New research on aquaculture industry reveals murky waters surrounding fish-feed issue

The products of aquaculture, the farming of sea creatures and plants, are often divided into “bad fish” — piscavores, like salmon, that eat more pounds of protein in the form of other fish than they yield — and “good fish,” omnivores like tilapia and carp that can survive on plant matter. A new paper from an international team of authors in the Proceedings of the National Academy of Sciences (PNAS, subscription only) and a recent paper in the journal Aquaculture (subscription only) show that the good/bad division isn’t quite so straightforward.

Photo of fish cages from SantiMB's flickr collection, subject to a Creative Commons License

Photo of fish cages from SantiMB's Flickr collection, subject to a Creative Commons License.

Aquaculture currently supplies more than 50% of the fish and shellfish consumed by humans worldwide. Part of the diet of farmed sea creatures (salmon being the most notable example) includes wild fish that have been converted into fish meal or fish oil.  But, as the two articles reveal, shrimp and “good fish” like carp and tilapia also eat fish. Not nearly as much as the piscavores do, but when the whole industry is considered, it adds up to a lot of fish.

With nearly all fisheries either fully exploited or over-exploited, and demand for farmed seafood likely to continue its rapid increase, preventing the aquaculture industry from causing massive disruption of ocean ecosystems in its search for fish food needs to be a major priority.

(Note: The 50% statistic comes from the PNAS article. For the remainder of this post, when a figure is from the PNAS article, I will cite it with [PNAS]. When a figure is from the Aquaculture article, I will cite it with [Aq]. The full citations for the two articles are at the bottom of this post.)

In the first of a two-part series on these two aquaculture articles, I’ll provide some background on the “feeding fish to fish” situation. In the second, I’ll present some of the alternatives to wild fish that are described in the PNAS article.

Aquaculture explodes: Up 6,000% since 1950

In just a few decades, aquaculture has gone from a mostly local activity to a globalized industry with feed, technology, and the final product traveling all over the world. The figure below shows the explosive growth in farmed fish production since 1950: from under a million metric tons in 1950 to almost 60 million metric tons in 2005. (One metric ton is 1,000 kilograms.  Data are from the FIGIS database in the UN’s Food and Agriculture Organization.) Most of the production — and probably most of the consumption — is in the Asia Pacific region: in 2006, that region produced 61.6 million metric tons of aquaculture products; Europe, 1.98  million metric tons; the Americas, 2.12 million metric tons; and Africa, 0.70  million metric tons [Aq].
aquaculture-fish-meal-and-fish-oil-from-faoFeeding the beasts: Almost 20% of captured wild fish go to aquaculture

Supplying more than half of the world’s seafood requires a lot of feed. For most fish — even “good fish” like tilapia and carp — the feed is a combination of plant-based ingredients and fish-based ingredients. For ease of handling and consistency, fish-based ingredients are typically in the form of fish meal, a uniform powder made from rendered fish parts, or fish oil, an oil extracted from fish parts.  In general, the raw ingredients for fish meal and oil are the class of small fish called “reduction fish” or “forage fish,” which includes sardines, herring, members of the anchovy family, and menhaden. (See related Ethicurean guest post on the importance of menhaden, by Alice Friedemann, and my post at Mental Masala about the California sardine fishery). As a rough approximation, a metric ton of raw fish yields about 225 kg of fish meal and 50 kg of fish oil [Aq].

Fulfilling the needs of the fish meal and fish oil factories requires a surprisingly high fraction of the overall catch. Almost one-third of the wild fish caught around the world, according to a 2005 review in Annual Review of Environment and Resources, goes to make the products. For 2001, the authors of that review estimated that of the 95 million metric tons  captured from wild fisheries, 68% of the fish went to direct human consumption, while 32% was converted to fish meal and oil (7% was discarded at sea.)

