Global Environmental Change 17 (2007) 238–249
Feeding aquaculture growth through globalization: Exploitation of
marine ecosystems for ﬁshmeal
, Sara Gra
, Carl Folke
, Max Troell
, Miriam Huitric
, Louis Lebel
Department of Systems Ecology, Natural Resource Management, Stockholm University, S-106 91 Stockholm, Sweden
Centre for Transdisciplinary Environmental Research, Stockholm University, S-106 91 Stockholm, Sweden
Beijer International Institute of Ecological Economics, Royal Swedish Academy of Sciences, Stockholm, Sweden
Unit for Social and Environmental Research, Chiang Mai University, Chiang Mai, Thailand
Received 10 May 2005; received in revised form 27 July 2006; accepted 16 August 2006
Like other animal production systems, aquaculture has developed into a highly globalized trade-dependent industry. A major part of
aquaculture technology requires ﬁshmeal to produce the feed for farmed species. By tracing and mapping patterns of trade ﬂows globally
for ﬁshmeal we show the aquaculture industry’s increasing use of marine ecosystems worldwide. We provide an in-depth analysis of the
growth decades (1980–2000) of salmon farming in Norway and shrimp farming in Thailand. Both countries, initially net exporters of
ﬁshmeal, increased the number of import source nations of ﬁshmeal, peaking in the mid-1990s. Thailand started locally and expanded
into sources from all over the globe, including stocks from the North Sea through imports from Denmark, while Norway predominantly
relied on northern region source nations to feed farmed salmon. In 2000, both have two geographically alternate sources of ﬁshmeal
supply: the combination of Chile and Peru in South America, and a regional complement. We ﬁnd that ﬁshmeal trade for aquaculture is
not an issue of using ecosystems of the South for production in the North, but of trade between nations with industrialized ﬁsheries
linked to productive marine ecosystems. We discuss the expansion of marine ecosystem appropriation for the global aquaculture industry
and observed shifts in the trade of ﬁshmeal between marine areas over time. Globalization, through information technology and
transport systems, has made it possible to rapidly switch between marine areas for ﬁshmeal supply in economically connected food
producing systems. But the stretching of the production chain from local to global and the ability to switch between marine areas
worldwide seem to undermine the industry’s incentives to respond to changes in the capacity of ecosystems to supply ﬁsh. For example,
trade information does not reveal the species of ﬁsh that the ﬁshmeal is made of much less its origins and there is lack of feedback
between economic performance and impacts on marine ecosystem services. Responding to environmental feedback is essential to avoid
the trap of mining the marine resources on which the aquaculture industry depends. There are grounds to suggest the need for some
global rules and institutions that create incentives for seafood markets to account for ecosystem support and capacity.
r2006 Elsevier Ltd. All rights reserved.
Keywords: Globalization; Aquaculture; Fishmeal trade; Ecosystem support; Sustainable ﬁsheries; Shrimp farming; Salmon farming; Seafood production
We are concerned with feeding our growing world
population and argue that to insure food security we must
focus on maintaining ecosystem performance (Folke et al.,
2004) as the basis of efforts to increase yield and
production outputs. Today, there is heightened concern
about the state of the world’s oceans when three-quarters
of global ﬁsh stocks are fully or over-exploited (Botsford
et al., 1997;Garcia and de Leiva Moreno, 2000). Fish
provides almost 20% of global animal protein consumed
by humans (FAO, 2003). Aquaculture is often discussed,
optimistically, as a method to augment dwindling ﬁsh
ARTICLE IN PRESS
0959-3780/$ - see front matter r2006 Elsevier Ltd. All rights reserved.
Corresponding author. Department of Systems Ecology, Natural
Resource Management, Stockholm University, S-106 91 Stockholm,
Sweden. Tel.: +468 16 17 77; fax: +468 15 84 17.
E-mail address: firstname.lastname@example.org (L. Deutsch).
catches. It is assumed that it will contribute heavily to the
global food supply as the human population continues to
grow. However, recent research has challenged whether
some marine aquaculture technologies can replace ecosys-
tem production and increase global food security or if it,
instead, not only increases demand on other ﬁsh species as
inputs to aquaculture feed, but also reduces overall protein
available for human consumption (Naylor et al., 2000). In
this paper, our discussion of intensive aquaculture refers to
such production, where high yields are generated through
the use of commercial ﬁsh-based feed inputs.
In the area of ﬁsheries exploitation, we currently are
presented with two options: (1) depend on intensive
aquaculture for marine food supplies, and/or (2) restore
and practice ecosystem management of the world’s ﬁsh
stocks. We assert that industry actions and consumer
acceptance supported by government policies have led to a
steady increase in aquaculture production. Considering
that we are already ﬁshing down the food web, and farming
and marketing up the food chain, we conjecture that
intensive aquaculture is the chosen alternative (Pauly et al.,
But this type of aquaculture is not our only option: there
are less resource intensive ways to produce ﬁsh and
shellﬁsh protein. Examples include culturing herbivorous
ﬁsh species or bivalves, and creating integrated aquaculture
systems. Further, we should acknowledge that the focus of
aquaculture of shrimp and salmon is to generate revenues,
not to directly provide food. Intensive aquaculture of cash
crop species should be discussed in such a context.
In order to analyze whether a given type of aquaculture
is a sustainable contributor to global food security, we need
to understand the production system and the underlying
resources upon which the industry depends. Aquaculture
has been practiced for over 2000 years (Tacon and De
Silva, 1997), as an integrated system, recycling wastes and
using those nutrients that humans cannot. But modern
intensive aquaculture systems are in many respects
comparable to terrestrial, high-intensity animal production
systems (Folke and Kautsky, 1989). Like the latter, an
important development in the aquaculture industry is
globalization of the production–distribution–consumption
Aquaculture production and consumption take place on
regional and global scales. For example, about 30% of
global shrimp consumption is supplied by aquaculture
(FishStat Plus, 2004). Modern intensive aquaculture also
depends on the global market for supplying production
inputs including fertilizer, commercial feed, antibiotics and
pesticides. Today, 40% of aquaculture production is
dependent on industrial feeds with a major part of their
origins in marine and coastal ecosystems (New and
Aquaculture has grown with enormous strides in the last
20 years. Cultured seafood production (not including
aquatic plants) has increased more than seven-fold by
weight (from 5 to 36 million tonnes (Mt)) between 1980
and 2000 and the value generated has grown from USD 9
billion in 1984, the ﬁrst year that statistics are available, to
USD 52 billion in 2000 (FIGIS, 2004). This economic
growth has lead to the adoption of aquaculture as a
preferred development path for many nations despite its
environmental and social shortcomings (Lebel et al., 2002;
Often, as this industry has grown, local supplies of
production inputs have proven inadequate with respect to
both quantities and qualities to support industry needs and
these inputs have then been imported (New and Wijk-
¨m, 1990). The success of the salmon and shrimp
aquaculture industries, combined with the existing supply
market developed for terrestrial meat production, resulted
in an easy entrance into the global feed market (Naylor et
al., 2003). This development has driven the lengthening of
the production chain that Lebel et al. (2002) term
‘stretching’. Stretching of the food production–distribu-
tion–consumption system refers not only to the lengthening
of geographical distance between input suppliers, produ-
cers and consumers, but also to the increase in number of
agents, and thus the overall complexity of the produc-
Globalization is certainly not the only driver of
aquaculture industry development. Policies and practices
of agencies and governmental organizations (i.e., liberal
investment policies, subsidies, development aid) encour-
aged development of the industry in the ﬁrst place but
failed in many cases to develop and/or implement
environmental policies (Eagle et al., 2004;Huitric et al.,
2002). Moreover, corporate interests have been the
dominant driver of industry development (Lebel et al.,
There are many issues to consider in an evaluation of
globalized aquaculture, including: scale of growth, i.e. local
activities are now global and increasing; rapidity of growth,
e.g. technology for ﬁshing, distributing, and selling;
implications for other food sectors, e.g. effects on produc-
tion, food availability and food security.
