Blue Frontiers: Managing the Environmental Costs of Aquaculture

Abstract

A comprehensive analysis of global aquaculture production across all major species and farm production systems. This global review aims to inform policy makers about the impacts of aquaculture on the environment and to stimulate debate on the optimal animal food production systems for tomorrow.

Full text

REPORT
BLUE
FRONTIERS
Managing the
environmental
costs of aquaculture

iManaging the environmental costs of aquaculture
Cover photo:
Mindanao, Philippines. Fish Farming on Lake Sebu
Image by © Philippe Lissac /Godong/Corbis
An electronic version of this
publication can be downloaded at:
http://www.worldfi shcenter.org/global_aquaculture/
www.conservation.org/marine
Preferred citation:
Hall, S.J., A. Delaporte, M. J. Phillips, M. Beveridge and M. O’Keefe. 2011. Blue Frontiers: Managing the
Environmental Costs of Aquaculture. The WorldFish Center, Penang, Malaysia.
Blue Frontiers
Managing the environmental
costs of aquaculture
Stephen J. Hall,
Anne Delaporte,
Michael J. Phillips,
Malcolm Beveridge,
Mark O’Keefe.
Authors

ii Managing the environmental costs of aquaculture
There is a pressing need to elevate the debate on
the future of aquaculture and to place this in the
context of other animal food production systems,
including wild capture fi sheries. Between 1970
and 2008 aquaculture production grew at an
annual average rate of 8.4% and remains among
the fastest growing food production sectors in the
world. But with global demand for aquatic food
products continuing apace, there are worries about
the development trajectory of aquaculture. Of
particular concern for Conservation International
and many others is whether and how further
growth can be met in ways that do not erode
biodiversity or place unacceptable demands on
ecological services. In this context, the potential
for aquaculture to reduce pressure on wild capture
fi sheries by meeting global demand for aquatic
food products is also important.
Directed towards helping inform and stimulate
policy debate, this report provides a global review
and analysis of these issues for both coastal and
freshwater aquaculture. Such debate is needed to
help ensure that the current and future potential
benefi ts of the burgeoning aquaculture sector are
captured and the associated costs minimized.
The report begins with an overview of the current
status of world aquaculture. It then goes on to
describe an approach for estimating the current
combined biophysical resource demands of
aquaculture for producer countries and regions.
Following a comparison of these results with those
available for other animal food production sectors
the report then examines the consequences
of likely future trends in production on the
environmental impacts of aquaculture. Finally,
the policy implications of the report’s fi ndings
are discussed along with the research agenda
that should be pursued to meet the challenge of
sustainable food production.
Acknowledgements
This report has benefi ted greatly from critiques by
several colleagues. We are especially grateful to
Professor Max Troell, Mr Patrik Henriksson and Dr
Patrick Dugan and colleagues at the World Bank
and Conservation International for their insightful
comments. We would also like to thank Professor
Trond Bjorndal for help with part of the text.
About this Report

iiiManaging the environmental costs of aquaculture
Table of contents
About this report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii
Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii
Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv
Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
Acronyms and abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi
Units of measure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1. Aquaculture today: Production and Production Trends . . . . . . . . . . . . . . . . 8
2. Aquaculture production: Biophysical demands and
ecological impacts
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3. The environmental effi ciencies of animal production
systems: How does aquaculture compare?
. . . . . . . . . . . . . . . . . . . . . . . . . 44
4. Looking Forward . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
5. Policy Implications and Recommendations. . . . . . . . . . . . . . . . . . . . . . . . . 70
Appendix. Systems modelled in this study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
Table of contents

iv Managing the environmental costs of aquaculture
1 Table 1.1: Food production statistics for major commodities. . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2 Table 1.2 The export value of selected agricultural commodities in 2007 . . . . . . . . . . . . . . . . . . . . 9
3 Table 1.3: The relative importance of aquaculture in global fi sh production per species group. . . . . . . . . . 13
4 Table 2.1: The generic species group - production systems used to assess environmental demands.. . . . . . 17
5 Table 2.2: The production intensity categories used in this analysis . . . . . . . . . . . . . . . . . . . . . . . 17
6 Table 2.3: The feed types used in this analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
7 Table 2.4: Summary of approaches to quantifying environmental impact. . . . . . . . . . . . . . . . . . . . . 20
8 Table 2.5: Parameter estimates and data sources for foreground data calculations. . . . . . . . . . . . . . . . 26
9 Table 2.6: (a) Total estimated impacts from the 75 production systems modeled in this study and an
estimate of the complete global impact assuming that, as with total aquaculture production, each
calculated estimate represents 82% of the total. (b) Sectoral comparison of CO2 emissions . . . . . . . . . . 28
10 Table 2.7: Summary of the models used to examine sensitivity relative to baseline results. . . . . . . . . . . . 36
11 Table 2.8: Comparison of results from other published studies. . . . . . . . . . . . . . . . . . . . . . . . . . 39
12 Table 3.1: Protein content of major animal foods and feed conversion effi ciencies for their production. . . . . . 45
13 Table 3.2: Percentage of world fi shmeal market use by sector.. . . . . . . . . . . . . . . . . . . . . . . . . . 46
14 Table 3.3: Summary of data on nitrogen and phosphorus emissions for animal production systems. . . . . . . 47
15 Table 3.4: Estimates of land demand (direct and indirect) for animal-source food production. . . . . . . . . . . 47
16 Table 4.1: Projected change in total environmental impact between 2008 and 2030 for the systems
modeled in this study, which produced 82% of world production in 2008 (data exclude seaweeds, and
assumes current production practices). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
17 Table 5.1: Recommendations summarized for key stakeholder groups. . . . . . . . . . . . . . . . . . . . . . 74
Tables

vManaging the environmental costs of aquaculture
Figures
1 Figure 1.1: World aquaculture production by continent in 2008 (China treated separately). . . . . . . . . . . . 10
2 Figure 1.2: Summary of 2007 aquaculture production by region. . . . . . . . . . . . . . . . . . . . . . . . . 11
3 Figure 1.3: Treemaps summarizing 2008 production by species group for each continent. . . . . . . . . . . . 12
4 Figure 2.1: Graphical summary of the system boundaries and model structure for the
Life Cycle Analyses undertaken in this study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
5 Figure 2.2: The relationship between overall production levels for each of the 75 unique
production combinations and level of impact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
6 Figure 2.3: Upper panel: The absolute environmental impact of 2008 aquaculture production
categorized by production system and habitat. Lower panel: The relative environmental impact,
per tonne of product categorized by production system and habitat . . . . . . . . . . . . . . . . . . . . . . 30
7 Figure 2.4: Upper panel: The absolute environmental impact of 2008 aquaculture production
categorized by species group. Lower panel: The relative environmental impact per tonne of product
categorize by species. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
8 Figure 2.5: The relative environmental impact of 2008 aquaculture production categorized
by habitat, production system and species group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
9 Figure 2.6: Maps showing the absolute size of total environmental impacts of 2008 production
for each of the 18 countries analyzed in this study. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
10 Figure 2.7: Maps showing the relative size of environmental effi ciencies (average environmental
impacts per tonne of production) for each of the 18 countries analyzed in this study. . . . . . . . . . . . . . .33
11 Figure 2.8: A comparison of environmental effi ciencies across countries growing the same
species group. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
12 Figure 2.9: The total proportional contribution to impact of the fi ve main processes
for each species group. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
13 Figure 2.10: Summary of sensitivity analysis results.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
14 Figure 4.1: The urban populations of countries in 2009 and the projected annual average
rate of growth in urbanization to 2050. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
15 Figure 4.2: The rise and decline of antibiotic use in the Norwegian salmon industry
compared to the trend of rising production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
16 Figure 4.3: Summary of non-native species production for the systems modeled in this study. . . . . . . . . . 58
17 Figure 4.4: The relationship between aquaculture and climate change. . . . . . . . . . . . . . . . . . . . . . 59
18 Figure 4.5: Comparison of historical trends in production of farmed fi sh with several projections
of future aquaculture production. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
19 Figure 4.6: Comparison of historical trends in farmed fi sh, pig and chicken meat production, the likely
production trajectory envelope and the combined aquaculture production targets envelope for nine
countries (Bangladesh, India, China, Indonesia, Philippines, Thailand, Vietnam, Brazil, Chile,
Canada, Egypt). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
20 Figure 4.7: Projected change in production distribution between 2008 and 2030 for the
systems modeled in this study, which produced 82% of world production in 2008 . . . . . . . . . . . . . . .
(data exclude seaweeds). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
21 Figure 4.8: Projected change in distribution of environmental impact between 2008 and 2030
for the systems modeled in this study (data exclude seaweeds). . . . . . . . . . . . . . . . . . . . . . . . . 66
22 Figure 5.1: Core recommendations for government and industry in all producer countries
and their relative importance for each region. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

vi Managing the environmental costs of aquaculture
Acronyms and
abbreviations
ARD Agriculture and Rural Development
CED Cumulative energy demand
CML Institute of Environmental Sciences
EU European Union
FAO Food and Agriculture Organization
FCR Feed Conversion Ratio
GM Genetically Modifi ed
ICES International Council for the Exploration of the Sea
IFPRI International Food policy Research Institute
IPCC Intergovernmental Panel on Climate Change
IUCN International Union for the Conservation of Nature
LCA Life Cycle Analysis
N Nitrogen
NGO Non-Governmental organization
OECD Organisation for Economic Co-operation and Development
OIE World Organization for Animal Health (formely Offi ce International des Epizooties)
P Phosphorus
RAS Recirculation Aquaculture Systems
TSP Triple Super Phosphate
USFDA U.S Food and Drug Administration
WWF World Wide Fund for Nature

viiManaging the environmental costs of aquaculture
Units of measure
ha hectare
Gj Giga joule
kg kilogram
Mj mega joule
m
3
cubic meter
t
metric ton (1000 kg)
US$ U.S dollar
yr year

viii Managing the environmental costs of aquaculture
Executive Summary

1Managing the environmental costs of aquaculture
EXECUTIVE SUMMARY
PHOTO CREDIT: He Qing Yunnan

TodayImpactsComparisonLooking ForwardPolicyAppendixGlossaryReferences
Summary
Executive Summary
2 Managing the environmental costs of aquaculture
Aquaculture is among the fastest growing food
production sectors in the world and this trend
is set to continue. However, with increasing
production comes increasing environmental
impact. For aquaculture to remain sustainable this
future growth must be met in ways that do not
erode natural biodiversity or place unacceptable
demands on ecological services.
This study is a review and analysis of global
aquaculture production across the major
species and production systems. It compares
the aggregate biophysical resource demands of
each system and their cumulative environmental
impacts. The study then compares these results
with those from other animal food production
systems before examining the consequences of
likely future trends. Finally, the policy implications
of the report’s fi ndings are discussed along with
the research agenda that should be pursued to
meet the challenges involved in producing food
sustainably.
Worldwide, aquaculture production has grown at
an average annual rate of 8.4% since 1970 and
reached 65.8 million tonnes in 2008. The growth
in farmed fi sh supply has signifi cantly outpaced
growth in world population. China supplies
61.5% of global aquaculture production; a further
29.5% comes from the rest of Asia, 3.6% from
Europe, 2.2% from South America, 1.5% from
North America, 1.4% from Africa and 0.3% from
Oceania. Production in China and the rest of Asia
is predominantly freshwater, from other continents
predominantly coastal. The annual average growth
rate in aquaculture between 2003 and 2005 in
North America and Europe is slow (1.4–1.6%); it is
rapid in China, Asia and South America (6, 11.2,
7.8% respectively) and explosive in Africa (16.2%),
albeit from a very low baseline.
Carp dominates production in both China and
the rest of Asia. In contrast, for Europe and South
America it is salmonids; African aquaculture
production is almost exclusively of fi nfi sh, primarily
tilapias. For Oceania, shrimps and prawns
dominate while in North America production is
more even across the species groups. Aquaculture
has growing signifi cance as a supplier of fi sh;
between 2003 and 2008 the proportion of
aquaculture in total fi sh production (i.e. for food and
industrial purposes) increased from 34 % to 42%.
The proportion of food fi sh supplied by aquaculture
in 2008 was 47%. Supply from aquaculture is now
dominant for seaweeds, carps and salmonids.
The rapid growth of aquaculture witnessed
over the last forty years has raised questions
concerning its environmental sustainability. To
answer those questions satisfactorily requires
quantitative analyses. This study, based on
2008 data, compares the global and regional
demands of aquaculture for a range of biophysical
resources across the dominant suite of species
and production systems in use today. The units
of analysis were the elements of a six dimensional
matrix comprising 13 species groups, 18 countries,
3 production intensities, 4 production systems, 2
habitats and 5 feed types. This gave 75 positive
matrix elements that accounted for 82% of
estimated total world aquaculture production in
that year.
The assessment method chosen to analyse the
data was Life Cycle Analysis (LCA). This method
required estimates of both the biophysical
resource inputs to and outputs from each of the
75 species-production systems identifi ed. The
input resources estimated were the amount of
land, water, feed, fertilizers and energy required
on-farm. The outputs (emissions) considered
were nitrogen, phosphorus and carbon dioxide.
From these data the LCA produced estimates of
the impact of these species-production systems
for each of six impact categories: eutrophication,
acidifi cation, climate change, cumulative energy
demand, land occupation and biotic depletion (use
of fi sh for fi shmeal and fi sh oil). Boundaries were
set to exclude environmental costs associated with
building infrastructure, seed production, packaging
and processing of produce, transport and other
factors.
Executive Summary

Executive summary
Today Impacts Comparison Looking Forward Policy Appendix Glossary References
Summary
3Managing the environmental costs of aquaculture
Sensitivity analyses were run to determine the
robustness of the fi ndings, and comparisons
were made with other LCA studies. Although
most variations tested gave results that differed
little from the model in use, some notable
deviations occurred. Most of these were related to
assumptions associated with on-farm energy use
and feed supply indicating that improved data in
these areas are required.
There is a growing demand for animal source
foods, driven partly by population growth but
mainly by rising standards of living and prosperity
in developing countries. The study continues with
a comparison of the environmental impacts of
aquaculture with those from other animal food
production sectors. This is important because
without a balanced picture of the environmental
impacts of producing animal source foods through
different systems, it is not possible for governments
or consumers to understand the true costs of
production.
The comparative analysis draws heavily on studies
of the environmental impact of livestock produced
by the FAO and considers four key aspects:
conversion effi ciencies, environmental emissions
(nitrogen, phosphorus and carbon dioxide), land
use and water use.
Fish convert a greater proportion of the food they
eat into body mass than livestock and therefore
the environmental demands per unit biomass or
protein produced are lower. The production of 1
kg of fi nfi sh protein requires less than 13.5 kg of
grain compared to 61.1 kg of grain for beef protein
and 38 kg for pork protein. However, although
farmed fi sh may convert food more effi ciently than
livestock there are important issues with respect
to carnivorous fi sh species, which place heavy
demands on the fi shmeal and fi sh oil industry—
the use of capture fi sheries for animal feeds.
Unfortunately, simply substituting a vegetable-
based food for fi shmeal is often not possible at
present.
Overall, and unsurprisingly, the data from the 75
species-production systems reviewed showed a
positive relationship between overall production
levels and impact. The levels of impact were then
compared across production system, species
group and country.
Inland pond culture is the predominant production
system and it contributes the greatest impact
across all the six impact categories, with demand
for wild fi sh (biotic depletion) also notable for
marine cage and pen culture. Similarly carps, as a
species group, dominate overall impacts refl ecting
the fact that carp production is greater than that of
other species groups. Eel production stands out
as highly environmentally demanding, largely due
to high energy consumption, and salmonid, and
shrimp and prawn production are notable for their
demand for wild fi sh. Bivalves and seaweeds place
low demands on the environment and actually
reduce eutrophication.
A comparison of environmental effi ciencies across
countries gave a variable picture. For example,
for the salmon producing nations of north Europe,
Canada and Chile, the impact from eutrophication
was moderate and biotic depletion high, but they
were more effi cient than China and Asia across
the other four environmental impacts. Perhaps
more interestingly however, were the differences in
effi ciencies within species-production categories
between countries suggesting scope for improving
environmental performance. For shrimp and prawn
culture, for example, China is much less effi cient, in
relative terms, than other producer countries when
considering impact on acidifi cation, climate change
and energy demand.
A look at the drivers of impact, i.e. those attributes
of the production system that contribute most to
environmental impact, showed that the aquaculture
production system itself contributed most to
eutrophication, but impacts on climate change and
acidifi cation were dependent on the nature of the
national energy supply; a factor outside the control
of the local operator.

TodayImpactsComparisonLooking ForwardPolicyAppendixGlossaryReferences
Summary
Executive Summary
4 Managing the environmental costs of aquaculture
Extensive livestock production places heavy
demands on land use through deforestation and
land degradation. However, land use demands
per unit of protein production appear broadly
similar across other animal food production
systems. Intensive livestock production is
noteworthy, however, for the high levels of nitrogen,
phosphorus, carbon dioxide and methane
produced. Comparatively, aquaculture systems
perform well with respect to the emissions
produced from beef and pork production.
Livestock rearing, especially in intensive systems,
also places heavier demands on the use of fresh
water.
There are, however, a number of issues concerning
the calculations which make true comparisons
diffi cult and there are insuffi cient data to properly
compare the different intensities and methods
of animal production, so the results must be
viewed as ‘broad-brush’. Certainly there are some
effi ciencies associated with farming a product that
is cold blooded and feeds near the bottom of the
food chain but much depends on the species,
production system and management used. And
there are trade-offs between extensive systems
that place higher demands on land use, and
ecological services such as water, fuel, nutrient
cycling, and intensive systems that require higher
levels of fossil fuels, feed, and produce more
effl uent.
In the fourth section the authors briefl y review the
drivers of demand and environmental constraints
to aquaculture production, along with published
predictions of future trends for the aquaculture
sector. Driven largely by increasing wealth and
urbanization, published estimates suggest
production will reach between 65 and 85 million
tonnes by 2020 and between 79 and 110 million
tonnes by 2030. As an illustration of the potential
environmental impact of this growth, in the absence
of signifi cant innovations and improvements in
management and technology, a production level of
100 million tonnes by 2030 (excluding seaweeds)
will lead to environmental demands that will be
between 2 and 2.5 times greater than 2008 levels
for all the impact categories studied.
A number of key conclusions and
recommendations arise from the analysis, and
point the way towards improved productivity for
aquaculture with reduced environmental impact.
These include the following points.
• As the degree of environmental impact is
largely determined by the level of production,
with carp production from inland ponds
in China and Asia creating the largest
environmental footprint, this is an important
fi eld where research needs to be undertaken
to develop measures to reduce overall
environmental impact.
• The variety in impact measured by the same
species-production system operating in
different countries suggests strongly that
the potential to improve performance exists,
such as through regional learning networks
for both policies and technologies. Much
of the aquaculture industry in developing
countries provides opportunities for improved
effi ciencies.
• Feed constraints are key to aquaculture
development. Reducing the dependency
on fi shmeal and fi sh oil will require new
innovations in technologies and management
but the payoffs may be spectacular both in
terms of profi tability, food and nutrition security
and reduced environmental impact.
• Analysis shows that reductions can be made
to the sector’s impact on both climate change
and acidifi cation by improving energy effi ciency
throughout the production and value chains.
The use of water and energy audits and better
practices should lead to reduced resource
demands.
• It is apparent from this study that aquaculture
has, from an ecological effi ciency and
environmental impact perspective, clear
benefi ts over other forms of animal source food
production for human consumption. In view of
this, where resources are stretched, the relative
benefi ts of policies that promote fi sh farming
over other forms of livestock production should
be considered.

