The claim that fish farming is efficient because it produces 1 kilogram of farmed salmon for 1.25 kilograms of fish (in the food they are fed) is factually correct but leaves so many other facts unstated that it is greenwash.
The fish farming industry gets this impressive ratio by:
1. Substituting vegetables and other food for some of the fish and then saying “Look we can get 1 kg of fish for 1 kg of fish” and not mentioning the weight of food from other sources contained in the feed that the laws of physics oblige them to feed the fish.
2. Not mentioning that the food has a much lower water content than the farmed salmon. Salmon contain around 78% water and the feed around 6.6% water using these numbers I get a feed conversion ratio of 5.3:1
The feed is carefully manufactured to be energy-dense which is sensible because less feed needs to be transported. Using the statistics published by the industry and Scottish government I find that:
In 2015 the feed used to feed the salmon contains 6.7 times the food energy contained in the harvested salmon.
In 2016 the feed used to feed the salmon contains 7.4 times the food energy contained in the harvested salmon
It follows we could feed 6.7 – 7.4 times the number of people at lower energy and carbon cost if we just eat the salmon food. (Not as salmon food of course but processed into something humans would find acceptable or perhaps used in other foods.)
It makes sense for Beef farmers, say, to compare their efficiencies by comparing the amount of feed needed to produce a cow, they are running their farms in more or less the same way using more or less the same food. At first sight, it also makes sense to compare different kinds of animal protein production by comparing their respective feed conversion ratios. So comparing intensive beef and pork production in this way is reasonable. However, as the systems become less alike it makes less and less sense to do this.
If you really want to compare farming systems you must calculate how much energy is used to construct and run the farm and to manufacture, or harvest, or fish for the food that is fed to the animals and to compare that with the energy that is contained in the fish or meat that humans eat.
In other words, we need to know the total energy cost of running a beef farm or a fish farm. How many litres of fuel are needed to make the steel for the moorings, boats, buildings and to run all the machinery. How much fuel does it take to catch or farm and transport the fish feed? And how much steel is in the farm machinery and buildings of the cattle farm. How much energy does it take to run all that and to grow the cattle food?
Tyedmers [1, 2] has calculated these energy ratios for several types of animal. He found that intensive beef farming (with the cattle in sheds fed on corn) and sea cage fish farming are the least efficient ways to produce food. (You get a range of numbers depending on the kind of farm you analyse. A small farm with cattle grazing in salt marshes will be much more efficient than industrial beef production with the cattle in sheds. I have tabulated his results below.
The total (mainly fossil fuel) energy cost of producing animal protein
Food production system Units of energy used per unit of edible food energy produced Third world farmers 0.01 shown for comparison with animal protein production First world vegetable production 1-3 shown for comparison with animal protein production Commercially caught coho salmon (B.C.) 13.5 Milk (USA) 14.1 Pork (USA) 17.9 Commercial cod fishery (USA) 20.0 Chicken (USA) 26.3 Eggs (USA) 26.3 Lamb (USA) 50.0 Intensively cultured Atlantic salmon 50.0 Beef (USA) 52.6
From Tyedmers,  table 12 p 33
For comparison first world vegetable farming rarely consumes more than 3 times the amount of energy contained in the food and sometimes breaks even. And third-world farmers, who feed a large proportion of the human population, get 100 times the amount of edible food energy for every calorie of human effort. In other words, they are 5,000 times more efficient at feeding people than intensive beef and salmon farmers are. And do so without using much / any fossil fuel.
And since salmon production requires fish or vegetable protein the maximum sustainable production is set by:
1. The maximum catch of wild fish that can be caught without collapsing the fishery.
2. The maximum land area and energy we are willing / able to use for growing salmon feed.
Thus there is a larger question: What sort of food production systems can we afford to run and what can we eat if we want to leave our grandchildren something to eat?
I would be more sympathetic to the fish farming industry if we were told a more honest story which goes something like: Salmon farming is comparable in its energy use to other intensive farming practices but it does create needed work.
1. Tyedmers, P. (D. Zeller, R. Watson, and D.and Pauly, Eds.) (2001) Energy consumed by North Atlantic Fisheries. In “Fisheries Impacts on North Atlantic Ecosystems: Catch, Effort and National/Regional Datasets”. Fisheries Centre Research Reports 2001 Vol 9 No. 3, pp. 12-34.
2. Peter Horst Tyedmers. (2000) Salmon and Sustainability: The Biophysical Cost of Producing Salmon Through the Commercial Salmon Fishery and the Intensive Salmon Culture Industry. Ph.D. Thesis, Faculty of Resource Management and Environmental Studies at The University of British Columbia., Submitted in 2000. (Available online if you search with this title.)