In this study, we focus solely on the scale of growth of
the intensive aquaculture industry. We discuss the globa-
lized production chain of the industry and analyze the
structure and trade ﬂows of the feed input necessary for
current production systems. We concentrate on one key
protein source as input to feed, ﬁshmeal.
We have not
attempted to analyze remaining portions of the chain, e.g.,
distribution and consumption, in detail. Through our
analysis, we illuminate some of the challenges facing
We begin by describing the connection between aqua-
culture production and use of ﬁshmeal through the
ARTICLE IN PRESS
Fish oil is a key input into feed that is often grouped together with
ﬁshmeal. Presently, it cannot be excluded or replaced, and, in fact, ﬁsh oil
supply appears to be even more limited than ﬁshmeal. However, since ﬁsh
oil is a co-product of ﬁshmeal production, i.e. it is derived from the same
resource as meal, we have chosen not address it separately in this paper.
L. Deutsch et al. / Global Environmental Change 17 (2007) 238–249 239
dependence of the industry on commercial feed and the
ﬁshmeal in this feed. We then identify a pattern of
increasing import dependence as a national industry grows
and quantify the level of imported ecosystem support
(dependency on ﬁshmeal imports) highlighting the indus-
try’s embeddedness in the global food production market.
Next, we discern the sources of the ﬁshmeal usage
(production, imports, exports and consumption) for the
two case studies of Thailand and Norway by mapping out
the trade ﬂows between 1980 and 2000. This is the period of
major expansion of the commercial aquaculture sector,
mainly shrimp and salmon, in the two countries, respec-
tively. We discuss the expansion of the size, number and
location of marine support areas, and whether shifts occur
over time. We were unable to obtain statistics for ﬁshmeal
usage by the aquaculture sector alone, i.e. data also include
ﬁshmeal usage by other industries. However, the evidence
we present herein in combination with other sources lead
us to believe that it is the aquaculture industry that
controls ﬁshmeal supplies, even at the expense of other
industries (Hardy and Tacon, 2002;Seafeeds, 2003)
because other sectors can substitute for ﬁshmeal.
Data on production of ﬁshmeal, as well as ﬁsheries and
aquaculture production were derived from the FishStat
Plus database of the FAO Fisheries Department (FishStat
Plus, 2004). The UN database FIGIS/FIDI was also
accessed for some trade values (FIGIS, 2004). Trade
statistics were derived from the Comtrade database
(COMTRADE, 2004). The Comtrade data are imports of
ﬁshmeal classiﬁed as unﬁt for human consumption (SITC
rev. 2 code 08.142).
The global scope of the ﬁndings is then discussed in the
context of sustainable seafood production. We suggest,
given the worldwide overexploitation of ﬁsh resources, that
seafood production is already following the path of an
expanding aquaculture sector aiming at substituting farm-
ing for ﬁshing. This development may be a form of a social
trap (Costanza, 1994) that unwittingly results in more,
rather than less, exploitation of marine ecosystems, with
unintended effects. We conclude that a global commitment
to revive ocean productivity and diversity requires institu-
tional mechanisms that take into account the aggregate
impacts of production–consumption systems and, through
negotiations, distributes clear responsibilities for improv-
ing practices of all actors involved including consumers.
2. Dependence of aquaculture on ﬁshmeal
Aquaculture (not including aquatic plants) has grown
from providing 6% of global ﬁsh supplies by weight in
1980 (FishStat Plus, 2004) to over 27% in 2000 (Tacon,
2003a). Projections indicate continued increases for all
types of aquaculture. Currently, carp dominates with 44%
of all aquaculture production by volume and 24% by value
(Tacon, 2003a). Shrimp aquaculture has grown from
supplying 4% of total shrimp production to 27% in
2000. Salmon aquaculture now provides almost 60% of
total salmon production, up from 1% in 1980. Total
salmon production has tripled since 1980, but salmon
aquaculture has increased 127 times.
Although all aquaculture sectors are growing, the shrimp
and salmon industries have undergone especially rapid
expansion because of economic incentives (Lebel et al.,
2002). During the 1980s, shrimp production grew at an
average of 25% annually; presently growth is around 5%
(Tacon, 2003a). Total shrimp aquaculture production
represents 3% of total aquaculture volumes, but 15% of
total value. In 2000, a single dominant species Peneaus
monodon was ranked as number 20 of all cultured species
by weight, but as number one by value, generating 8% of
total ﬁsh production value worldwide (Tacon, 2003a).
Meanwhile, the value of farmed salmon increased 16 times
since 1984 from USD 195 million to USD 3.3 billion
(FIGIS, 2004), and between 1980 and 2000, global annual
output growth averaged 27% (Guttormsen, 2002).
2.1. Commercial feeds in aquaculture
Almost 40% of all aquaculture production is now ﬁrmly
dependent on commercial feed. This is especially true of
high value carnivorous species, like shrimp, salmon and
trout whose feed contains large portions of marine inputs
in the form of ﬁshmeal (Tacon, 2002). The percentage of
farms using commercial feeds varies from 100% for salmon
and trout to 83% in marine shrimp to 38% in carp farms
(Table 1). The trend towards ever-increasing usage of
commercial feeds took place more rapidly than anticipated
by the industry. In 1990, it was estimated that the
percentage of shrimp farms using commercial feed in
2000 would be 52% (New and Wijkstro
¨m, 1990). Presently,
75–80% of all farmed shrimp are grown on commercial
feed and it is proposed that commercial feeds will soon
replace farm-made feeds in most shrimp farming (Tacon,
2002). Moreover, major volume producers, particularly
carp, are also increasing their usage of commercial feed.
This is not because these largely herbivorous ﬁsh need it,
but because the improved growth rate raises farmers’
proﬁts. That this is possible is a good indication that
ﬁshmeal may be too cheap and/or that some regulation of
its use may be needed. With the exception for the El Nin
year of 1998, prices for ﬁshmeal have been stable around
USD 400/tonne between 1994 and 2005 (FAO, 2006). Until
recently, Asian carp farmers used only natural foods in
ponds (Hardy and Tacon, 2002), whereas in 2000 they used
almost 7 Mt of feed (Hardy and Tacon, 2002;Tacon, 2002).
Since carp feed accounted for almost 60% of all ﬁsh feed
production, this change could have the greatest overall
impact on ﬁshmeal quantities demanded due to the sheer
volume of production.