Pulau Ketam fish farms in Malaysia (iStockphoto)

Pulau Ketam fish farms in Malaysia (iStockphoto)

Most of that fish meal and fish oil goes to aquaculture. In 2006, of the 3.724 million metric tons of fish meal produced, 68.2% went to aquaculture; of 0.835 million metric tons of fish oil produced, aquaculture used 88.5% [Aq]. The remaining fish meal and fish oil goes to a variety of products including fertilizer, livestock feed, and nutritional supplements.

Even “good fish” eat a lot of fish

The top fish meal consumers in 2006 were marine shrimp, salmon, and Chinese carp. In that same year, the top fish oil consumers were salmon, marine fish, and trout [Aq]. The graphs below show the total amount of raw fish needed to produce the fish meal and oil that’s fed to farmed seafood, first the total and then broken out by species (mmt in the y-axis label is million metric tons).*

When we look at the total amount of wild fish turned into food for their farmed brethren, we see a doubling between 1995 and 2005, followed by a moderate decrease that was primarily caused by rising fish meal and oil prices. The overall production of fish meal and fish oil was relatively constant during that period [PNAS], so aquaculture’s increase is the result of its growing share of total use. The farmed salmon industry consumes the most fish, followed by shrimp, which has been increasing its share in recent years. Other big users are trout and “marine fish” such as Japanese amberjack, Seriola quinqueradiata; gilt-head bream, Sparus aurata; cobia, Rachycentron canadum; and numerous other species.

Carp and tilapia together, which are often considered “good” fish because they can be raised on plants alone, consume about 50% less than salmon. Adding fish products to carp and tilapia feed increases growth rates, so it has become part of some feed systems, but the practice is fairly sensitive to price, as the drop off in the most recent years corresponds to a sharp increase in fish meal and oil prices [Aq].


What’s the catch?

Overall, the two articles tell a “good news, bad news” story. The good news is that the aquaculture industry is learning how to produce more farmed seafood while using less wild fish in their feed. A prime motivation for this is higher prices for fish meal and fish oil. For many reasons — increased demand from a growing aquaculture industry, higher energy costs, a fixed supply of reduction fish, to name a few — prices have almost quadrupled since 1994.

A commercial Atlantic Salmon farm in Tasmania, Australia (iStockphoto)

A commercial Atlantic Salmon farm in Tasmania, Australia (iStockphoto)

And yet, high prices for fish meal and fish oil raise other concerns, most notably the specter of overfishing: the incentive is to catch as much as possible while prices are good without regard to the future of the fishery — or the other creatures that depend on the fish, like birds and marine mammals. Furthermore, since the timing of price increases corresponds to a period during which a variety of fisheries are collapsing or have collapsed (such as Atlantic cod, Pacific salmon, bluefin tuna), it is not inconceivable that certain governments will provide subsidies to retool the nation’s ships to catch small fish like sardines, herring, or anchoveta to keep fishing crews working and avoid political problems.

The globalization of the aquaculture food chain is another bit of bad news. With fish caught in one place, processed in plants in another location, and then fed to farmed fish in yet another location, it becomes difficult to trace the fish-to-fish food chain back to the beginning. Additionally, reduction-fish factories can be put on ships in international waters — during the heyday of the California sardine fishery in the 1930s, reduction ships operated three miles off-shore to avoid California law.

Transparency is not the farmed-fish industry’s strong point. Many people believe that tilapia and other non-piscavores are fed a completely fish-free diet; sustainable fishing organizations and the fish companies themselves should provide us with information about the makeup and source of farmed fish feed.  We definitely have the technology to keep track of what a fish was fed — imagine bar codes on each pen of fish, bar codes on each bag of feed, and so on, all feeding into a database that creates a report that travels with the packaged fish.

Then there’s the meat industry’s use of fish products. The 2005 review cited above reported that the livestock industry consumed nearly 20% of all fish caught for feed additives in 2001. But have you ever seen a label on a package of bacon that says “our hog feed has 1% fish meal,” or conversely, “grown without wild fish”? I don’t think so. And so, with the oceans under intense stress from all directions, it is time for a marine advocacy organization to create a certification campaign around fish products in livestock feed.

