Fish In, Fish Out: Perception of Sustainability and Contribution to Public Health

ABSTRACT

This article contributes to the ongoing debate regarding the sustainability of fishmeal and fish oil in aquaculture diets. It demonstrates why the “Fish In, Fish Out” metric, which is frequently used to show how many units of wild fish is needed to produce one unit of farmed fish, is not a valid tool for measuring the sustainability or efficiency of aquaculture production. Additionally, the metric diverts attention away from the human health implications of how we raise fish. It substitutes the mass of seafood for the arguably more important value-added dimension – the long chain omega-3 content per unit mass, which is low in fish raised on diets low in marine ingredients. Fishmeal and fish oil produced by sustainable fisheries remain some of the most ecologically efficient ingredients that contribute to the overall gain of seafood biomass. Because many aspects of our health and wellbeing depend on wild fisheries, we must insist on well-managed fish harvest for the health of the world's population

RESUMEN

El presente artículo es una contribución al debate actual sobre la sustentabilidad y el aceite de pescado utilizado para elaborar dietas en acuacultura. Se demuestra cómo la métrica llamada “pez entra, pez sale”, la cual es frecuentemente utilizada para calcular cuántas unidades de pescado extraído del medio natural se necesitan para producir una unidad de pescado cultivado, no es una herramienta válida para dimensionar la sustentabilidad o eficiencia de la producción acuícola. Adicionalmente, esta métrica desvía la atención de las implicaciones que tiene en la salud humana, la forma en la que se crían los peces. Se sustituye la masa de alimentos de origen marino con la supuestamente más importante dimensión con valor agregado –contenido del aceite omega 3 por unidad de masa– cuya concentración es poca en peces criados a base de dietas pobres en ingredientes marinos. La carne y el aceite de pescado producidos por pesquerías sustentables siguen siendo uno de los ingredientes ecológicamente más sustentables que contribuyen a la ganancia de biomasa marina. Ya que muchos aspectos de nuestra salud y bienestar dependen de las pesquerías, debemos insistir en promover una captura bien manejada en pos de la salud de la población mundial.

INTRODUCTION

Cardiovascular diseases continue to be the most prevalent cause of death in the United States (Lichtenstein et al. 2006). Concurrently, farm-raised seafood is becoming less heart-healthy (Seierstad et al. 2005). Why? At least partially because demand for seafood and aquaculture production continues to increase, and the supply of omega-3-rich, marine aquafeed ingredients remains flat. As a result, these ingredients are being displaced by land-based ingredients (Torrissen et al. 2011). Not surprisingly, the long-term availability of marine resources became one of the most important and complex issues affecting environmental well-being and public health. But unfortunately, the “Fish In, Fish Out” concept, which intended to help make fish production more sustainable and is used as a guide for the environmentally conscious consumer (Tacon and Metian 2008), is fundamentally flawed. Those who popularize it may mislead the consumer.

FORAGE FISH PRODUCTS IN AQUAFEED

The metabolism of carnivorous marine species evolved on nutrients from natural marine sources, which cannot be economically replaced in total with land-based ingredients at this time. However, market forces and advancements in nutrition and genetics continue to facilitate partial and increased replacement of fish meal and fish oil in compound feeds for these species. Currently, marine ingredients often represent a minor fraction of feed formulations in diets for carnivorous fish. The rest of the ingredients come from terrestrial sources.

Since the mid-1990s (when comprehensive data first became available), fish meal and fish oil inclusion rates have substantially decreased (Tacon et al. 2011; Food and Agriculture Organization [FAO] 2014c). Concurrently, the diets became more nutrient dense and the feed conversion ratio has improved. This trend is projected to continue (Tacon et al. 2011; FAO 2014c). In 2013, diets of farm-raised salmon contained 15% fish meal and 8% to 9% fish oil (Marine Harvest 2014). Other ingredients of compound aquafeed for these groups of fish mostly come from terrestrial sources, such as soy, sunflower, wheat, corn, and rapeseed (Marine Harvest 2014; FAO 2014c). In addition, the use of animal by-product meals and fats continues to increase (Tacon et al. 2011). However, in the natural environment, salmon feed primarily on other fish and crustaceans (Jacobsen and Hansen 2001). When raised on entirely fish-based diets, some carnivorous species such as Yellowfin Tuna Thunnus albacares have been reported to consume up to 34 kg of feed fish to gain 1 kg of biomass (Wexler et al. 2003). As a general rule, only 10% of the energy from a trophic level is captured as biomass in the next higher trophic level (Welch et al. 2010). Therefore, most farmed carnivorous fish consume much less forage biomass than their wild counterparts.