2.2. Fishmeal in commercial feeds
Presently, capture ﬁsheries yield 110–130 Mt of seafood
annually (FishStat Plus, 2004). Of this total, 70 Mt goes
ARTICLE IN PRESS
L. Deutsch et al. / Global Environmental Change 17 (2007) 238–249240
directly to human consumption, 30 Mt is discarded and
30 Mt becomes ﬁshmeal (Naylor et al., 2000). In 1988,
commercial aquaculture feeds used approximately 8% of
global ﬁshmeal supplies, in 2000 consumption was over
one-third (35%), and use is estimated to approach 70% by
2010 (New and Wijkstro
¨m, 2002;Tacon, 2003c). The
remainder is used in livestock feed, particularly for
chickens and pigs (Barlow, 2002). The proportion of
ﬁshmeal used for feed for different aquaculture species
varies highly (see Table 1). Thailand is the single largest
producer of cultured shrimp, with over 20% of world
shrimp aquaculture production (FishStat Plus, 2004).
Approximately, 90% of the country’s farmed shrimp
production is P. monodon and it is estimated that
30–50% of their feed is ﬁshmeal (Hardy and Tacon,
2002;Tacon, 2002). Presently, salmon, shrimp and trout
aquaculture alone account for almost 50% of all ﬁshmeal
use in aquaculture (Hardy and Tacon, 2002), but provide
less than 10% of ﬁsh production volumes.
While signiﬁcant research is underway to reduce the
percentage of ﬁshmeal in feed, the success of these efforts is
unclear (Hardy, 1999;Tacon, 2004). In general, ﬁshmeal
protein has not proven highly substitutable (Sugiura et al.,
2000 cited in Hardy and Tacon, 2002;Webster et al., 1999).
Various alternatives to ﬁsh protein in feeds are being
evaluated, including waste from seafood processing plants;
terrestrial animal by-product meals (Tacon, 2002); syn-
thetic amino acids (as used in livestock feed (Deutsch and
¨rklund, unpublished manuscript); agricultural by-pro-
ducts, such as palm kernal expellents (Tacon, 2002); or
unicellular bacteria, fungi and algae (Tacon, 2002).
However, it remains to be seen whether these alternatives
are economical and can actually be used in commercial
aquaculture; some present potential human health risks,
for example ﬁsh wastes often contain toxic contaminants
(Hites et al., 2004). While industry acknowledges the
problem and the portion of ﬁshmeal in feed is in fact
decreasing in several species—increases in production
volumes, especially for such dominant species as carp,
has meant that efﬁciency increases have been more than
counterbalanced by growth in production (Goldburg et al.,
The aquaculture industry does not perceive increased
demands for ﬁshmeal as a potentially insurmountable
problem. It is predicted instead that aquaculture will
increase its use of ﬁshmeal at the expense of pig and
poultry production because these animals can substitute
vegetable proteins, such as soybeans, in their diets
(Seafeeds, 2003) and use synthetic amino acids. This has
indeed been the pattern of development historically, since
the amount of ﬁshmeal used in the animal feed industries
has remained relatively constant between 25 and 34 Mt
(Tacon, 2003c), while the aquaculture sector has continu-
ously increased its use of ﬁshmeal (see Box 1).
3. Sources of ﬁshmeal for aquaculture
Annual global ﬁshmeal production was below 5 Mt in
1980. Since 1985, production has remained between 6 and
7 Mt/year, with the notable exceptions during El Nin
years, which in 1987 and 1998 caused signiﬁcant produc-
tion decreases (FishStat Plus, 2004).
In the 1980s, Japan, Chile, Peru and the USSR
dominated production. High Chinese production ﬁgures
in the 1980s may reﬂect incomplete statistics rather than
actual high production levels (Watson and Pauly, 2001).
During the 1990s, Peru, Chile and China were the largest
producers. In 2000, Peru was the dominant ﬁshmeal
producer, providing as much as one-third of global
production; other large producers are Thailand, Denmark,
USA, Norway and Iceland.
About 50–60% of ﬁshmeal production was exported
during 1980–2000. Chile was the largest exporter through-
out the 1980s, after which Peru became dominant.
ARTICLE IN PRESS
Estimated ﬁshmeal and commercial ﬁsh feed usage for selected aquaculture species in 2000
production 1000 t
consumption 1000 t
Fishmeal content in
Marine shrimp 1143 1670 83 23 1.7–2.1 372
413 388 42 23 1–1.3
Marine ﬁsh 603 902 62 42 2.9–3.7 415
Salmon 1009 1636 100 40 2.6–3.3 454
Trout 603 551 100 30 1.5 176
Milkﬁsh 462 313 42 9 0.33–0.42
Carp, using feed 15,525 6991 38 5 0.15–0.19 350
Tilapia 1257 776 42 6 0.24–0.29
Catﬁsh 415 505 86 3 0.28–0.35 15
Eels 233 348 80 50 3.4–4.2 173
Total incl minor
35,487 12,527 2115
Sources: Hardy and Tacon (2002), Pike and Barlow (2002), Tacon (2002, 2003a, c), FishStat Plus (2004) and Tacon, Aquatic Farms, pers. comm.
Predicted ﬁshmeal usage for 2010 estimated at 2.831 Mt (Barlow cited in Hardy and Tacon, 2002) if use at current FCR and content levels ﬁshmeal usage
would be at 4.081 Mt by 2010.
L. Deutsch et al. / Global Environmental Change 17 (2007) 238–249 241
Denmark, Germany, Japan and Norway were also among
the largest exporters during the 1980s, and Iceland joined
their ranks as Japan ceased export production in the 1990s
(FishStat Plus, 2004).
Despite high levels of ﬁshmeal production, some
countries have relatively low export levels, notably the
large aquaculture nations of Japan, Thailand and Norway.
These countries drastically reduced the percentage ex-
ported during 1980–2000. If we examine the largest
producers of shrimp and salmon, Thailand and Norway,
respectively, we see that demand for ﬁshmeal has increased
substantially over the period (Figs. 1 and 2). These
increases in ﬁshmeal consumption closely follow increases
in aquaculture production (excluding plants and bivalves).
Furthermore, since the 1990s, Thailand has shifted from
exporting to importing ﬁshmeal to supply its growing
aquaculture production (Fig. 1). As aquaculture produc-
tion increased in Thailand, levels of domestic ﬁshmeal
production did not rise initially. Instead, meal exports
gradually declined to nearly zero while imports began to
rise. For example, Thailand exported 60% of their ﬁshmeal
in 1980, thereafter steadily decreasing the export quantities
to less than 1% by 1992 and thereafter. Thailand became a
net importer of meal in 1992. The major decreases in
imports during 1997 may have been due to the baht
devaluation, followed in 1998 by El Nin
˜o related reduc-
tions in ﬁshmeal production on a global scale and raised
ﬁshmeal prices (FAO, 1999). In 1998, Thailand actually
exported 6% of its ﬁshmeal production (FishStat Plus,
Norway’s trade patterns are similar in that we see a
decrease in export levels and a deﬁnite increase in imports
as aquaculture production grew (Fig. 2). This shift
occurred after 1985. However, Norway maintained export
trade during the rest of the period and was only a net
importer during the years 1995–1997 and 2000. We also see
ARTICLE IN PRESS
Fishmeal use in shrimp farming.