Executive summary
Today Impacts Comparison Looking Forward Policy Appendix Glossary References
Summary
5Managing the environmental costs of aquaculture
• The growing need for aquaculture to contribute
to food security, especially in African and Asian
countries will require governments to actively
support growth of the sector and stimulate
private sector investment.
• Aquaculture affects climate change and
climate change will affect aquaculture. To
minimise the potential for climate change,
energy consumption should be kept as low
as possible and new aquaculture enterprises
should not be located in regions that are
already high in sequestered carbon such as
mangroves, seagrass or forest areas.
• There are measures that policy makers can
take which include providing support to
innovative and technological developments,
ensuring a suitable regulatory framework
that captures environmental costs within
aquaculture processes, building capacity for
monitoring and compliance, and encouraging
research on the supply and demand for fi sh
and fi sh products.
This study is the fi rst to provide a global picture of
the demands fi sh farming makes on environmental
resources using Life Cycle Analysis. It illustrates
the opportunities and challenges that lie ahead for
aquaculture. The key messages for policy makers,
NGOs, entrepreneurs and researchers are that
there must be a wider exchange of knowledge and
technology, with policies and action to promote
sustainability and investment in research to fi ll the
knowledge gaps. These efforts can lead to a more
ecologically sustainable industry—an important
goal, given the likely rapid growth in aquaculture
production. They will also help ensure that
aquaculture contributes fully to meeting our future
needs for fi sh.

6 Managing the environmental costs of aquaculture
1. Aquaculture Today

7
1. AQUACULTURE TODAY
PHOTO CREDIT: The WorldFish Center

1. Aquaculture Today
Today
ImpactsComparisonLooking ForwardPolicyAppendixGlossaryReferences Summary
8 Managing the environmental costs of aquaculture
Aquaculture production in
context
For several decades aquaculture has been the
fastest growing food production sector in the world.
Five year averages for global production increases in
major food commodities rank aquaculture number
one for every period since 1974. Worldwide,
aquaculture production has grown at an average
annual rate of 8.4%, since 1970 (Table 1.1). With
poultry showing the next largest rate of increase over
this period at 5%, aquaculture’s dynamism stands
out clearly.
This rate of production growth has ensured that, as
a global average, farmed fi sh supply has outpaced
population growth. From a per capita value of 0.7
kg in 1970, global supply of farmed fi sh rose to 7.8
kg in 2006. The estimated average per capita fi sh
consumption for wild and farmed combined was
16.8 kg in 2006, indicating that about 47% of fi sh for
human consumption was supplied by aquaculture
at that time. Given the unlikely prospect of increased
yields from wild capture fi sheries, this value will
increase as aquaculture production grows.
1. Aquaculture Today:
Production and Production
Trends
Average annual
production increase
(1970–2008)
Average annual
production increase
(2004–2008)
2008 Production
(tonnes x 1000)
Plant Food Commodities
Cereals 2.1% 3.9% 2,525,107
Pulses 1.1% 0.6% 60,929
Roots and Tubers 0.9% 0.9% 729,583
Vegetables and Melons 3.4% 1.7% 916,102
Animal Food Commodities
Beef and Buffalo 1.3% 1.6% 65,722
Eggs 3.2% 2.2% 65,586
Milk 1.5% 2.4% 693,707
Poultry 5.0% 3.9% 91,699
Sheep and Goats 1.8% 2.4% 13,174
Fish
8.4% 6.2% 52,568
Table 1.1: Food production statistics for major commodities. (Source: FAOStat and FishStat)

1. Aquaculture Today
Today
Impacts Comparison Looking Forward Policy Appendix Glossary ReferencesSummary
9Managing the environmental costs of aquaculture
Fish is also pre-eminent as an internationally traded
animal source food. Representing about 10% of
total exports of agricultural products by value,
seafood exports from wild fi sheries and aquaculture
in 2008 had a combined value of US$102 billion
(FAO, 2010), an 83% increase from 2000. The share
of exports from developing countries is close to
50% by value and 60% by volume. Of internationally
traded agricultural commodities seafood export
value is exceeded only by fruits and vegetables
(Table 1.2). The European Union is the world’s
largest seafood importer, followed by the United
States and Japan.
Table 1.2: The export value of selected agricultural
commodities in 2007. (Source: FAOStat and FAO
TradeStat 2007)
Unfortunately national trade statistics do not
distinguish between aquaculture and wild capture
as the source of imports. It is, therefore, diffi cult
to draw fi rm conclusions at a global level about
the proportion of total international fi sh trade
volume that aquaculture provides. A 2006 estimate
for China, however, was that 39% by volume
and 49% by value of the country’s aquaculture
production was exported (Fang, 2007). A high level
of international trade in aquaculture products is
important because it offers a potentially powerful
entry point for harmonizing and improving
environmental standards of production.Several
recent reviews of global aquaculture production are
readily available (e.g., Muir et al., 2009; Bostock et
al., 2010), and the FAO provides biannual updates in
its Status of Fisheries and Aquaculture series (FAO,
2009b). We have built on these to offer a concise
global overview of current aquaculture production
that helps put into context the analyses and results
that follow. It also serves to introduce the reader
to the data categorization approach we used for
analyses described later in the report.
Using FAO data
1
, our starting point is the overall
global picture (Figure 1.1). This fi gure summarizes
how the world’s total aquaculture production of
65.8 million tonnes in 2008 was distributed across
continents by adjusting continental areas to refl ect
production volume. Following convention, we have
treated China separately from the rest of Asia—a
decision that is clearly appropriate given its pre-
eminence as a producer.
With 61.5% of global production (40,508,119
tonnes) China deserves special attention. The further
29.5% of global production (19,401,808 tonnes)
supplied by the rest of Asia places the continent as
a whole in an overwhelmingly dominant position. By
contrast, production in Europe with 3.6% (2,341,646
tonnes), South America with 2.2% (1,461,061
tonnes), North America with 1.5% (965,792 tonnes),
Africa with 1.4% (952,133 tonnes) and Oceania with
0.3% (176,181 tonnes) is trivial in overall terms.
Trade Value US$ billions
2007
Plant Commodities
Fruit and Vegetables 150.89
Wheat 36.40
Tobacco 29.06
Sugar 18.58
Coffee 17.67
Rice 13.48
Pulses 4.82
Animal commodities
Fish 92.80
Pigs 30.21
Cattle 28.99
Poultry 22.10
Sheep and Goats 4.35
1
All data are from FAO FishStat unless otherwise stated.

1. Aquaculture Today
Today
ImpactsComparisonLooking ForwardPolicyAppendixGlossaryReferences Summary
10 Managing the environmental costs of aquaculture
Figure 1.1: World aquaculture production by continent in 2008 (China treated separately). Land areas are adjusted
proportionally to refl ect production volumes.
But, despite the overall dominance of Asia,
aquaculture is an important economic activity on
most continents and its importance is growing
almost everywhere. To illustrate how production
is distributed within regions Figure 1.2 lists the
countries that account for at least 90% of production
on each continent. Production is spread most widely
among countries in Europe and Asia with over 90%
of production accounted for by 11 and 9 countries,
respectively. In contrast, most African and South
American production is accounted for by only three
countries on each continent.
Figure 1.2 also shows how production is distributed
in each country between coastal
2
and freshwater
systems. Overall, 60% of global production occurs
in freshwater. China and the rest of Asia contribute
most to this average value, producing over 59 and
64% in freshwater, respectively. In contrast, coastal
production dominates in South America, Europe and
Oceania with respective values of 78, 80 and 98%
from coastal areas. Production in North America is
almost evenly split between coastal and freshwater
habitats, while FAO reports there is a 60:40 split
between coastal and freshwater in Africa. This picture
is dominated by production from Egypt, which
accounts for 73% of total aquaculture production
on the continent. Data for Egypt are somewhat
misleading, however, because although the FAO
classifi es the majority of production as coming from
brackishwater, almost all of this is from very low
salinity ponds in the Nile Delta.
2
For this analysis we combined data classifi ed in the FAOStat database for brackishwater and marine production into a single coastal production category.
Continent Production 2008 Proportion
China
Asia
Europe
South America
North America
Africa
Oceania
40,508,119 61.5
19,401,808 29.5
2,341,646 3.6
1,461,061 2.2
965,792 1.5
952,133 1.4
176,181 0.3
China
South America
Asia
(excluding China)
North America
Africa
Europe
Oceania
Korea
Japan

1. Aquaculture Today
Today
Impacts Comparison Looking Forward Policy Appendix Glossary ReferencesSummary
11Managing the environmental costs of aquaculture
Freshwater : Coastal
production by Region
FW = Production in Fresh Water
C = Production in Coastal areas
Region
Production
2008 (T) FW:C
USA 500,114
Mexico 151,065
Canada 144,099
Honduras 47,080
Cuba 33,039
N. America 965,791
Region
Production
2008 (T) FW:C
Chile 849,159
Brazil 290,186
Ecuador 172,120
S.America 1,445,392
Region
Production
2008 (T) FW:C
Norway 843,730
Spain 249,062
France 237,833
Italy 181,469
UK 179,187
Russia 115,420
Greece 114,888
Ireland 57,210
Netherlands 46,622
Faroe Islands 45,929
Germany 43,977
Europe 2,341,339
Region
Production
2008 (T) FW:C
Egypt 693,815
Nigeria 143,207
Uganda 52,250
Africa 942,044
Region
Production
2008 (T) FW:C
India 3,478,692
Vietnam 2,461,700
Indonesia 1,709,783
Thailand 1,374,024
Bangladesh 1,005,542
Japan 748,474
Philippines 741,142
Myanmar 674,776
Korea,
Republic of
477,389
Asia 13,677,725
Region
Production
2008 (T) FW:C
China 35,233,199
Taiwan 323,982
Hong Kong 4,754
China 35,561,935
Region
Production
2008 (T) FW:C
New Zealand 112,358
Australia 57,152
Oceania 174,115
Korea
Japan
Europe
Asia
Africa
North America
China
Oceania
South America
To summarize the distribution of production with
respect to species we have constructed treemaps
that show the relative proportion of production by
continent for each of 12 species groups (excluding
seaweed, Figure 1.3). These maps show how carp
dominates production in both China and the rest
of Asia. In contrast, for Europe and South America
salmonids dominate and account for more than
70% of worldwide salmonid production (capture
and culture). African aquaculture production is
almost exclusively of fi nfi sh, of which tilapias are the
most important. For Oceania, shrimps and prawns
dominate while in North America the pattern of
production is somewhat more evenly distributed
among species with shrimps and prawns, catfi sh,
bivalves and salmonids accounting for the majority.
Rates of change in production (indicated by color
in Figure 1.3) show several patterns. The fi rst is that
China and Asia continue to grow apace. Overall
growth rates were 30% and 56% over fi ve years,
respectively. Growth in Oceania at 37% and South
America at 39% is also high. The continent with the
highest growth rate over the period, however, was
Africa at 81%. Admittedly, this growth was from a
very low baseline, but these “blue shoots” provide
an indication that Africa may be poised for further
dramatic production increases. In contrast, growth
patterns in Europe and North America were the
lowest at 8% and 7%, respectively.
The second is the explosive growth of catfi sh
culture in Asia (307%) and Africa (496%) during
the period. Albeit from a low base, these fi gures
show how quickly a sub-sector can develop. While
not so spectacular, growth for many other species
groups is also high. In Asia, for example, tilapia
production increased by 121%, carp production
by 67% and shrimps and prawns by 53% over
the fi ve year period. Similarly large growth rates
for several species groups can be found on all
continents.
Another feature of these production growth data
is that the only regions where production changes
were positive for all species groups cultured were
China and Oceania. In contrast, the rest of Asia
saw declines for bivalves and the “other fi nfi sh”
category, Europe for bivalve and carps and North
America for catfi sh, carps and salmonids. Declines
in Africa and South America were restricted to
groups that contribute relatively little to the total
continental production.
Figure 1.2: Summary of 2007 aquaculture production by region.

1. Aquaculture Today
Today
ImpactsComparisonLooking ForwardPolicyAppendixGlossaryReferences Summary
12 Managing the environmental costs of aquaculture
There are also signifi cant differences in the relative importance of various species groups for wild capture and
farmed fi sh production. Table 1.3 shows that between 2003 and 2008 the proportion of aquaculture in total
fi sh production (i.e., for food and industrial purposes) increased from 34% to 42%. Supply from aquaculture
is now dominant for seaweeds (99.5%) carps (89.9%) and salmonids (72.8%). At around 50% of total supply,
cultured tilapia, catfi sh, mollusks, crabs and lobsters are now reaching prominence. This is especially true
of tilapia and catfi sh where aquaculture production has increased dramatically against a backdrop of almost
stagnation in wild capture. As a result, the share of production of farmed catfi sh and tilapia rose by 19.3 and
18.4%, respectively.
Carps
China
= 500,000 Tonnes
Asia Europe
South America
North America
Africa Oceania
Carps
Carps
Carps
Carps
Carps
Bivalves
Bivalves
Bivalves
Bivalves
Bivalves
Bivalves
Other
Finsh
Other
Finsh
Other
Finsh
Other
Finsh
Other
Finsh
Other
Finsh
Other
Finsh
Shrimps
& Prawns
Shrimps
&
Prawns
Shrimps
&
Prawns
Shrimps
& Prawns
Shrimps
&
Prawns
Shrimps
&
Prawns
Catsh
Catsh
Catsh
Tilapias
Salmonids
Eels
Tilapias
Catsh
Eels
Gastropods
Crabs &
Lobsters
Tilapias
Tilapias
Tilapias
OI
OV
Salmonids
Salmonids
Salmonids
Salmonids
-10-0
0.1-10
10.1-20
20.1-40
40.1-70
70.1-90
90.1-100
OI =
Other Invertebrates
OV=Other Vertebrates
Annual
growth rate
(%)
= 500,000 Tonnes
= 100,000 Tonnes
= 25,000 Tonnes
= 25,000 Tonnes
= 5,000 Tonnes
= 50,000 Tonnes
Figure 1.3: Treemaps summarizing 2008 production by species group for each continent (excluding seaweed). The area
for each species in a map is proportional to the tonnage produced (Note differing scale for each map). The color of
each block indicates the rate of increase between 2003 and 2008.

1. Aquaculture Today
Today
Impacts Comparison Looking Forward Policy Appendix Glossary ReferencesSummary
13Managing the environmental costs of aquaculture
Table 1.3: The relative importance of aquaculture in global fi sh production per species group.
(Source: FAO FishStat)
Capture production (Mt) Aquaculture production
(Mt)
Proportion of total production from
aquaculture (%)
Species Group 2003 2008 2003 2008 2003 2008 Difference
Carps 2.02 2.21 15.04 19.72 88.2 89.9 1.8
Catfi sh 2.33 2.77 1.03 2.78 30.8 50.1 19.3
Tilapias 3.95 3.14 1.59 2.80 28.6 47.1 18.4
Eels 0.65 0.62 0.32 0.48 32.9 43.4 10.5
Salmonids 1.16 0.84 1.85 2.26 61.5 72.8 11.3
Other Finfi sh 50.81 51.79 4.40 5.79 8.0 10.0 2.1
Bivalves 18.43 19.72 11.06 12.65 37.5 39.1 1.6
Gastropods 0.30 0.32 0.21 0.37 41.4 53.7 12.3
Crabs and Lobsters 0.93 0.78 0.49 0.76 34.4 49.4 15.0
Shrimps and Prawns 8.85 8.47 2.59 4.35 22.7 33.9 11.3
Other Invertebrates 1.14 1.18 0.12 0.31 9.7 20.5 10.8
Seaweeds 0.34 0.07 9.02 13.24 96.3 99.5 3.1
TOTAL 91.31 92.3 47.9 65.81 34.4 41.6 7.2
Conclusion
This brief overview highlights several key features of the aquaculture sector: high overall growth in
production, rapid emergence of species that meet market demand (e.g., striped catfi sh (Pangasianodon
hypophthalmus) from Vietnam), growing signifi cance as a supplier of food fi sh, and dominance by China.
But growth in production has not come without environmental cost. In the next section we examine how
these costs compare across the sector.

14 Managing the environmental costs of aquaculture
2. Impacts

1515
2. IMPACTS
PHOTO CREDIT: The WorldFish Center

2. Impacts
Today
Impacts
ComparisonLooking ForwardPolicyAppendixGlossaryReferences Summary
16 Managing the environmental costs of aquaculture
Preliminary data analysis
We have based our assessment of environmental
demands on the 2008 estimates of aquaculture
production summarized in Section 1. To produce
a manageable data set for analysis, however,
some data reduction and aggregation of the full
disaggregated data set was necessary. This was
achieved using the following steps. First, we
identifi ed those species, excluding seaweeds,
which cumulatively accounted for 90% of total
world production. This list comprised 71 species.
Extracting records for these species revealed that
29 countries contributed to this total. Using this
data set, each of the individual species was then
allocated to one of twelve separate species groups.
Production for a given species by a given country
was then further categorized into one of four
separate production systems, resulting in 16 species
group—production system combinations (Table
2.1). For each production system we made a further
distinction between production in inland (freshwater)
and coastal (marine and brackishwater) habitat,
recognizing that some production systems are used
in both (Table 2.1).
The rapid growth of aquaculture described in the
previous section raises questions concerning
the environmental sustainability of future industry
growth. Central to these concerns is the demand
aquaculture places on biophysical resources.
Unsustainable consumption of these resources will
ultimately undermine the productivity of the sector
and bring it into competition for resources with other
sectors (Gowing et al., 2006; Primavera, 2006).
Balanced against these concerns is the fact that
farming aquatic animals that feed low in the food
chain can be an ecologically effi cient means
of producing animal proteins. Some forms of
aquaculture can also help mitigate environmental
impacts. For example seaweed and mollusk farming
are known to mitigate the effects of eutrophication
(Troell et al., 1999; Neori et al., 2004; Nellemann et
al., 2009).
To better understand the effects of aquaculture on
the environment and its demands on biophysical
resources, we need quantitative analyses. These are
needed at several scales, from detailed studies for
production of a particular species through to larger
scale studies across regions and species-production
systems. This study focuses on the larger scale,
comparing and contrasting the global and regional
environmental demands of aquaculture for a range
of biophysical resources across the dominant suite
of species and production systems in use today. It
then goes on to examine their ecological impacts.
This section describes our approach for achieving
this.
2. Aquaculture production:
Biophysical demands and
ecological impacts

2. Impacts
Today
Impacts
Comparison Looking Forward Policy Appendix Glossary ReferencesSummary
17Managing the environmental costs of aquaculture
From the resulting data set we then extracted
the species-country production records that
cumulatively accounted for 90% of the production
for each species group. To this we added the
records accounting for 90% of seaweed production,
all of which we classifi ed as off-bottom marine
culture.
In total, these combined records accounted for just
over 82% of total world aquaculture production
in 2008 and reduced the number of countries in
our data set to 18. Further data reduction was
then achieved by summing production within each
unique species group, country, production system
and habitat combination.
For the relevant production systems (e.g., coastal
pond culture) we also considered the intensity
of production for each species group—country
combination in our data set. This is important
because intensity of production determines the
amount and type of feed and fertilizer regime
required and the consequent level of emissions
(Table 2.2).
Species Group Bottom Culture Off-Bottom Culture Cages & Pens Ponds
Bivalves
3c 3c 3ci
Carps
3i
Catfi sh
3i
Crabs and Lobsters
3c 3c
Eels
3i
Gastropods
3ci
Other Finfi sh
3ci 3ci
Other Invertebrates
3ci
Other Vertebrates
3i
Salmonids
3c
Shrimps and Prawns
3ci
Tilapias
3ci
Table 2.1:
The generic species group—production systems used to assess environmental demands. The subscript
c denotes a coastal system and i denotes an inland (freshwater) system. ci indicates that the system occurs in both
inland and coastal systems. (Note: Although carps are also cultured in cages and pens, this accounts for a small
proportion of production and has, therefore, been omitted).
Production Intensity Description
Extensive
Systems requiring large areas of earthen ponds or water area; primarily for
species in the fi nfi sh, mollusk, seaweeds, and shrimps and prawns species
groups. Extensive production relies on natural productivity, but in ponds it is often
supplemented by locally available crop wastes and other material. Little or no
processed feed is used.
Semi-intensive
Primarily freshwater but also some coastal earthen pond systems in which natural
productivity is augmented with fertilizers and farm made or industrially produced
feeds. The majority of Asian fi nfi sh aquaculture is produced in freshwater, semi-
intensive earthen pond culture systems.
Intensive
Some highly productive pond systems (e.g., shrimp, striped catfi sh), fi nfi sh cage
culture and some high value species, such as eels in China. Intensive systems are
mostly supplied with complete industrially produced pellet feeds that meet all of
the nutritional requirements of the culture species.
Table 2.2: The production intensity categories used in this analysis. (After de Silva and Hasan, 2007).

2. Impacts
Today
Impacts
ComparisonLooking ForwardPolicyAppendixGlossaryReferences Summary
18 Managing the environmental costs of aquaculture
Assessment method
The objective of this study is to compare and
contrast the global and regional demands of
aquaculture for a range of biophysical resources
across the entire suite of species and production
systems in use today. Examples of the sorts of
questions we wish to ask include:
• How do countries or regions differ in their
resource demands for aquaculture produc-
tion?
• Which species groups or production
systems are especially demanding, or ef-
fi cient, and in what respect?
• Are there particular areas of the produc-
tion process to which attention might most
profi tably be paid to reduce environmental
demands?
To assign intensities to the data records we
examined the available literature and consulted
experts on species production methods within each
species group within a country. For countries where
species within a species group were produced at
more than one intensity we duplicated the data
record and adjusted production values for each
record to refl ect the proportion produced under
each production intensity.
Finally, we considered the types of feed used for
each species group, country, production system,
habitat and intensity combination. Drawing on Neori
et al. (2004) and de Silva and Hasan (2007) we
distinguished fi ve primary feed categories (Table
2.3). We then examined the literature and combined
this with expert opinion where necessary (6% of
records) to estimate the dominant feed type for
each data record.
Feed Category Description
Natural Feeds
Plant materials, mainly crop waste, used in combination with other material but with little or
no processing. The feeds vary in nutrient quality.
Trash Feeds
Small or lower value fi sh used for aquaculture feeds and fed directly into aquaculture
systems. This practice is common for marine fi sh cage production in Asia. Trash fi sh require
no processing energy (except occasionally for chopping before feeding).
Mash Feeds
Mixed materials with some processing; processing is on farm and specifi c to farmers’
requirements. These are ‘farm-made’ feeds and the major feed input for semi-intensive
aquaculture.
Pellet Feeds
Feed pellets are manufactured in industrial feed plants and distributed through conventional
market chains. The pellets are expected to completely fulfi ll all nutritional requirements of
species. The pellets are mainly used in intensive aquaculture operations.
Extracted Food
Organic matter and nutrients for growth are assimilated from the environment through
autotrophic processes or fi lter feeding. This category applies largely to bivalves, aquatic
plants and some fi lter feeding fi shes (e.g., silver carp).
With the data reduction described above our
fundamental units of analysis are the elements
of a sparse six dimensional matrix comprising:
13 species groups x 18 countries x 3 production
intensities x 4 production systems x 2 habitats
x 5 feed types. This resulted in 75 positive
matrix elements, accounting for 82% of total
world production in 2008 (Appendix). These 75
unique production elements form the basis of our
assessment.
To facilitate meaningful comparisons of this
sort, we require a method that can be ap-
plied in a standardized way across all units of
analysis. Several approaches have been used
previously to examine the sustainability of aqua-
culture and we were faced with a choice of the
most appropriate method for this study. Table
2.4 summarizes the key features of several of
these approaches.
Table 2.3: The feed types used in this analysis. (After Neori et al., 2004 and de Silva and Hasan, 2007)

2. Impacts
Today
Impacts
Comparison Looking Forward Policy Appendix Glossary ReferencesSummary
19Managing the environmental costs of aquaculture
Photo by Kam Suan Pheng
CHINA

2. Impacts
Today
Impacts
ComparisonLooking ForwardPolicyAppendixGlossaryReferences Summary
20 Managing the environmental costs of aquaculture
Table 2.4: Summary of approaches to quantifying environmental impact. (Adapted from Bartley et al., 2007)
Method Key attributes Advantages Disadvantages Scientifi c rigor Ease of application and
communicability
Environmental Impact
Assessment
Project-based
Descriptive
Site-specifi c
Public planning and transparent
process
Based on multiple criteria and can be
used in sensitivity analysis
Identifi es hazards and impacts
Allows redesign of project to reduce
impacts
Does not quantify trade-offs or effects
Does not provide a single performance
indicator for comparisons
Problems with how to interpret data
Variable (very high to
low)
Lots of uncertainty due
to lack of data
Often time-constrained
due to development
deadlines
Good
Often fi gures prominently in
decision-making
Risk Assessment or
Analysis
Tool for understanding
environmental processes
Contributes to better understanding
of environmental fl ows and impacts
Attempts to be quantitative but can
also be qualitative
Identifi es hazards and impacts
Relies on qualitative judgments and
estimates due to knowledge gaps
Limited comparative use (some risks apply
to some sectors, others not)
Variable at present
Quantitative measures
need to be developed
(environmental
indicators)
Good
Formalized in legislation as
decision-making tool
Material Flows
Accounting, Mass
Balance, Input/Output
models
Examines input and
output of key materials
Accounts for biological
fl ows associated with
economic activities
Applicable to systems at
many scales
Quantifi es levels of inputs and
outputs
Can produce comparable information
over time and space
Used to improve ecological effi ciency
Well-known tool with standard
protocols
Does not refl ect environmental effects
Snapshot picture of fl ows at a specifi c point
in time and place
High Very good
Cost Benefi t Analysis,
including environmental
costs
Uses valuation techniques
for non-marketable goods
to compare net results
of activities of different
sectors (e.g., contingent
valuation, willingness to
pay, hedonic pricing)
Can compare production systems
Can be very inclusive of many types
of information, including non-
marketable goods
Long history and familiarity with
concept; decision-makers need and
want to know this information
Provides aggregate measures of
the relative performance of various
production systems
Environmental values hard to determine
Ecological function changes hard to predict
Often environment is not included
Normally long term sustainability issues not
addressed
Discount rates are arbitrary and may be
political
Loses information during aggregation
High Results easily communicated and
understood
Including valuation of
environmental goods and services
and non-marketable goods
makes application diffi cult

2. Impacts
Today
Impacts
Comparison Looking Forward Policy Appendix Glossary ReferencesSummary
21Managing the environmental costs of aquaculture
Method Key attributes Advantages Disadvantages Scientifi c rigor Ease of application and
communicability
Ecological Footprint Method to aggregate
impacts into a single
statistic to address
eco-effi ciency of human
activities
Converts all impacts to a
measure of area needed
to support a given activity
Provides a single indicator for
comparison
Can be applied to many levels
and scales (e.g., a footprint for
an individual to one for a national
economy)
Provides accumulative/aggregated
effects
Does not include all fl ows
Applications to food production systems are
not obvious
Method does not deal well with water
Does not provide specifi c information about
impacts or effects
Does not address specifi c effects in specifi c
environments
Aggregated statistic treats all environments
as homogenous and equal
Low Easy to communicate, but
statistic is often misused or can
be misinterpreted
Application is constrained
by knowledge gaps on
environmental differences among
habitats
Life Cycle Analysis (LCA) Examines a range
of impacts of food
production systems
Product-oriented
environmental impact
assessment, with a cradle
to grave perspective,
multiple criteria analysis
Quantifi es potential
contribution to global
impacts
Allows hazards to be identifi ed and
prioritized
Can build on previous work/data
Can compare between products/
processes/alternatives and different
scenarios
Basic method to develop eco-
labeling criteria to support
purchasing decisions for consumers
Can provide policy-relevant insights
Large data requirements
Some studies use different functional units
Results address global impacts at expense
of local impacts
Some indicators may not be appropriate for
specifi c cases
Results are not directly applicable unless
conducted for specifi c comparison
Some standard impact categories may not
be relevant to food product systems, thus
need to develop new ones
High Can “streamline LCA” for specifi c
comparisons
Communication on multiple
criteria may be diffi cult

2. Impacts
Today
Impacts
ComparisonLooking ForwardPolicyAppendixGlossaryReferences Summary
22 Managing the environmental costs of aquaculture
From our review we concluded that the Life Cycle
Analysis (LCA) approach provides the strongest
platform to conduct analysis over a range of
different production systems, and at different scales
of analysis. The approach is also readily amenable
to updating or refi ning with new information.
LCA approaches are now in widespread use
and are conducted at a variety of scales. There
is an emerging body of LCAs that examines
the environmental resources and emissions of
aquaculture production systems (Pelletier and
Tyedmers, 2007; Ayer and Tyedmers, 2009;
Ellingsen et al., 2009). To date, however, the bulk
of LCAs have been undertaken for single species
and production systems (e.g., Mungkung et al.,
2006; Pelletier et al., 2009) and comparability
among studies remains a signifi cant issue owing
to the very wide range of choices available for
describing LCA processes. There has been no effort
to undertake a systematic global and regional level
LCA comparison for aquaculture production of the
type presented here.
LCA is a systematic four phase process comprising:
1. Goal Defi nition and Scoping — To a) defi ne
and describe the product, process or
activity, b) establish the context in which
the assessment is to be made and c)
identify the boundaries and environmental
effects to be reviewed for the assessment.
2. Inventory Analysis — To identify and
quantify energy, water and materials
usage and environmental releases (e.g.,
air emissions, solid waste disposal, waste
water discharges).
3. Impact Assessment — To assess the
potential human and ecological effects of
energy, water, and material usage and the
environmental releases identifi ed in the
inventory analysis.
4. Interpretation — To evaluate the results
of the inventory analysis and impact
assessment to select the preferred
product, process or service with a clear
understanding of the uncertainty and the
assumptions used to generate the results.
LCA practitioners make a distinction between
screening studies that use readily available data and
extensive studies that require a major investment of
resources to gather new data. This study lies fi rmly
at the screening end of this continuum and aims
to provide a robust approach for answering the
questions we pose. It also provides a foundation for
further debate and refi nement.
Our next requirement is to defi ne the system
boundaries for our analysis. In its full form LCA is
a cradle-to-grave approach that begins with the
gathering of raw materials from the earth to create
the product and ends at the point when all materials
are returned to the earth. When complete, an LCA
estimates the cumulative environmental impacts
resulting from all stages in a product’s life cycle.
This often includes factors such as raw material
extraction, material transportation, ultimate product
disposal, that are often ignored by other methods.
In common with others studying aquaculture,
however, we have adopted a more bounded
approach (Figure 2.1) that excludes environmental
costs associated with building infrastructure, seed
production, packaging and processing of produce,
transport of feed or produce, cooking the produce
and disposing of the waste. Previous studies
suggest that setting limits as shown in Figure 2.1
is defensible because the bulk of environmental
resources and environmental emissions lies within
these bounds (Pelletier and Tyedmers, 2007;
Pelletier and Tyedmers, 2010). The biggest energy
demands for aquaculture production systems occur
on farm, for processing feed, for reduction of wild
fi sh into fi shmeal and fi sh oil and in the capture of
wild fi sh to feed into the production process.
The main sources of eutrophying emissions
(nitrogen and phosphorus) are those released from
the farm (Pelletier and Tyedmers, 2007; Pelletier and
Tyedmers, 2010).
The system shown in Figure 2.1 is generic and
was used to analyze each of the 75 unique
combinations of species group, country, production
intensity, production systems, habitat and feed
type. For some combinations particular processes
become irrelevant or are reversed. With seaweed or
bivalve culture, for example, nutrients are taken up
from the environment rather than released. Similarly,

2. Impacts
Today
Impacts
Comparison Looking Forward Policy Appendix Glossary ReferencesSummary
23Managing the environmental costs of aquaculture
with bivalves, since these extract food from the
environment we set the feed production process to
make no demands on energy, crop meal, fi shmeal
or fi sh oil.
Unit Processes
Data collection is the most time demanding task of
LCAs. There are two types of LCA data required;
foreground data and background data. Foreground
data is the specifi c data required to model the
systems (Goedkoop et al., 2008). This data refers
to the biophysical resources required during
aquaculture production, specifi cally, the amount
of land, water, feed, fertilizers and energy required
on farm. This data was collected from a variety of
sources during a literature review.
Background data refers to predefi ned unit
processes available in the standardized databases
used by LCA practitioners and provided with
several LCA software tools. Background data have
been defi ned for a variety of agricultural production
and energy production processes.
Figure 2.1 illustrates the system boundary of the
model, distinguishing between the biosphere
inputs (raw materials) and the technosphere
inputs (any material transformed by human action)
and indicating where emissions are released.
The fi gure also distinguishes where foreground
and background data has been used. By linking
the foreground data to the background unit
processes we capture upstream processes and
their associated inputs from the biosphere and
thetechnosphere (Goedkoop et al., 2008).
Aquaculture
Production
Wild Fish Fish Capture
Energy Energy Energy
Crop
Meal
Energy
Feed Production
Inorganic
Fertilizer
Production
NP
Land
Water
Organic
Fertilizer
Energy
Foreground data
Background data
Technosphere Input
Oil
Meal
Exclusions: Transport, seed production, processing, packaging, waste disposal
Environmental Emission
Resource Flow
Biosphere Input
Primary Production Process
Fish Reduction
Figure 2.1: Graphical summary of the system boundaries and model structure for the Life Cycle Analyses undertaken in
this study. Note: in the case of seaweeds, the fl ows to nitrogen (N) and phosphorus (P) would be negative (reversed).

2. Impacts
Today
Impacts
ComparisonLooking ForwardPolicyAppendixGlossaryReferences Summary
24 Managing the environmental costs of aquaculture
In LCA parlance, the following demands on resources become our inventory categories:
1. The area of land required to grow fi sh.
2. The amount of wild fi sh used as fi sh feed.
3. The amount of organic and inorganic fertilizer required to grow fi sh.
4. The energy required for the various production processes involved (shown in Figure 2.1).
5. The amount of carbon dioxide the environment must assimilate from the production processes.
6. The amount of waste nitrogen and phosphorus the environment must assimilate from fi sh
production.
As noted above, these six categories of demand were chosen because they are most likely to constrain the
potential for sustainable aquaculture growth (Rockström et al., 2009; Duarte et al., 2009; Muir et al., 2009).
Process defi nition and model parameterization
Having identifi ed the categories for inventory, we must now specify how inputs to the LCA are calculated.
The following section describes the basis for this. Literature sources and the approach used to estimate
model parameters are given in Table 2.5.
The foundation of our approach is to work back from aquaculture production
for each species group i
within production system j in habitat k at intensity l with feed m for country n. (Note: These subscripts remain
constant throughout this paper, unless otherwise stated). Using these data we fi rst used the following
equations to calculate the land or sea area required for production and the volume of freshwater required for
inland systems:
Where is the production effi ciency per unit production area, and β is the production effi ciency per unit
water volume. For production from coastal systems (marine and brackishwater) the freshwater requirement
was set to zero.
To calculate total on farm energy use we modeled country specifi c energy mixes (IEA, 2010) to estimate
the energy use effi ciency γ such that:
Organic fertilizers are defi ned here as on farm wastes that enhance the natural productivity of the culture
system. We distinguished four categories: cow, chicken and pig manure and plant compost and calculate
organic fertilizer input as the sum of inputs into a given system from these sources i.e.:
Where is the application rate of fertilizer p per unit aquaculture production area for a given
production system.

2. Impacts
Today
Impacts
Comparison Looking Forward Policy Appendix Glossary ReferencesSummary
25Managing the environmental costs of aquaculture
Similarly, for inorganic fertilizer inputs we distinguished two sources, urea and Triple Super Phosphate
(TSP), and calculate total application as the sum of these two inputs:
Where and are the application rates per unit area for urea and TSP, respectively for each
production system.
Aquaculture feeds are a combination of fi shmeal, fi sh oil, and crop meal. We estimated the total quantity
of fi sh required to provide the necessary fi shmeal and fi sh oil to meet observed fi sh production using the
following equations:
Where FCR is the Food Conversion Ratio, defi ned as the amount of processed feed required for every unit
weight of fi sh produced, is the proportion of fi shmeal or oil in feeds and is the yield of meal or oil per
unit of wild fi sh from the fi sh reduction process. Because a given quantity of wild fi sh produces both meal
and oil, we take the larger of the two values to represent total wild fi sh demand (Kaushik and Troell, 2010).
Energy requirements for the fi sh reduction process were copied from the unit process ‘Fishmeal’ from the
DK data library supplied with SimaPro, the software used for our LCA analyses. This unit process states
that reducing 1 kg of sandeel to fi shmeal and fi sh oil requires 1332 kj of heat energy and 0.04 kwh of
electricity. We assume here that the costs of reduction for sandeel apply to costs of reduction for other fi sh
species. The energy needed for wild fi sh capture was based on estimates of the fuel oil required for fi shing
provided by Ellingsen and Aanondsen (2006). During fi sh reduction two products, fi shmeal and fi sh oil, are
produced. We allocated environmental burdens for each product based on the weight of each produced.
Total crop meal required was estimated from:
Once the main crop types were identifi ed through literature review, unit processes were identifi ed within
the EcoInvent library that represented these crops. This was then used to estimate the energy needed to
produce it. We defi ned main crop types as those that accounted for approximately 70% of all feed used in
the grow-out of a unique species combination.
To calculate nitrogen and phosphorus emissions from aquaculture production we used a simple mass
balance approach where the total weight of N or P from processed feed and fertilizer inputs was calculated
and subtracted from the total N or P content of the fi sh produced. These quantities were calculated from
the following equations:
Where is the percentage nitrogen by weight in feed.