Farm-raised omnivorous fish consume little fish meal and fish oil (Tacon et al. 2011). For example, in 2010, compound aquafeeds for carp and tilapia contained about 2% and 3% fish meal, respectively (Tacon et al. 2011). Diets of these species usually do not include fish oil (Tacon et al. 2011). Thus, on a global scale, considering quantities of fish meal, fish oil, and farmed fish and shrimp production, “feeding fish to fish” is part of a process that results in the overall gain of seafood biomass (FAO 2014a, 2014b).

MEETING TODAY'S NEEDS WITHOUT COMPROMISING NEEDS OF FUTURE GENERATIONS

Consumers increasingly demand that seafood products come from sustainable sources. The concept of sustainability has been defined in many ways, but one of the definitions possibly most relevant to this discussion is “the ability of an ecosystem to maintain ecological processes and functions, biological diversity, and productivity over time” (Draper 1998, 13).

Consistent with this definition, the menhaden Brevoortia spp. purse-seine fishery, for instance, produces fish meal and fish oil with minimal environmental impact. In fact, Pelletier et al. (2010) reported that menhaden meal and oil are among the least impact-intensive aquafeed ingredients, whereas crop-derived ingredients such as wheat gluten meal have the most impact on the environment, based on the cumulative energy use, biotic resource use, emissions, and other factors. With regard to the robustness of fish stocks used in production of fish meal and fish oil globally, there are valid concerns pertaining to Southeast Asian, Chinese, and some other fisheries. However, most fish meal and fish oil on the global market come from well-managed fisheries, like that of Peruvian Anchovy Engraulis ringens, the world's largest source of fish meal and fish oil (FAO 2014c).

An assessment of 53 maritime countries, which account for most of the global catch, ranked the United States among the top three with the most sustainable fisheries (Mondoux et al. 2008). Although small on a global scale, the menhaden fishery is the second largest and, perhaps, one of the most regulated fisheries (Everett 2008) in the United States.

Furthermore, many fish meal products do not rely on traditional fishmeal fisheries (FAO 2014c). In 2013, about 35% of the world's fishmeal was made from by-products of commercially processed food-grade fish (FAO 2014c). In addition, the amount of fish by-products used in production of meal and oil continues to increase (FAO 2014c). To summarize, the use of fish meal and fish oil, derived from well-managed and renewable fisheries, can be considered to be sustainable.

FISH IN, FISH OUT

In 2009, global production of farm-raised salmonids was 3.5% of total aquaculture production (FAO 2014b). Nonetheless, salmon has become a poster child of the inefficiency of “feeding fish to fish” and a victim of its own popularity among consumers. A few years ago, a peer reviewed paper laid a foundation to a popular belief that it takes five units of forage fish to produce one unit of salmon (Tacon and Metian 2008). Their argument was that typical salmon diets contained 20% fish oil and 30% fish meal, and one metric ton of unidentified small pelagic fish (forage fish) produces 50 kg of fish oil and 225 kg of fish meal. Thus, 150 kg of meal is wasted for each ton of forage fish used to make salmon feed, because 50 kg of fish oil makes 250 kg of salmon feed (i.e., 50 kg/0.2 = 250 kg), which then requires the addition of only 75 kg of fish meal (i.e., 250 kg × 0.3 = 75 kg). Under this scenario, 150 kg of fish meal (i.e., 225 kg − 75 kg = 150 kg) from each ton of forage fish is lost, and each ton of forage fish will only produce 250 kg of salmon feed. Furthermore, assuming a typical feed conversion ratio of 1.25, this amount of feed can produce only 200 kg of salmon biomass. So, according to Tacon and Metian (2008), the Fish In, Fish Out ratio of forage fish to salmon is 5:1 (i.e., 1,000 kg:200 kg). However, this model is grossly incomplete. As an example, in addition to fish oil, one metric ton of forage fish yields 225 kg of fish meal. In the scenario above, only 75 kg of fish meal is used for production of salmon feed, and the remaining 150 kg of fish meal is wasted. In actuality, market forces do not allow fish meal to go to waste. The conversion of forage fish into salmon does not occur in closed systems (Jackson 2009). The fish meal, which was unaccounted for in Tacon and Metian's (2008) calculations, is used in the formulation of feeds for other species that do not require high inclusion rates of fish oil. For example, farm-raised crustaceans constitute 7.1% of aquaculture production (FAO 2014b). In 2006, a typical shrimp feed contained 15% fish meal and only 1.5% fish oil (Tacon and Metian 2008). Because crustacean production is twice that of salmon, salmon diets could be argued to contain fish oil leftover from shrimp diet formulations, a scenario that was not contemplated by the authors.