The aquaculture sectors growing the fastest are those requiring the highest input of ﬁsh resources. We
examine the consumption of ﬁshmeal by the shrimp farming sector. Of total commercial aquaculture feed
production in 2000 (12.5 Mt), 13% (1.6 Mt) was used for shrimp farming (Hardy and Tacon, 2002). On
average approximately one-quarter of shrimp feed is ﬁshmeal (Tacon, 2003b). The three main cultivated
species account for over 86% of total shrimp aquaculture in 2000 (Tacon, 2003a), with Penaeus monodon
over 50% of this. There is a higher protein content in Asian shrimp feed than in American, corresponding to
the differing protein needs of the species farmed (Tacon, 2002).
Some predict that ﬁshmeal consumption will decrease during 2001–2010 due to a lowered feed
conversion ratio (FCR) as well as a decrease in the portion of ﬁshmeal in shrimp feed (Pike and Barlow,
2002;Tacon, 2003b). We have grounds to question whether these predictions are realistic. A more realistic
estimate of feed use acknowledges the difﬁculties in improving FCR and substituting away ﬁshmeal. In
fact, we would argue that the ﬁshmeal content is likely to be greater than one-quarter because: (1) the
percentage of ﬁshmeal in the feed produced by the largest feed producer, Charoen Pokphand, in the
largest shrimp producing nation, Thailand, is 35–40% (Po Garden, pers. comm.) and (2) 50% of global
farmed shrimp production is Penaeus monodon, which still uses 35–50% ﬁshmeal in feed (Tacon, 2002). To
this should be added increasing shrimp production levels and the rising portion of farmers using feeds
(Hardy and Tacon, 2002).
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000
FM production FM import FM export FM consumption Aquaculture production
Fig. 1. Fishmeal consumption and trade compared to aquaculture production (not including plants and bivalves) in Thailand 1980–2000.
L. Deutsch et al. / Global Environmental Change 17 (2007) 238–249242
a doubling of exports from the previous year during the El
˜o year of 1998. Chile also decreased the exported
portion of its ﬁshmeal production at the same time as its
salmon aquaculture production grew (FishStat Plus, 2004).
Hence, as intensive aquaculture production within these
nations expanded, the demands for ﬁshmeal grew as well.
The prediction of reducing ﬁshmeal consumption in
aquaculture does not yet ﬁt reality.
4. Expansion of ecosystem support through trade
Similar to other animal production systems, such as
pork, poultry, and eggs, aquaculture has grown into a
highly globalized industry. Intensiﬁed aquaculture has
increased the need for feed inputs. There is already an
existing market for livestock feed that is based on imported
inputs (Deutsch and Folke, 2005). Thus, as demands for
ﬁshmeal exceed national production capacity, nations can
enter the feed commodities market and start to import.
Today, major ﬁshmeal importers include China (with 27%
of total imports), Japan (8%), Germany (7%), Taiwan
(7%), UK (5%), Norway (4%) and Thailand (2%)
(FishStat Plus, 2004).
We examine ﬁve questions with respect to the growth of
the ﬁshmeal trade and entrance into the global ﬁshmeal
market by aquaculture producers:
1. How is increased ﬁshmeal consumption achieved?—Are
nations able to increase their own production, do they
have to reduce exports, do they have to import?
2. Do nations increase the number of import sources?
3. Do importers switch between input sources over time?
4. Is there a dependency on the ﬁshing grounds of low-
income countries for input supplies?
5. How is the industry affected by the supply of ﬁshmeal?
As illustrated in the previous section, even nations that
have established ﬁsheries sectors and have historically been
net exporters of ﬁshmeal, such as Thailand and Norway,
eventually need to import if demand is great enough. This
has clearly been the case for ﬁshmeal trade as intensiﬁed
aquaculture developed in these countries (Figs. 1 and 2).
Thailand increased domestic production, and Norway
maintained similar production levels over the period.
However, in both nations, increased ﬁshmeal demand
was satisﬁed by both decreasing exports and increasing
We then examine the sources of these imports to see how
they have changed in number and origin over the period
(Figs. 3 and 4). We chose 4 years to represent different
periods in the development of the industry: a pre-import
period (1988 for Thailand and 1985 for Norway); 1990;
1995, and 2000.
We begin by examining Thailand’s import trade. In the
ﬁrst period, there is little trade with basically one partner,
Japan (Fig. 3, 1988). In the second period (Fig. 3, 1990),
import sources have grown to 9 in total, with 95% of
imports from Denmark, Chile, Japan and Republic of
Korea and Thailand becomes a net importer. In 1995 (Fig.
3, 1995), imports peak at 32% of ﬁshmeal consumption.
Further, the total number of import sources has doubled to
18 and the number of key importers is only ﬁve (key
importers provide over 90% of imports). Chile dominates
with almost 60% of volumes, Denmark drops to 14% (less
than half of its contribution in the previous period), and
Peru enters the market at 10%, equal to the combined
contributions of Japan and Republic of Korea. The last
period (Fig. 3, 2000) is after both the baht devaluation in
1997 and the El Nin
˜o event in 1998. We see import and
consumption levels recovering, but the number of sources
has decreased from the previous period. The level of import
dependency has declined slightly to 19% and we see a
switch in sources. Denmark now provides only 2% of
imports, Peru replaces Chile with 72% of volumes and
Republic of Korea’s share has grown to 14%.
In summary, Thailand’s level of import dependency rises
as well as the number of import sources as the aquaculture
industry develops between 1980 and 2000. We see that
Thailand utilizes sources all over the globe, importing from
the established major exporters Denmark and Chile.
ARTICLE IN PRESS
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000
FM production FM import FM export FM consumption Aquaculture production
Fig. 2. Fishmeal consumption and trade compared to aquaculture production (not including plants and bivalves) in Norway 1980–2000.
L. Deutsch et al. / Global Environmental Change 17 (2007) 238–249 243
However, ties are maintained to regional sources through-
out the period in both number of sources (7 of 14) and in
volumes (22% of total imports). In 2000, Thailand is highly
dependent on one major supplier, Peru, for 72% of
The development of Norwegian imports of ﬁshmeal is
similar in several respects. There is little import and only
three sources until after 1985 (Fig. 4, 1985). However, we
can see that consumption levels are already increasing and
export levels are decreasing. In the second period (Fig. 4,
1990), we note an increase to nine import sources. All new
sources are European except for a negligible amount from
USA and Chile. Chile enters the market with 4% of
imports. In the next period (Fig. 4, 1995), Norway has
become dependent on imports for 20% of ﬁshmeal
consumption. There is a large increase in the amounts of
imports and the number of sources increases to 13. There is
a partial shift towards the southeast Paciﬁc, speciﬁcally
Chile and Peru that provide one-quarter of imports,
combined. However, Denmark and Iceland still dominate
with 75% of Norwegian imports. Although there are many
remaining sources, together these do not even equal 1% of
imported volumes. In the last period (Fig. 4, 2000), the
number of import sources declines to 9, but import
dependence continues to climb to 27%. Iceland and
Denmark still dominate trade providing 67%, while Peru
alone now provides 29% of imports.
Finally, although import dependency increases over the
entire period, Norway maintains exports throughout the
period. Northern neighbors dominate trade ﬂows during
the entire period, particularly Denmark and Iceland who
still provide 67% even at their lowest level in 2000.
Basically, there are no signiﬁcant amounts of ﬁshmeal
imports other than those from the major suppliers of
Iceland, Denmark, Chile and Peru, although there is an
increase in the number of sources.
In summary, we see that both nations expand their level
of imports and dependence on marine ecosystem support
of other nations over the period from an initial level of
domestic self-sufﬁciency. Initially, both increase the num-
ber of import sources of ﬁshmeal, peaking in the mid-
1990s, but reduce this number somewhat by 2000. Once
these nations begin to trade ﬁshmeal in earnest, we see that
Thailand utilizes a larger amount of sources than Norway.
ARTICLE IN PRESS
Fig. 3. Imported ﬁshmeal to Thailand 1988, 1990, 1995 and 2000 and import sources. Fishmeal amounts are metric tonnes and the numbers in parentheses
are the percentage of total imports.
L. Deutsch et al. / Global Environmental Change 17 (2007) 238–249244
Thai trade with other nations than their main ﬁshmeal
suppliers is between 5% and 8%, when Norwegian levels
are 1% at most. Yet, in 2000, both nations have only 3–4
key source nations that provide at least 90% of all imports.
Further, both nations expand ﬁshmeal provision with the
same 3 sources (Denmark, Chile and Peru) complemented
by a regional supplier each (Thailand uses the Republic of
Korea and Norway uses Iceland). Thus, very broadly, in
2000 both have two geographically alternate sources of
supply: (1) the combination of Chile and Peru in South
America, and (2) a regional complement, Thailand imports
from Japan and Republic of Korea and Norway imports
from Iceland and Denmark.
Over the time period under study, Thailand shifted its
support from local areas to sources all over the globe.
However, over time, imports from Denmark gradually
became negligible. In the late 1990s, Thailand returned to
its own region, Asia, for a portion of its meal and has
retained these import sources, but in 2000, it was over-
whelmingly dependent on Peru for meal. However, we note
that Japan and Republic of Korea are not considered to be
local sources to Thailand as they are quite distant and
probably different ﬁshing areas entirely. In contrast,
Norway retained its dominant ties to its local northern
region throughout the period, choosing to supply only one-
third of its market from Chile and then Peru. In both
Thailand and Norway, import support from Chile came
ﬁrst and then was almost completely replaced by Peruvian
Over the time period observed, the effects of El Nin
ﬁshmeal consumption were only really obvious in 1998. El
˜o had an apparent negative effect on ﬁshmeal
consumption in Thailand, but this was probably a result
of the combination of the currency devaluation the year
before as well. Imports to Norway also decreased, but not
to the same extent, indicating that the Thai response was
connected to the devaluation of the baht and the resulting
price sensitivity of its ﬁshmeal consumption. In fact, in
1998, Thailand increased its own production and even
exported, perhaps due to 4 factors: lowered availability of
meal, increased prices made meal expensive, while at the
same time higher prices encouraged exports as potential
revenues increased and foreign currency was relatively
more valuable due to the recent baht devaluation. During
this crisis, Thailand depended heavily on its regional
sources, since Asian neighbors supplied 80% of imports
ARTICLE IN PRESS
Fig. 4. Imported ﬁshmeal to Norway 1985, 1990, 1995 and 2000 and import sources. Fishmeal amounts are metric tonnes and the numbers in parentheses
are the percentage of total imports.
L. Deutsch et al. / Global Environmental Change 17 (2007) 238–249 245
that year. Apparently, there exists some capacity for
increased production and exports if prices increase.
However, this may not be a buffer supply of ﬁshmeal
stocks that are normally not utilized, but rather that the
decreased supply on the world market made it possible to
sell lower quality products. For example, it could be
tempting to resume use of ﬁshmeal derived from the Baltic
Sea, which is now not allowed for use within the EU due to
contamination from dioxins and PCBs (EU regulation in
2001: EG nr. 2375/2001). Recent reports of high levels of
dioxin and PCBs in European farmed salmon (Hites et al.,
2004) may well be due to the fact that Baltic ﬁsh are used
for ﬁshmeal despite these regulations, but there may also be
ﬁshmeal from other ﬁshing areas that should not be used.
Such new ﬁndings of contaminated areas could drastically
reduce available ﬁshmeal sources.
5. Discussion and implications
A characteristic of ﬁsheries products in contrast to other
agricultural commodities is the high percentage of inter-
national trade. Over 75% of global ﬁsheries catch is traded
(Watson and Pauly, 2001) and in 2000, over 60% of
ﬁshmeal was traded (Seafeeds, 2003;FishStat Plus, 2004).
As a comparison, only 7% of meat and meat products,
17% of wheat and 5% of rice are exported (World Trade
We assert that the rapid growth in production, as well as
consumption, of many aquaculture products, especially
shrimp and salmon, would not have been possible without
the pre-existing global trade system. Without the ability to
trade neither Thailand nor Norway would have been able
to expand aquaculture production to the extent they have
and become the world’s largest producers of shrimp and
salmon. As aquaculture production grew, exports of
ﬁshmeal decreased, and an increasing portion of demand
was supplied by imports. While production levels were
already high in Norway, in Thailand domestic production
levels increased markedly initially, but neither country has
increased its domestic production since the early 1990s
(with the notable exception of the El Nin
˜o year 1998). Both
nations experienced a point at which local production was
insufﬁcient and were able to increase ﬁshmeal consumption
by expanding their supply network into the ecosystems of
other regions and increasing the amount of ecosystem
support supplied from abroad.
We do not propose that trade is a bad thing. On the
contrary, the international market is a highly efﬁcient
mechanism for supplying goods and services unavailable to
a large portion of the world’s population and is potentially
beneﬁcial even from an ecological point of view. In this
sense, trade ﬂows can build social and economic resilience.
Globalization can mean lesser dependence on local
ecosystems as supply can be increased as needed, if trade
is feasible. However, the ecological economic interdepen-
dency should be recognized (Anderson et al., 1995),
including the implications of ﬁshmeal exploitation on
foodwebs and ﬁsh stocks for human consumption.
Recognition that the aquaculture industry has become
highly globalized also in its ecological dimension has not
been sufﬁciently explored. It is no longer enough to discuss
the local effects of aquaculture production in an analysis of
the sustainability of the industry, as trade decoupled from
its origin effectively masks environmental feedbacks
(Berkes and Folke, 1998).
In our analysis, we see an increasing dependence of both
Thailand and Norway on one marine ecosystem, the
southeastern portion of the Paciﬁc Ocean, represented by
trade with Chile and then Peru. For the aquaculture
industry, ﬁshmeal trade does not seem to be an issue of
North–South, but of trade between nations with indus-
trialized production systems. Both Thailand and Norway
began importing from Denmark, then as its industries
develop Thailand switches its major supplier to Chile and
subsequently Peru. Norway maintained trade imports from
Denmark, but increased trade with Iceland, Chile and
Peru. These four supply nations all have well-developed
industrial ﬁsheries. Thus, it is not a matter of exploiting the
oceans of South, as Chile and Peru do not represent the
Southern Hemisphere, nor developing countries, but rather
the world’s most productive marine ecosystem. We see that
the development of a level of dependency on a single region
decreases supply options instead of increasing them.
The observed shift in ﬁshmeal export from Chile to Peru
is largely due to Chile’s own developing aquaculture
industry, which is now the most rapidly growing salmon
producer in the world. Based upon our case studies of
Norway and Thailand, one might project that Chile’s
ﬁshmeal exports will decline further as the majority of its
ﬁshmeal is used for its own expanding regional/local
aquaculture industry. This could have large impacts on
global ﬁshmeal availability.
It is critical to be aware of import dependency if supply
sources are vulnerable. The southeastern Paciﬁc Ocean is
frequently perturbed by the natural occurrence of the El
˜o Southern Oscillation (ENSO). This needs to be
considered in the context of increasing ﬁshing pressure in
the area as aquaculture production increases and with it a
predicted rise in the demand for ﬁshmeal (Barlow, 2002;
Hardy and Tacon, 2002).
We hope that aquaculture nations recognize the ob-
served and potential alternate states of marine waters of
their ﬁshmeal suppliers (Beamish et al., 2004;Knowlton,
2004) and have incorporated these variations into their
production strategies. From our analysis, it appears that
Thailand was much more susceptible to ﬂuctuations in
ﬁshmeal availability due to trading with countries affected
by ENSO events. However, both nations are increasing
their dependence on a relatively few number of marine
Owing to information gaps, we cannot analyze the
effects of ﬁshmeal production on ﬁsh stocks. Speciﬁcally,
trade information does not reveal the species of ﬁsh that
the ﬁshmeal is made of much less its origins so that we
ARTICLE IN PRESS
L. Deutsch et al. / Global Environmental Change 17 (2007) 238–249246
could determine the particular stock and then examine the
status of this stock over time. The present international
market system has few receptors to capture changes in the
capacity of ecosystems to supply ﬁsh and ﬁshmeal.
Technological developments, like information technology
and transport systems, that characterize the global market,
have made it possible to connect the economic part of food
producing systems worldwide, but have not included
ecosystem signals (e.g., ﬁsh stock declines due to over-
harvesting) in this system. This hampers the ability to
respond to environmental feedbacks and implement the
ecosystem approach to ﬁsheries management. We use the
term ‘‘masking’’ to refer to the delinking of social
feedbacks from change in ecosystem dynamics (Berkes
and Folke, 1998).
Global trade can mask the constraints of local ecosys-
tems and thus allow producers and consumers to ignore
them by enabling substitution of input sources, or even
sequential exploitation (Grima and Berkes, 1989;Berkes
et al., 2006). The stretched commodity chain, particularly
because of the scale and speed at which it operates when
modern technology is used, can mask a seemingly obvious
ecosystem signal, such as the collapse of a local ﬁsh
population. For many natural resources, including ﬁsh-
meal, supply and demand is mediated through long-
distance commodities traded based on market prices.
Demand for ﬁshmeal from one market (and its associated
set of producing ecosystems) can change to another with a
fax, a phone call, or the glance at a computer screen and
the ﬂick of a ﬁnger. This global high-tech trade moves us
mentally and physically further from our life-support base.
A lack of ecological understanding, coupled with the
masking of environmental feedback, reinforces the decou-
pling of people from their supporting environments.
Ecosystem resilience is being lost in the very ecosystems
upon which we depend (Holling and Meffe, 1996).
We do not advocate halting resource extraction from
ecosystems (it is an inescapable livelihood source), but
point out that the effects of incorrect usage of ecosystem
goods and services spill out over other regional or even
global areas (Holling, 1994). In the case where imports
come from distant sources, ecosystem services and support
are not only taken for granted, but consumers are far
removed from the immediate consequences and impacts of
their purchases (Ekins et al., 1994). Ecosystem support is
not apparent, and therefore ignored by the consumers
whose individual purchasing choices affect the price and
production levels of farmed salmon and shrimp (Folke,
2003). Hence, stretching of the production system seems to
have decreased the social capacity to respond to environ-
mental feedback (Berkes and Folke, 1998).
This is happening at a time when there is increasing
evidence, including historical reconstructions, of the
ecological effects of human ﬁshing pressures and practices
(Jackson et al., 2001;Worm et al., 2003). That the world’s
ﬁsheries are presently in a state of crisis is nothing new.
Although some may argue that the effects are natural
ﬂuctuations (Chavez et al., 2003) or that the effects of
overﬁshing may be reversible (Myers et al., 1995) or even
that they are ‘‘ecologically acceptable’’ (Steele and Hoa-
glund, 2003) we do not yet know the long-term ecological
impacts of our choices today.
Many of the main species of ﬁsh used in ﬁshmeal have
experienced collapse, some are not yet recovered or have an
unknown status, while others seem to have recovered
(Folke and Kautsky, 1989;Burke et al., 2000;Hjermann
et al., 2004;ICES, 2004;Matishov et al., 2004). The most
well-known example is the Peruvian anchoveta population,
which has collapsed repeatedly (1972, 1977, 1987, 1992,
1998, and 2002). These severe declines have been explained
by the El Nin
˜o phenomena, but considering the extremely
high ﬁshing pressure, overﬁshing is likely to have deepened
the crashes and delayed the recoveries (Pauly et al., 2002;
Tuominen and Esmark, 2003). Furthermore, we can see
that species used for ﬁshmeal production in the 1970s: the
Japanese pilchard, South American pilchard and Chilean
Jack mackerel, were replaced by the chub mackerel and the
Atlantic mackerel by the 1990s (NRC, 1999). It has been
proposed that all of the stocks used for ﬁshmeal are fully
exploited (Hardy and Tacon, 2002). Heavily exploited
populations tend to be more vulnerable to stress and
sustained environmental variability (Sharp, 1995) than are
lightly exploited stocks. Maintaining heavy ﬁshing pressure
at the lower levels of the food web, spurred in part by ever
increasing demand for ﬁshmeal in the growing aquaculture
sector, may make it difﬁcult for marine ﬁsh species at
higher trophic levels to recover even if ﬁshing pressure on
these stocks was signiﬁcantly decreased.
Moreover, the aquaculture industry has a poor record of
success in replacing ﬁshmeal with other ingredients for
shrimp and salmon feeds (Hardy and Tacon, 2002).
Furthermore, we see that decreases in overall national
ﬁshmeal consumption do not result in corresponding
declines in aquaculture production levels (Figs. 1 and 2).
We suggest that the aquaculture industry may not be as
sensitive to market ﬂuctuations as other animal production
sources that may then be forced to ﬁnd alternatives to
ﬁshmeal use (Hardy and Tacon, 2002). Currently, it is
difﬁcult to see how the global aquaculture industry will
provide space for a reduction of ﬁshing pressure.
We need to expand the discussion and management of
aquaculture production from the local farm site to include
its use and dependence on the global marine production
system supporting the farm. As shown in the examples
presented in this paper, a stretching of the supply system of
intensive aquaculture production systems has already
taken place through the global market, made possible by
information technology and transport. Aquaculture pro-
ducers are seldom constrained by local resource inputs but
operate on a global scale with exploitation of rapidly
varying locations across the globe. Scientists and policy
ARTICLE IN PRESS
L. Deutsch et al. / Global Environmental Change 17 (2007) 238–249 247
makers have not kept up with these market developments
and changed their conceptual frameworks accordingly, but
have instead to a large extent remained in a pre-
globalization worldview of environmental resource man-
Given the increasing scale and speed of this human
activity, we conclude that the industry has truly global
effects today. In the future, the industry should (1)
acknowledge that intensive aquaculture uses a global
marine resource base in a stretched production system,
(2) increase the capacity to trace the marine resource base
of aquaculture products, and (3) develop measures to
detect feedback from the marine ecosystems that provide
consumers with these resources, even when they are distant
and used only for a short period of time. There is a need for
global guidelines that reconnect users to input sources.
Responding to environmental feedback is essential to avoid
further mining of marine resources. There are even grounds
to suggest the need for some global rules and institutions
that create incentives for markets to account for ecosystem
support and capacity (Costanza et al., 1995;Naylor et al.,
1998). Governance of this capacity could be included as the
fourth link in the traditional chain, so that we analyze an
We would like to thank: Robert Kautsky at azote images
for his masterful ﬁgures; Albert Tacon for his open sharing
of data and expertise; two anonymous reviewers for their
constructive comments; and Po Garden for his input
during development of the manuscript.
Anderson, T., Folke, C., Nystro
¨m, S., 1995. Trading with the Environ-
ment. Earthscan, London.
Barlow, D.S.M., 2002. The world market overview of ﬁsh meal and ﬁsh
oil. Second Seafood By-Products Conference, International Fishmeal
& Fish Oil Organisation, Alaska, USA. Available at: /http://
Beamish, R.J., Benson, A.J., Sweeting, R.M., Neville, C.M., 2004.
Regimes and the history of the major ﬁsheries off Canada’s west
coast. Progress in Oceanography 60, 355–385.
Berkes, F., Folke, C. (Eds.), 1998. Linking Social and Ecological Systems:
Management Practices and Social Mechanisms for Building Resilience.
Cambridge University Press, Cambridge.
Berkes, F., Hughes, T.P., Steneck, R.S., Wilson, J.A., Bellwood, D.R.,
Crona, B., Folke, C., Gunderson, L.H., Leslie, H.M., Norberg, J.,
¨m, M., Olsson, P., O
¨sterblom, H., Scheffer, M., Worm, B.,
2006. Globalization, roving bandits, and marine resources. Science
Botsford, L.W., Castilla, J.C., Peterson, C.H., 1997. The management of
ﬁsheries and marine ecosystems. Science 277, 509–515.
Burke, L., Kura, Y., Kassem, K., Revenga, C., Spalding, M., Mcallister,
D., 2000. Pilot Analysis of Global Ecosystems (PAGE): Coastal
ecosystems, World Resources Institute, Washington, DC. Available at:
Chavez, F.P., Ryan, J., Lluch-Cota, S.E., Niquen C, M., 2003. From
anchovies to sardines and back: multidecadal change in the Paciﬁc
Ocean. Science 299, 217–221.
COMTRADE, 2004. UN Commodity Trade Statistics Database. Avail-
able at: /http://unstats.un.org/unsd/comtrade/dqBasicQueryResult-
Nations Statistics Division.
Costanza, R., 1994. Three general policies to achieve sustainability. In:
Jansson, A., Hammer, M., Folke, C., Costanza, R. (Eds.), Investing in
Natural Capital—The Ecological Economics Approach to Sustain-
ability. Island Press, Washington, DC, pp. 392–407.
Costanza, R., Audley, J., Borden, R., Ekins, P., Folke, C., Funtowicz,
S.O., Harris, J., 1995. Sustainable trade: a new paradigm for world
trade. Environment 37, 16–44.
Deutsch, L., Bjo
¨rklund, J., unpublished manuscript. Made in Sweden? Re-
deﬁning the Swedish animal production system.
Deutsch, L., Folke, C., 2005. Ecosystem subsidies to Swedish food
consumption from 1962–1994. Ecosystems 8, 512–528.
Eagle, J., Naylor, R., Smith, W., 2004. Why farm salmon outcompete
ﬁshery salmon. Marine Policy 28, 259–270.
Ekins, P., Folke, C., Costanza, R., 1994. Trade, environment &
development: the issues in perspective. Ecological Economics 9, 1–12.
FAO, 1999. Commodity market review 1998–1999. Food and Agriculture
Organization of the United Nations—Commodities and Trade
FAO, 2003. Aquaculture production trends analysis. In: Tacon, A.J. (Ed.),
Review of the State of World Aquaculture. FAO Fisheries Circular.
No. 886, Rev. 2. FAO Inland Water Resources and Aquaculture
Service, Rome, pp. 5–30. Available at: /http://www.fao.org/DOC-
FAO, 2006. GLOBEFISH. Fishmeal Market Report—June 2006. Avail-
able at: /http://www.globeﬁsh.org/index.php?id=3015S.
FIGIS, 2004. FAO Fisheries Global Information System. FIDI Dataset.
FAO. Available at: /http://www.fao.org/ﬁgis/servlet/static?dom=-
FishStat Plus, 2004. Universal Software for Fishery Statistical Time Series,
version 2.3.2000. FAO Fisheries Department, Fisheries Informations,
Data and Statistics Unit.
Folke, C., 2003. Freshwater for resilience: a shift in thinking. Philosophi-
cal Transactions of the Royal Society of London Series B—Biological
Sciences 358, 2027–2036.
Folke, C., Kautsky, N., 1989. The role of ecosystems for a sustainable
development of aquaculture. Ambio 18, 234–243.
Folke, C., Carpenter, S., Walker, B., Scheffer, M., Elmqvist, T.,
Gunderson, L., Holling, C., 2004. Regime shifts, resilience and
biodiversity in ecosystem management. Annual Review of Ecology,
Evolution and Systematics 35, 557–581.
Garcia, S., de Leiva Moreno, I., 2000. Trends in World Fisheries and their
Resources. The State of World Fisheries and Aquaculture. FAO,
Goldburg, R., Matthew, M., Elliott, S., Naylor, R., 2001. Marine
aquaculture in the United States: environmental impacts and policy
options. Available at: /http://www.pewoceans.org/reports/137PE-
WAquacultureF.pdfS, Pew Oceans Commission, Arlington, Virginia.
Grima, A., Berkes, F., 1989. Natural resources: access, rights to use and
management. In: Berkes, F. (Ed.), Common Property Resources.
Ecology and Community-Based Sustainable Development. Belhaven
Press, London, pp. 33–54.
Guttormsen, A., 2002. Input factor substitutability in salmon aquaculture.
Marine Resource Economics 17, 91–102.
Hardy, R.W., 1999. Collaborative opportunities between ﬁsh nutrition
and other disciplines in aquaculture: an overview. Aquaculture 177,
Hardy, R.W., Tacon, A.G.J., 2002. Fish meal: historical uses, production
trends and future outlook for sustainable supplies. In: Stickney, R.R.,
McVey, J.P. (Eds.), Responsible Marine Aquaculture. CAB Interna-
tional, Wallingford, pp. 311–325.
ARTICLE IN PRESS
L. Deutsch et al. / Global Environmental Change 17 (2007) 238–249248
Hites, R.A., Foran, J.A., Carpenter, D.O., Hamilton, M.C., Knuth, B.A.,
Schwager, S.J., 2004. Global assessment of organic contaminants in
farmed salmon. Science 303, 226–229.
