Article by Dr Neil Auchterlonie, published in International Aquafeed, February edition

Introduction

As is the case for the aquaculture industry itself, the marine ingredients industry has been the focus of attention on its use of forage fish stocks as raw material for aquafeed ingredients. In an effort to adopt some method for accounting for the volumes of whole fish being used in fishmeal and fish oil production, metrics have been proposed for calculating the use of wild fish in aquaculture production. Within that approach the Fish In: Fish Out (FIFO) ratio, and, with possibly less emphasis, the Forage Fish Dependency Ratio (FFDR) are to the fore.  Although at least one group of authors has challenged the use of FIFO on the basis that it is unclear whether it is an ecological, or an ethical indicator of fish resource use (Taylor et al., 2011), the stories behind these acronyms have long become an accepted way of looking at the fishmeal, aquafeed and aquaculture industries’ environmental performances.  Behind that acceptance there has been rather little discussion and debate over the usefulness of the application of the concept. It seems straightforward to assume that a consideration of the amount of wild fish used in the production of farmed fish would be a true reflection of environmental sustainability, but is that really correct?

Nutritional contributions

Fishmeal is a nutritionally complete ingredient for carnivorous fish species. This is a fact that should be unsurprising given the evolution with, and adaptation of, the carnivorous species to a piscivorous diet. The nutritional benefits are well documented and include for example: high relative digestibility, excellent amino acid profiles and few issues with anti-nutritional factors (Lane, et al., 2014) as well as being rich in certain vitamins  (e.g. A, B-group and D) and minerals (e.g. Ca, P, Fe, Zn, Se, I) (Olsen & Hasan, 2012). All these nutrients are known to benefit physiology not only from the perspective of growth (obviously a key consideration in aquaculture production systems), but also from the perspective of fish health, and the nutritional composition of the end product. As IFFO has argued previously (Auchterlonie, 2016), modern fed aquaculture is successful partly by virtue of the foundation that fishmeal and fish oil provided in nutritionally complete diets for carnivorous species (e.g. salmon, shrimp) in the early years, freeing up the industry to make the advances in systems technology and health which have been so important to its progress. 

More fish, more feed

Aquaculture is the fastest growing protein sector and although that rate of growth was estimated by the FAO estimated to decline to 5.8% (over 2005-2014) from 7.2% (over 1995-2004) (Food and Agriculture Organisation, 2016), its contribution to food security is significant, and it is one of the few protein sectors showing growth. Although all that growth is not supported entirely from the fed species (it also includes contributions from shellfish and seaweed cultivation for instance), fed species are considered to be about half of the total but increasing at a faster rate relative to the unfed species (Food and Agriculture Organisation, 2016).  With that growth comes an obvious increasing need for feed volume.  A finite supply of fishmeal and fish oil every year is set against a background of increasing feed supply, as has been pointed out by authors such as Fry et al., (2016), showing predicted growth between 2000 and 2020 (Fig.1.).  It is clear that the early feeds would have to change in composition to meet the growing demand, as there just wasn’t enough fishmeal and fish oil to continue to manufacture diets with the early formulations.

Fig.1. Predicted requirement for aquafeed 2000-2020, from Fry et al., (2016)

The aquafeed industry has consistently substituted some of the fishmeal and fish oil in feeds for salmon and other farmed species over time. This has been recognised as a response to the availability of the marine ingredients (Ytrestøyl, Aas, & Åsgård, 2015), which is especially an issue during years where there are reductions in supply such as when there may be an El Niňo event in the Pacific Ocean. The feed industry sought ways to dampen the peaks and troughs attached to ingredient supply through the provision of alternative ingredients. That decline in use has been catalogued in the scientific literature occasionally (Ibid.), although much of the work to achieve those reductions is a result of those feed companies’ long-term investment in research, and so is by definition commercial information.  With all that effort, it actually seems that for many fish species marine ingredient substitution has a lower limit based on current feed technology, and may occur down only as far as a threshold level because of the specific nutritional attributes of fishmeal and fish oil, viewed against the nutritional needs of those species.  In some instances we may be very close to those thresholds such as in, for example, fish oil inclusions in grower diets for salmon (Sprague, Dick, & Tocher, 2016). In other species there is also an acknowledgement of the role that fishmeal plays in palatability of feeds (Glencross et al., 2016), an often overlooked characteristic of this ingredient, and perhaps increasingly important when viewed in the context of continuing increasing partial replacement by commodity protein ingredients.  Substitution of ingredients as nutritionally complex as fishmeal and fish oil is not straightforward.

FIFO Numbers

IFFO has been tracking the FIFO ratio in fed aquaculture since 2000. Figures for 2000 were calculated by IFFO using the FAO production data and the IFFO annual estimates, updated in 2010 and a further recent calculation made using the 2015 data. These comparisons are presented in Fig.2., below.  Not unexpectedly, these figures show a declining trend – the background of increased aquaculture production, an increased volume of feed set against a finite supply of fishmeal and fish oil makes this no surprise.

	2000	2010	2015
Crustaceans	0.91	0.45	0.46
Marine Fish	1.48	0.88	0.53
Salmon & Trout	2.57	1.38	0.82
Eels	2.98	1.81	1.75
Cyprinids	0.07	0.03	0.02
Tilapias	0.27	0.18	0.15
Other Freshwater	0.60	0.15	0.13
Aquaculture total	0.63	0.33	0.22

 

Fig.2.  FIFO calculations for 2000, 2010 and 2015

That the FIFO ratio calculations have continued to show a decline has been generally well received, even though the concept continues to draw attention away from the actual contribution that the marine ingredients make to aquaculture production. Some authors have explored the relevance of the approach (Welch et al., 2010), but it is still widely acknowledged as a metric for aquaculture performance even though there are layers of complexity below the apparent comparison in the FIFO ratio. That ratio simply compares crude protein and energy supply in aquafeed to the amount of edible fish supplied via fed aquaculture. It does not take account of the other factors supplied in fishmeal and fish oil as high value aquafeed ingredients, and therefore underestimates the actual contribution made.  IFFO has been presenting fishmeal and fish oil as strategic ingredients in aquafeeds for at least the last decade, and the view has been shared by investors (Rabobank, 2015). It is quite clear that such a strategic positioning is a reflection of the materials’ superior nutritional profile, a factor that is neglected within the FIFO calculation. That strategic use accompanies a change in incorporation in feeds towards optimal benefits, such as in feeds for juveniles and broodstock, for example.  What comes “out” is more dependent on the nutritional qualities of what goes “in” than FIFO seems to portray. Utilisation of marine ingredients in feeds to optimise performance in juvenile stocks so the fish can get the nutritional benefits when the young animals need it most, is analogous to the materials use in land animal feeds such as for pigs and poultry. 

Meeting the criticisms – were they ever actually valid?

Over the period where the feed companies have been working hard to make the available supplies of fishmeal and fish oil go further, one thing that didn’t change was the perennial criticism from some groups of the aquaculture sector’s use of wild fish as raw material for ingredients used in feed formulations.  That criticism erupted in the early 1990s with the general question “how many kilos of wild fish does it take to produce a kilo of farmed salmon” and reflected what appeared to some to be a sensible question to ask, which was why feed fish to farmed fish when humans could eat those wild fish directly anyway. For those critics, it seems that the effort of the whole process of making fishmeal and fish oil, fish feed, and then farming salmon and other fish species took fish away from the environment and local communities, an expense that was – apparently - to the benefit of commercial interests thinly veiled as corporate greed.

Attempts to address the use of wild fish in aquafeeds were encapsulated in the term FIFO, and to a lesser extent FFDR, but the subtext in both essentially mean the same thing: reduce the amount of (whole) wild fish used in aquafeed formulations. The rationale appeared to be straightforward, a continual downward pressure on fishmeal and fish oil use in aquafeeds would reduce the pressure on the fisheries which are being used to supply the raw material for fishmeal and fish oil, hence improving their sustainability and marine ecosystem health. The thesis was that these fish would also go to direct human consumption, supporting local fishers, their communities and nutrition (often) in developing countries. Unfortunately, that view is not quite sophisticated enough to take into account the realities of the situation, and there are at least six points that should be considered when reviewing the success, or otherwise, of the FIFO/FFDR approach. These are summarised below.

Point 1. A reduction in marine ingredient inclusion in aquafeeds has been achieved by the feed sector without the influence of the FIFO/FFDR agenda. As we have seen, the reduction in marine ingredients and the corresponding increase in available volume of aquafeeds has been driven largely by the feed companies, rather than any adoption of FIFO/FFDR principles. 

Point 2.  The reduction fisheries that are used for fishmeal and fish oil production are a natural resource that would otherwise not contribute significantly to global food production. The idea of using these fish stocks to support direct human nutrition is based on several incorrect assumptions. In the vast majority of instances, the stocks do not have a market for their product as food for direct human consumption, even where there has been considerable investment by governments, in for example Peru, to promote those markets. Although a small move in direct human consumption has been observed in some of these stocks (and a corresponding increase in the proportional use of byproduct in fishmeal and fish oil manufacture), this is not significant in volume. There are some obvious species that are used for reduction that are unlikely ever to have a direct human consumption market because they are unpalatable, such as the Peruvian anchovy (Engraulis ringens), Atlantic (Brevoortia tyrannus) and Gulf (B. patronus) menhaden species in North America, and the boarfish (Capros aper) in Europe.