2. Impacts
Today
Impacts
ComparisonLooking ForwardPolicyAppendixGlossaryReferences Summary
26 Managing the environmental costs of aquaculture
Where is the percentage nitrogen by weight in cow, chicken and pig manure and plant compost,
respectively.
Where is the percentage nitrogen by weight in urea.
Where is the percentage nitrogen in fi sh tissue.
Phosphorus was calculated in the same way except that the percentage phosphorus in TSP replaced
percentage nitrogen in urea to calculate the contribution from inorganic fertilizers. Although this approach is
reasonable as a fi rst approximation, we recognize that not all nutrients are lost. In some pond systems, for
example, up to half of the nutrients may end up in sediments which can be re-used for agriculture (Islam,
2005).
Table 2.5: Parameter estimates and data sources for foreground data calculations. In cases where parameter estimates
for a particular system could not be obtained directly from the literature, values for the system with the closest similarity
or expert opinion was used. The proportion of records determined by expert opinion are shown in parentheses at the
end of each list of data sources.
Parameter Description Units Data Sources
1 Production per unit area. t.ha
-1
Atmomarsono and Nikijulluw, 2004; Barman
and Karim, 2007; Biao and Kaijin, 2007; Brown,
2000; Cao et al., 2007; Chen, 2003; CIFA,
accessed in 2010; Cruz, 1997; El-Sayed, 2007;
Gupta and Acosta, 2004; Losinger et al., 2000;
Nakada, 2002; Nur, 2007; Phan et al, 2009;
Phuong, 2010; Rosenberry, 1999; Sturrock et al.,
2008; Sumagaysay-Chavoso, 2007; Unknown,
accessed in 2010; Weimin and Mengqing, 2007.
(9%)
2 Production per unit water volume. t.m
3
Dugan et al., 2007; Muir et al., 2009 (18%)
3
On farm energy use effi ciency per
unit fi sh production.
Mj.t
-1
ADB, 2005; Bosma et al., 2009; Bunting and
Pretty, 2007; Henriksson, 2009; Olah and Sinha,
1986; Pelletier and Tyedmers, 2010; Tlusty and
Langueux, 2009; Troell et al., 2004. (25%)
4 Application rate of cow, chicken
and pig manure and plant
compost for each production
system
kg.ha
-1
Barman and Karim, 2007; Cruz, 1997; Cruz-
Lacierda et al., 2008; de Silva and Hasan, 2007;
El-Sayed, 2006; El-Sayed, 2007; FAO, accessed
in 2010; Flores-Nava, 2007; Hung and Huy, 2007;
Weimin and Mengqing, 2007. (73%)
5
Application rate per unit area of
urea and TSP, respectively for
each production system.
kg.ha
-1
Atmomarsono and Nikijulluw, 2004; Barman and
Karim, 2007; Cruz, 1997; Cruz-Lacierda et al.,
2008; El-Sayed, 2006; El-Sayed, 2007; Flores-
Nava, 2007; Hung and Huy, 2007; Pelletier et al.,
2009. (70%)
6
Food conversion ratio. (Food
required: Fish produced, by wet
weight)
- Tacon and Metian, 2008; FAO, 2004. (10%)
7 The proportion by weight of
fi shmeal and oil in pellet feeds.
- Barman and Karim, 2007; Tacon and Metian,
2008. (10%)

2. Impacts
Today
Impacts
Comparison Looking Forward Policy Appendix Glossary ReferencesSummary
27Managing the environmental costs of aquaculture
Parameter Description Units Data Sources
8
The yield of fi shmeal or oil per unit
wet weight of fi sh.
- Péron et al., 2010.
9 The proportion by weight of
nitrogen and phosphorus, in fi sh
feed.
- Craig and Helfrich, 2009.
10 The proportion by weight of
nitrogen and phosphorus (i
= 1,..,2, respectively) in cow,
chicken and pig manure and plant
compost (j = 1,..,4, respectively).
- Barman and Karim, 2007.
11 The proportion by weight of
nitrogen and phosphorus in urea
and TSP, respectively.
- Graslund and Bengtsson, 2001.
12 The proportion by weight of
nitrogen and phosphorus, in fi sh
tissues.
- Ramseyer, 2002; Tanner et al., 2000.
Note: In all cases subscripts denote: species group i within production system j in habitat k at intensity l with feed m for country n.
From Inventory to Impact Categories
From the estimates derived using the methodology
described above we ran an LCA analysis for each
of the 75 unique combinations. All analyses were
conducted using SimaPro V 7.0 (Goedkoop et al.,
2008). In common with other LCAs impacts were
assessed using a mid-point approach, which takes
the inventory results and translates them into impact
measures that fall somewhere short of the ultimate
impacts (end points) of interest. With acidifi cation,
for example, one might choose an impact end point
as area of forest lost through acid rain. This will be
diffi cult to estimate, however, so researchers usually
use the inventory data to estimate the aggregate
acidifi cation burden on forests as a mid-point
measure. For this study, the following six impact
categories were used:
Eutrophication: includes all impacts due
to excessive levels of macronutrients in the
environment caused by emissions of nutrients to air,
water and soil. Expressed as t PO
4
equivalents
3
.
Acidifi cation: acidifying substances impact on the
functioning of ecosystems and human well-being.
Acidifi cation potentials are expressed in t SO
2
equivalents.
Climate Change: refl ects the characterization
model developed by the Intergovernmental Panel on
Climate Change (IPCC). Results are expressed as
climate change potential in t CO
2
equivalents.
Cumulative Energy Demand (CED): represents
the direct and indirect use of industrial energy,
expressed in Gj, required throughout the production
process.
Land Occupation: calculated as the sum of direct
and indirect land occupation, using equivalence
factors adjusted for each type of land (e.g., arable,
pasture, sea) for relative levels of bioproductivity.
The higher the bioproductivity of the land, the higher
the equivalent factor becomes (Wackernagel et
al., 2005)
4
. Land occupation is expressed in ha
equivalents.
Biotic Depletion (Fish): the amount (t) of wild
fi sh required to support observed aquaculture
production. There was no differentiation of the type
of fi sh used during the production process, but
we assume that all the fi sh used for feed are small
pelagic fi sh species.
3
Although nitrogen is often the limiting nutrient in marine systems, it is convenient to express eutrophication potential in terms of PO4
throughout and does not affect the conclusions.
4
All species within ‘coastal’ habitats were classifi ed as occupying sea (equivalence factor 0.36). Species cultivated in ‘inland’ habitats
were assumed to occupy arable land (equivalence factor 2.19). Thus, if cultivation of a species group required 1 hectare of sea
area it was characterized as requiring 0.36 hectares. In contrast, species requiring 1 hectare of arable land (e.g., carp, tilapia) was
characterized as requiring 2.19 hectares of land.

2. Impacts
Today
Impacts
ComparisonLooking ForwardPolicyAppendixGlossaryReferences Summary
28 Managing the environmental costs of aquaculture
The defi nition and approach used for estimating eutrophication, acidifi cation and climate change was the
‘CML Baseline 2001’ impact assessment methodology of The Institute of Environmental Sciences of Leiden
University (CML) (Guinée et al., 2002). The standard method to calculate Cumulative Energy Demand (CED)
was based on the method published by EcoInvent version 1.05 and expanded by PRé Consultants for
energy resources available in the SimaPro database (VDI, 1997).
Results
Table 2.6a summarizes the overall impact of the 82% of 2008 production that was modeled in this study
along with a projection of the impacts for the total production that year. Putting such fi gures in context
is, of course, challenging, but one indication of the relative signifi cance of these values can be obtained if
one compares estimates for CO
2
emissions with those available for other sectors (Table 2.6b). This table
suggests that aquaculture contributes about 0.96% to total CO
2
emissions and between 6.3 and 7.5%
of agriculture emissions. This is based on IPCC estimates of total agricultural emissions ranging between
5120 MtCO
2
-eq/yr (Denman et al. 2007) and 6116 MtCO
2
-eq/yr (US-EPA, 2006) in 2005. If one were to
offset the CO
2
contribution from all aquaculture production it would cost about US$ 52.5 billion at the
current market price for CO
2
in offset markets of around US$ 15 per tonne (World Bank, 2010).
a)
Eutrophication
(Mt PO
4
eq)
Acidifi cation
(Mt SO
2
eq)
Climate Change
(Mt CO
2
eq)
Land
Occupation
(Mha)
Energy
Demand
(Tj eq)
Biotic
Depletion
(Mt)
Modeled 3.33 2.60 298.26 55.77 3,431,361 15.11
Total 3.92 3.06 350.89 65.61 4,036,895 17.78
b) Sectoral Source
Total Emission
(M tonnes CO
2
eq)
Energy 22,952
Transport 4,815
Industrial Processes 2,105
Agriculture 4,650
Waste 1,057
Aquaculture (this study) 385
Total 30,824
Relationships with aquaculture production
As expected, for the most part, data for all impact categories show a positive relationship between overall
production levels and impact (Figure 2.2). The only exceptions to this are for the subset of the data
representing species that extract food from the natural environment. With the exception of a relatively minor
contribution (on a global scale) to eutrophication through pseudo-feces deposits to bottom sediments by
mollusks, these make no contribution to eutrophication or biotic (fi sh) depletion. This is apparent from the
horizontal line of data points at the bottom of these panels in Figure 2.2. Despite these linear relationships,
however, there is clearly considerable variance in impact for a given level of production. This is especially
true for acidifi cation, climate change, cumulative energy demand and land occupation.
Table 2.6: (a) Total estimated impacts from the 75 production systems modeled in this study and an estimate of the
complete global impact assuming that, as with total aquaculture production, each calculated estimate represents 88%
of the total. (b) Sectoral comparison of CO
2
emissions. (Note: not all categories are mutually exclusive so fi gures do not
add up to the total estimate). Source: UNSTATS Environmental Indicators, accessed December, 2010.

2. Impacts
Today
Impacts
Comparison Looking Forward Policy Appendix Glossary ReferencesSummary
29Managing the environmental costs of aquaculture
Impacts by habitat and production system
Given the positive relationship between production
and absolute levels of impact described above
it is unsurprising that, with its dominance as a
production system, inland pond culture contributes
the greatest impact overall for all impact categories
(Figure 2.3, upper panel). Nevertheless, despite
this overall dominance, demand for wild fi sh (biotic
depletion) is also notable for marine cage and pen
production. Negative values for eutrophication
in bottom and off-bottom culture refl ect bivalve
farming where nutrients are taken up from the
environment. However, although we can rightly
view this as a regional removal, we must recognize
that at a more local scale impact through the
deposition of pseudo-feces will occur.
When one considers effi ciency of production,
and compares levels of impact for a given unit
of product, impacts from pond and cage and
pen production dominate in both freshwater and
marine systems (Figure 2.3, lower panel). With
the exception of land occupation, however, cage
and pen culture has consistently greater impact.
Overall, however, cage and pen production in
inland waters appears to cause the greatest
impact. One must also bear in mind that deposits
into freshwater pond sediments are also often used
for agriculture.
100,000 200,000 500,000 1,000,000 2,000,000 5,000,000
Production 2008 (t)
1
10
100
1,000
10,000
100,000
1,000,000
Eutrophicati on
1
10
100
1,000
10,000
100,000
1,000,000
10,000,000
Acidification
100
1,000
10,000
100,000
1,000,000
10,000,000
100,000,000
1,000,000,000
100,000 200,000 500,000 1,000,000 2,000,000 5,000,000
Production 2008 (t)
100
1,000
10,000
100,000
1,000,000
10,000,000
Land Occupation
10,000
100,000
1,000,000
10,000,000
100,000,000
1,000,000,000
10,000,000,000
Energy Demand
1
10
100
1,000
10,000
100,000
1,000,000
Biotic Depletion
Climate change
Figure 2.2: The relationship between overall production levels for each of the 75 unique production combinations and
level of impact: Eutrophication (t PO
4
eq); Acidifi cation (t SO
2
eq); Climate Change (t CO
2
eq); Land Occupation (ha eq);
Cumulative Energy Demand (Gj); Biotic Depletion(t).

2. Impacts
Today
Impacts
ComparisonLooking ForwardPolicyAppendixGlossaryReferences Summary
30 Managing the environmental costs of aquaculture
Impacts by species group
In absolute terms, we see that carps dominate
overall impact (Figure 2.4, upper panel), refl ecting
the fact that carp production is greater than
that of other species groups. Production in the
“Other fi nfi sh” category is also notable, however,
particularly for acidifi cation, climate change and
energy demand, three measures that are correlated
with one another. A recent review of environmental
impacts of marine fi nfi sh culture provides further
perspectives on this production category (Volpe et
al., 2010). For the biotic depletion category, total
demand for fi sh to produce shrimps and prawns
and salmonids almost reaches that for carps.
In relative terms, eel production stands out as
being especially environmentally demanding (Figure
2.4, lower panel), refl ecting the highly intensive
and energy demanding nature of eel production
systems. No other species group dominates
impact categories to the same extent, although
shrimps and prawns tend to be among those
causing the most impact, while salmonids are
notable for their demand for fi sh. Figure 2.5 further
summarizes the relative effi ciency of production
for species groups categorized by habitat and
production system.
Land occupation impacts vary with species group
and system, but largest impacts are not surprisingly
associated with pond farming, particularly in
Asia and South America. One should recognize,
however, that LCA does not fully capture
biodiversity and other values associated with land
use for aquaculture. More local analysis will be
is required to determine such impacts. Impacts
of concern may relate to loss of biodiversity
associated with replacement of habitat by ponds,
or loss of ecosystem functions such as those
associated with carbon sequestration or provision
of nursery areas for wild fi sh populations.
Habitat Production System
0M 1M 2M
Eutrophication
0K 500K 1000K
Acidification
0M 50M 100M 150M
Climate Change
0M 10M 20M 30M
Land Occupation
0M 500M 1000M 1500M
Energy Demand
0M 2M 4M 6M
Biotic Depletion
Coastal
Bottom Culture
Cages & Pens
Off-Bottom Culture
Ponds
Inland
Cages & Pens
Off-Bottom Culture
Ponds
A
bsolute Values
Habitat Production System
050100
Eutrophication
0 50 100 150
Acidification
0K 5K 10K 15K
Climate Change
0.0 0.5 1.0
Land Occupation
0K 50K 100K 150K 200K
Energy Demand
0 500 1000 1500 2000
Biotic Depletion
Coastal
Bottom Culture
Cages & Pens
Off-Bottom Culture
Ponds
Inland
Cages & Pens
Off-Bottom Culture
Ponds
Relative Values (per tonne production)
Figure 2.3: Upper panel: The absolute environmental impact of 2008 aquaculture production categorized by production
system and habitat: Eutrophication (t PO
4
eq); Acidifi cation (t SO
2
eq); Climate Change (t CO
2
eq); Land Occupation (ha
eq); Cumulative Energy Demand (Gj); Biotic Depletion (t). Lower panel: The relative environmental impact, per tonne of
product categorized by production system and habitat: Eutrophication (kg PO
4
eq); Acidifi cation (kg SO
2
eq); Climate
Change (kg CO
2
eq); Land Occupation (ha eq); Cumulative Energy Demand (Mj); Biotic Depletion (kg).

2. Impacts
Today
Impacts
Comparison Looking Forward Policy Appendix Glossary ReferencesSummary
31Managing the environmental costs of aquaculture
Figure 2.4: Upper panel: The absolute environmental impact of 2008 aquaculture production categorized by species
group; units as for Figure 2.3 (upper panel). Lower panel: The relative environmental impact per tonne of product
categorized by species; units as for Figure 2.3 (lower panel).
Figure 2.5: The relative environmental impact of 2008 aquaculture production categorized by habitat, production
system and species group; units as for Figure 2.3 (lower panel).
0 50 100 150
Eutrophication
0 50 100 150 200
Acidification
0K 5K 10K 15K 20K
Climate Change
0.0 0.2 0.4 0.6 0.8
Land Occupation
0K 100K 200K 300K
Energy Demand
0 500 1000 1500 2000
Biotic Depletion
Bottom Culture Bivalves
Cages & Pens
Crabs and Lobsters
Other finfish
Salmonids
Off-Bottom
Culture
Bivalves
Gastropods
Seaweeds and Aquatic
plants
Ponds
Bivalves
Other finfish
Other Invertebrates
Shrimps and Prawns
Tilapias
Coastal
0 50 100 150
Eutrophication
0 100 200 300
Acidification
0K 10K 20K 30K
Climate Change
0.0 0.5 1.0 1.5
Land Occupation
0K 100K 200K 300K 400K
Energy Demand
0K 1K 2K
Biotic Depletion
Cages & Pens
Crabs and Lobsters
Other finfish
Off-Bottom
Culture
Gastropods
Ponds
Bivalves
Carps
Catfish
Eels
Other finfish
Other Vertebrates
Shrimps and Prawns
Tilapias
Inland
Species Group
0K 500K 1000K 1500K 2000K
Eutrophication
0K 200K 400K 600K 800K
Acidification
0M 20M 40M 60M 80M
Climate Change
0M 10M 20M 30M
Land Occupation
0M 200M 400M 600M 800M1000M
Energy Demand
0M 1M 2M 3M
Biotic Depletion
Carps
Catfish
Eels
Salmonids
Other finfish
Bivalves
Gastropods
Crabs and Lobsters
Shrimps and Prawns
Other Invertebrates
Other Vertebrates
Seaweeds and
Aquatic plants
A
bsolute Valu
e
Species Group
050100150
Eutrophication
0 100 200 300
Acidification
0K 10K 20K 30K
Climate Change
0.0 0.5 1.0 1.5
Land Occupation
0K 100K 200K 300K 400K
Energy Demand
0 500 1000 1500 2000 2500
Biotic Depletion
Carps
Catfish
Eels
Salmonids
Other finfish
Bivalves
Gastropods
Crabs and Lobsters
Shrimps and Prawns
Other Invertebrates
Other Vertebrates
Seaweeds and
Aquatic plants
Relative Values
Talipias
Talipias

2. Impacts
Today
Impacts
ComparisonLooking ForwardPolicyAppendixGlossaryReferences Summary
32 Managing the environmental costs of aquaculture
Impacts by country
Figures 2.6 and 2.7 summarize the absolute and relative impacts of aquaculture production for the 18
countries in our analysis
5
. Figure 2.6 gives a clear sense of the overall dominance of China, but also
illustrates how absolute demand for fi sh is somewhat more evenly distributed, refl ecting the mix of
species that are produced in different regions. The demands of salmonids and shrimps and prawns, for
example, explain the bulk of the fi sh demand for Europe and the Americas.
Figure 2.6: Maps showing the absolute size of total environmental impacts of 2008 production for each of the 18
countries analyzed in this study. Scales have been omitted from these fi gures for clarity.
In terms of effi ciency of production with respect to environmental impacts, the picture is rather more
variable (Figure 2.7). For eutrophication, for example, results are broadly comparable across all
countries, whereas for four of the remaining impact categories, aquaculture production is markedly
more “effi cient” in the salmon producing nations of north Europe, Canada and Chile, and for Japan.
Not surprisingly, however, this picture reverses for effi ciency in production with respect to wild fi sh
consumption (biotic depletion) where the salmon producing countries, are joined by those where
shrimps and prawns dominate the production mix.
Ma p data © OpenS treetMa p (and) contributors, CC -BY- SA.
Eutrophication
Map data © OpenStreetMap (and) contributors, CC-BY-SA. M ap data © OpenStre etMap ( and) contributors , CC- BY- SA.
*limate *hange
Ma p data © OpenS treetMa p (and) contributors, CC -BY- SA.
Land Occupation
Map data © OpenStreetMap (and) contributors, CC-BY-SA.
Energy Demand
Ma p data © OpenStre etMa p (and) contributors, CC -BY- SA.
Biotic Depletion
Acidification
5
Scales have been omitted from these fi gures for clarity.

2. Impacts
Today
Impacts
Comparison Looking Forward Policy Appendix Glossary ReferencesSummary
33Managing the environmental costs of aquaculture
Figure 2.7: Maps showing the relative size of environmental effi ciencies (indicated by the average environmental
impacts per tonne of production) for each of the 18 countries analyzed in this study.
Further insight into how these values are derived can be obtained by looking in more detail at how
environmental effi ciencies differ across countries that culture the same species groups (Figure 2.8). Taking
shrimp and prawn culture in coastal systems we can see, for example, that China is much less effi cient,
in relative terms, than other producers with respect to acidifi cation, climate change potential and energy
demand (Figure 2.8 upper panel). By contrast, the eutrophication burden through production of other
fi nfi sh is markedly greater in Indonesia and the Philippines than it is for other producers. For salmonid
production, environmental performance is broadly similar across countries, but Canada appears to
have greater relative demands for fi sh-based feeds (Figure 2.8 upper panel). For inland carps and tilapia
production, no single country stands out across all impact categories, but for catfi sh the United States
and Vietnam are among the least effi cient in most cases.
Of particular interest in Figure 2.8 is the variation between countries for a given species. In 22 of the 36
comparisons shown, the best performers had impacts per tonne produced that were more than 50%
lower than the worst performers. This variation indicates that large effi ciency gaps in environmental
performance exist between countries, indicating great potential for improvement (see discussion).
Ma p data © OpenS treetMa p (and) contributors, CC -BY- SA. M ap data © O penStreetMap (and) contributors, C C- BY-S A. Map data © OpenStre etMap (a nd) contributors, C C- BY- SA.
Ma p data © OpenS treetMa p (and) contributors, CC -BY- SA. M ap data © O penStreetMap (and) contributors, C C- BY-S A. Map data © OpenStre etMap (a nd) contributors, C C- BY- SA.
Eutrophication
*limate *hange
Land Occupation
Energy Demand
Biotic Depletion
Acidification

2. Impacts
Today
Impacts
ComparisonLooking ForwardPolicyAppendixGlossaryReferences Summary
34 Managing the environmental costs of aquaculture
Drivers of impact
An important tool in understanding our results is
contribution analysis. This shows which processes
are playing a signifi cant role in the impact results.
Often, even in an LCA containing hundreds of
different processes more than 95% of the results
are determined by just ten or fewer. Figure 2.9
summarizes the contributions to impact of the
fi ve main processes in our models for each of the
species groups
6
.
This shows clearly that it is the fi sh production
process itself which contributes most to
eutrophication, whereas, for most groups,
acidifi cation and climate change impacts are
contributed primarily by the national energy
production process. This indicates that much of the
variation in acidifi cation and climate change impacts
across countries for a given production system
will be driven by the energy mix that supplies that
country. Production in a country such as China that
is dominated by coal production, therefore, will be
greater than in a country with a large proportion of
energy coming from nuclear or hydro power.
As we would expect biotic (fi sh) depletion is driven
primarily by the feed production process. Fertilizer
production processes for urea and TSP, generally
contribute little to the total impact.
An interesting feature of this analysis is the
exceptions to the general pattern. It is notable,
for example, how the feed production process
dominates most impact categories for salmon
aquaculture and, to a lesser extent, for tilapia and
carps.
Country
0 50 100 150
Eutrophication
0 50 100 1500K 5K 10K 15K 20K
Climate Chan ge
0.0 0.5 1.0 1.5 2.0 2.5
Land Occupation
0K 100K 200K 300K
Energy Demand
0 200 400 600
Bio tic Dep letio n
Carps
Bangladesh
China
India
Catfish
China
Indonesia
USA
Viet Nam
Tilapias
China
Indonesia
Philippines
Th ailand
Inland
Country
0 50 100 150 200
Eutrophication
0 100 200 300 0K 10K 20K 30K
Climate Chan ge
0.0 0.5 1.0 1.5
Land Occupation
0K 100K 200K 300K
Energy Demand
0K 1K 2K 3K
Bio tic Dep letio n
Other
finfish
China
Egy pt
Indonesia
Japan
Philippines
Salmonids
Canada
Chile
Norway
UK
Shrips
And
Prawns
China
Ecuad or
Indonesia
Mexico
Thailand
Viet Nam
Coastal
Acidification
Acidification
Coastal
Inland
Species Group Species Group
6
One feature of this analysis that it is important to bear in mind is that a given process may occur in several places in the model; energy production, for
example, will contribute to both feed and fertilizer production processes. Figure 2.9 shows the sum of all these contributions from a given process.
Figure 2.8: A comparison of environmental effi ciencies across countries growing the same species group.

2. Impacts
Today
Impacts
Comparison Looking Forward Policy Appendix Glossary ReferencesSummary
35Managing the environmental costs of aquaculture
Sensitivity analysis
With 75 separate LCAs a complete analysis of
both within and between model sensitivities would
be an enormous and impractical undertaking. In
view of this, we focused on those models where
we felt the greatest uncertainties existed. The
results of our analysis can be sensitive to both
the functional form (structure) of our model and
its parameterization. Assumptions made during
the goal setting and scoping phases affect
model structure and the quality of available data
determines the uncertainty in input parameters.
Our primary uncertainties concerning both model
structure and parameterization are with feed and
fertilizers.
For feed, we used 5 categories and assigned each
of our 75 production systems to one of these.
Natural feeds provided by the inherent productivity
of the system were not considered as having
any negative environmental effect and were not,
therefore, included in the inventory stage of the
LCA. Mash feeds are farm-made and require little
processing. Where the databases provided with
Simapro allowed, we chose crops ‘at farm’ to
represent the lesser degree of processing of mash
compared to pellet feeds. Pellet feeds were treated
as industrial feed, meaning that processes were
chosen from the database to better represent the
higher degree of processing needed for this feed
type.
For fertilizers we assumed that organic fertilizers
are only used in extensive and semi-intensive
systems, inorganic fertilizers only in semi-intensive
systems and none of them in intensive systems
(unless otherwise stated). As noted earlier, we
encountered some diffi culties in fi nding data on
fertilizer use and had to appeal to expert opinion to
fi ll in the gaps, especially for China.
For some systems where data were poor, we
also examined sensitivity to the food conversion
effi ciency and assumptions about on-farm energy
use.
Impact
Category
Habitat / Species Group
Coastal
Tilapias
Salmonids
Other finfish
Bivalves
Gastropods
Crabs and
Lobsters
Shrimps and
Prawns
Other Inverts
Seaweeds
Inland
Carps
Catfish
Tilapias
Eels
Other finfish
Bivalves
Gastropods
Crabs and
Lobsters
Shrimps and
Prawns
Other Verts
,\[YVWOPJH[P
(JPKPMPJH[PVU
*SPTH[L
*OHUNL
3HUK
6JJ\WH[PVU
,ULYN`
+LTHUK
)PV[PJ
+LWSL[PVU
+YP]LY
<YLH7YVK\J[PV U
(87YVK\J[PVU7YVJLZZ
,SLJ[YPJP[`.LULYH[PVU
-LLK 7YV K\J[PV U
;:77YVK\J[PVU
Figure 2.9: The total proportional contribution to impact of the fi ve main processes for each species group.

2. Impacts
Today
Impacts
ComparisonLooking ForwardPolicyAppendixGlossaryReferences Summary
36 Managing the environmental costs of aquaculture
To explore the sensitivity of impact results to these issues we examined models for 3 species groups
(carps, shrimps and prawns, tilapias) and for each species group we compared the results for 2 countries
(China + 1). We changed the assumptions on feed, by either modifying the feed source, by assuming
that there is only one crop in the diet (the one having the biggest share in the feed composition) or by
substituting one crop by another when it couldn’t be found in the EcoInvent database (e.g., coconut
(=husked nut) for groundnut). We only changed one parameter at a time unless otherwise stated. Table
2.7 summarizes the set of contrasts we examined. In essence, these can be considered plausible, but less
likely options compared to our baseline choices.
Table 2.7: Summary of the models used to examine sensitivity relative to baseline results.
Country Intensity Uncertainty Variation from Baseline
Carps
India semi-intensive Feed source Replaced husked nuts PH by rapeseed extensive at farm CH
Feed source Rice only (main crop)
Food conversion FCR 2 instead of 1.5 (i.e. same as for intensive)
India intensive Feed source Replaced husked nuts PH by rapeseed extensive at farm CH
Feed source Replaced husked nuts by rapeseed conventional FR
Feed source Rice only (main crop)
On-farm energy Changed on farm energy (=20,000 instead of 65,000)
On-farm energy Changed on farm energy + rapeseed extensive
China semi-intensive Feed source Rapeseed only (main crop)
Food conversion FCR 2 instead of 1.5 (i.e. same as for intensive)
Fertilizer Added inorganic fertilizers (150/150)
Fertilizer Removed organic fertilizers
China intensive Feed source Rapeseed only (main crop)
China extensive Fertilizer Added inorganic fertilizers (50/50)
Tilapia
Thailand semi-intensive Feed source Cassava only (main feed)
Food conversion FCR 1.7
Thailand intensive Feed source Cassava only (main feed)
Food conversion FCR 1.3
China intensive Feed source Wheat grains extensive at farm/CH cf livestock feed wheat
Feed source Livestock feed soy instead of soybeans at farm US
Feed source Soybeans at farm US only (main feed)
Shrimps and Prawns
China extensive inland Fertilizer Removed urea and TSP
semi-intensive Feed source Wheat only (main crop)
inland Feed source Replaced wheat grain organic CH by livestock feed wheat
Fertilizer Added urea and TSP (50-50)
intensive inland Feed source Replaced wheat grain organic CH by livestock feed wheat
Feed source Wheat only (main crop)
semi-intensive
coastal
Feed source Wheat only (main crop)
intensive coastal Feed source Wheat only (main crop)
Feed source Soy meal instead of husked nuts
On-farm energy Change on farm energy to be same as Thailand
Thailand intensive coastal Feed source Replace soybean meal Brazil at farm by soy meal
CH = Switzerland; FR = France; PH = Philippines; US = United States.

2. Impacts
Today
Impacts
Comparison Looking Forward Policy Appendix Glossary ReferencesSummary
37Managing the environmental costs of aquaculture
Results
Most of the results for our alternative models
differed relatively little from their baseline
counterparts (Figure 2.10). Of the 180 comparisons
that were made, 113 (63%) were within ± 10% of
their baseline value. Given that these comparisons
were chosen as those most likely to be sensitive to
our assumptions, this is encouraging.
There were, however, some notable deviations.
The most striking of these concern assumptions
about on-farm energy use in China for shrimp and
prawn farming. Using energy-use values equivalent
to those used for Thailand reduced impacts on
acidifi cation, climate change, land impact and
energy demand by between 50 and 60% over
baseline estimates. Other comparisons for shrimp
and prawn farmed were very similar to one another.
For tilapias, the only major deviations occurred
with respect to estimates of land occupancy for
intensive farming in China, which increased from
between 110 and 140% with altered assumptions
about feeds. For carps, changed assumptions
concerning on-farm energy use in India reduced
estimates of acidifi cation and climate change by
between 50 and 60%. A large (50%) increase in
estimates of land occupation also occurred when
feed supply assumptions were altered for intensive
carp production in China.
Overall, we conclude that our baseline models are
generally robust and are not overly sensitive to
model assumptions. In common with the fi ndings
of others, however, signifi cant sensitivities do
exist and can markedly affect results. This helps
point towards those areas for greatest immediate
attention. Improving estimates of on-farm energy
use in emerging economies, developing new
process descriptions for crop production in
developing countries and improving data on the
exact feed sources used for aquaculture are
particularly important.

2. Impacts
Today
Impacts
ComparisonLooking ForwardPolicyAppendixGlossaryReferences Summary
38 Managing the environmental costs of aquaculture
Figure 2.10: Summary of sensitivity analysis results. Details of comparison are given in Table 2.7 Red dots denote large deviations from baseline estimates.

2. Impacts
Today
Impacts
Comparison Looking Forward Policy Appendix Glossary ReferencesSummary
39Managing the environmental costs of aquaculture
Table 2.8: Comparison of results from other published studies. All values are per tonne live weight of product. Data in
parentheses are from the current study. Literature sources: 1. Pelletier et al., 2009; 2. Pelletier and Tyedmers, 2010; 3.
Bosma et al., 2009.
Study Source Energy Demand
(MJ-eq)
Climate Change (kg
CO
2
-eq)
Eutrophication (kg
PO
4
-eq)
Acidifi cation
(kg SO
2
-eq)
Salmon Norway 1 26,200 (23,300) 1,793 (1,290) 41.0 (66.1) 17.1 (6.38)
Salmon Chile 1 33,200 (26,700) 2,300 (1,520) 51.3 (73.5) 20.4 (7.26)
Salmon Canada 1 31,200 (22,300) 2,370 (1,850) 74.9 (75.0) 28.4 (13.5)
Salmon UK 1 47,900 (21,500) 3,270 (1,390) 62.4 (73.0) 29.7 (8.17)
Tilapia Indonesia 2 26,500 (33,300) 2,100 (2,010) 45.7 (131.0) 23.8 (70.4)
Catfi sh Vietnam 3 13,200 (215,000) 8,930 (23,100) 40.0 (89.0) 459.0 (150.0)
Comparisons with other LCA studies
As well as exploring the sensitivity of our results
to model assumptions and parameter estimates,
we can also ask how our results compare with
those from other studies. We can get some
insight into this question by comparing them with
those of the more detailed LCA studies that have
been undertaken for selected systems. Table 2.8
summarizes comparable fi ndings for studies on
salmon, tilapia and catfi sh.
In drawing these comparisons, we stress that
our system boundaries exclude medicine, seed
and fi ngerling production, and construction and
other processes. In contrast, the data we are
comparing them with come from cradle-to-farm-
gate LCAs, which include some or all of these
processes. These considerations, combined
with the high degree of complexity and choice
available when constructing LCAs, render ‘like
with like’, or benchmark comparisons with other
studies impossible. The value of our study is in
the comparative analysis across systems globally,
using a consistent, albeit coarse approach. The
comparisons below are offered, therefore, to
stimulate debate, rather than validate estimates.
Comparing data from these studies with our own
fi ndings (in parentheses in Table 2.8) we fi nd
considerable variation in the level of agreement
across impact categories and systems. While
broadly comparable, estimates from our four
salmon studies for energy use, climate change
and acidifi cation are consistently lower than those
published by Pelletier and co-workers. In contrast,
our estimates for eutrophication are consistently
higher. Examination of the inventory data for these
studies show that our input values for feed, on-
farm energy use, and nitrogen and phosphorus
emissions are very similar to these earlier studies.
This suggests, therefore, that the discrepancy is
largely due to the less comprehensive treatment of
feed formulation in our study.

2. Impacts
Today
Impacts
ComparisonLooking ForwardPolicyAppendixGlossaryReferences Summary
40 Managing the environmental costs of aquaculture
On a comparative basis the more detailed LCAs of
Pelletier and colleagues rank the UK as being the
least effi cient across all categories. In contrast, our
own analysis is much more variable. Again this may
refl ect the way feed issues have been treated in the
various studies, but it may also be a function of how
nitrogen and phosphorus emissions are treated.
For tilapia in semi-intensive systems in Indonesia,
our estimates for eutrophication and acidifi cation
are consistently and considerably higher than those
of Pelletier and Tyedmers (2010), but the largest
single difference is between the estimates of energy
demand for catfi sh in Vietnam.
Discussion
Life Cycle Analysis in aquaculture is in its early
stages and, of the few case studies available, most
focus on salmon. This is, perhaps, unsurprising
given the relatively dispersed and small, to medium,
scale nature of much of the industry and the fact
that so much of aquaculture production occurs in
developing countries.
The objective of the analysis described in this
section was to compare and contrast the global
and regional demands of aquaculture for a range
of biophysical resources across the suite of major
species and production systems in use today.
This complements the more detailed studies for
production of particular species. By undertaking
a broader scale scoping comparison we are
able to identify more clearly, and on a standard
methodological foundation:
1. How environmental impact compares across
systems and geographies.
2. Which species groups or production systems
are especially demanding on biophysical
resources.
3. How environmental performance differs among
countries for similar systems.
The distribution of absolute impact values
shows where greatest attention should be
paid for achieving environmental performance
improvements.
In many respects, our results are broadly consistent
with expectations. First, with explainable departures,
such as for bivalve and seaweed culture, absolute
impact levels correlate with overall levels of
production. As a consequence, when one looks at
the global picture in absolute terms, the impact of
Chinese aquaculture, and carp culture in particular,
stands out.
In contrast, relative effi ciencies in production by
species, system or country provide an indication
of the potential for performance improvement.
Of particular signifi cance in this regard are the
comparisons between species cultured in the
same system in different countries. Here we fi nd
considerable variance refl ecting a combination of
differences, both in production practices where
farm level choices and management may exert
signifi cant infl uence on ecological impacts, and in
systemic country specifi c conditions over which
fi sh farmers may have little control. One factor that
farmers cannot control, for example, is the mix
of energy sources used by a country to generate
electricity, which has impacts on climate change and
acidifi cation estimates.
To the extent that observed variances refl ect
differences in species and system choices and
management practices, we have an indication of the
potential for large improvements in effi ciency. Shared
learning of best practice across the industry should
provide signifi cant opportunities to close effi ciency
(productivity) gaps. It is perhaps unsurprising that
the salmon industry shows least variation across
both countries and impact categories (see Figure
2.8). The explanation for this almost certainly lies in
the greater investments in salmon farming research,
the global nature and competitiveness of the
industry and the fact that the sector is dominated
by a few large companies. This suggests that similar
research investments, combined with the right
institutional, policy and market drivers, could lead to
dramatic performance improvement in many other
aquaculture sub-sectors.
We return to these issues when we consider the
policy implications of this study. Before doing
so, however, we explore how production in the
aquaculture sector compares with that for other
animal food sources.

2. Impacts
Today
Impacts
Comparison Looking Forward Policy Appendix Glossary ReferencesSummary
41Managing the environmental costs of aquaculture
Photo by Francis Murray
CHINA

42 Managing the environmental costs of aquaculture
3. Comparison

4343
3. COMPARISON
PHOTO CREDIT: The WorldFish Center

3. Comparison
TodayImpacts
Comparison
Looking ForwardPolicyAppendixGlossaryReferences Summary
44 Managing the environmental costs of aquaculture
“there isn’t any more land. We are exploiting the
available production factors to a great extent. The
environment is becoming more polluted. Increased
production has to come from high-yielding
farming.” (Jacques Diouf, 2006 in Flachowsky,
2007)
The growing demand to consume animal products
continues to rise. This is particularly true of the
developing world where, between 1980 and
2005, the consumption of terrestrial animal
meat increased from 14.1 to 30.9 kg/capita; it is
predicted to increase further to 36.7 kg/capita by
2030 (FAO, 2009a, WHO, 2010). This growing
demand for animal products risks increasing
undesirable impacts on the environment.
Livestock meat production can be grouped into two
categories: ruminant species (such as cattle, sheep
and goats) and monogastric species (such as pigs
and poultry). Generally speaking, ruminant species
are either produced intensively or in extensive
grazing systems, while monogastrics are produced
in traditional or industrial systems (FAO, 2009a).
Four production systems, however, dominate
the sector: grazing, rain fed mixed (defi ned as a
combination of rain-fed crop and livestock farming),
irrigated mixed, and landless/industrial systems
(Steinfeld et al., 2006).
These species categories and production systems
place different demands on ecological goods and
services. For example, the traditional monogastric
production systems for chickens and pigs are
considered overall to have negligible environmental
impact due to their extensive nature, limited
manufactured feed demand and their dominant
position in small-scale household oriented
production systems. Intensive systems for pigs and
poultry, however, lead to greater impacts, although
they are less damaging than beef production
(see below). As detailed in Table 2.2, aquaculture
production systems also fall into several categories:
extensive, semi-intensive and intensive. As with
livestock these systems differ in the environmental
impacts they impose.
Because livestock farming is more established
as a major food production sector its impact on
the environment has received more attention than
aquaculture. In recent years, for example, a large
number of studies on the environmental impact
of livestock have been produced (FAO, 2009a).
In 2006, however, an early effort to compare the
environmental costs of aquaculture with those of
livestock was undertaken by the FAO (Bartley et
al., 2007). Such comparisons are important to
help ensure that the animal food production sector
develops in ways that use available resources
wisely. As the authors of the FAO report point out,
there is thus “a need to present a balanced picture
of the environmental costs of all food-producing
sectors and to formulate environmental policies
that deal with the impacts of all sectors... So long
as this balanced picture of environmental costs is
absent, policy does not refl ect farming realities, the
prices of food products cannot refl ect the real costs
of their production, especially for ecosystems and
communities, and both the public and government
receive very mixed messages [regarding policy
options]”. (ibid., p.5).
3. The environmental
effi ciencies of animal
production systems:
How does aquaculture
compare?