Hjermann, D.O., Ottersen, G., Stenseth, N.C., 2004. Competition among
ﬁshermen and ﬁsh causes the collapse of Barents Sea capelin. PNAS
Holling, C., 1994. An ecologist’s view of the Malthusian conﬂict. In:
Lindahl-Kiessling, K., Landberg, H. (Eds.), Population, Economic
Development and the Environment. Oxford University Press, Oxford.
Holling, C., Meffe, G., 1996. Command and control and the pathology of
natural resource management. Conservation Biology 10, 2.
Huitric, M., Folke, C., Kautsky, N., 2002. Development and government
policies of the shrimp farming industry in Thailand in relation to
mangrove ecosystems. Ecological Economics 40, 441–455.
ICES, 2004. Report of the Baltic Fisheries Assessment Working Group.
Jackson, J.B.C., Kirby, M.X., Berger, W.H., Bjorndal, K.A., Botsford,
L.W., Bourque, B.J., Bradbury, R.H., Cooke, R., Erlandson, J., Estes,
J.A., Hughes, T.P., Kidwell, S., Lange, C.B., Lenihan, H.S., Pandolﬁ,
J.M., Peterson, C.H., Steneck, R.S., Tegner, M.J., Warner, R.R., 2001.
Historical overﬁshing and the recent collapse of coastal ecosystems.
Science 293, 629–638.
Knowlton, N., 2004. Multiple ‘‘stable’’ states and the conservation of
marine ecosystems. Progress in Oceanography 60, 387–396.
Lebel, L., Tri, N.H., Saengnoree, A., Pasong, S., Buatama, U., Thoa,
L.K., 2002. Industrial transformation and shrimp aquaculture in
Thailand and Vietnam: pathways to ecological, social, and economic
sustainability? Ambio 31, 311–323.
Matishov, G.G., Denisov, V.V., Dzhenyuk, S.L., Karamushko, O.V.,
Daler, D., 2004. The impact of ﬁsheries on the dynamics of commercial
ﬁsh species in Barents Sea and the Sea of Azov, Russia: a historical
perspective. Ambio 33, 63–67.
Myers, R.A., Barrowman, N.J., Hutchings, J.A., Rosenberg, A.A., 1995.
Population dynamics of exploited ﬁsh stocks at low population levels.
Science 269, 1106–1108.
Naylor, R.L., Goldburg, R.J., Mooney, H., Beveridge, M., Clay, J., Folke,
C., Kautsky, N., Lubchenco, J., Primavera, J., Williams, M., 1998.
Ecology—nature’s subsidies to shrimp and salmon farming. Science
Naylor, R.L., Goldburg, R.J., Primavera, J.H., Kautsky, N., Beveridge,
M.C.M., Clay, J., Folke, C., Lubchenco, J., Mooney, H., Troell, M.,
2000. Effect of aquaculture on world ﬁsh supplies. Nature 405,
Naylor, R.L., Eagle, J., Smith, W.L., 2003. A global industry with local
impacts. Environment 45, 18–39.
New, M.B., Wijkstro
¨m, U.N., 1990. Feed for thought: some observations
on aquaculture feed production in Asia. World Aquaculture 21
pp. 17–19, 22–23.
New, M.B., Wijkstro
¨m, U.N., 2002. Use of ﬁshmeal and ﬁsh oil in
aquafeeds: further thoughts on the ﬁshmeal trap. FAO Fisheries
Circular No. 975 FIPP/C975, Rome, 61pp.
NRC, 1999. Sustaining Marine Fisheries. National Academy Press,
National Research Council, Washington, DC.
Pauly, D., Christensen, V., Dalsgaard, J., Froese, R., Torres Jr., F., 1998.
Fishing down marine food webs. Science 279, 860–863.
Pauly, D., Christensen, V., Gue
´nette, S., Pitcher, T.J., Sumaila, U.R.,
Walters, C.J., Watson, R., Zeller, D., 2002. Towards sustainability in
world ﬁsheries. Nature 418, 689–695.
Pike, I.H., Barlow, S.M., 2002. Impact of ﬁsh farming on ﬁsh stocks.
Bordeaux Aquaculture and Environment Symposium. International
Fishmeal & Fish Oil Organisation. Available at: /http://www.iffo.or-
Primavera, J.H., 2000. Development and conservation of Philippine
mangroves: institutional issues. Ecological Economics 35, 91–106.
Seafeeds, 2003. SEAfeeds 2003 Final Workshop Report. Available at:
Sharp, G., 1995. It’s about time: new beginnings and old good ideas in
ﬁsheries science. Fisheries Oceanography 4, 324–341.
Steele, J., Hoaglund, P., 2003. Are ﬁsheries ‘‘sustainable’’. Fisheries
Research 64, 1–3.
Tacon, A.G.J., De Silva, S.S., 1997. Feed preparation and feed manage-
ment strategies within semi-intensive ﬁsh farming systems in the
tropics. Aquaculture 151, 379–404.
Tacon, A.J., 2002. Thematic Review of Feeds and Feed Management
Practices in Shrimp Aquaculture. Report prepared under the World
Bank, NACA, WWF and FAO Consortium Program on Shrimp
Farming and the Environment, Rome, 69pp.
Tacon, A.J., 2003a. Aquaculture production trends analysis. Review of the
State of World Aquaculture. FAO Fisheries Circular. No. 886, Rev. 2.
FAO Inland Water Resources and Aquaculture Service, Rome,
pp. 5–30. Available at: /http://www.fao.org/DOCREP/005/Y4490E/
Tacon, A.J., 2003b. Presentation: Sustainable aquaculture feeds: an
overview and global perspective in Seafeeds. SEAfeeds 2003 Work-
shop. Available at: /http://www.seafeeds.net/Files/Albert%20Ta-
Tacon, A.J., 2003c. Sustainable aquaculture feeds: an overview and global
perspective in Seafeeds. SEAfeeds 2003 Final Workshop Report.
Available at: /http://www.seafeeds.net/Files/Final%20Report.pdfS.
Tacon, A.J., 2004. Fish meal and ﬁsh oil use in aquaculture: a global
perspective. Aquatic Resources, Culture and Development 1, 3–14.
Tuominen, T.-R., Esmark, M., 2003. Food for Thought: the Use of
Marine Resources in Fish Feed. WWF-Norway, Norway, 52pp.
Watson, R., Pauly, D., 2001. Systematic distortions in world ﬁsheries
catch trends. Nature 414, 534–536.
Webster, C.D., Tiu, L.G., Morgan, A.M., Gannam, A., 1999. Effect of
partial and total replacement of ﬁsh meal on growth and body
composition of sunshine bass Morone chrysops x M. saxatilis fed
practical diets. Journal of the World Aquaculture Society 30, 443–453.
World_Trade_Organization, 2003. International Trade Statistics 2001.
WTO Publications, Geneva, Switzerland. Available at: /http://
Worm, B., Lotze, H.K., Myers, R.A., 2003. Predator diversity hotspots in
the blue ocean. PNAS 100, 9884–9888.
ARTICLE IN PRESS
L. Deutsch et al. / Global Environmental Change 17 (2007) 238–249 249