MEDIA COVERAGE AND PUBLIC PERCEPTION

Shortly after the publication of Tacon and Metian (2008), Jackson (2009) published a rebuttal and explained why their approach was incorrect. He demonstrated that when the leftover fish meal was properly accounted for, and considering fish meal and fish oil that come from marine by-products, the actual Fish In, Fish Out ratio of forage fish to salmon was three times less than previously reported. He also demonstrated that on the global scale, one unit of forage fish supports the production of two units of farmed fish, shrimp, and freshwater crustaceans (Jackson 2009).

In 2010, Tacon coauthored another article and partially accepted the criticism, admitting that the metric was a “relatively narrow analytic tool” (Welch et al. 2010, 236). However, by that time, the Fish In, Fish Out concept had been picked up by the popular media and some retailers and producers. It continues to be quoted by some bloggers and op-ed writers, who make assumptions and claims based on inaccurate metrics. For instance, one author multiplied the numerator of Fish In, Fish Out by a factor of five and confused the Fish In, Fish Out described for salmon with a Fish In, Fish Out for most farm-raised fish. “It takes roughly five pounds of small fish to produce one pound of dry fishmeal, and for most farmed species of fish, it takes about five pounds of fishmeal to produce 1 pound of finished product” (Fish In, Fish Out of 25:1), wrote Charlie Levine, now former senior editor of Marlin Magazine (Levine 2009).

Unfortunately, but not surprisingly, Jackson's rebuttal that exposed some of the fundamental flaws of the Fish In, Fish Out concept did not receive the same level of media coverage and public attention as the original publication. Banobi et al. (2011) reported that high-profile environmentalist articles were cited 17 times more frequently than their peer-reviewed critiques, even when the originals were challenged by independent scientists on several different occasions. Furthermore, articles that did not cite the critiques almost always accepted the results of the original reports without adequate evaluation of the claims.

This was not the first time some environmentalists and nongovernmental organizations have taken advantage of their status as guardians of environmental sustainability and benefited from our cognitive bias, known as the “halo effect” (Balanson 2008). The halo effect often allows such reports to escape the critical examination by the media that occasionally repackages and recirculates erroneous information (Balanson 2008). As a result, the headline-driven research trickles down to producers and retailers, who sometimes find a way to benefit from dubious claims. For example, the president of a U.S. fish farming company that intends to market marine fish fed primarily plant-derived feed refers to a Fish In, Fish Out ratio of 1:1 as the “Holy Grail of marine fish feed research” (Coleman 2012). However, in nature, the Fish In, Fish Out ratio is very far from perfect (Welch et al. 2010). For example, Atlantic Salmon Salmo salar, at a trophic level of 4.4 ± 0.1, is more than one trophic level above the Atlantic Herring Clupea harengus (fishbase.mnhn.fr/Summary/SpeciesSummary.php?ID=236&genusname=Salmo&speciesname=salar&AT=Salmo+salar&lang=English and www.fishbase.org/summary/Clupea-harengus.html in Froese and Pauly 2013). Because the conversion efficiency between trophic levels is about 10% (Welch et al. 2010), the production of one unit of salmon biomass in nature requires more than 10 units of herring.

IMPLICATIONS FOR PUBLIC HEALTH

Large-scale epidemiologic investigations demonstrated that people at risk for coronary heart disease and ischemic stroke benefit from consumption of fish rich in omega-3 fatty acids (Mozaffarian et al. 2011). Consistent with these findings, the American Heart Association recommends the consumption of oily fish at least twice a week (Lichtenstein et al. 2006). Although fish consumption provides protein, selenium, magnesium, vitamin D, and other nutrients (Mozaffarian et al. 2011), the American Heart Association suggests that health benefits of fish consumption are primarily attributable to marine omega-3s (Lichtenstein et al. 2006). However, there are no standards for omega-3 content or for the omega-3 to omega-6 fatty acid ratio of farmed salmon. This ratio depends on the fish oil levels in salmon diets and ultimately determines the health benefits of seafood consumption (Seierstad et al. 2005). For example, in a double-blinded intervention study, patients with coronary heart disease were divided into three groups consuming Atlantic Salmon, which was raised on diets containing 100% fish oil, 100% rapeseed oil, or a blend containing equal quantities of these two oils (Seierstad et al. 2005). It is worth noting that all fish groups were raised on diets that satisfied their minimum requirement of omega-3 fatty acids for salmonids. Lipids extracted from fillets of salmon fed on 100% fish oil, rapeseed oil, or the equal-blend diets contained 30.2%, 11.7%, or 20.5% of total omega-3 fatty acids, respectively. The ratio of omega-3 to omega-6 lipids in salmon fed fish oil, rapeseed oil, or the equal-blend diet was 6.5, 0.6, and 1.7, respectively. As expected, patients who consumed salmon raised on the 100% fish oil diet exhibited reduced serum triglycerides and other improved indicators of health, whereas these effects were not significant in patients consuming salmon raised on 100% rapeseed or equal-blend diets (Seierstad et al. 2005).