3.Comparison
Today Impacts
Comparison
Looking Forward Policy Appendix Glossary ReferencesSummary
45Managing the environmental costs of aquaculture
Although largely focused on methodological issues,
the FAO study provides some initial comparative
understanding. Here we briefl y summarize the
fi ndings from the FAO study along with other
available literature. We stress, however, that the
methodological foundations for such comparisons
remain under-developed and appropriate data are
sorely lacking.
Comparative analysis of
impacts
Conversion Effi ciencies
An important (and perhaps the clearest)
perspective on relative impacts of animal-source
food production can be obtained by considering
feed conversion ratios. From this perspective fi sh
come out well because, in general, they convert
more of the food they eat into body mass than
livestock. Poultry for example, convert about 18%
of their food and pigs about 13%; in contrast, fi sh
convert about 30% (Hasan and Halwart, 2009).
Much of this difference refl ects the fact that fi sh are
poikilotherms (cold blooded) and do not expend
energy maintaining a constant body temperature.
Moreover, because aquatic animals, especially
fi nfi sh, are physically supported by the aquatic
medium few resources are expended on bony
skeletal tissues. As a result the usable portions
of fi nfi sh are high compared to those of terrestrial
animals, especially cattle (Moffi tt, 2006). From
such principles, therefore, it would appear that the
environmental demands of fi nfi sh production will be
lower. This certainly appears to be the case when
comparing fi nfi sh with beef or pork. Looked at in
another way, the production of 1 kg beef protein
requires 61.1 kg of grain while 1 kg pork protein
requires 38 kg and 1 kg fi sh protein requires less
than 13.5 kg (calculated from White, 2000).
Of course, for species such as mussels and
oysters that grow on the natural productivity of
the ecosystem, the question of food conversion
effi ciency becomes moot. Although unlikely
to be a mainstream food commodity, in many
respects, these animal food sources are among
the most desirable from an ecological sustainability
perspective.
A complementary perspective on the question of
effi ciency is provided by Smil (2001) who compared
feed and protein conversion effi ciencies for several
animal based foods (Table 3.1). As with other
analyses, fi nfi sh come out favorably compared with
pork and beef, and are broadly comparable with
poultry and dairy products. With these superior
conversion ratios aquaculture may become a
signifi cant competitor to monogastric species in
regions such as South East Asia and sub-Saharan
Africa (Bartley et al., 2007).
Table 3.1: Protein content of major animal foods and feed conversion effi ciencies for their production. (Based on Figure
5 of Smil, 2001). Calculations of feed conversion effi ciencies based on average US feed requirements in 1999.
Commodity Milk Carp Eggs Chicken Pork Beef
Feed Conversion
(kg of feed/kg live weight)
0.7 1.5 3.8 2.3 5.9 12.7
Feed Conversion
(kg of feed/kg edible weight)
0.7 2.3 4.2 4.2 10.7 31.7
Protein Content
(% of edible weight)
3.5 18 13 20 14 15
Protein Conversion Effi ciency (%) 40 30 30 25 13 5

3. Comparison
TodayImpacts
Comparison
Looking ForwardPolicyAppendixGlossaryReferences Summary
46 Managing the environmental costs of aquaculture
Table 3.2: Percentage of world fi shmeal market use by sector. (Source: Fishmeal Information Network (FIN, accessed in
2010)).
2002 2007 2008 2010
Ruminants 1 - - <1
Pigs 24 24 31 20
Poultry 22 7 9 12
Fish 46 65 59 56
Others 7 4 1 12
A key concern with the intensifi cation of both the
fi sh and the livestock sectors is demand for fi shmeal
and fi sh oil in feed formulations (see Section 4).
Although farmed fi sh convert feeds more effi ciently
than livestock (Moffi tt, 2006; Brummett, 2007; FAO,
2009a), aquaculture is presently more dependent
on fi shmeal and fi sh oil than other animal production
sectors (Table 3.2). The share of fi shmeal used by
aquaculture grew from 8% in 1988 to about 35% in
2000 (Delgado et al., 2003) to 45% in 2005 (World
Bank, 2006) and estimated to be 56% in 2010.
Species such as salmon are particularly dependent,
because the main source for several essential fatty
acids is oily fi sh. Indeed, it is this dependency by
aquaculture and the growth of the aquaculture
sector that is believed to have forced the livestock
sector to search for other protein substitutes in
livestock feed (Bartley et al., 2007). Prohibiting
the use of animal offal in livestock feed to reduce
the risk of mad-cow disease, has also increased
pressure to produce vegetable protein for animal
feed. Recent estimates by the Fishmeal Information
Network indicate that 56% of world fi shmeal
production is now consumed by fi sh with 20% for
pigs and 12% for poultry (Table 3.2).
Although fi shmeal use is controversial in some
quarters, one must also recognize that substitution
with suitable land-based crops brings with it
demands on land and water use and perhaps the
production of a nutritionally inferior product to its
wild counterpart (Karapanagiotidis et al., 2006,
2010). As production methods intensify, and the
animal derives more of its nutritional requirements
from crop-based feedstuffs, total lipid levels tend to
rise and lipid profi les shift to become dominated by
less desirable omega-6 fatty acids.
Despite such concerns, however, the high cost
and limits to supply of fi shmeal and fi sh oil are
likely to drive the current trend of increased use of
crop substitutes in animal-source food production.
Soybean meal use rose from around 20 million
tonnes in the 1970s to over 120 million tonnes in
the early 2000s (Bartley et al., 2007) and further
increases in its use seem assured.
Environmental Emissions
With respect to environmental emissions, the
livestock sector is often characterized as having a
“severe impact on air, water and soil quality because
of its emissions” (de Vries and de Boer, 2010). It has
also received considerable attention as a contributor
of greenhouse gases (Steinfeld et al., 2006).
Extensive livestock systems contribute indirectly
through land degradation and deforestation, while
in intensive systems, the application of manure that
emits methane and enteric fermentation directly
contributes to climate change. All this said there
is considerable variation among meat production
systems and comparisons are fraught with diffi culty.
With the exception of poultry, however, it seems
likely that aquatic animal products have rather
less impact than other animal production systems
from an environmental emissions perspective.
This conclusion is further supported by the data
on nitrogen emissions shown in Table 3.3, which
show that, while emissions of waste nitrogen and
phosphorus vary considerably, aquaculture systems
generally perform well compared to beef and pork.

3.Comparison
Today Impacts
Comparison
Looking Forward Policy Appendix Glossary ReferencesSummary
47Managing the environmental costs of aquaculture
Table 3.3: Summary of data on nitrogen and phosphorus emissions for animal production systems. Data for beef, pork
and chicken are derived from Flachowsky (2002) in Postrk, 2003. Data for fi sh are derived from this study.
Commodity Nitrogen emissions (kg/tonne
protein produced)
Phosphorus emissions (kg/tonne
protein produced)
Beef 1200 180
Pork 800 120
Chicken 300 40
Fish (average) 360 102
Bivalves -27 -29
Carps 471 148
Catfi sh 415 122
Other fi nfi sh 474 153
Salmonids 284 71
Shrimps and prawns 309 78
Tilapia 593 172
Land Use
To compare land use we took our data on the land required to produce 1 tonne of edible fi sh product and
compared this with data provided by de Vries and de Boer (2010) who summarized the land required to
produce 1 tonne of edible beef, pork and chicken (Table 3.4). These data suggest that land use demands
are broadly comparable.
Table 3.4: Estimates of land demand (direct and indirect) for animal-source food production.
Commodity Yield tonne/ha (edible product)
Livestock
Beef 0.24 – 0.37
Chicken 1.0 – 1.20
Pork 0.83 – 1.10
Aquaculture
Bivalves
0.28 – 20
Carps 0.16 – 0.90
Catfi sh 0.20 – 1.23
Other fi nfi sh 0.38 – 3.70
Shrimps and prawns 0.34 – 1.56
Tilapia 0.15 – 3.30
Alternative approaches to calculating land use, however, come up with markedly different conclusions.
Based on an analysis for British Columbia summarized in Box 3.1, for example, Brooks (2007) concluded
that “the landscape directly affected for cattle production is several hundred times greater than it is for
production of the same amount of food in salmon aquaculture”. Such contrasting conclusions serve
to illustrate the complications of comparative analysis and point towards the importance of adopting a
standardized methodology that is explicit about the basis for calculation.
Environmental impacts associated with land use will also vary with the ecological values of land used, for
example grasslands, wetlands, mangroves and seagrass beds all providing different ecological services.
More detailed analysis is required to account for these differences.

3. Comparison
TodayImpacts
Comparison
Looking ForwardPolicyAppendixGlossaryReferences Summary
48 Managing the environmental costs of aquaculture
Water Use
Livestock production is a signifi cant user of freshwater
resources, with an estimated 8% of global human
water use devoted to the sector. While around 2% is
consumed through direct consumption the majority
(more than 98%) is primarily associated with the
production of feed crops (Verdegem et al., 2006). In
intensive systems where livestock are concentrated in
feedlots, water use is particularly high because of the
high demand for concentrated feed and additives that
require an increased production of raw materials such
as cereals and oil crops (Steinfeld et al., 2006). Current
published estimates suggest that producing 1 kg of
edible beef requires 15,500 l of water compared to
3,900 l for 1 kg of edible chicken (World Bank, 2010
7
)
and varies between 11,500 and 45,000 l for 1 kg of fi sh.
There are, however, a number of issues concerning
calculations of water consumption in food production
that make evaluation and comparisons diffi cult. For
example, much of the water used to produce crops is
‘green’ rather than ‘blue’ water; i.e. infi ltration and not
surface water from lakes or rivers is used (see Molden et
al., 2003; Verdegem and Bosma, 2009). The exception
is, of course, irrigated crop production.
Another complication arises because the bulk of
global aquaculture production is from semi-intensively
managed ponds. The majority of these ponds tend to
be fi lled and drained once per year with water added
periodically to counterbalance water lost through
seepage and evaporation. While one might consider
this water use, because it is needed for physical
support, to supply dissolved oxygen and for dispersal
and assimilation of wastes, one could also argue it to
be a form of water storage and that seepage losses
from ponds represent an ecosystem service, serving
to recharge groundwater reserves. The latter argument
only holds, however, if seepage is uncontaminated
by nitrogen and phosphorus wastes and preliminary
experiments suggest that nutrient uptake by sediments
is enhanced as seepage water moves through the pond
bottom interface (Verdegem et al., 2006). Of course,
coastal aquaculture has a further major advantage in
this respect in that it makes use of seawater.
Feed associated water use in aquaculture comes mainly
from the production of feed crops and grains.
Use associated with fi shmeal and fi sh oil and with other
feed sources (e.g. meat and bone meal) are negligible
(Verdegem and Bosma, 2009).
Conclusion
Because vegetarianism is unlikely to ever be a voluntary
choice for the overwhelming majority of people, as
global demand for food rises, fi nding ways to be more
ecologically effi cient consumers of animal food will
become increasingly important. Indeed, many would
7
Data in the literature usually refer to Pimentel et al. (2004) who assumes that the production of 1 kg of beef requires 100,000 l of water. These fi gures seem a little bit outdated.
The World Development Report provides more recent fi gures taken from www.waterfootprint.org (incl. direct and indirect water consumption).
Box 3.1
Brooks (2007) compared land use by salmon farming and
cattle rearing in the following way:
• The edible meat yield from an Angus steer is 42% of live
weight
• The yield of salmon fi lets is approximately 50% of the live
weight
• A salmon farm producing 2500 tonnes of live salmon
would supply 1250 tonnes of edible fi lets which is
equivalent to 5411 steers weighing 550 kg each.
• In the Pacifi c Northwest, one acre of actively managed
pasture supports one cow for 7.5 months (7.5 animal
month units or AMUs) and it takes approximately 30
months to produce a marketable steer.
• 5411 steers require 162338 AMUs or 8658 acres (3504
hectares) for 2.5 years.
• The substrate under well sited salmon farms chemically
remediates in six months to a year and biologically
remediates in another year showing a full return of the
normal benthic community.
• In contrast, in the Pacifi c Northwest, it will take hundreds
or a thousand years for the pastures to return to the
original old growth forest.
Edible
Portion
(kg)
Yield Footprint
(ha)
Remediation
Time (y)
Salmon 1,250,000 0.5 1.6 2
Angus
Beef
Cattle
1,250,000 0.42 6,982 200+

3.Comparison
Today Impacts
Comparison
Looking Forward Policy Appendix Glossary ReferencesSummary
49Managing the environmental costs of aquaculture
argue that it is essential if the ecological demands
of our food production systems are to remain within
acceptable bounds (e.g., Rockström et al., 2010).
Comparisons indicate that dairy foods can be
produced most effi ciently in terms of ‘feed protein
to food-protein conversion effi ciency’, but that
herbivorous fi sh from aquaculture, eggs, and chicken
come close. In contrast, pork production converts
feed protein to meat only about half as effi ciently.
Examining these issues from a nitrogen budget
perspective Smil (2001) concludes that American
beef cattle herds require at least fi ve to six times
the feed energy per unit of lean meat compared to
the country’s broiler population. As a consequence
its production also requires 5 to 6 times as much
nitrogen fertilizer to produce the requisite feed. Smil
estimates that the United States would have to
use less than half its concentrate feed, and hence
less than half of the N-fertilizer used to grow it, if its
protein-rich diet were composed of equal shares of
dairy products, eggs, chicken, pork and farmed fi sh.
Beyond the clear issues concerning beef production,
however, analyses indicate that there is no simple
answer to the question of which animal production
system has least environmental impact. Each system
makes different demands on environmental services
and the appropriate trade-offs between them relative
to the benefi ts of providing a particular form of animal
source food will be context specifi c. Clearly, aquatic
products have some advantages, not least the
effi ciency gains possible from farming a cold blooded
animal, but much depends on the species, systems
and management practices.
Available analyses also rarely make reference
to the variability that is found in the effi ciencies
associated with the various intensities and methods
of production used for the various animal products.
This is clearly an important consideration that bears
further examination, particularly because, with the
high demand put on resources, there is a trend in
intensifying animal farming rather than extensifying
it (Gerber et al., 2007). There are clearly trade-offs
between alternative approaches. Extensive systems
require more land and are more dependent on
ecosystem services for their productivity (freshwater,
fuel, food, water purifi cation, nutrient cycling,
etc.), while intensifi cation means more inputs and
effl uents and also more (fossil fuel) energy (Prein,
2007). We need to better understand and quantify
these trade-offs in order to better manage and
mitigate environmental impacts. Pathways for future
development of these sectors will clearly have a
signifi cant infl uence on future impacts, and targets
for management interventions.
In this context it is important to appreciate that, in
contrast to livestock, from a biophysical perspective
there remains considerable scope for aquaculture
expansion. Limits to land availability mean that
livestock production will only intensify, while
aquaculture will both intensify within the existing area
under production and grow into new areas.
Another issue one must consider is the potential for
integrated agriculture-aquaculture systems (e.g.,
poultry and carp) which, although not examined
using life cycle approaches, have been considered
more ecologically effi cient than monoculture systems
(Prein, 2007; Gabriel et al., 2007). There has been
a trend away from such systems in China, the
traditional home of integrated farming, due largely to
economic drivers, and the inability to recover value
from the ecosystem services they provide. A new
look at such systems using LCA tools is warranted,
but above a threshold size such systems may
become ineffi cient and diffi cult to manage. This may
limit the growth potential of these integrated systems.
Finally, while not a focus for this study, and not really
amenable to analysis using an LCA framework,
it is also important to recognize concerns over
biodiversity loss. The loss of biodiversity is a
signifi cant concern with livestock, with major issues
of overgrazing leading to erosion, desertifi cation
and tropical deforestation for conversion to pasture
(Brown, 2000). But, while the scale of habitat loss
in the livestock sector, with massive conversion of
habitat to extensive grazing, far outweighs that of the
aquaculture sector, aquaculture development can still
threaten biodiversity. These threats include habitat
loss in fi sh and shrimp nursery areas (e.g., Primavera,
2006), use of inland wetlands for conversion to
ponds, as seen in India and Bangladesh and risk
of genetic pollution from escape of farmed fi sh (see
also Section 4). Conversion to ponds in wetland
areas such as mangroves in particular can lead
to loss of ecosystem services, including loss of
carbon sequestration properties. For the most part,
managing these threats will require local studies
coupled with sound planning processes.

50 Managing the environmental costs of aquaculture
4. Looking Forward

5151
4. LOOKING FORWARD
PHOTO CREDIT: Randall Brummett

4. Looking Forward
TodayImpactsComparison
Looking Forward
PolicyAppendixGlossaryReferences Summary
52 Managing the environmental costs of aquaculture
With the stagnation or, optimistically, only limited
growth in wild catches any increase in demand for
fi sh can only be met by aquaculture (Delgado et
al., 2003; Bostock et al., 2010). But how big is the
aquaculture sector likely to become and what are
the environmental implications? In this section we
explore this question by fi rst examining the drivers
of increased demand for aquaculture products and
how are these likely to evolve in the coming years.
We then go on to briefl y review the sector’s efforts
to overcome some of the environmental constraints
to meeting this demand. Finally, we examine
published projections for how production by the
sector may evolve and examine the implications of
such growth for biophysical resource demands.
Demand drivers
Growth in population, wealth and urbanization
At fi rst sight, one would imagine that population
growth would be a major driver of increased fi sh
production. At present, however, world population
growth averages 1.17% per annum according to
The World Bank. This represents less than one fi fth
of the current rate of increase in global farmed fi sh
production. As a result, increased demand resulting
from population growth is currently a relatively minor
driver of fi sh production, at least in global terms.
A more important determinant of demand for fi sh
and other animal source foods is wealth (Speedy,
2003)
8
.
Increases in per capita consumption of animal
source foods are fastest where food consumption
levels are low, wealth and urbanization is increasing
rapidly, and domestic supply is also increasing
(Delgado et al., 1997). It is these factors that explain
the explosion of demand for meat, milk and fi sh
in the emerging economies of Asia. In China, for
example, the annual rate of population growth
is currently around 0.51%, adding an estimated
6.6 million people to its population each year.
And, although the growth of Chinese aquaculture
production is many times this rate, Speedy (2003)
estimates that, as a result of increased personal
wealth, demand is likely to increase from 25 kg
per person per year in 2005 to 35 kg per person
per year by 2020. And it may not just be wealth.
Although increased wealth is closely associated with
increased urbanization, urbanization per se may
also contribute to increases in animal source food
consumption. Delgado et al. (1997), for example,
suggests that changes in food preference driven by
urbanization alone has in the past accounted for an
extra 5.7–9.3 kg per capita consumption of meat
and fi sh per annum. Similarly, Betru and Kawashima
(2009) present data from Ethiopia indicating
urbanization affects animal food consumption rates
independently of income. In contrast, however,
Stage et al. (2010) present data from India and
China and cite studies from Vietnam and Tanzania
indicating that families with equivalent incomes
in rural and urban settings do not differ in their
consumption of animal source foods.
With growing wealth and urbanization as key
drivers of change in fi sh demand we can expect
the largest growing market over at least the next
decade to come from emerging economies. More
generally, global trends in urbanization, which
generally correlates with increased wealth, suggest
that developing country demands for fi sh will
increasingly dominate. By 2025, almost six out
of ten people on earth are likely to live in urban
centers, and over half of these will live in the cities of
developing countries. In 2009 there were 2.5 billion
urban dwellers in the developing world, compared
to 0.92 billion in the developed. By 2025 those
fi gures are expected to rise to 3.52 and 1 billion
respectively. This represents a shift in numerical
dominance from 72% of the world’s urban dwellers
4. Looking Forward
8
In economics parlance the demand for many animal source food products is ‘income elastic’, meaning that income growth increases
demand. Indeed, some animal source foods can even be considered luxury goods, meaning that a 1 % increase in income will lead to
an increase in demand of more than 1 %.

4. Looking Forward
Today Impacts Comparison Policy Appendix Glossary ReferencesSummary
Looking Forward
53Managing the environmental costs of aquaculture
4ap data  Open: treet4 ap and contriIutorZ, **B@ :A
0000 000
.Yo^tO9atL
living in the developing world today to 80% in 2030.
By 2050 the projections are for 5.19 billion in the
less developed regions and about 1.1 billion in the
developed world. Figure 4.1 summarizes the current
levels of urbanization and the projected annual
average growth rate to 2050.
Cultural factors and product attributes
Fish product attributes must also be considered in
the context of other foods. Growing recognition of
the health benefi ts of fi sh consumption, for example,
can alter patterns of demand relative to meat
products for some consumers, although the overall
importance of health information may be relatively
limited (Shroeter and Foster, 2004). Conversely,
concerns about mercury levels in carnivorous fi sh
such as salmon and tuna, have depressed demand
in some markets (Oken et al., 2003).
Product issues for other foods, also affect demand.
For example, Egypt has experienced a substitution
effect, in part a result of what happened to the
poultry sector. Poultry lost signifi cant market share
after 2006 because of fears of avian fl u, which
caused some 30 deaths in the country (WHO,
2010). Similarly, in Nigeria the avian fl u outbreak
led to a shift in consumer preference away from
poultry towards beef, pork and fi sh (Obayelu, 2007).
Future zoonotic or other animal health issues,
widely anticipated by experts due to increasing
intensifi cation of production methods and trade
liberalization, may have dramatic effects on markets
for animal derived foods. Depending on where
disease strikes this may either stimulate or reduce
demand for fi sh.
In the coming years we can expect demand side
processes such as seafood awareness, food safety,
quality convenience, sustainability and ethics to
become even more important. Trends will be driven
not only by developed country consumers, but
also by the growing middle class in the developing
world. While the signifi cance of such issues took
decades to appear among developed world
consumers it seems likely that the attitudes of
wealthier consumers in the developing world will
evolve much faster. Consumer trends in major Asian
markets, particularly China and Southeast Asia, are
currently poorly understood, but will have a major
infl uence on aquaculture production trends.
For developed countries, while overall demand
seems unlikely to change markedly, the value of
purchases is expected to rise through value addition
(Cressey, 2009) and aquaculture products will
continue to substitute for both expensive and cheap
wild fi sh products (see for example Beveridge et al.,
2010). The rise of supermarket chains in Asia, and
elsewhere in the developing world, will also have
Figure 4.1: The relative size of urban populations of countries in 2009 (indicated by circle size) and the projected annual
average rate of growth in urbanization to 2050 (indicated by shading). Data extracted from UN World Urbanization
Prospects 2009 Revision (UN, 2010).