Point 3. Under modern fishery management approaches, the variability observed in forage fish stock populations is influenced by environmental factors rather than fishing pressure. The concept of leaving these fish stocks to support marine ecosystem health is also based on a false premise, that the fisheries models predicting simplified relationships across trophic levels were accurate. Science continually advances and we now understand that the variability in these low trophic level fish populations is broad and influenced rather more by environmental factors than fishing pressure (Hilborn et al., 2017). That variability has been predicted to have occurred before the advent of exploitation of the resource by humans in the case of some fisheries including that for the important Peruvian anchovy (Finney et al., 2010). An oft-quoted argument related to the harvesting of these stocks is the one of maintaining a prey source for marine predators (especially mammalian and avian). In fact, developments in the science of these stocks and their interactions with predators indicate a great degree of size selectivity, spatial distribution, and opportunism in predator populations, and a consequent lack of evidence for a strong connection between forage fish abundance and the abundance of their predators (Hilborn et al., 2017).  

Point 4.  The management of forage fish stocks is relatively successful. The idea that FIFO/FFDR would support improved management of the small pelagic fish species’ stocks that supply the bulk of the raw material for fishmeal and fish oil production is unfounded. It assumes in the first instance that these stocks are not well managed, which is a broad generalisation. Unlike some of the food species, the forage fish species are typically small, fast-growing, early maturing fish, with a tendency towards single stock populations, all of which make modelling and management more straightforward than many of the more complex fisheries that supply food for direct consumption. The Sustainable Fisheries Partnership notes in their review of reduction fisheries in 2017, that “more than three quarters (81%) of the total catch volume in this analysis comes from stocks that are reasonably well managed (or better)” in an investigation that looked at 20 stocks with a volume equivalent to approximately 74% of the whole fish reduction annual total volume. [The SFP also notes that there is an outlier to this good news with a particular regional issue related to SE Asia.  At IFFO we have also noted this and together with the Global Aquaculture Alliance (GAA) we are funding a study that will look at where improvements could be made.] The fishmeal industry itself has achieved much to improve the health of these stocks through the adoption of certification schemes such as the IFFO RS (Responsible Supply) scheme (see: www.iffors.com). Volumes of IFFO RS certified fishmeal achieve proportionate volumes of certified product well in excess of comparable feed ingredients, estimated to cover 49% of global production in 2017.

Point 5. Driving lower inclusion rates for fishmeal and fish oil potentially influences the nutritional profile of aquaculture product. An arbitrary setting of marine ingredient inclusion rates by external groups to the marine ingredients, aquafeed and aquaculture industries has the potential to have a direct effect on the nutritional qualities of the farmed fish product. This is a direct consequence of the nutritional richness of the micronutrient composition of fishmeal, a richness that is unparalleled in alternative ingredients. In farmed salmon this could have an effect on, for example, the long chain omega-3 fatty acid concentration, as has been documented in Scotland (Sprague et al., 2016). It also potentially carries a direct conflict with those producers who may wish to produce a high omega-3 product for the premium market. It is feasible that there may be other effects of reduced micronutrients, and at least one EU project has been funded to investigate this important aspect of the nutritional content of farmed fish. 

Point 6.  Reducing fishmeal and fish oil may impact fish health and survival of farmed fish stocks. As well as product quality, setting lower marine ingredient inclusions may also have an impact on fish health and the ability of the farmed animals to cope with disease challenge. In modern aquaculture systems this is amongst the most important of issues relating to sustainability, and readers will be only too aware of the criticisms received by the salmon farming industry about levels of sea lice, and volumes of veterinary medicines used for treatments. We know that some of the micronutrients in fishmeal and fish oil are essential to fish physiology, and some notably for immune system function and competence. Some of this science isstill developing, but we can see that omega-3 concentrations in farmed salmon are important in relation to “robustness” and the ability of the fish to cope with physical handling and stress, from work undertaken by Nofima recently. We also know that in terrestrial species some gut microflora populations in some farmed animals are influenced by diet, which again has an impact on animal health. At least one project (at the University of Aberdeen) is already investigating this subject in relation to farmed salmon. There is the very real possibility that pressurising the inclusion rates for fishmeal and fish oil downwards actually has an impact on health, survival and therefore sustainability of farmed fish stock in aquaculture systems. This must have been very far from the minds of those who advocated a reduction of marine ingredient inclusions in fish feeds in the first place, and although perhaps follows the concept of unintended consequences, is a very real-life example of how the complexities of aquaculture production cannot be viewed as single issues in isolation.

Conclusion

Ultimately, then, we must question the adoption of the principle of limiting of marine ingredients in aquafeeds since largely it doesn’t seem to have achieved what it set out to do. It was a procedure that was designed to influence the sustainability agenda of the forage fish stocks that provide the raw material for fishmeal and fish oil production, but actually there is little evidence that this has happened. In reality improvements have been driven by the fishmeal industry itself through the adoption of certification schemes and responsible supply practices. Where reductions in marine ingredients have occurred, these have actually been driven by the aquafeed companies in response to the availability of marine ingredients. FIFO and FFDR seem to have very little bearing on the improvements in sustainability over time, contrasting very heavily with the real advancements made by the fishmeal and feed industries.