An executive in the aquaculture industry publicly stated that consumers will not get the same benefits from salmon consumption if fish oil inclusion continues to decrease (Hage and Fiskaren 2012). According to the executive, consumers in the future may have to double their fish consumption in order to achieve the same intake of omega-3s. But simply increasing the consumption of vegetable-fed fish will not result in the same benefits as those derived from eating fish fed with fish oil. The decreased omega-3s are accompanied by increased omega6s, which are already overly abundant in the typical Western diet and thought to be responsible for numerous inflammatory health issues that are prevalent in the Western world (Schmitz and Ecker 2008). Hence, standards based on incomplete science, media bias, and ill-defined sustainable aquaculture may lead consumers to accept seafood that is becoming less healthy.

THE NUMERATOR OF FISH IN, FISH OUT

Interestingly, no one seems to have questioned the numerator of the ratio, citing Fish In, Fish Out for salmon, tilapia, carp, shrimp, and other farmed species, while ignoring the “Fish In.” Let's consider Fish In, Fish Out calculations by Welch et al. (2010). Like Tacon and Metian (2008), the authors considered a closed system, where forage fish were converted into salmon, and did not account for the unused fishmeal. Similar to Tacon and Metian (2008), these authors concluded that the Fish In, Fish Out ratio of an unidentified lean species, perhaps Peruvian Anchovy, to salmon was 4:1 (Welch et al. 2010). But now, let us use Gulf Menhaden B. patronus, the most common source of fish oil from the U.S. and a commonly used oil in salmon feeds, for the numerator. The fish oil yield of this species is reported to be 12% to 19%, and the fish meal yield is 19% to 23% (Parker and Tyedmers 2012). Let us use the 12% yield of oil and 23% yield for meal for this exercise.

So, one metric ton of Gulf Menhaden yielded 120 kg of fish oil and 230 kg of fish meal, which could produce 600 kg of salmon feed, because salmon feed contained 20% fish oil (i.e., 120 kg/0.2 = 600 kg). Under this scenario, because only 180 kg of fish meal is needed for 600 kg of salmon feed (i.e., 600 kg × 0.3 = 180 kg); 50 kg of fish meal (i.e., 230 kg – 180 kg) from each metric ton of fish is wasted. Using the feed conversion ratio of 1.25, 600 kg of feed can be converted into 480 kg of salmon. In other words, in this example, 2.1 units of forage fish produce one unit of salmon, which is one half the ratio reported by Welch et al. (2010).

In 2013 the inclusion rates of fish oil in diets of Atlantic Salmon declined to 8% (Chilean-raised fish), and the inclusion rate of fish meal declined to 15% (Marine Harvest 2014). The feed conversion ratio improved to 1.17 (Marine Harvest 2014).

Thus, in 2013 one metric ton of Gulf Menhaden could produce 1,500 kg of feed for Chilean salmon (i.e., 120 kg/0.08 = 1,500 kg). Because only 225 kg of fish meal is needed for 1,500 kg of salmon feed (i.e., 1,500 kg × 0.15 = 225 kg), 5 kg of fish meal (i.e., 230 kg – 225 kg) from each metric ton of fish is lost from the model. Using the feed conversion ratio of 1.17, 1,500 kg of feed can be converted into 1,282 kg of salmon. Thus, in 2013 the theoretical Gulf Menhaden-In:Atlantic Salmon-Out ratio has declined to 0.78:1, whereas excess and unaccounted for meal was used elsewhere. Nevertheless, some media sources still cite Fish In, Fish Out of 5:1.

THE DENOMINATOR

These examples assume a simplistic theoretical system, where forage fish is converted into salmon and the leftover fish meal is wasted. Our calculations are not meant to provide a meaningful or accurate metric but to point out that theoretic arithmetic exercises on forage fish conversion into farmed fish should be based on specific species for both the numerator and denominator; for example, menhaden/salmon or anchovy/tilapia. For a more realistic approach, collective terms should be used for the numerator and denominator. The estimates of conversion of wild-caught biomass to farmed biomass should consider all major sources of marine meal, oil, and all major farmed species to determine the efficiency of converting wild fish into farmed fish. In the interest of full disclosure, it is worth noting that little menhaden fish meal is currently being used in salmon diets. Instead, it is used in various animal diets, including but not limited to various non-salmonid fish, dogs, cats, baby pigs, shrimp, laboratory and zoo animals, and dairy cows. In fact, approximately 20% of the world's fish meal is used in diets of baby pigs before they grow to consume all-vegetable feed (Jackson 2009). In addition, some coproduct of fish meal processing, fish solubles, is a fertilizer for organic fruits and vegetables, and a significant amount of fish oil goes to direct human consumption, primarily in the form of fish oil supplements and pharmaceuticals. Thus, “Fish In” produces a variety of “Outs”: “Fish Out,” “Pets Out,” “Plants Out,” “Zoo Animals Out,” “Laboratory Animals Out,” “Fish Oil Supplements Out,” “Pharmaceuticals Out,” etc.

MASS BALANCE

Since the early 1950s, total aquaculture production has increased dramatically and reached about 89.6 million metric ton in 2012 (FAO 2014b). Neither fish meal nor fish oil is used in the production of approximately 39.0 million metric tons of this total, which consists mostly of molluscs (such as scallops, mussels, clams, and oysters) and seaweeds. However, both fish meal and fish oil are used in the production of the remaining farmed organisms, which mostly consist of crustaceans and fishes. Cyprinids (mostly carp) have been produced since the early 1950s, whereas salmonids (salmon and trout), cichlids (mostly tilapia), and crustaceans (mostly shrimp and prawn) were not produced in large quantities until the mid-1980s (FAO 2014c). Although feed for carp and tilapia contains relatively low levels of fish meal, because these fishes account for 67% of all farmed fishes, they consume relatively large quantities of fish meal on a global scale.

According to the FAO (2014b), the aquaculture sector produced approximately 27.2 and 50.6 million metric tons of fish and shrimp in 2003 and 2012, respectively. Concurrently, due to the increasing amount of forage fish going directly to human consumption (FAO 2014b) and improvements in fisheries management, the global quantities of fish harvested for fish meal and fish oil productions slightly decreased from over 19.3 million metric tons in 2003 to 16.3 million metric tons in 2012 (FAO 2014a). Thus, data from the FAO (2014a, 2014b) indicate that the ratio of fish harvested for fish meal and fish oil production to quantities of farm-raised fish and shrimp decreased steadily from 0.7 (Fish In, Fish Out ratio of 0.7:1) in year 2003 to 0.3 (Fish In, Fish Out ratio of 0.3:1) in 2012 (FAO 2014b). It is also worth noting that due to increasing utilization of marine by-products, the global supply of fish meal and fish oil has remained relatively stable. Over the last decade, the world's annual production of fish meal was about 5 million metric tons and the production of fish oil was about 1 million metric tons (FAO 2014b).

SUMMARY

Although some fisheries were poorly managed and as a result collapsed, many fisheries in the developed world have been well managed for decades (Hilborn 2007). These fisheries are sustained not only by government monitoring and regulations but also by long-term business decisions, ethical principles, and market forces (www.msc.org; www.friendofthesea.org). Some of the largest Salmon feed producers implement specific rules for purchase of fish meal and fish oil and follow guidelines and regulations from their national fisheries authorities. For example, Skretting, which produces approximately 1.5 million metric tons of feed annually for farmed fish and shrimp, has a strong commitment to sustainable fish feed production (Skretting 2014).

The use of forage fish in aquaculture production now results in the net gain of fish and crustacean biomass. Furthermore, other important products are also made from forage fish (for example, feed for conventional livestock, pets, and other captive animals; organic fertilizers; omega-3 oil for human consumption) but are not accounted for by the Fish In, Fish Out metric. Presently, feeding fish to fish is the only practical way to ensure that farmed seafood has health benefits comparable to those of fish harvested from the ocean. Thus, if the goal is to provide healthy, sustainable, good-quality nutrition for humans, fish farmers will choose to continue supplementing land-based fish diets with marine ingredients, at least until new cost-effective sources of omega-3s are developed, engineered, or discovered. Because many aspects of our economy and wellbeing depend on wild fisheries, ensuring their sustainability is one of the most important and complex issues today. Therefore, people's views regarding what is sustainable should not be manipulated through incomplete research and flawed metrics.