Fortunately, there are a number of feed alternatives for farmed seafood that do not rely on forage fish. The PNAS article provides a succinct review of some of them, including their pros and cons, which I’ll review in part 2 of this series.

Postscript: Far more factors than fish feed have a negative impact

In this entire post, I haven’t mentioned any other consequences from aquaculture, but they can be as significant as the “fish eating fish” component. Farmed salmon, for example, which gets the “Avoid” rating from Seafood Watch, has a lot of negative impacts on ecosystems, including fish escape (which can spread disease and interfere with the local gene pool), huge amounts of fish waste in small areas, parasites, and more.  The New York Times recently covered Chile’s current problems, and other salmon issues are painstakingly documented in the Monterey Bay Aquarium Seafood Watch report on farmed salmon (PDF). Shrimp also cause plenty of trouble, from destruction of mangrove swamps, which act as “nurseries of the ocean” and protect land from storms, to terrestrial pollution — plenty of nasty chemicals are used to grow shrimp and improve their appearance, some of which can carry through to the final product — to harm to local farmers, as the book “Bottomfeeder,” by Taras Grescoe (see my review for  Ethicurean) compellingly describes. Other species have similar issues. Oysters and other filter-feeding mollusks, however, can actually clean the water, as they feed on algae and nutrients that negatively affect water quality.


“Feeding aquaculture in an era of finite resources,” Rosamond L. Naylor, Ronald W. Hardy, Dominique P. Bureau, and others, Proceedings of the National Academy of Sciences 106 (36), 15103–15110, doi: 10.1073/pnas.0905235106. Summaries available from Stanford University and Environmental Research Web.

“Global overview on the use of fish meal and fish oil in industrially compounded aquafeeds: Trends and future prospects,” Albert G.J. Tacon and Marc Metian, Aquaculture 285 (2008) 146–158, doi:10.1016/j.aquaculture.2008.08.015

*I followed an approach in the Aquaculture article to estimate the mass of fish required to produce the fish oil and fish meal for several types of fish and in total.

4 Responsesto “New research on aquaculture industry reveals murky waters surrounding fish-feed issue”

  1. Thank you for more great information.  This is yet another arm of our industrialized food system.  Salmon eating plants, cows eating corn, pigs eating….(I don’t want to know).  It’s just against nature.  The only way to turn this around is to stop buying food just because it’s cheap.  We’ve all got to ask ourselves and our providers,  “What am I sacrificing by paying such a low price for my food?”  If you love fish, get the best fish you can.  Go to your LOCAL fish store, support the guy that’s trying to make a living in your community, not the fish farmer, packer, distributor, grocery store that gives you corn-fed talapia.  The fish will taste so much better, you won’t believe it. Chef Todd Mohr

  2. Reading this I am struck by an irony. We raise pigs on pasture and hay without commercial feed/corn/soy. We turn sunshine into high quality pork. I’ve been experimenting on a small scale with feeding the slaughter scraps of meat to trout. The fish look great although I have yet to try eating any of them – they’re still out swimming in the upper pond reservoir. So we’re running it backwards, feeding pigs to fish instead of fish to pigs. :)

    I love fish, but don’t eat much for the simple reason that it is expensive and shipped in from a long distance. I’ve got other local, grown right on our farm, sources of meat. But still, there’s this idea in the back of my mind that maybe I could grow shrimp and fish…

  3. Thanks for blogging about this. I’m looking forward to reading part 2!

  4. One thought to add is that while some fish farms are working to reduce the percentage of fishmeal and fishoil in fish diets, the overall growth of the aquaculture industry results in those reductions being swamped by an overall increase in demand for fishmeal/fishoil based feeds.

    This topic highlights the some of the complexities in the simple act of choosing seafood. The Monterey Bay Aquarium Seafood Watch program takes these environmental impacts into account when creating our seafood recommendations. I encourage people to carry a Seafood Watch pocket guide as a great reference for which seafoods to avoid due to their high level of impact and which are best choices thanks to the work of producers and fish farmers to reduce their environmental impacts. It’s easy to download the version for your region from our website: