Reviews in Fisheries Science, 17(3):305–317, 2009 Copyright C Taylor and Francis Group, LLC ISSN: 1064-1262 print DOI: 10.1080/10641260802677074

Fishing for Aquaculture: Non-Food Use of Small Pelagic —A Global Perspective

ALBERT G. J. TACON1 and MARC METIAN2 1The University of Las Palmas de Gran Canaria, Las Palmas de Gran Canaria, Spain 2Hawaii Institute of Marine Biology, University of Hawaii, Kaneohe, Hawaii, USA

About 33.29 million tonnes or 36.2% of the total world fisheries catch was destined for non-food uses in 2006, either targeted for reduction into dry meal (fish meal) and oil (fish oil) for use within industrially compounded feeds, or used directly as animal feed in fresh, frozen, or wet processed form. However, whereas the proportion of non-food landings destined for reduction has been relatively constant since 1970 (mean ± SD: 23.17 ± 3.74 million tonnes), “other” non-food use landings have risen markedly from 0.90 million tonnes in 1970 (3.7% total non-food use landings) to 13.14 million tonnes in 2006 or 39.5% total non-food use landings. At present, small pelagic forage fish species, including “low-value/trash fish,” form the bulk of the fisheries catch destined for non-food uses, with the aquaculture sector currently being the largest consumer. In 2006, it is estimated that the aquaculture sector consumed an estimated 23.8 million tonnes of small pelagic forage fish in the form of feed inputs, including 3.72 million tonnes used to make fish meal, 0.83 million tonnes to make fish oil used in compounded aquafeeds, and an additional 7.2 million tonnes of low value/trash fish as a direct feed or within farm-made aquafeeds.

Keywords fisheries, aquaculture, fishmeal, fish oil, trash fish, aquafeeds

OVER-FISHING AND DISPOSITION OF THE overexploited (or depleted and recovering from depletion; Fig- FISHERIES CATCH ure 1), reinforcing earlier observations that the maximum wild Downloaded By: [Tacon, Albert G. J.] At: 03:44 29 March 2009 capture fisheries potential from the world’s oceans has probably Fishing has been practiced since mankind first started hunt- been reached and calls for a more cautious and closely con- ing and searching for food. From its humble beginnings as a trolled development and management of world fisheries (FAO, subsistence-based food fish hunting and gathering activity, fish- 2007). ing has grown into a multibillion dollar industry, with total cap- Global trends in over-fishing have been fueled to a large ex- ture fisheries’ landings of fish and shellfish in 2006 reported at tent by market demands within economically developed coun- 92.0 million tonnes (FAO, 2008a) and valued at over US $91.24 tries for the consumption and importation of high market value billion in 2006 (FAO, 2008b). However, rising global population carnivorous finfish and crustacean species (Alder and Sumaila, and increasing market demand for fish has resulted in increasing 2005; Montaigne, 2007; Rosenthal, 2008), with developed coun- fishing pressure and consequent over-fishing in many parts of tries importing 80% of total internationally traded fisheries the world. According to the latest FAO assessments concerning products in 2006 valued at US $72.6 billion (FAO, 2008a). the health and exploitation of known capture fishery resources, At present, over 76.9% of total finfish landings from capture more than 75% of the world fish stocks for which assessment in- fisheries are species positioned high in the aquatic food chain, formation is available are reported as already fully exploited or with a mean trophic level of 3 and above (Figure 2). In fact, it has been reported by several authors that over-fishing of large- sized high-value carnivorous finfish species has resulted in fish Address correspondence to Dr. Albert G. J. Tacon, Visiting Professor, the species being targeted lower on the aquatic food chain by fish- University of Las Palmas de Gran Canaria (ULPGC), Las Palmas de Gran eries (Caddy and Garibaldi, 2000; Hannesson, 2002; Pauly et al., Canaria, 35001, Spain. E-mail: [email protected] 1998).

305 306 A. G. J. TACON AND M. METIAN

Figure 1 Global trends in the state of fish stocks since 1974 (FAO, 2007).

Notwithstanding the above, about 33.29 million tonnes or year (FAO, 2008a). For the purposes of this article, “other” 36.2% of the total landed fisheries catch was destined for non- non-food use landings include fresh, frozen, or wet-processed food uses in 2006, either targeted for reduction into dry meal fish used for direct animal feeding (including farmed fish and (fish meal) and oil (fish oil), or used directly for animal feeding crustaceans), for use within canned semi-moist pet foods or used (Figure 3). Although the proportion of the total fisheries catch as fish bait. destined for non-food uses has remained relatively constant over For comparative purposes, the growth of the aquaculture sec- the past 35 years (fluctuating from 39.0% of the total catch in tor (an important user of non-food landed fish products as feed 1970 to 36.2% of the total catch in 2006), total reported non-food inputs) and reported landings of small pelagic forage fish, which fish landings have increased in real terms by 8.79 million tonnes constitute the bulk of fish usually targeted for non-food uses or 35.9%, from 24.5 million tonnes in 1970 to 33.29 million (Alder and Pauly, 2006) are also shown in Figure 3. Of particu- tonnes in 2006 (FAO, 2008a, 2008b). Moreover, although the lar note is the strong correlation between the reported non-food proportion of non-food landings destined for reduction has been use trend lines with that of the estimated small pelagic forage relatively constant in global terms over this period (mean ± SD: fish landings over the same period (total landings estimated at 23.17 ± 3.74 million tonnes), “other” non-food use landings 27.26 million tonnes in 2006; FAO, 2008a). have risen markedly over this period, from 0.90 million tonnes For the purposes of this article, small pelagic forage fish in- in 1970 (3.7% total non-food use landings) to 13.14 million cludes those usually lower trophic level finfish species, which, tonnes in 2006 or 39.5% total non-food use landings for that because of their size and position in the upper layers of the water Downloaded By: [Tacon, Albert G. J.] At: 03:44 29 March 2009

Figure 2 Global trends in weighted mean trophic level of capture fisheries’ landings by major trophic grouping (excludes mammals, amphibians, reptiles, and other miscellaneous invertebrates). Trophic levels of individual species taken from FishBase (Froese and Pauly, 2008), where mean trophic level range (TL): aquatic plants: 1; molluscs: 2.0–4.1; crustaceans: 2.2–3.4; finfish: 2.0–2.99 (top 10 by landings in 2006) Peruvian and Japanese , European and Californian pilchard, Sardinellas nei, Araucanian , Gulf , Tilapia nei, Cyprinids nei, Indian oil ; finfish: 3.0–3.99–marine and freshwater fishes nei, Alaska pollock, , chub , Chilean jack mackerel, Scads nei, croakers and drums nei, European , threadfin breams nei, Atlantic mackerel; finfish: 4.0–4.99–skipjack tuna, blue whiting, largehead hairtail, yellowfin tuna, Atlantic cod, saithe, daggertooth pike conger, bigeye tuna, Argentine hake, North Pacific hake (FAO, 2008a). reviews in fisheries science vol. 17 3 2009 NON–FOOD USE OF SMALL PELAGIC FORAGE FISH 307

Figure 3 Total landings from capture fisheries and aquaculture, reported proportion of fish catch destined for non-food uses, and total landings of small pelagic forage fish species (capture fisheries and aquaculture production values exclude aquatic plants and mammals; FAO, 2008a, 2008b).

column, usually serve as prey for higher trophic level be processed industrially for reduction into fish meal and fish to forage on, including larger piscivorous finfish, seabirds, and oil (http://www.fao.org/fishery/species/3021). marine mammals. Also included within the forage fish grouping The order represents the bulk of total small are sandeels (= sandlances: Ammodytes sp.), which, although pelagic forage fish landings (19.76 million tonnes or 72.5% demersal in habit, aggregate in large schools and are targeted of total small pelagic forage fish landings in 2006) and in- for reduction in Europe (Masuda et al., 1984). As mentioned cludes , herring, pilchards, sardinellas, sprat, , previously, in contrast to targeted food-fish species landings, menhaden, shad, , and dagaas (FAO, 2008a). Other ma- small pelagic forage fish species are usually (with a few ex- jor forage fish include species from the order Scombroidei ceptions) positioned lower on the aquatic food chain, with a (3.83 million tonnes or 14.0% total forage fish landings in weighted mean annual trophic level ranging between 2.80 and 2006, including chub mackerel, Atlantic mackerel, Indian mack- 3.0 (Figure 4). Gadiformes (cods) and, specifically, blue whit- erels nei, short mackerel, Indian mackerel, narrow-barred Span- ing, hakes, and grenadiers have been excluded from the listing ish mackerel), Percoidei (2.39 million tonnes or 8.8%, includ- of small pelagic forage fish species, as they are primarily fished ing Chilean jack mackerel, Japanese jack mackerel, Atlantic for human consumption (Cohen et al., 1990). Notwithstanding horse mackerel), Trachinoidei (605,612 tonnes or 2.2%, includ- the above, it is often claimed that it is because of the poor ing Pacific sandlance and sandeels nei), Beloniformes (396,175 flesh-keeping qualities of small “oily” pelagic forage fish and tonnes or 1.5%, including Pacific saury and Atlantic saury) and certain demersal species such as blue whiting that they have to Salmoniformes (278,289 tonnes or 1.0%, including ). Downloaded By: [Tacon, Albert G. J.] At: 03:44 29 March 2009

Figure 4 Global trends in mean trophic level of capture fisheries landings of finfish by major trophic grouping (trophic levels of individual finfish species taken from FishBase; Froese and Pauly, 2008). reviews in fisheries science vol. 17 3 2009 308 A. G. J. TACON AND M. METIAN

As mentioned previously, over 75% of the world fish stocks reduction to fish meal/oil (FAO/RAP, 2005; Funge-Smith are currently considered as being fully exploited or overex- et al., 2005). It has been estimated that “low value/trash fish” ploited and, as such, are not expected to yield further increases. represent about 35% of the total marine fish catch (weighted Thus, in the specific case of the key small pelagic forage fish average for six countries: Bangladesh, China, India, Philip- species, according to FAO (2007), the two main stocks of an- pines, Thailand, Vietnam; Funge-Smith et al., 2005). Apart from choveta (Engraulis rigens) in the Southeast Pacific are consid- small pelagic forage fish species, “low value/trash fish” may ered fully exploited and overexploited; Atlantic herring (Cleu- also contain (depending upon the fishery) discarded fish (by- pea harengus) stocks are fully exploited or recovering from de- catch) and juveniles of other commercially important coastal and pletion in the North Atlantic; and Japanese anchovy (Engraulis demersal fish species. For example, it has been estimated that japonicus) stocks are fully exploited in the Northeast Pacific. “low value/trash fish” represents about 60% of the trawl catch in The major fishing areas with the highest proportions of fully the Gulf of Thailand, and that between 18 and 32% of trash-fish exploited stocks (69–77%) are the Western Central Atlantic, the were juveniles of commercially important species (FAO/RAP, Eastern Central Atlantic, the Northwest Atlantic, the Western 2005; Funge-Smith et al., 2005). Indian Ocean, and the Northwest Pacific, while the areas with the highest proportions (46–60%) of overexploited, depleted, and recovering stocks are the Southeast Atlantic, the Southeast REDUCTION OF SMALL PELAGIC FORAGE FISH AND Pacific, the Northeast Atlantic (FAO, 2007). USE OF AND FISH OIL Moreover, within many Asian countries, overfishing of “low value/trash fish” species for use as feed for the local finfish The estimated total production of fish meal and fish oil from and shellfish aquaculture sector has fueled increased fishing small pelagic forage fish by a major producing country is shown pressure on already degraded local fishery resources (FAO, in Figure 5; total fishmeal and fish oil production in 2006 re- 2007). “Low value/trash fish” is generally defined as fish that ported as 5.46 million tonnes and 943,167 tonnes, respectively have a low commercial value by virtue of their low qual- (FAO, 2008a). Fish meal production has fluctuated from a min- ity, small size, or lack of consumer preference. They are ei- imum of 4.57 million tonnes to a maximum of 7.48 million ther used for human consumption (often processed or pre- tonnes over the past 30 years (mean ± SD = 6.10 ± 0.77 mil- served) or used for livestock/fish, either directly or through lion tonnes), whereas fish oil production has fluctuated from a Downloaded By: [Tacon, Albert G. J.] At: 03:44 29 March 2009

Figure 5 World fish meal and fish oil production from 1976 to 2006, and total reported fish meal and fish oil production in 2006 (dry as-fed basis; FAO, 2008a). reviews in fisheries science vol. 17 3 2009 NON–FOOD USE OF SMALL PELAGIC FORAGE FISH 309

minimum of 0.85 million tonnes to a maximum of 1.67 million Prior to mid 1950s, the global production of fish meal was tonnes (mean ± SD = 1.24 ± 0.23 million tonnes). Moreover, only about 1 million tonnes per year, reaching 2 million tonnes production is currently dominated by Peru and Chile, with these by 1960, and then production rapidly expanded over the next two countries alone producing 38% and 49% of total global fish 20 years (within countries such as Peru, Chile, Norway, Ice- meal and fish oil production in 2006, respectively (Figure 5; land, and Denmark), up to 6–7 million tonnes per year. With FAO, 2008a). the increase in global fish meal production and quality, through Despite the fact that fish meal and fish oil are usually pro- technological improvements in production methods, fish meal duced from landings of single fish species (especially within emerged as the premier protein source in animal feeds, ini- the Americas and Europe), over 85% of the total fish meal tially almost exclusively for use within terrestrial livestock feeds production and 58% of the total fish oil production reported (poultry, swine, cattle), and from the 1990s increasingly within by FAO is non-species specific (FAO, 2008a), with only a few aquaculture feeds (Hardy and Tacon, 2002). countries being species specific in their reporting, including Table 1 shows the reported use of fish meal and fish oil Chile (for meals and oils), Canada, Iceland, Peru (for oils), and from published data derived from industry sources within the the USA. It is not known why many major producing coun- fish meal and fish oil manufacturing sector. From the data pre- tries do not always report their fish meal and fish production sented, it can be seen that, in the case of fishmeal, the largest statistics to FAO down to the species level. A case in point is consumer in 1988 was within poultry feeds (60%), followed Peru, where fish meal is not reported down to the species level, by pig feeds (20%), aquaculture feeds (10%), ruminant feeds whereas fish oil, which is a byproduct of the fish meal manu- (3%), and others (7%). However, by 2006, the situation changed facturing process, is reported as Anchoveta oil (FAO, 2008a). markedly, with aquaculture being the largest consumer (57%), Clearly, this discrepancy requires more attention by major fish followed by pig feeds (21%), poultry (13%) and others (6%). meal and fish oil reporting countries for better transparency Similarly, in the case of fish oil, in 1994 (the earliest year for and credibility. Moreover, in some Asian countries, fish meal is which complete data is available), the largest consumer was for manufactured from “low value/trash fish,” including Thailand edible food purposes (hydrogenation and use in margarines and (Kaewnern and Wangvoralak, 2005) and Vietnam (Edwards et shortenings, 68.5%), followed by feed (primarily aquaculture, al., 2004). Moreover, the above global figures usually only re- 24.7%), and pharmaceutical/industrial uses (6.8%). However, fer to whole fish destined for reduction, and so generally ex- by 2002 (the latest year for which complete information exists), clude fish processing wastes and other fish scraps. For example, the major use of fish oil had changed to aquaculture (81%), fol- within the European Union (EU), it was estimated that about lowed by edible food purposes (14%), and industrial (5%). The 33% of the fish meal produced in the EU-15 in 2002 was man- main reason for the above shifts in fish meal and fish oil use ufactured from trimmings from food fish processing, including has been mainly economic, namely the market price and avail- Spain 100% trimmings, France 100%, Germany 100%, Italy ability of fish meal and fish oil (depending upon quality), and 100%, UK 84%, Ireland 60%, Sweden 25%, and Denmark 10% the generally higher market prices paid for these much sought (Huntington et al., 2004). Similarly, Hardy and Shepherd (2007) after nutrient-rich commodities by the aquaculture feed man- report that 84,579 tonnes of fish meal and 21,916 tonnes of fish ufacturing sector. However, the future use of these finite and oil were produced in Alaska from Alaskan pollock and cod valuable commodities is likely to change yet again in view of

Downloaded By: [Tacon, Albert G. J.] At: 03:44 29 March 2009 processing wastes, although this information is not usually ev- the dramatic price increases observed in the case of fish meal ident within annual fishery statistical reviews (NMFS, 2007a, between 2005 and 2006 and, more recently, for fish oil between 2007b). 2007 and 2008 (Figure 6). Notwithstanding the above reporting difficulties, the use of In the case of fish meal, prices more than doubled from US fish meal and fish oil has varied considerably since the begin- $694 per tonne to US $1379 per tonne between July 2005 and ning of the 20th century, when industrial fish meal production July 2006 (Figure 6). The reason for this price increase was due first emerged in the market place (Kreuzer, 1974). Thus, in the to numerous interconnected factors, including: 1800s in the United States, menhaden were fished primarily for oil production, and the cooked press-cake residue after oil ex- traction then usually used as fertilizer. At the time, fish oil was 1. Reduced landings of pelagics within the major fish meal- a major industrial commodity used in paints, lubricants, soaps, producing countries and consequent reduced fish meal pro- printing ink, and in the tanning of hides (Hardy and Tacon, duction and availability (Peruvian anchoveta landings des- 2002). By 1920, hydrogenated or hardened fish oils were heav- tined for reduction decreasing by 31.7% from 8.6 to 5.9 ily used in margarine and shortenings in Europe (Gauglitz et al., million tonnes between 1995 and 2006; PRODUCE, 2007). 1974). During the early mid to late 1950s, the global production 2. Increasing price paid for pelagics for fish meal production and use of fish oil had increased considerably (especially within (prices paid by fish meal factories for anchoveta from fish- countries such as Peru, Norway, and Japan), and a substantial ermen in Peru reportedly increasing from an average of US part of the oil was converted into edible food products such as $80 per tonne to as high as US $220 per tonne). margarine, shortening, and cooking fats (with the exception of 3. The strong market demand for fish meal by the resident the United States; Gauglitz et al., 1974). aquaculture and livestock sector within the major importing reviews in fisheries science vol. 17 3 2009 310 A. G. J. TACON AND M. METIAN

Table 1 Reported global usage of fish meal and fish oil by major user (values given in %)

Year Use Aquaculture Poultry feed Pig feed Other feeds References

1988 Fishmeal 10 60 20 101 New and Csavas (1995) 1990 Fishmeal 14 58 20 82 Pike (1991) 1994 Fishmeal 17 55 20 83 Pike (1998) 1995 Fishmeal 15 50 25 104 Kilpatrick (2003) 2000 Fishmeal 35 24 29 125 Barlow and Pike (2001) 2001 Fishmeal 40 39 14 76 Kilpatrick (2003) 2002 Fishmeal 46 22 24 87 Pike (2005) 2006 Fishmeal 57 13 21 6 Jackson (2007a)

Year Use Aquaculture Edible8 Indust/Pharm9 Animal feed

1994 Fish oil 24.710 68.5 6.8 — Pike (1998) 1995 Fish oil 18 70 711 5 Kilpatrick (2003) 2000 Fish oil 54 34 2612 — Barlow and Pike (2001) 2001 Fish oil 70 19 813 3 Kilpatrick (2003) 2002 Fish oil 81 14 5 — Pike (2005) 2006 Fish oil 8714 — — — Jackson (2007a)

1Includes 3.0% ruminants; 2Includes 3.0% ruminants, 1% fur;3Includes 2.5% ruminants, 1.5% fur; 4Includes 5.0% ruminants; 5Includes 3.0% ruminants; 6Includes 2.0% ruminants; 7Includes 1.0% ruminants; 8Includes refined and hydrogenated or hardened fish oil used in margarines and shortenings; 9Includes industrial and pharmaceutical uses; 10Small amounts of hydrogenated fish oils reportedly used in calf milk replacer feeds (Pike, 1998); 11Includes 1.0% refining/pharmaceutical; 12Includes 2.0% refining/pharmaceutical; 13Includes 5.0% industrial, 1.0% refining, 2.0% nutraceutical; 14Total use of fish oil in aquafeeds in 2006 reported as 783,000 tonnes, and 117,000 tonnes or 13% used direct human consumption, land animal feeds and industrial purposes: Jackson, 2007a).

countries and, in particular, within China 2005 at 603,000 tonnes (33.35% total), became the largest con- (FAO/GLOBEFISH, 2007; GAIN, 2007; Hongjle, 2007; sumer of fish meal in 2006 at 526,000 tonnes (40% total; Figure Tacon and Nates, 2007). 7). In the specific case of aquaculture in China, increasing fish meal prices resulted in dietary fish meal decreasing from 70% The net effect of increasing global fish meal prices on sub- to 55% for eel, 40% to 30% for marine finfish, 35% to 25% for sequent dietary fish meal usage and substitution in compound shrimp, and 20% to 10% for freshwater fish from 2005 to 2006 animal feeds was both profound and immediate. For example, (Tacon and Metian, 2008). in China (the largest pig and aquaculture producer in the world), Similarly, the recent dramatic increase in fish oil prices to new fish meal usage fell by 28% from 1.80 million tonnes in 2005 to highs of over US $2000 per tonne (fish oil prices increasing from 1.29 million tonnes in 2006, with pig feeds falling from being US $894 to $1700 per tonne between March 2007 and March the largest consumer of fish meal in China at 692,000 tonnes in 2008; Figure 6) have been due to a realignment of fish oil prices Downloaded By: [Tacon, Albert G. J.] At: 03:44 29 March 2009 2005 (38.4% total) to the second largest consumer at 386,000 with other key vegetable oil sources (rapeseed oil, soybean oil, tonnes in 2006 (30% total; Huang, 2007). By contrast, aquacul- palm oil) and increasing global petroleum prices, and a new and ture, which had been the second largest consumer of fish meal in rising demand of omega-3 oils for direct human consumption

Figure 6 International market price for fish oil and fish meal (monthly average, 64/65% crude protein), any origin, wholesale, CIF Hamburg (US$ per tonne; Helga Josupeit, FAO GLOBEFISH Database–personal communication, May 2008). reviews in fisheries science vol. 17 3 2009 NON–FOOD USE OF SMALL PELAGIC FORAGE FISH 311

Figure 7 Fish meal consumption by major livestock category in China in 2005 and 2006 (values in metric tonnes and % total; Huang, 2007).