4. Looking Forward
TodayImpactsComparison
Looking Forward
PolicyAppendixGlossaryReferences Summary
54 Managing the environmental costs of aquaculture
major implications for the many small producers
currently engaged in aquaculture production
(Reardon et al., 2010).
OECD countries represent a relatively small but
nonetheless important sector of the global market
for aquatic foods in view of their purchasing
power and demand. Increasingly, they not only
consume their own farmed aquatic foods but
also those of many developing countries (OECD,
2008, 2010). Much of the production of farmed
Vietnamese striped catfi sh, for example, is targeted
at EU member states where it has gained rapid
market penetration as a cheap substitute for the
increasingly expensive marine white fi sh traditionally
supplied from domestic fi sheries. Striped catfi sh is
often promoted by supermarkets and sold as highly
profi table convenience products such as seafood
pies or ready-to-cook breaded fi lets. We can also
expect other inexpensive farmed species such
as tilapia to penetrate wealthy western markets
provided the following conditions are met:
• Fish continues to be considered as a healthy
option to other animal food sources
• Trade policies that affect farmed fi sh continue
to be liberalized
• Developing country aquaculture producers can
continue to meet wealthy country food safety
standards
• Supermarkets continue to capture signifi cant
economic benefi t from the value chains and
thus continue to develop and market value-
added convenience products
• Farmed aquatic foods can be produced and
brought to markets in environmentally sound
ways
• Pricing continues to make aquaculture a com-
petitive animal source food.
Price
Demand for fi sh depends on the price of the
product. Most often fi sh products are what the
economists term own-price elastic, meaning that
when the price falls, people buy more. However,
it is not only changes in the price of fi sh that
matter, but also the changes in the prices of
competing (substitute) food products. The trend
in prices over the past 15-20 years has been for
food fi sh prices to rise, although not for several
aquaculture products, such as salmon. In contrast,
red meat prices have fallen by approximately
50% over the same period. Although data are
scant, it would appear that the prices for capture
fi sheries products have increased, but those of
aquaculture products have decreased. Salmon
and shrimp for example, previously considered
high value products, are now signifi cantly lower in
price, and have broadened their consumer base
tremendously.
Although predicting how absolute and relative
prices of meat, fi sh and milk will evolve and affect
consumer choice is diffi cult, some quantitative
projections have been attempted. The Fish to
2020 analysis by Delgado et al. (2003) provides
perhaps the most comprehensive recent attempt.
This analysis concluded, as one would expect
given urbanization and economic growth trends,
that China and India will lead the global growth in
per capita consumption, with 1.3 and 0.9% per
year, respectively. Other developing countries of
Southeast Asia and Latin America are in the middle
rank with 0.4 and 0.5% growth respectively. The
rest of the world is likely to see static or declining
per capita consumption. Supported by the World
Bank, efforts are now underway by to update these
projections and forecast trends out to 2030.
Environmental constraints to
sector growth
The last decade has seen a dominant narrative
arguing that aquaculture growth will be constrained
by local environmental factors and the carrying
capacity of the environments where production
occurs (Hempel, 1993; WRI, 1998). This view
has been re-enforced by evidence from several
intensive production sectors. We have seen major
disease outbreaks in the prawn and salmon
industries (Flegel, 1997; Wiwchar, 2005; Kautsky
et al., 2000), evidence of genetic pollution and
transmission of parasites and disease to wild
salmon stocks (Pearson and Black, 2001), and
habitat destruction, eutrophication and antibiotic
pollution in many systems (Emerson, 1999).

4. Looking Forward
Today Impacts Comparison Policy Appendix Glossary ReferencesSummary
Looking Forward
55Managing the environmental costs of aquaculture
The second is government regulation, which is
essential for limiting the impact of those effects
that do not affect the productivity of the industry
itself (e.g., limits to pollutants in effl uents). A third
driver, currently favored by NGOs such as WWF in
western markets, is to move the sector towards
environmental improvements by raising retailer and
consumer awareness of environmental impacts.
Driven by profi t, intensifi cation has only been
possible because prevailing economics have
allowed increased reliance on nutritionally complete
feeds and energy-intensive technologies, such as
aeration and oxygen injection. These production
innovations have depended largely on private
sector investment. This trend is likely to continue.
For many parts of the industry, we are likely to
see considerable increases in intensifi cation in the
coming decades and new approaches for handling
environmental concerns.
One innovation that is, at fi rst glance particularly
attractive from an environmental standpoint, is the
development of Recirculation Aquaculture Systems
(RAS). Such systems offer a high degree of control
over environmental variables, and high levels of
biosecurity and waste treatment. They are of
particular interest for locations close to consumer
markets. However, while the virtues of urban RAS
have been promoted for some time (Costa Pierce
et al., 2005) they have yet to fulfi ll their potential.
RAS are highly complex with high capital and
operational expenditure and have not always
operated reliably or profi tably. They also have high
energy demands and carbon footprints although
these could be reduced by use of non fossil fuel
energy sources (wind energy, solar, etc). With
However, while these concerns are undoubtedly
legitimate, there are signs that such problems
are commonly confi ned to the early stages of
intensifi cation and can be overcome as the sector
matures (Asche, 2008). Reduction in pollution
with organic wastes (per tonne of fi sh produced)
in the Norwegian salmon industry, for example,
appears to be related to industry growth (Tveterås,
2002). With the development of new vaccines, the
absolute volume of antibiotics used in Norwegian
salmon production also declined markedly despite
continuing production increases (Figure 4.2). i
In most cases there are two drivers that stimulate
an aquaculture sector to address environmental
constraints (Asche, 2008). The fi rst is the reduction
in productivity and hence profi t that results from
the negative feedbacks from the effects of a
deteriorating production environment on fi sh
health and increased risk of disease outbreaks.
Year
Production (Tonnes x 1000)
Antibiotics
Production
Antibiotic Use (kg x 1000)
500
400
300
200
100
1980 1985
1990 1995 2000
20
40
60
Figure 4.2: The rise and decline of antibiotic use in the Norwegian salmon industry compared to the trend of rising
production (adapted from Asche, 2008).

4. Looking Forward
TodayImpactsComparison
Looking Forward
PolicyAppendixGlossaryReferences Summary
56 Managing the environmental costs of aquaculture
little take-up of the technology, there is minimal
incentive or revenue stream for suppliers to invest
in the necessary development and manufacturing
capacity for standard mass-produced low-cost
systems.
While intensifi cation of the currently dominant
systems will undoubtedly continue, there is also
interest in using the abundant areas off-shore to
reduce environmental pressure. Cage (synonymous
with ‘pen’) systems dominate the production
of high value marine fi sh species, especially in
Europe, North and South America. As a result of
climate change, and competition for near-shore
coastal areas (with accompanying concerns about
their local environmental impact in some parts of
the world), some investment has been made in the
design of offshore cage systems able to withstand
the extreme wave and wind climates associated
with more exposed environments. Such systems
rely on stronger materials, more robust designs and
integrated cage and mooring systems that allow
cages to be submerged below the water surface to
avoid hostile weather conditions (Beveridge, 2004;
Grøttum and Beveridge, 2007) . Although these
technologies will continue to be developed they
are unlikely to result in any signifi cant expansion of
production in view of the high capital and operating
costs and the limited market for the high value
farmed fi sh that can be produced in such systems.
Feeds
Despite the trend in intensifi cation of production
methods the majority of aquaculture production
is still derived from extensive and semi-intensive
aquaculture of omnivores and herbivores. There
are powerful economic incentives to intensify
production, however, and we can expect to see
increasing dependence on feeds. This brings
with it concerns about the resultant demands
on biophysical resources and impacts on food
security.
The bulk of aquaculture feedstuffs are of crop
origin—maize, soya, wheat—and crop production
makes substantial demands on ecosystem
services (Tilman et al., 2002). Using such materials
to feed fi sh and shrimp may lead to competition
for use of the same materials for human food or
bio-fuels, with consequent implications for prices
and affordability. It may also lead to changes in
crop production (e.g., change in land use from
growing human food staples to production of
aquaculture feedstuffs). Demand on ecosystem
services may be further exacerbated by the
global trade in the feeds and feedstuffs that
sustain aquaculture production. For example, the
Egyptian aquaculture industry uses an estimated
1 million tonnes of aquaculture feed per annum.
All feedstuff ingredients are imported, primarily
from North America, which may add to the overall
environmental cost of production.
Other important aquaculture feedstuffs include
‘trash’ fi sh, fi shmeal and fi sh oil, derived from
industrial and artisanal fi sheries, and widely
used to sustain shrimp and carnivorous fi sh
production (Tilman et al., 2002). Fishmeal and oil
are particularly important for these species groups
because they require long-chain fatty acids that are
only found in high amounts in these feed sources.
There are concerns that these ‘feedfi sh fi sheries’
aggravate food security in parts of the world by
diverting fi sh from direct human consumption to
aquaculture. It appears, however, that, while there
is considerable scope to increase the proportion of
feedfi sh for human consumption in Latin America,
the situation is more ambiguous in Asia where
use of such feedstuffs in small-scale aquaculture
disadvantages some but has considerable
livelihood benefi ts for others (Huntington and
Hasan, 2010).
Notwithstanding these concerns the track record of
innovation to deal with these resource constraints
is impressive in those parts of the aquaculture
sector where industry competition has driven
effi ciency increases. This is most evident in the
salmon industry where production costs have
declined dramatically. In Norway, for example,
production costs have decreased by 60% in
the last 20 years. Although reductions in labor
demand account for a substantial proportion of
this, technical innovation to improve, for example,
feeding effi ciencies is also signifi cant (Subasinghe
et al., 2003). Decreasing dietary fi shmeal and fi sh
oil inclusion in aquaculture feeds and limiting their
use to starter, broodstock and ‘fi nisher’ feeds
are among the most immediately implementable
strategies for further effi ciency improvements
(Tacon and Metian, 2008). This may in time be
complemented by selective breeding. Fish have the
ability—albeit limited—to de-saturate and elongate
lipids, which varies not only among species but

4. Looking Forward
Today Impacts Comparison Policy Appendix Glossary ReferencesSummary
Looking Forward
57Managing the environmental costs of aquaculture
also families. Identifying the genes that control this
and determining the heritability of the trait may
facilitate selective breeding of strains with reduced
dependence on fi sh oils (Aquaculture News, 2009).
Last, long promised microalgal based technologies
capable of producing commercial quantities of
affordable material that can substitute for fi shmeal
and fi sh oils in aquaculture feedstuffs may be
beginning to become commercially viable (Durham,
2010).
Aquaculture will increasingly have to compete
with other animal production sectors for use of
feedstuff crops and agricultural by-products. The
sector will be able to continue to secure access
only if it can afford to pay the going rate and if the
roles of aquaculture in food security and economic
development are suffi ciently recognized to have
resulted in an enabling policy environment.
Genetics, selective breeding and Genetically
Modifi ed Organisms
Aquaculture production is almost entirely
comprised of plants and animals derived from
broodstock that have been in captivity for only a
few generations. As a result, growth of farmed
aquatic organisms is similar to, or because of poor
management of captive breeding systems, worse
than that of their wild counterparts (Brummett
et al., 2004). Domestication, in which life history
traits are altered through selective breeding to
meet human needs, affords the possibility to
develop more productive (i.e., fast growing, disease
resistant, high fl esh yield) strains. The development
of faster growing strains reduces demands on
some ecosystem services, such as land and water.
However, although yet to be thoroughly studied it
is probable that the development of faster growing
strains, as being pursued at present, will have only
little effect on the demand for feed. In essence
current breeding programs primarily select for fi sh
that eat more, not explicitly for fi sh that convert
food more effi ciently into fl esh. It may, however, be
possible to widen breeding objectives to select for
both faster growth and better feed utilization.
Farming provides the opportunity to infl uence every
aspect of the life cycle of an animal, including many
of the attributes that might appeal to consumers:
color, size, shape, nutritional composition. The
relative importance of genes in determining many
of these attributes, however, is as yet unknown as
is our understanding of the genes involved or the
heritability of these traits. Powerful new tools, such
as genetic markers, are expected to increasingly
assist us in identifying these genes and gene
complexes.
At present, genetic improvement programs
are underway for a dozen or so widely farmed
species, including both marine shrimps and
freshwater prawns, common and Indian major
carps, tilapias, African and channel catfi sh, rainbow
trout and Atlantic salmon. Results from such
selective breeding programs can be impressive:
the selectively bred Jayanti strain of La beo rohita
(‘rohu’), for example, widely used by Indian farmers,
grew up to 17% faster per generation over fi ve
generations compared with local strains, across a
range of production environments (Ponzoni et al.,
2009).
The fi rst genetically modifi ed (GM) farmed fi sh is
a strain of Atlantic salmon that grows twice as
fast as other domesticated strains. Produced by
AquaBounty Technologies, it is currently awaiting
approval for commercial production by the U.S.
Food and Drug Administration (USFDA). The
animal has a single copy of a DNA sequence that
includes code for a Chinook growth gene as well
as regulatory sequences derived from Chinook
salmon and ocean pout (Marris, 2010). Several
other aquaculture species await permission for
commercial use, including common carp in China
(Aldhous, 2010). The permitting process has until
recently taken many years, but in 2009 the USFDA
announced that they intended to treat GM traits
in farmed animals as veterinary drugs, potentially
speeding up the licensing process. Nevertheless,
strong public concern about the potential for
adverse environmental effects should fi sh escape
and breed with wild fi sh is likely to infl uence
licensing arrangements. GM technology will only
be adopted in aquaculture if it results in lower
production costs, greater profi ts and expanded
markets. Market size will, however, ultimately
depend on the perceived safety of the product to
consumers and, indeed, with the brand image of
GM foods in general.

4. Looking Forward
TodayImpactsComparison
Looking Forward
PolicyAppendixGlossaryReferences Summary
58 Managing the environmental costs of aquaculture
Another issue with respect to genetics concerns
non-native species. A precautionary approach
would, of course, severely restrict the use of alien
species in aquaculture and rely instead on the
development of native stocks. Currently, however, a
considerable proportion of aquaculture production
comes from non-natives (Figure 4.3). Even in China,
where native carps dominate production, 12% of
production comes from non-natives.
Recognizing that the current incentives for use of
alien species in aquaculture remain high, particularly
for developing countries, future efforts will need
to be directed towards improving risk assessment
and mitigation measures. Based on the FAO Code
of Conduct for Responsible Fisheries (1995) and
the ICES Code of Practice on the Introductions
and Transfers of Marine Organisms (2005), IUCN
provides a useful series of recommendations for
national governments to implement responsible use
of alien species in aquaculture (Hewitt et al., 2006).
Tools for risk analysis associated with introductions
of aquatic animals are also available (Kapuscinski,
2007; Arthur et al, 2009).
Figure 4.3: Summary of non-native species production for the systems modeled in this study. This calculation
excludes seaweeds and accounts for 90% of global production in 2008. Values under each country are production
( x 1000 t).

4. Looking Forward
Today Impacts Comparison Policy Appendix Glossary ReferencesSummary
Looking Forward
59Managing the environmental costs of aquaculture
Fish Health
Aquaculture production methods are increasingly
intensifying and farms are getting larger and more
spatially concentrated. Because of this, there is a
growing concern about increasing risks from the
spread of pathogens and infectious aquatic animal
diseases and the increased movement of aquatic
animals. Inter-regional trade and the introduction
of new species and strains to meet economic and
market demands both pose signifi cant risks. The
use of trash fi sh is also a risk factor in the transfer
of pathogens. Current estimates suggest that
between one third to a half of fi sh and shrimps
put into cages or ponds are lost to poor health
management before they reach marketable size
(Tan et al., 2006).
Although technologies and measures for aquatic
animal disease prevention, control and treatment
have improved signifi cantly in recent years, abuse
of antimicrobials and other veterinary drugs and
associated environmental and human health
risks remain a major concern. Antimicrobials and
other medicines are of particular concern given
their importance for human health. Uneaten
feed provides a source of these contaminants to
the environment, while ingested medicines are
metabolized, excreted or voided in feces.
Accumulation of residues from these sources can
increase antimicrobial resistance in farmed fi sh.
Impaired decomposition of organic material in the
environment because of declines in bacterial fl ora
can also occur. Disease prevention often proves
diffi cult and many farmers currently focus more
on treatment than prevention, but increased use
of antimicrobials as prophylactics and as growth
promoters is possible in future. This will further
increase the risks of developing new, drug-resistant
strains of pathogens. Developing vaccines is
one route to reducing use of veterinary drugs,
but research in this area is currently restricted to
relatively few species (e.g., salmon, trout, grouper)
and vaccines are only effective against certain
types of disease.
Environmental stressors, such as poor water
quality, acting alone or in conjunction with
other stressors such as over-crowding, poor
handling or inadequate nutrition, compromise the
immunity of farmed aquatic animals, increasing
their susceptibility to attacks by pathogens
present in the farmed environment. Increasingly,
the aquaculture industry and others—national
governments, the FAO, the OIE—recognize that
effective biosecurity measures are needed to
reduce the spread of pathogens. Adequate welfare
standards are also required to minimize stress and
reduce the incidence of disease and its consequent
impacts on production and profi ts. Two other
factors are also important. First, environmental
standards have been developed for many of the
compounds used as medicines by aquaculture,
and have been widely disseminated, if perhaps less
widely enforced. Second, food safety standards,
designed to protect consumers from exposure to
potentially harmful medicinal and other chemical
residues, are driving more responsible use. Such
standards are more widely used by developed
countries, and for products from developing
countries for export to them, but many developing
countries will need to apply the same or similar
regulations to protect their domestic consumers.
Industry codes of practice may help, but legislation
and its implementation, combined with capacity
building, are also needed.
Climate change
Climate change – aquaculture interactions are
two-way: climate change affects aquaculture,
and aquaculture contributes to climate change
(Figure 4.4). The fi gure below illustrates that the
impact of climate change on the sector and those
who depend on it and vice versa is moderated by
a range of other external factors which may be
occurring at the same time (Beveridge and Phillips,
2010).
Figure 4.4: The relationship between aquaculture and
climate change. (From Beveridge and Phillips, 2010)
CLIMATE CHANGE AQUACULTURE
population
growth
energy prices
environmental
deterioration
health
inter-sectoral
competition
economic
recession
conflict
impacts on
impacts on
trade
restrictions

4. Looking Forward
TodayImpactsComparison
Looking Forward
PolicyAppendixGlossaryReferences Summary
60 Managing the environmental costs of aquaculture
Climate change is likely to increase global seawater
temperatures. Combined with sea level rises,
changes can be expected in inshore salinities,
currents and seawater mixing patterns, and in wind
speeds and direction. The changes in the physico-
chemical environment will impact on ecosystem
structure and function—the distribution of species,
aquatic productivity and the incidence of harmful
algal blooms. Coastal areas and estuaries are likely
to experience the greatest changes in biophysical
conditions and ecology. Inland, changes in the
levels and pattern of precipitation are likely to
increase the incidence of fl ooding in some areas
and drought in others and impact on groundwater
and surface water reserves. Temperature rises
will increase evaporative water losses, change
stratifi cation and mixing patterns of lakes, aquatic
community composition and aquatic productivity
(for reviews see Handisyde et al., 2006; Allison et
al., 2009; Brierley and Kingsford, 2009; Cheung et
al., 2009; Beveridge et al., 2010).
Temperature changes can be expected to impact
not only on the aquatic environments that support
aquaculture production but also on the farming
operations themselves. Temperature increases
will increase productivity especially in areas where
anthropogenic nutrient inputs are increasing. The
incidence of harmful algal blooms, however, is
also likely to increase, limiting bivalve and other
types of culture. Moreover, above some critical
point elevated temperatures stress farmed aquatic
animals suffi ciently to markedly impact survival,
reproduction, growth, production, and profi ts.
Climate change will thus directly affect aquaculture
production through choice of species, location,
technology and production costs. Development
of heat tolerant strains is likely to be limited given
the complex interactions between temperature
and physiology. In short, adaptation strategies to
climate change are likely to be limited. Instead,
we can expect geographic winners and losers.
Aquaculture production will disappear from areas
that become too hot, dry or stormy while areas
presently considered as excessively cold may
benefi t, as is anticipated in coastal Norway.
With respect to the impact of aquaculture on
climate change, perhaps the most specifi c
effect concerns the use of wetlands and coastal
mangroves . These habitats sequester high levels
of carbon, and efforts are needed to ensure that
any aquaculture should be sited in areas which
such areas does not compromise such natural
carbon sinks.
Production projections
“Aquaculture production has continually
outstripped projections, and there is little reason
to believe that it will not continue to do so.” (ARD,
2006)
The global picture
Notwithstanding our historic tendency to under-
estimate the rise of aquaculture, several projections
of future production are available. We have drawn
on these to examine likely future trends. Figure 4.5
shows actual aquaculture production up to 2008
(excluding seaweeds) against the values projected
under various scenarios from published studies
summarized in an analysis for the FAO (Brugère
and Ridler, 2004). The various projections have
been made under somewhat different assumptions
and approaches. Two of the forecasts (Ye, 1999;
Wijkström, 2003) assume constant fi sh prices and
are based solely on demand driven by population
growth and per capita consumption. In contrast,
both supply and demand considerations and their
effects on prices are included in the analysis by
IFPRI (Delgado et al., 2003), which disaggregated
food fi sh into high and low value categories on the
basis of their markets and price elasticities.

