m very proud to say that for the first time we have produced reliable genomic breeding values for feed intake in Atlantic salmon”, says Difford.
The scientists can therefore calculate what would happen if we selected fish that grow well but require less feed. And after three generations, a conservative estimate is that feed conversion ratio can be improved by 10 percent, as mentioned.
But there is a catch. We’ve done this in land based facilities with fresh water parr up to sea water post smolt. For this technology to be ready and relevant, the Nofima scientist will seek opportunities to do experiments in sea.
The research has been carried out in the projects PrecisionVision, NewTechAqua and AquaImpact, and have been financed by the EU program Horizon 2020.
Contact persons
Research facilities