(IFFO, 2007b, 2008a, 2008b), with fish oil currently being a these landings are small pelagic forage fish species and, in par- more valuable commodity than the protein meal. Clearly, with ticular, “low value/trash fish species” from Asia (Funge-Smith fish oil prices now being double what they were a year ago, the et al., 2005). continued use of fish oil as a relatively inexpensive source of Moreover, the rapid growth of “other” non-food fish land- dietary energy within compound aquafeeds (as in the case of ings since 1985 mimics very closely the rise in production of salmonid diets which consumed over 55% of the fish oil used those higher value (in marketing terms) aquaculture species cur- by the aquaculture sector in 2006; Jackson, 2007) will no longer rently dependent upon the use of “low value/trash fish” and other be economically sustainable in the long run. small pelagic forage fish as direct feed inputs (Figure 8; Funge- Smith et al., 2005; Edwards et al., 2004; Tacon et al., 2006). For example, finfish species whose production is currently largely OTHER NON-FOOD USES OF SMALL PELAGIC based upon the direct use of small pelagic forage fish, either FORAGE FISH fed alone as a complete natural diet or fed in minced/processed form within farm-made aquafeeds, include most wild-caught In contrast to capture fisheries landings destined for reduc- marine carnivorous aquaculture species, including tuna (Aus- tion, which have not increased since the mid 1980s, the propor- tralia, Croatia, Cyprus, Italy, Lybia, Mexico, Spain, Tunisia; tion of the catch destined for other non-food uses in fresh and/or Allan, 2004; Ottolenghi et al., 2004; Van Barneveld and wet processed form as animal feed (and to a lesser extent fishing Vandepeer, 2007; Zertuche-Gonzalez et al., 2008), yellowtail bait) has increased significantly (Figure 3); “other” non-food use (Japan: Seriola sp.; Kolkovski and Sakakura, 2007; Nakada, landings increased from almost zero in 1985 to 13.14 million 2000), grouper (China and most Southeast Asian countries:

Downloaded By: [Tacon, Albert G. J.] At: 03:44 29 March 2009 tonnes in 2006 or 39.5% of total reported non-food use landings Epinephelus sp.; Ottolenghi et al., 2004; Sim et al., 2005), for that year (FAO, 2008b). As mentioned previously, the bulk of barramundi (Thailand: Lates calcarifer; Thongrod, 2007; Sim,

Figure 8 Relationship between reported landings of “other” non-edible small pelagic forage fish and the growth of “low value/trash fish” and forage fish- dependent high-value aquaculture fish and shellfish production (values given in million tonnes; FAO, 2008a, 2008b). For the purposes of this article, high-value aquaculture species includes Pangas and other Asian catfishes, crabs, other miscellaneous freshwater fish species (Mandarin fish, swamp eel, pike, gobies, etc.), Asian marine fishes (marine fishes nei, seabass, amberjacks, groupers, seabreams, croakers, Cobia, flounders, and other flatfishes, rockfish, puffers, etc.), turtles, lobsters, and tunas (total high-value species production 4.5 million tonnes in 2006 or 6.7% of total aquaculture production by weight, valued at US $12.4 billion or 14.4% total aquaculture production by value in 2006; FAO, 2008a). reviews in fisheries science vol. 17 3 2009 312 A. G. J. TACON AND M. METIAN

2008), seabream (Pagrus sp.), snapper (Lutjanus sp.), flounder is currently realized; the major producers of high-value aqua- (Paralichthys sp.), croaker (Larimichthys sp.), pompano (Tra- culture species in 2006 include China (65.5% total), followed chinotus sp.), and cobia (Rachycentron canadum, China and by Vietnam (10.1%), Japan (5.5%), Thailand (4.6%), Indonesia most Southeast Asian countries; De Silva et al., 2008; Hung and (3.2%), India (2.4%) and the republic of Korea (1.9%; FAO, Huy, 2007; Kim et al., 2007; Mao et al., 2007; Sim, 2008). 2008a). In the case of Pangasius catfish farming in Vietnam (with Notwithstanding the lack of official published information production at 825,000 tonnes in 2006, and expected to reach concerning the direct use of “low value/trash fish” and other over 1.0 million tonnes in 2007), it is estimated that 30% to small pelagic forage fish species as aquaculture feed, it is esti- 50% of farmers use farm-made aquafeeds, of which 80% use mated that the total use in aquaculture was between 5.6 and 8.8 “low value/trash fish” as the main ingredient (Nguyen, 2007). million tonnes in 2006 (mean 7.2 million tonnes); China alone Similarly, the production of cultured ornate spiny lobster (Pan- reportedly consumed 4 to 5 million tonnes in 2005 (Jin, 2006), ulirus ornatus) in Vietnam is expected to have exceeded 1500 which may explain, in part, the reported current lack of raw tonnes in 2005/2006 and was fed exclusively on forage feed fish material for processing currently experienced by fish meal man- from 0.3 g (wild-caught seed) to market size (950 g) over a 18– ufacturing plants along the coast of China (IFFO, 2007a). Sim- 22 culture period, with a reported feed conversion ratio (FCR) of ilarly, between 176,000 to 364,000 tonnes of low “value/trash 40–45:1 (Thuy et al., 2007). On the basis of this reported FCR, fish” was estimated to be used in Vietnam as aquaculture feed in the lobster industry in Vietnam consume approximately 60,000 2003 (Edwards et al., 2004), with 131,931 tonnes of trash fish to 67,500 tonnes of forage feed fish. De Silva and Phillips (2007) consumed by seabass and grouper culture in Thailand in 2004 estimated that over 90% of the marine fish farms in Vietnam use (Thongrod, 2007). forage feed fish as the main direct feed (with less than 10% using Interestingly, the above global estimate is within the total farm-made feeds), and calculate that nearly one million tonnes landings reported by FAO for “other” non-food fish landings of trash fish is currently being used as direct feed in aquaculture of 13.14 million tonnes in 2006 (FAO, 2008a). However, the in Vietnam. current estimates are significantly higher than those reported Other aquaculture species that are also currently routinely by De Silva et al. (2008), who estimated that the total use of being fed on forage feed fish include wild-caught spiny lob- “low value/trash fish” within the Asia Pacific region in 2004 ster (China, Vietnam: Panulirus sp. Stimsoni; Chen et al., 2000; was between 1.60 and 2.77 million tonnes. To a large extent, Hung and Huy, 2007; Mao and Tuan, 2007; Thuy et al., 2007), these differences are due to data reported for China: the value mangrove crab (most Southeast Asian countries, Kenya: Scyla estimated for China in the current article estimated at between serrata; De Silva et al., 2008; Marichamy and Rajapackiam, 4 and 6 million tonnes, and being in line with the 4 to 5 million 1999; Mwaluma, 2002), spotted babylon (China, Thailand, Viet- tonne estimate for 2005 reported by the representative of the nam: Babylonia areolata; De Silva et al., 2008), snakehead International Fish meal and Fish Oil Organization (IFFO) in (Chana sp.; De Silva et al., 2008), and catfish (Vietnam: Pan- China (Jin, 2006). gasius sp.; Hung and Merican, 2006; Nguyen, 2007; Phu et al., In addition to aquaculture, a significant proportion of the 2007). “other” non-food catch is also destined for use as fishing bait Among the different aquaculture species currently dependent and within other animal feeds, including pet foods. However,

Downloaded By: [Tacon, Albert G. J.] At: 03:44 29 March 2009 upon the use of forage feed fish, wild-caught tuna (Thunnus sp.) as with aquaculture, no official statistical information exists stand out as being particularly dependent, and at the same time concerning the total global usage by these sectors. Notwith- one of the most lucrative to farm. For example, Huntington standing, a limited amount of information does exist for certain (2008) estimated that the tuna fattening farms in the Mediter- countries. For example, in South Africa it has been estimated ranean area consumed 225,000 tonnes of forage feed fish in that about 60,000 to 70,000 tonnes of the Southern African 2004, with the global usage, including Mexico and Australia, pilchard catch is canned for human consumption, and the re- estimated at between 200,000 to 300,000 tonnes in 2006 for mainder (130,000 tonnes) is packed and used as bait in the the production of about 15,000 tonnes of tuna. According to tuna pole fishery (local and foreign), in the recreational fishery, FAO, total tuna aquaculture production was reported as 14,624 and also exported for use as feed for tuna fattening operations tonnes in 2006 (Mexico 32.4%, Australia 24.7%, Croatia 15.8%, (Hecht and Jones, 2008). Similarly, in the U.S., 99% of the France 8.5%, Spain 6.6%, others 12.0%) and valued at over US total reported anchovy catch (Anchoa mitchilli: 25,163 tonnes $163.3 million, with tuna production being the top aquaculture landed in 2006) was reportedly used for bait (NMFS, 2007b), commodity by value in Croatia in 2006, and the second most and a total 210,000 tonnes of total U.S. fisheries landings (4.9% valuable aquaculture commodity after shrimp and salmon in of domestic landings) destined in fresh or frozen form for use Mexico and Australia, respectively (FAO, 2008a). as bait and animal food. A further 24,000 tonnes of canned With the exception of the tuna fattening operations in Mex- products were also produced in 2006 for use as bait and animal ico, the Mediterranean, and the North African region, the direct feed (presumably pet food). Moreover, although no statistical use of “low value/trash fish” and other small pelagic fish as data is available, Atlantic menhaden (Brevoortia tyrannus)is aquaculture feed is currently restricted to the Asia and Pacific also reportedly used as bait in commercial blue crab, lobster, region, where over 92% of total global aquaculture production crayfish, and eel fisheries. In addition, Elliot (2006) reports that reviews in fisheries science vol. 17 3 2009 NON–FOOD USE OF SMALL PELAGIC FORAGE FISH 313

the Atlantic herring ( harengus) is the major source of animal feeding (including use within aquafeeds; shaded in red), lobster bait in the U.S., and 90% of the bait used in Maine, whereas values to the right of the dotted line refer to direct food with an estimated 50,000 to 60,000 tonnes used annually for use of captured fish and landed aquaculture products (shaded in the harvest of approximately 35,000 tonnes of adult lobsters. yellow). Other important U.S. commercial fisheries employing bait fish While there is no doubt of the economic viability of aquacul- include the Alaska King crab fishery and spiny lobster fishery ture production systems based on the use of fishery resources as (O’Malley, 2004). feed inputs, there is growing concern about the long-term sus- Finally, another rapidly growing market for small pelagic tainability of aquaculture production systems dependent upon forage fish and fishery byproducts is the pet food industry; Aus- the use of wild fishery resources as feed inputs (Deutsch et tralia alone in 2003 imported 33,600 tonnes of fish for the pro- al., 2007; FAO, 2008c; Naylor et al., 1998, 2000; Tacon et al., duction of canned and dry pet foods, primarily for cats. More- 2006). The segments of the aquaculture sector currently most over, De Silva et al. (2008) have estimated, on the basis of a dependent upon fishery resources as feed inputs are primarily series of generalized assumptions, that the pet food sector glob- those engaged in the production of higher-value (in market- ally currently uses the equivalent of 2.43 million tonnes of fish ing terms) crustacean and carnivorous finfish species, either within canned pet foods and 2.9 million tonnes of fish within through the use of industrially compounded aquafeeds contain- dry pet foods and for fur animals. However, these estimates ing high levels of fish meal and fish oil (including salmonids, appear to be high and remain to be confirmed through direct marine shrimp, marine finfish, eels; Tacon and Metian, 2008), or surveys. through the direct use of “low-value/trash fish” either fed alone or within farm-made aquafeeds (including marine fish and some freshwater fish, primarily within the Asian region; FAO, 2008c; AQUACULTURE FISH-IN FISH-OUT BALANCE SHEET, Funge-Smith et al., 2005; Hung and Huy, 2007; Merican, 2005; AND CONCLUDING REMARKS Sim, 2008). In the case of fish meal and fish oil, it is generally accepted In terms of landed fish supply for direct human consumption, that future dietary inclusion levels within compound aquafeeds capture fisheries and aquaculture produced 58.71 and 51.67 mil- (and animal feeds in general) will decrease in the long run lion tonnes of food fish in 2006, respectively (live weight equiva- as wild supplies diminish and become tighter and the mar- lent; Figure 3). However, aquaculture’s production was achieved ket price for these commodities increases (Tacon and Metian, through the consumption of about 23.8 million tonnes of small 2008), with dietary inclusion levels being reduced to supply- pelagic forage fish in the form of feed inputs in 2006, including ing the necessary minimum essential dietary nutrients for the 3.72 million tonnes of fish meal and 0.83 million tonnes of fish target species as high-value key nutrient additives rather than oil within industrially compounded aquafeeds (equivalent to the as major dietary sources of proteins and lipids, respectively. consumption of 16.6 million tonnes of small pelagic forage fish; Dietary substitution of fish meal and fish oil with alternative Tacon and Metian, 2008) and a mean value of 7.2 million tonnes feed ingredient sources will be considerably easier for herbiv- of “low value/trash fish” as a direct feed or within farm-made orous/omnivorous aquaculture and terrestrial species than for aquafeeds (Figure 9). In this figure, values to the left of the the more nutritionally demanding carnivorous aquaculture and

Downloaded By: [Tacon, Albert G. J.] At: 03:44 29 March 2009 central dotted line refer to non-food uses of captured fish for terrestrial animal species. Notwithstanding, fish meal and fish

Figure 9 Flow of fish products through direct and indirect human use in 2006 (values given in million tonnes and calculated from FAO, 2008a). Compound aquafeed-based aquaculture production (with feeds containing fish meal and fish oil) include all marine shrimp and finfish, salmonids, non-filter feeding Chinese , eels, tilapia, milkfish, freshwater crustaceans, catfish, and other miscellaneous freshwater fish species. Low value/trash fish-based aquaculture production includes species listed in Figure 8. All other finfish and shellfish aquaculture production includes non-capture fisheries-dependent species, including filter feeding fish (silver , bighead carp, catla, rohu), molluscs, and all other production not identified down to the species level. reviews in fisheries science vol. 17 3 2009 314 A. G. J. TACON AND M. METIAN

Figure 10 Historical development of feeds in the Norwegian salmon industry in relation to prevailing technologies and typical inclusion levels of crude proteins (CP), digestible energy (DE, MJ/kg) and digestible protein (DP, g/kg) levels (adapted from Talbot and Rosenlund, 2002).

oil are not essential feed ingredients per se, but rather have rep- some farmers have resorted to the use of poorer quality forage resented cost-effective providers of high-quality animal protein fish sources (sometimes 7–10 days old) and byproducts from and marine lipids packaged in near ideal nutritional proportions fish processing factories (presumably from the same species). for most carnivorous and omnivorous high-value aquaculture As with the development of feeds for the salmon industry, it species. is believed that wet forage feed fish-based feeding regimes will Concerning the use of “low-value/trash fish,” substitution be gradually replaced with dry compound pelleted and/or ex- with alternative feed ingredients and dry diet-based feeding truded feed-based feeding regimes, with transfer from wet/moist regimes will be more troublesome and will take longer. How- feeds to dry feeds being most rapid for those species where cost- ever, the current widespread use of forage feed fish-based feed- effective hatchery feeding technologies have already been de- ing regimes in the Asian region, particularly for the higher value veloped for seed production, and the species can be conditioned carnivorous marine fish and crustacean species, is very similar to the use of dry feeds at an early age. As observed with fish to how the salmon farming industry started in Norway in the meal and fish oil, the speed of dietary substitution or transfer early 1970s (Talbot and Rosenlund, 2002), the first farmed At- over to dry feeds in the case of forage feed fish will be governed lantic salmon (Salmo salar) being fed raw fish in the 1970s, and in the short term mainly by economic factors, and primarily by the industry then progressing to the development of semi-moist increasing fuel and forage fish prices—the latter due to dimin- and dry pelleted feeds in the 1980s, to the use of high-energy ishing supplies on the one hand and increasing competition for

Downloaded By: [Tacon, Albert G. J.] At: 03:44 29 March 2009 extruded pelleted feeds in the 1990s and 2000s (Figure 10). Of a shrinking resource on the other, and consequent increasing particular importance is the fact that, as a result of these feed price volatility. technology advancements (see also Kearns, 2005; Larrain et al., In the long term, however, other factors could also play a 2005), fish growth has increased, and fish production costs and role in dictating the speed of transfer over to dry pellet feed- FCRs have reduced for the farmer. ing and to reduced dependence upon forage feed fish, including The main reasons why farmers in China and the Asian region (1) increased environmental pollution effects due to the use of reportedly prefer to use forage feed fish as a direct feed is the forage feed fish-based feeding regimes (Chen et al., 2007; Mer- belief that the cost is lower than that of using pelleted feed; ican, 2005; Ottolenghi et al., 2004; Sim et al., 2005; Xu et al., the reported cost of feeding with trash fish (forage feed fish) 2007), (2) increased biosecurity and disease risks of feeding in China is approximately US $1.5/kg fish produced (Chen et unpasteurized forage fish products back to cultured fish (Anon, al., 2007). Similarly, according to Hung and Merican (2006), 2005; Gill, 2000; Kim et al., 2007; Merican, 2005; Sim et al., average production costs for catfish in Vietnam are higher if fish 2005), (3) increased competition with other non-food users for are fed pelleted feed (VND 8000–11,000/kg) and lower if fed available forage fish resources, including the pet food and fish- farm-made feeds (VND 400–1000/kg). In fact, the same authors ing/bait sector, and (4) increased competition with humans as believe that the increased use of trash fish (forage fish) in farm- an affordable food source (Edwards et al., 2004). made feeds is driven to a large extent by the drive of farmers Moreover, since the dietary nutrient requirements of many to reduce production costs so as to remain profitable, especially of the cultured marine fish and crustacean species in Southeast when catfish farm prices are low and demand from fish proces- Asia are still poorly understood (Boonyaratpalin, 1997), the de- sors is low. In fact, the demand for forage feed fish is such that velopment and market availability of appropriate cost-effective

reviews in fisheries science vol. 17 3 2009 NON–FOOD USE OF SMALL PELAGIC FORAGE FISH 315

compound feeds for many of these species (including crabs and Deutsch, L., S. Graslund, C. Folke, M. Troell, M. Huitric, N. Kautsky, lobsters) is still in its infancy, and, as such, the continued use of and L. Lebel. Feeding aquaculture growth through globalization: forage feed fish as the preferred feed by farmers will continue Exploitation of marine ecosystems for fishmeal. Global Environ. in the short term until this is resolved. Chang., 17: 238–249 (2007). De Silva, S. S., and M. J. Phillips. A review of cage aquaculture: Asia (excluding China). In: Cage Aquaculture–Regional Reviews and ACKNOWLEDGMENTS Global Overview. FAO Fisheries Technical Paper, No. 498 (M. L. Halwart, D. Soto, and J. R. Arthur, Eds., pp. 18–48), Rome, FAO The first author would like to thank the Lenfest Ocean Pro- (2007). De Silva, S. S. Use of wild fish and other aquatic organisms as feed gram of the Pew Charitable Trust for funding, and the second in aquaculture—a review of practices and implications in the Asia- author was supported by a Hoover Foundation Brussels Fellow- Pacific: key issues to be addressed. In: Report of the FAO Expert ship (Belgian American Educational Foundation). The authors Workshop on the Use of Wild Fish and/or Other Aquatic Species also thank Matt Elliot for reviewing earlier versions of this as Feed in Aquaculture and its Implications to Food Security and manuscript. Poverty Alleviation. Kochi, India, 16–18 November 2007, FAO Fish- eries Report. No. 867 (FAO, Ed., pp. 21–22), Rome, FAO (2008). Edwards,P.,L.A.Tuan,andG.A.Allan.A Survey of Marine Trash REFERENCES Fish and Fish Meal as Aquaculture Feed Ingredients in Vietnam. Australian Centre for International Agriculture Research, Canberra, Alder, J., and D. Pauly. On the Multiple Uses of Forage Fish: From Australia, 56 pp. (2004). Ecosystems to Markets. Fisheries Centre Research Reports 14(3), Elliot, M. American Lobster (Homarus americanus). Seafood Watch Fisheries Centre, University of British Colombia, Canada, 109 pp. Seafood Report. Monterey Bay Aquarium, 66 pp. (2006). (2006). FAO/Regional Office for Asia and the Pacific. APFIC Regional Work- Alder, J., and R. Sumaila. Western Africa: A fish basket of Europe past shop on low value and “trash fish” in the Asia-Pacific region, Hanoi, and present. J. Environ. Dev., 13: 156–178 (2005). Vietnam, June 7–9, 2005. RAP Publication 2005/21, Bangkok, Thai- Allan, G. Fish for feed vs fish for food. In: Fish, Aquaculture, land, 32 pp. (2005). and Food Security: Sustaining Fish as a Food Supply (A. G. FAO/GLOBEFISH. Fishmeal: Higher Fishmeal Prices Result in Brown, Ed., pp. 20–26). Record of a conference conducted by Good Business. GLOBEFISH Seafood Highlights 2007, pp. 19–20 ATSE Crawford Fund, Parliament House, Cambera, August 11, (www.infofish.org/pdf/gsh/GSH 2007.pdf) (2007). 2004 (http://www.crawfordfund.org/ awareness/fish aqua food.pdf) FAO. The State of World Fisheries and Aquaculture 2006. FAO Fish- (2004). eries and Aquaculture Department, FAO, Rome, 162 pp. (2007a). Anon. Risk on local fish populations and ecosystems posed by the FAO. FAO Fisheries Department, Fishery Information, Data and Statis- use of imported feed fish by the tuna farming industry in the tics Unit. Fishstat Plus: Universal Software for Fishery Statistical Miditerranean. WWF Mediterranean Programme, April 2005 Time Series. Aquaculture production: Quantities 1950–2006; Aqua- (http://www.wwf.or.jp/activity/marine/lib/wwfonenvironmentalris- culture production: Values 1984–2006; Capture production: 1950– koftunafarming-e.pdf) (2005). 2006; Commodities production and trade: 1950–2006; Total pro- Barlow, S. M., and I. Pike. Sustentabilidad en el abastecimiento de duction: 1970–2006, Vers. 2.30. (2008a). harina y aceite de pescado. Panorama Acuicola, Sept/Oct: 38–41 FAO. Yearbooks of Fishery Statistics 2006. Summary Tables of Fish- (2001). eries Statistics, Capture–Aquaculture–Commodities, Latest Sum- Downloaded By: [Tacon, Albert G. J.] At: 03:44 29 March 2009 Boonyaratpalin, M. Nutrient requirements of marine food fish in South- mary Tables, Fishery production: Estimated value by groups of east Asia. Aquaculture, 151: 283–313 (1997). species: ftp://ftp.fao.org/fi/stat/summ tab.htm. (2008b). Caddy, J. F., and L. Garibaldi. Apparent changes in the trophic com- FAO. Report of the FAO Expert Workshopon the Use of Wild Fish and/or position of world marine harvests: The perspective from the FAO Other Aquatic Species as Feed in Aquaculture and its Implications capture database. Ocean Coast. Manage., 43: 615–655 (2000). to Food Security and Poverty Alleviation. November, 16–18, 2007, Chen, Z., C. Chen, Z. Wu, X. Bei, and Y. Huang. Studies on the rearing Kochi, India. FAO Fisheries Report, No. 867. Rome, FAO, 29 pp. of effects of various diets for the spiny lobster, Panulirus stimsoni. (2008c). J. Shanghai Fish. Univ., 9: 308–312 (2000). Froese, R., and D. Pauly. FishBase. Worldwide web electronic publi- Chen, J., C. Guang, H. Xu, Z. Chen, P. Xu, X. Yan, Y. Wang, and J. Liu. cation. www.fishbase.org, version (12/2007) (2007). A review of cage and pen aquaculture: China. In: Cage Aquaculture– Funge-Smith, S., E. Lindebo, and D. Staples. Asian Fisheries Today: Regional Reviews and Global Overview, FAO Fisheries Technical The Production and Use of Low Value/Trash Fish from Marine Fish- Paper, No. 498 (M. L. Halwart, D. Soto, and J. R. Arthur, Eds., pp. eries in the Asia-Pacific Region. FAO Regional Office for Asia and 50–68), Rome, FAO (2007). the Pacific (RAP), RAP Publication 2005/16, Bangkok, Thailand, Cohen, D. M., T. Inada, T. Iwamoto, and N. Scialabba. An Annotated 38 pp. (2005). and Illustrated Catalogue of Cods, Hakes, Grenadiers, and other GAIN (Global Agriculture Information Network). People’s Repub- Gadiform Fishes Known to Date. FAO Fisheries Synopsis. No. 125, lic of China–Fishery Products Annual 2007. USDA Foreign Agri- Vol. 10. Rome, FAO. 442 pp. (1990). cultural Service, GAIN Report No. CH7094, 12/31/2007, 33 pp Dalton, R. Fishing for trouble. Nature, 431: 502–504 (2004). (http://www.fas.usda.gov/gainfiles/200712/146293362.pdf) (2007). De Monbrison, D., and B. Guillaume. Preliminary studies for DOTT Gauglitz, E. J., V. F. Stout, and J. C. Wekell. Application of fish oils in Symposium–BFT impacts on local development a socio-economic the food industry. In: Fishery Products (R. Kreuzer, Ed., pp. 132– approach. Cahiers Options Mediterraneenes, 60: 127–138 (2003). 136). Surrey, England: Fishing News (Books) Ltd. (1974). reviews in fisheries science vol. 17 3 2009 316 A. G. J. TACON AND M. METIAN

Gelman, A., U. Cogan, S. Mokady, V. Drabkin, and L. Glatman. Ob- Jackson, A. J. Challenges and opportunities for the fishmeal and fish taining human food from whole underutilized fish. In: Advances in oil industry. Feed Technol. Update, 2: 9 (2007a). Seafood Byproducts: 2002 Conference Proceedings (P. J. Bechtel, Jin, W. Fishmeal as a dietary ingredient in China–First impressions. Ed., pp. 403–420), Alaska Sea Grant College Program, University Paper presented at the International Fishmeal and Fish Oil Organi- of Alaska Fairbanks (2003). aation 2006 Annual Conference, October 23–26, Barcelona, Spain Gill, T. A. Waste from Processing Aquatic Animals and Animal Prod- (2006). ucts: Implications on Aquatic Pathogen Transfer. FAO Fisheries Kaewnern, M., and S. Wangvoralak . Status of trash fish utilization of Circular, No. 956. Rome, FAO. 26 pp. (2000). aquaculture in Thailand. Proceedings of the 43rd Kasesart Univer- Hannesson, R. The economics of fishing down the food chain. Can. J. sity Annual Conference on Fisheries, 1–4 February 2005, Bangkok, Fish. Aquat. Sci., 59: 755–758 (2002). Thailand (2005). Hardy, R. W., and J. Shepherd. Sustainable Marine Resources for Or- Kearns, J. P. Current production methods for the production of salmon ganic Aquafeeds. Paper presented at the National Organic Stan- feeds. Int. Aquafeed, 8: 28–33 (2005). dards Board Organic Aquaculture Symposium, November 27, 2007, Kilpatrick, J. S. Fish processing waste: Opportunity or liability. In: Arlington, VA (2007). Advances in Seafood Byproducts: 2002 Conference Proceedings Hardy, R. W., and A. G. J. Tacon. Fish meal: Historical uses, production (P. J. Bechtel, Ed., pp. 1–10), Alaska Sea Grant College Program, trends, and future outlook for supplies. In: Responsible Marine University of Alaska Fairbanks, 566 pp. (2003). Aquaculture (R. R. Stickney and J. P. MacVey, Eds., pp. 311–325). Kim, J.-H., D. K. Gomez, C. H. Choresca, Jr., and S. C. Park. De- New York: CABI Publishing, 391 pp. (2002). tection of major bacterial and viral pathogens in trash fish used Hecht, T. Use of wild fish and other aquatic organisms as feed in to feed cultured flounder in Korea. Aquaculture, 272: 105–110 aquaculture—a review of practices and implications in Africa and (2007). the Near East: key issues to be addressed. In: Report of the FAO Kolkovski, S., and Y. Sakakura. Yellowtail kingfish culture– Expert Workshop on the Use of Wild Fish and/or Other Aquatic Opportunities and problems. World Aquacult., 38: 44–71 (2007). Species as Feed in Aquaculture and its Implications to Food Secu- Kreuzer, R. Fish and its place in culture. In: Fishery Products (R. rity and Poverty Alleviation. Kochi, India, 16–18 November 2007, Kreuzer, Ed., pp. 22–47), Surrey, England: Fishing News (Books) FAO Fisheries Report. No. 867 (FAO, Ed., pp. 21), Rome, FAO Ltd. (1974). (2008). Larra´ın, C., P. Leyton, and F. Almendras. Aquafeed country profile– Hongjle, Y. China: Sleeping giant rapidly awakens. Feed Technol., 7: Chile and salmon farming. Int. Aquafeed, 8: 22–27 (2005). 18–21 (2007). Mao, N. D., and L. A. Tuan. Trash fish species and vitamin supplemen- Huang, H. China fishmeal consumption status and outlook for 2008. Pa- tation affects productivity of juvenile spiny rock lobster Panulirus per presented at the International Fishmeal and Fish Oil Organiza- ornatus fed moist diets. In: Book of Abstracts, Prospering from Dy- tion 2007 Annual Conference, October 17, 2007, Sydney, Australia namic Growth–Asian-Pacific Aquaculture 2007 International Sym- (2007). posium, August 5–8, Hanoi, Vietnam (2007). Hung, L. T., and H. P. V. Huy. Analysis of feeds and fertilizers for sus- Mao, N. D., P. D. Hung, and L. A. Tuan. Effects of different trash fish tainable aquaculture development in Vietnam. In: Study and analysis on growth and body composition of juvenile cobia Rachycentron of Feeds and Fertilizers for Sustainable Aquaculture Development candum. In: Book of Abstracts, Prospering from Dynamic Growth– (M. R. Hasan, T. Hecht, S. S. De Silva, and A. G. J. Tacon, Eds., Asian-Pacific Aquaculture 2007 International Symposium, August pp. 331–361). FAO Fisheries Technical Paper, No. 497. Rome, FAO 5–8, Hanoi, Vietnam (2007). (2007). Marichamy, R., and S. Rajapackiam. Commercial farming of mud crab Hung, L. T., and Z. Merican. Freshwater fish culture in Vietnam for in coastal ponds at Tuticorin. Proceedings of the Fourth Indian Fish- Downloaded By: [Tacon, Albert G. J.] At: 03:44 29 March 2009 the global white fish market. AQUA Culture AsiaPacific Magazine, eries Forum, November 24–28, 1996, Kochi, India, pp. 215–218 2: 14–16 (2006). (1999). Huntington, T., C. Frid, R. Banks, C. Scott, and O. Paramor. As- Masuda, H., K. Amaoka, C. Araga, T. Uyeno, and T. Yoshino. The sessment of the Sustainability of Industrial Fisheries Producing Fishes of the Japanese Archipelago. Vol. 1. Tokyo, Japan: Tokai Fish Meal and Fish Oil. Report to the Royal Society for the University Press, 437 pp. (1984). Protection of Birds (RSPB). Poseidon Aquatic Resource Manage- Merican, Z. Marine finish cage culture. AQUA Culture Asia Pacific ment Ltd, Lymington, Hampshire, UK. June (http://www.rspb. org. Magazine, 2: 22–24 (2005). uk/Images/fishmeal tcm5-58613.pdf) (2004). Mwaluma, J. Pen culture of the mud crab Scylla serrata in Mtwapa Huntington, T. Use of wild fish and other aquatic organisms as feed mangrove system, Kenya. Western Indian J. Mar. Sci., 1: 127–133 in aquaculture—a review of practices and implications in Europe: (2002). key issues to be addressed. In: Report of the FAO Expert Workshop Nakada, M. Yellowtail and related species culture. In: Encyclopedia on the Use of Wild Fish and/or Other Aquatic Species as Feed of Aquaculture. (R. R. Stickney, Ed., pp. 1007–1036). New York: J. in Aquaculture and its Implications to Food Security and Poverty Wiley and Sons, Inc. (2000). Alleviation. Kochi, India, 16–18 November 2007, FAO Fisheries Naylor, R. L., R. J. Goldberg, H. Mooney, M. Beveridge, J. Clay, C. Report. No. 867 (FAO, Ed., pp. 22–23), Rome, FAO (2008). Folke, N. Kautsky, J. Lubchenco, J. Primavera, and M. Williams. IFFO (International Fishmeal and Fish Oil Organization). IFFO Weekly Nature’s subsidies to shrimp and salmon farming. Science, 282: Report for Associate Members–Week 44, October 26–November 4, 883–884 (1998). 2007. Friday, November 9, 2007. 8 pp. (2007a). Naylor, R. L., R. J. Goldberg, J. H. Primavera, N. Kautsky, M. C. IFFO. IFFO Update No. 183, Sept/Oct 2007, 10 pp. (2007b). M. Beveridge, J. Clay, C. Folke, J. Lubchenco, H. Mooney,and M. IFFO. IFFO Update No. 186, January 2008, 10 pp. (2008a). Troell. Effects of aquaculture on world fish supplies. Nature, 405: IFFO. IFFO Update No. 187, February 2008, 11 pp. (2008b). 1017–1024 (2000). reviews in fisheries science vol. 17 3 2009 NON–FOOD USE OF SMALL PELAGIC FORAGE FISH 317

NMFS (National Marine Fisheries Service). Fisheries of the United Report of the FAO Expert Workshop on the Use of Wild Fish and/or States 2005. Current Fisheries Statistics No. 2005. National Oceanic Other Aquatic Species as Feed in Aquaculture and its Implications to and Atmospheric Administration, Silver Spring, MD, 104 pp. Food Security and Poverty Alleviation. Kochi, India, 16–18 Novem- (2007a). ber 2007, FAO Fisheries Report. No. 867 (FAO, Ed., pp. 26–27), NMFS (National Marine Fisheries Service). Fisheries of the United Rome, FAO (2008). States 2006. Current Fisheries Statistics No. 2005. National Oceanic Sim, S. Y., M. A. Rimmer, J. D. Toledo, K. Sugama, I. Rumengan, and Atmospheric Administration, Silver Spring, MD (2007b). K. C. Williams, and M. J. Phillips. A Practical Guide to Feeds and New, M. B., and I. Csavas. The use of marine resources in aquafeeds. Feed Management for Culture Groupers. Network of Aquaculture In: Sustainable Fish Farming (H. Reinertsen and H. Haaland, Eds., Centers in Asia and the Pacific (NACA), Bangkok, Thailand, 18 pp. pp. 43–78), Rotterdam, The Netherlands: A. A. Balkema (1995). (2005). Nguyen, V. H. Vietnam’s catfish and marine shrimp production: An ex- Tacon, A. G. J., and M. Metian (2008). Global overview on the use of ample of growth and sustainability issues. AQUA Culture AsiaPacific fish meal and fish oil in dinustrially compounded aquafeeds: Trends Magazine, 3: 36–38 (2007). and future prospects. Aquaculture, 285: 146–158 (2008). O’Malley, J. M. Trap-weight influence on catches of Hawaiian Spiny Tacon, A. G. J., and S. F. Nates. Meeting the feed supply challenges. In: lobster (Panulirus marginatus) and Scaly slipper lobster (Scyllar- Global Trade Conference on Aquaculture (R. Arthur and J. Nierentz, ides squammosus) from the Northwestern Hawaiian Islands. Pacific Eds). Qingdao, China, May 29–31. FAO Fisheries Proceedings. No. Islands Fisheries Science Center, Administrative Report H-04-06. 9. Rome, FAO, 271 pp. (2007). Joint Institute for Marine and Atmospheric Research, University of Tacon, A. G. J., M. R. Hasan, and R. P. Subasinghe. Use of Fishery Hawaii, Honolulu (2004). Resources as Feed Inputs to Aquaculture Development: Trends and Ottolenghi, F., C. Silvestri, P. Giordano, A. Lovatelli, and M. B. New. Policy Implications. FAO Fisheries Circular No. 1018, Rome, Italy. Capture-Based Aquaculture. The Fattening of Eels, Groupers, Tu- 99 pp. (2006). nas, and Yellowtails. Rome, FAO, 308 pp. (2004). Talbot, C., and G. Rosenlund. Learning from the salmonid industry Pauly, D., N. Christensen, J. Dalsgaard, R. Froese, and F. Torres. Fishing growing fish on nutrient-dense diets. Aquafeed Int., 5: 7–10 (2002). down marine food webs. Science, 279: 860–863 (1998). Thongrod, S. Analysis of feeds and fertilizers for sustainable aquacul- Phu, T. M., T. L. C. Tu, T. T. T. Hien, N. T. Phuong, and B. D. Glen- ture development in Thailand. In: Study and Analysis of Feeds and cross. Trash fish use in the feed production for tra catfish (Pangasius Fertilizers for Sustainable Aquaculture Development (M. R. Hasan, hypophthalmus). In: Book of Abstracts, Prospering from Dynamic T. Hecht, S. S. De Silva, and A. G. J. Tacon, Eds., pp. 309–330) FAO Growth–Asian-Pacific Aquaculture 2007 International Symposium, Fisheries Technical Paper, No. 497. Rome, FAO (2007). August 5–8, Hanoi, Vietnam, p. 293 (2007). Thuy, N. T. B., D. T. Thien, and L. M. Vuong. Performance of cultured Pike,I.H.Increasing Importance of Aquaculture to the Fish Meal ornate spiny lobster (Panuliris ornatus) fed trash fish in Vietnam. In: Industry. Technical Market Report, International Fish Meal Manu- Book of Abstracts, Prospering from Dynamic Growth–Asian-Pacific facturers, Potters Bar, Hertfordshire, UK, 5 pp. (1991). Aquaculture 2007 International Symposium, August 5–8, Hanoi, Pike, I. H. Future supplies of fishmeal and fish oil. In: International Vietnam, p. 201 (2007). Aquafeed Directory and Buyers Guide 1998, Uxbridge, UK: Turret Van Barneveld, R. J., and M. E. Vandepeer. Practical nutrition of farmed RAI PLC., pp. 39–49 (1998). tuna and pelagic finfish. AQUA Culture Asia Pacific Magazine, 3: Pike, I. H. Eco-efficiency in aquaculture: Global catch of wild fish used 22–23 (2007). in aquaculture. Int. Aquafeed, 8: 38–40 (2005). Volpe, J. F. Dollars without sense: The bait for big-money tuna ranching PRODUCE. Anuario Estadistico 2006. Ministerio de la Produccion,´ around the world. BioScience, 55: 301–302 (2005). Oficina General de Tecnolgia de la Informacion´ y Estadistica, San Xu, Z., X. Lin, Q. Lin, Y. Yang, and Y. Wang. Nitrogen, phosphorus, Downloaded By: [Tacon, Albert G. J.] At: 03:44 29 March 2009 Isidro, Lima, Peru, 223 pp. (www.produce.gob.pe) (2007). and energy waste outputs of four marine cage-cultured fish fed with Rosenthal, E. Empty seas: Europe’s appetite for seafood propels trash fish. Aquaculture, 263: 130–141 (2007). illegal trade. The New York Times, January 15, 2008. Available at Zertuche-Gonzalez,´ J. A., O. Sosa-Nishizaki, J. G. Vaca Rodriguez, http://www.nytimes.com/2008/01/15/world/europe/15fish.html?ei R. del Moral Simanek, C. Yarish, and B. A. Costa-Pierce. Ma- =5070&en=4781 fac5ca9. Montaigne, F. Still waters: The global rine science assessment of capture-based Tuna (Thunnus orientalis) fish crisis. National Geographic, 211: 33 (2007). aquaculture in Ensenada region of northern Baja California, Mex- Sim, S. Y. Trends in the use of trash fish/low value fish in three Asian ico. Final Report to The David and Lucile Packard Foundation, Los countries: The primary food source in marine finfish farming. In: Altos, CA, 95 pp. (2008).

reviews in fisheries science vol. 17 3 2009