Is sustainable aquaculture an oxymoron?



A number of comments to my previous post on the role of aquaculture as a milestone in the history of humanity have clearly identified one of the key problems for the growth of aquaculture: that some of marine aquaculture today can hardly be considered sustainable, particularly where the production targets predatory fish high-up in the food web.

Indeed, in an assessment of the bottlenecks to progress towards a sustainable aquaculture, my coworkers and I identified the high trophic level at which aquaculture is exploited as a major driver of the environmental impact of aquaculture (Duarte et al. 2009).

A comparison between the way we exploit marine and terrestrial food webs can illustrate the problem. The efficiency in the transference of organic matter up the food webs is typically below 10% (i.e. an organism grows in weight by less than 100 g for each kg of food ingested) (note 1). This means that production is dissipated as it moves up in the food web, so for every ton of plant production introduced in the food web, we can harvest up to 100 Kg of herbivores, 10 Kg of carnivores feeding on herbivores, and so on. On land, we eat largely plants and herbivores, with a few omnivores and very few carnivores (e.g. dogs in some Asian countries). Accordingly, the mean food production on land has a mean weighted trophic level of 1.008, where 1 is a plant, 2 is a herbivore, etc.

In contrast, we eat many large predatory fish, such as tuna or sharks, that sit high up in the marine food web, which has many more steps than the terrestrial food web does. For instance, tuna has a trophic level of about 5 (i.e. four other steps in between plankton production and tuna), which is unparalleled in terrestrial food webs, equivalent to imaginary monsters eating wolf-eaters (Duarte et al. 2009).

This means that the production of 1 Kg of tuna (trophic level 5) requires about 100,000 kg of plankton production, which is equivalent to the annual primary production of 5 hectares of ocean surface. If we, however, consume 1 kg of small pelagic fish, such as anchovies (trophic level 3), we are effectively harvesting the annual production of an ocean surface 100 fold smaller.

Hence, the production of predatory fish high up in the food web requires the appropriation of massive amounts of ocean production, typically as fish converted in flour and oil for fish feed production.

Yet, aquaculture, as practiced today produces food at a lower trophic level (1.84) compared to that of fisheries (trophic level 3.2, Duarte et al. 2009), indicating that, for a given unit production, aquaculture co-opts 50 times less ocean production than fisheries.


Understanding the efficiency of food webs, and introducing food web concepts in planning aquaculture production and human diets is important. For instance, a food web can use any given primary production far more effectively than a single species can. Accordingly, polycultures, where different species are cultured jointly composing a small food web rather than in isolation, can increase the yield of aquaculture by 30 %, for a give use of feed, while reducing environmental impacts (Duarte et al. 2009). For instance, polycultures combine fish cultures with a belt of filter-feeders, such as mussels or oysters, that filter out the excess particles in the water, and an outer belt of algae that strip the nutrients (nitrogen and phosphorus) released by excretory and decomposition processes out of the water. Cages with bottom detritivores can also remove excess feed and feces reaching the sea floor and turn them into valuable production.

However, the ultimate solution to the sustainability of aquaculture relies in the mass production of macroalgae and filter-feeding and herbivore organisms, bringing the trophic level of production down to levels comparable to those of food production on land. This shift will also allow aquaculture to close their production cycle, producing, in the farm, the fish feed required, thereby releasing aquaculture from its present dependence of wild fisheries catches.

In doing so, aquaculture can shift from being a source of (comparatively minor) problems to being a positive force in the marine environment. Large scale production of macroalgae, such as those already existing in China and Korea, can help rehabilitate degraded coastal waters by stripping excess nutrients from the water, injecting photosynthetic oxygen into hypoxic waters, and providing habitat to increase biodiversity. Much of that production can be used to produce biofuels, free of the problems (competition with crops for water and fertile land) that affect biofuel production on land, thereby helping mitigate climate change (Duarte et al. 2009).

The capacity to control the life cycles of marine organisms can also be instrumental as a tool in conservation biology, where populations of endangered marine organisms can be subsidized by the release – with proper consideration to avoid genetic dilution – of organisms grown in culture, catalyzing the recovery of endangered wild populations. This use would be comparable to successful breeding programs on endangered terrestrial birds and mammals, such as the Tasmanian devil. Indeed, the Pacific Salmon fishery off Alaska is already subsidized by the release of fry from aquaculture.

As evidenced by some comments to my earlier post, the perception that aquaculture is detrimental to the environment is widespread. Whereas impacts do exist in many operations, these do not fully justify the negative perceptions and biases the public often has against aquaculture.

For instance, experiments in Scotland have shown that when wild and aquaculture salmon are offered to the public, with each labelled both wild and aquaculture, a significant fraction of the test subjects consider the wild salmon to be superior in taste to the aquaculture one. However, this is the same fraction of respondents for the wild salmon labelled as “wild” and the aquaculture salmon also labelled as “wild” (Holmer et al. 200x). This clearly illustrates societal biases against agriculture that need be addressed.

In fact, I became first involved with aquaculture through research on its environmental impacts in Europe, across the Mediterranean and in SE Asia (The Philippines, Vietnam and Thailand). As I learned more about aquaculture and its impacts I realized that the impacts were relatively small and easily addressed, and that the approach taken to the assessment of the impacts of aquaculture are intrinsically unfair.

Provided we agree (and I hope we do!) that we ought to produce food to feed humans, then the relevant question is not only what is the environmental value of a pristine coastal area vs. one supporting aquaculture – the approach typically used in evaluation the impacts of aquaculture – but what is the environmental cost of producing food on land and at sea. I submit that this comparison clearly indicates that aquaculture is a relatively benign form of food production, in terms of its environmental impacts as well as risks to human health, than food production on land.

Many past pests with catastrophic consequences on human populations and contemporary risks to human health (mad cow disease, avian flue, porcine flue, etc.) derive from the fact that the animals we grow on land and evolutionary close to humans, so that pathogens and parasites may jump from them onto humans (Duarte et al. 2007). In contrast, marine organisms and, with the exception of mammals, too distant in evolution for their pathogens to easily jump across to humans. The conversion of wild ecosystems onto cropland and pastures is still responsible for much of deforestation, mass application of agricultural fertilizers have deteriorated aquatic ecosystems, both marine and inland, globally and contributed to climate change, and many hazardous persistent organic pollutants have been introduced to protect crops from insect pests and weeds. Yet, we have come to accept that about two thirds of our landscape be used for food production on land, while many people are appalled if they see an aquaculture form protruding far into the horizon.

The solutions to sustainable aquaculture are relatively simple, but will sustainable aquaculture be economically feasible? Can the changes I recommend above be implemented while still delivering benefits? Is this an industry for poor nations, with low labour costs, only or can aquaculture be a successful source of food, jobs, and still deliver benefits to the environment of developed countries with high labour costs such as Australia?

I invite you to offer your views on these questions, and – after listening – I will provide mine in my next post.

Note 1: Fortunately our own growth efficiency is well below this! Try calculating your own growth efficiency to figure out what would happen if your body weight was to increase by 10% of all the weight of the food you ingest in a year… Scary!


Duarte, C.M., N. Marbà, and M. Holmer. 2007. Rapid Domestication of Marine Species. Science 316: 382-383.

Duarte, C.M., M. Holmer, Y. Olsen, D. Soto, N. Marbà, J. Guiu, K. Black and I. Karakassis. 2009. Will the Oceans Help Feed Humanity? BioScience 59: 967–976.

Holmer, M., Black, K., C.M. Duarte, Marbà, N., Karakassis, I. (Eds.) 2008. Aquaculture in the Ecosystem. X, 326 p., Springer Netherlands.