4. Looking Forward
Today Impacts Comparison Policy Appendix Glossary ReferencesSummary
Looking Forward
61Managing the environmental costs of aquaculture
The studies by Delgado et al. (2003) and Ye (1999)
consider alternative scenarios for the future.
The IFPRI study explored six scenarios, three of
which are considered here: a baseline scenario
that embodied the authors “most plausible” set
of assumptions, an extreme scenario where
capture fi sheries production, including fi shmeal
fi sheries collapse with a minus 1% annual growth
in production, and an aquaculture development
scenario where technological progress increases
production growth by 50% relative to the baseline
scenario. Ye (1999) considered two scenarios:
the fi rst assumed per capita consumption would
remain at 1996 levels, the second that it would rise
to 22.5 kg/y, based on a combination of historical
time trends and modeled relationships between
GDP growth and consumption. Further richness to
these predictions was added by Brugère and Ridler
(2004) who considered how these projections
might be affected by either no growth in wild
capture fi sheries or by a modest 0.7% growth.
Examining these various projections in relation
to observed trends in production we derive
an uncertainty envelope for total aquaculture
production out to 2030 in the following way (Figure
4.6). Because the three projections up to 2015
fall broadly on the current growth trajectory for
production, there is consensus among the studies
that global production growth will continue along a
similar trajectory to the recent past for the next fi ve
years or so.
Figure 4.5: Comparison of historical trends in production of farmed fi sh with several projections of future aquaculture
production. Circles denote projections based on supply and demand considerations under various assumptions, as
summarized in Table 3 of Brugère and Ridler (2004). Historical production data are from FAOStat.
Year
Production (million tonnes)
20
40
60
1950 1960 1970 1980 1990
2000 2010 2020 2030
80
100
120
FAO (2004)
Wijkstrom (2003)
IFPRI (2003)
Ye (1999)
Fish
•Baseline scenario
•Technological advances in aquaculture
•Ecological collapse of fisheries
• Global consumption remains at
1996 levels (15.6 kg/y)
• Global consumption rises to 22.5 kg/y
Growing fisheries (0.7% per annum)
Stagnant fisheries
Production (million tones)

4. Looking Forward
TodayImpactsComparison
Looking Forward
PolicyAppendixGlossaryReferences Summary
62 Managing the environmental costs of aquaculture
20
40
60
1950
1960
1970
1980
1990
2000
2010
2020
2030
80
100
120
Pig
Chicken
Fish
Production forecast (this study)
Production targets (national data)
• Global consumption rises to 22.5 kg/y
• Technological advances in aquaculture
• Baseline scenario
• Ecological collapse of fisheries
• Global consumption remains at
1996 levels (15.6 kg/y)
Production (million tonnes)
Year
Wijkstrom (2003)
FAO (2004)
IFPRI (2003)
Ye (1999)
Predictions for the latter half of the decade are
variable, but if continued growth to 2015 holds
we will have surpassed all but the most optimistic
of the IFPRI scenarios to 2020. Thus, assuming
that we do not see the catastrophic collapse of
wild fi sheries assumed by the most pessimistic
scenario, but that we also see no growth in
this sector
9
(Mills et al., 2010), the envelope for
production by 2020 is between 65 and 85 million
tonnes. The lower bound of this range corresponds
to the IFPRI baseline scenario under a stagnant
fi sheries assumption and the upper bound
refl ects the continuation of the current production
trend and the prediction for IFPRI technological
innovation scenario under a stagnant fi sheries
assumption.
The bounds of uncertainty become even greater
as we look out to 2030. For this time horizon,
and in the absence of a new modeling effort, a
conservative envelope is probably between 79 and
110 million tonnes. The lower bound represents
a growth pattern that continues the trajectory for
the IFPRI baseline scenario prediction for 2020.
The upper bound represents the continuation
of the current production trend and the IFPRI
technological innovation scenario under a stagnant
fi sheries assumption. It also corresponds to the
midpoint between the two projections by Ye for
global consumption of 22.5kg.
One indication of the reasonableness of this likely
envelope for the aquaculture production trajectory
comes from a comparison with the targets for
Figure 4.6: Comparison of historical trends in farmed fi sh, pig and chicken meat production, the likely production trajec-
tory envelope and the combined aquaculture production targets envelope for nine countries (Bangladesh, India, China,
Indonesia, Philippines, Thailand, Vietnam, Brazil, Chile, Canada, Egypt). Historical production data are from FAOStat,
production target data are from Table 9 of Brugère and Ridler (2004). Aquaculture production predictions from Figure
4.5 are also shown.
9
Although we assume no growth in the real supply of fi sh from the wild capture sector, we do envisage an increase in the supply reported in offi cial statistics in coming years, in
particular as better data on small-scale fi sheries becomes available (Mills et al., 2010).

4. Looking Forward
Today Impacts Comparison Policy Appendix Glossary ReferencesSummary
Looking Forward
63Managing the environmental costs of aquaculture
production that were identifi ed in the national plans
of nine countries (Brugère and Ridler, 2004, Table
9). Figure 4.6 compares the envelopes for these
projections and shows that our estimated range
falls below the collective ambitions of these nine
countries. The envelope for production targets
was created based on two scenarios—an annual
growth rate for China of 3.5%, or a more modest
rate of 2% (Brugère and Ridler, 2004). Although
national targets are often over-optimistic there is
little to indicate that the aquaculture sector as a
whole will be unable to meet demand should it
eventuate.
It is also interesting to examine how pig and
chicken meat production has evolved and to
observe the remarkably similar growth rates for
production over the last decade (Figure 4.6). This
suggests, perhaps, that all three sectors have been
driven by similar demand drivers during this period
and that all three production systems have been
able to meet this demand.
Geographic distribution
The global distribution of production described here
for 2008 is likely to still hold in 2010, moderated
somewhat by some recent large changes (e.g.,
marked declines in Chile; marked increases in
some sub-Saharan African states). For the next
fi ve years, therefore, we may further assume that
the present global pattern of production will remain
largely unchanged: i.e. that Asia will account for
more than 90% of production, Europe for around
3–4% and South America, North America and
Africa for 2% each, and Oceania for a fraction of a
percentage point. Indeed, one can expect Asia to
further consolidate its position by a few percentage
points at the expense of the rest of the world.
The regional distribution of aquaculture production
growth beyond the next fi ve years is more diffi cult
to predict. Three factors are particularly signifi cant.
First, the industry is now a major global provider of
food which increasingly must compete for markets
with other sources of animal-derived foods, all
of which are changing too in response to market
globalization. Second, like other food production
sectors, aquaculture depends on a range of scarce
or fi nite resources for which it must increasingly
compete with others. Third, the sector is fi nally
beginning to be taken seriously at policy level;
governments are starting to develop and apply
incentives and penalties to facilitate or regulate
sectoral growth, the methods by which it is
achieved, and trade. They are doing this to ensure
that the sector makes appropriate contributions
to social, economic and environmental objectives.
Given these considerations and the complicated
relations these factors will have with production
costs and price to consumers one must be
cautious with defi nitive statements about how the
sector will evolve geographically.
There are, however, several conclusions that are
probably robust. First, despite the investment,
aquaculture production in Europe and North
America has remained largely static over the past
decade and is unlikely to grow substantially. This is
primarily due to lack of available sites, competition
from other producing countries and substitution of
comparatively expensive, domestically produced
fi sh such as cod by cheaper products from other
parts of the world (striped catfi sh from Vietnam,
tilapia from China). Marine production in the United
States remains constrained by lack of an enabling
legal framework, competition for coastal resources
and competition from overseas producers (e.g.,
Latin America and Asia for shrimp). Similarly,
freshwater production in the United States is limited
by overseas producers able to produce identical
(tilapias, carps) or substitute products (striped
catfi sh) at highly competitive prices.
Second, production in Africa is very low but is
growing fast in some countries, unconstrained
by resources that are often underutilized. Despite
the fact that fi sh is the most important source of
animal protein per capita for many countries in
this region and provides several essential vitamins
and nutrients, fi sh consumption is the lowest in
the world. Here it is projected that simply to keep
pace with population growth a further 1.6 million
tonnes—almost 10 times the current production
levels—will be needed by 2016 (Beveridge et al.,
2010). Growth in sub-Saharan Africa is increasingly
being driven by investors in countries such as
Uganda, Nigeria and Ghana, keen to develop
enterprise type operations that target both
domestic and regional markets (OECD, 2010).
However, because of the very low production base
and because of ineffi cient and poorly developed
value chains, it is likely to take at least a decade
before substantial increases in production in sub-
Saharan Africa are realized. If this is correct, local

4. Looking Forward
TodayImpactsComparison
Looking Forward
PolicyAppendixGlossaryReferences Summary
64 Managing the environmental costs of aquaculture
aquaculture production will be unable to fi ll the gap
between fi sh supply and demand that Africa faces
over the next decade. Despite this overall picture,
however, there will be large local increases in some
countries and this will likely bring with it substantial
resource demands.
Third, the current trends indicate that the majority
of increases in global production to 2030 will come
from South and Southeast Asia and China, with a
continued drive by major producer countries such
as China and Vietnam towards export to the strong
European and North American markets. Increased
import taxation, such as that currently being
imposed by the United States against Vietnamese
farmed striped catfi sh, can be expected to
periodically moderate this trade (Worldfi shing
and Aquaculture, 2010), but the general trend
is clear. The principal constraint to growth in
production in the region, other than markets, is
likely to be availability of resources (land, water) and
environmental change.
Finally, of the countries in the Asian region, it is
China where biophysical constraints seem most
likely to slow the rate of production growth. While
China is likely to further consolidate its position
as the world’s largest producer and consumer of
farmed aquatic products, the resource base upon
which this production depends will come under
increasing pressure. As a consequence, it is diffi cult
to imagine how current production growth rates
can be maintained in the longer term. Balanced
against this, however, will be considerable pressure
to satisfy internal demand through domestic
aquaculture production. While domestic production
will meet some of this need, increasing imports can
also be expected, some of which may be supplied
by Chinese overseas aquaculture investments.
The implications of sector growth for
biophysical resource demands
To explore and illustrate the consequences of
current production practices for future biophysical
demands of aquaculture might develop we have
constructed a scenario in which production from
our modeled systems (excluding seaweeds) will
reach 100 million tons by 2030. We chose 100
million tonnes as a landmark fi gure and because
it falls on approximately the upper quartile of
our uncertainty envelope. Given the tendency
of previous work to under-estimate aquaculture
growth choosing a fi gure in the upper part of the
range seems reasonable. We also made two other
assumptions to avoid projecting forward trends
that we believe are unlikely to persist and which
have high leverage on the predicted environmental
demands:
1. Production in China and striped catfi sh
production in Vietnam will slow faster than in
other countries owing to pressure on natural
resources
10
.
2. Whitefi sh production will grow relatively faster
than other forms owing to increasing demand
for this product category.
To estimate the distribution of global production,
a scaled estimate of the recent (2003 – 2008)
compound annual production growth rate was
used to project forward production from the 2008
starting value for each production system. For
all production systems the same scaling factor
of approximately 0.42 was used for all years and
systems. For China, we reduced production growth
rates by a further 50% and for catfi sh in Vietnam
by 90%. For all white fi sh products we increased
growth rates by 20%.
Results
Figure 4.7 summarizes the change in geographic
distribution of overall production between 2008
and 2030 under our growth scenario. The key
feature of this result is the continued dominance by
Asia, but the emergence of several other countries
(India, Indonesia and Thailand) as key players.
For Asia as a whole, this conclusion is almost
certainly robust, although how production will be
distributed across countries is far less certain given
the dynamic nature of the sector. The spectacular
rise to dominance in catfi sh production by Vietnam
in recent years is a testament to how quickly things
can change.
10
Although catfi sh demand may well be met by producers in countries such as Myanmar, India and Bangladesh, we have not included this is our projections.

4. Looking Forward
Today Impacts Comparison Policy Appendix Glossary ReferencesSummary
Looking Forward
65Managing the environmental costs of aquaculture
Figure 4.7: Projected change in production distribution between 2008 and 2030 for the systems modeled in this study,
which produced 82% of world production in 2008 (data exclude seaweeds). Blue circles: 2008 production; orange
circles: 2030 production.
Table 4.1 summarizes the change in overall environmental impact for each of our six categories. Increases
in impact are between 99 and 168% over the 22 year period. Precisely what this will mean for countries
and regions, is of course diffi cult to imagine but to put it in perspective, if the climate change contribution
from aquaculture were offset at current market price of $15 per tonne of CO
2
, the cost would rise from US$
43.6 billion in 2008 to US$ 101.19 billion in 2030. The largest projected change is for eutrophication, which
rose by 168%. This suggests that meeting demands for fi sh products into the future will require particular
attention to issues of waste disposal. Of course, these projections assume current (2008) practices,
whereas improved technologies, regulatory regimes and production practices should modify this trend; see
earlier discussions on intensifi cation.
Table 4.1: Projected change in total environmental impact between 2008 and 2030 for the systems modeled in this
study, which produced 82% of world production in 2008 (data exclude seaweeds, and assumes current production
practices).
Year Eutrophication
(Mt PO
4
eq)
Acidifi cation
(Mt SO
2
eq)
Climate
Change
(Mt CO
2
eq)
Land
Occupation
(Mha)
Energy
Demand
(Tj eq)
Biotic
Depletion
(Mt)
2008 3.57 2.54 291.2 50.61 3,358,468 15.11
2030 9.55 5.05 674. 6 113.63 7,622,647 37.88
% Change 168% 99% 132% 125% 127% 151%
2008
2030
Year
5
0
,
00
0
10
,
000
,
000
2
0
,
000
,
000
3
0
,
000
,
000
4
0
,
000
,
000
5
0
,
000
,
000
(
)
(
)
(
)
(
)
(
)
(
)

4. Looking Forward
TodayImpactsComparison
Looking Forward
PolicyAppendixGlossaryReferences Summary
66 Managing the environmental costs of aquaculture
Figure 4.8 shows the distribution of impact for each of our impact categories in 2008 and 2030. As we
would expect these distributions map broadly to overall production levels, re-affi rming the importance of
focused support to Asian producers to mitigate the environmental impacts of aquaculture.
Conclusions
In this section we have explored the drivers of demand for aquaculture products and the environmental
constraints to meeting this demand. We then examined published projections of future growth. These
suggest that aquaculture production is likely to increase at a rapid pace. Finally, we explored the future
environmental demands of aquaculture if it reached 100 million tonnes (excluding seaweeds) and in the
absence of signifi cant innovation and improvements in techniques and technology. Under this scenario
we estimate that the environmental demands will be between 2 and 2.5 times greater than 2008 levels by
2030 for all the impact categories studied.
Figure 4.8: Projected change in distribution of environmental impact between 2008 and 2030 for the systems modeled
in this study (data exclude seaweeds). Blue circles: 2008 production; orange circles: 2030 production.
4ap data  Open:treet4ap and contriIutorZ, **B@ :A 4 ap data  Open: treet4ap and contriIutorZ, **B@ :A  4 ap data  Open: treet4 ap and contriIutorZ, ** B@ :A
4ap data  Open:treet4ap and contriIutorZ, **B@ :A 4 ap data  Open: treet4ap and contriIutorZ, **B@ :A  4 ap data  Open: treet4 ap and contriIutorZ, ** B@ :A
Eutrophication
*limate *hange
Land Occupation
Energy Demand
Biotic Depletion
Acidification

4. Looking Forward
Today Impacts Comparison Policy Appendix Glossary ReferencesSummary
Looking Forward
67Managing the environmental costs of aquaculture
Photo by Stevie Mann
MALAWI

68 Managing the environmental costs of aquaculture
5. Policy

69
5. POLICY
PHOTO CREDIT: The WorldFish Center

5. Policy
TodayImpactsComparisonLooking Forward
Policy
AppendixGlossaryReferences Summary
70 Managing the environmental costs of aquaculture
Understanding, quantifying and explaining the
environmental impacts of aquaculture is essential
for sound decision making. Policy-makers need
this information to establish evidence based and
fair environmental regulations. Fish farmers need
it to implement better management practices
and understand and comply with environmental
regulations. And retailers and consumers need it
to make informed choices and drive appropriate
policy and farming practices.
In this section we distill the results of our LCA study
into seven policy relevant fi ndings. For each of
these fi ndings we then offer one or more specifi c
recommendations for action. Following this we offer
a more general conclusion and recommendations
regarding the future of aquaculture. We then
combine and further amplify our recommendations
for key stakeholder groups (Table 5.1) before
considering the future research investments that
are needed to support sector development.
Study fi ndings
Finding 1. Environmental impact is strongly
correlated with overall production levels.
The absolute levels of environmental impact
revealed by this study indicate those regions and
production systems where efforts to regulate and
reduce global environmental demands are best
targeted. Based on these fi ndings international
agencies and institutions should:
• Develop approaches to encourage and sup-
port China and other Asian and Latin American
countries to analyze impacts and better man-
age the sector towards improved environmen-
tal performance.
• Focus especially on improving produc-
tion practices in inland pond, pen and cage
aquaculture because these dominate global
production.
• Focus especially on carps, shrimps and
prawns as these are among the sectors which
have the largest overall impacts in absolute
terms.
The study also shows that the “other fi nfi sh”
sector has high aggregate impact. Unfortunately
this sector comprises many species, making a
common approach diffi cult to develop. Recent
comparative analyses of impacts in the marine
fi nfi sh sector, however, have begun to tease this
issue apart (Volpe et al., 2010).
Finding 2. Aquaculture systems vary markedly
in their environmental performance, offering
great potential for improvement.
The highly regulated nature of the salmon farming
industry in some countries has led to considerable
technical innovation that has both driven down
costs and reduced environmental impact. This
sector offers some lessons for the rest of the
industry, as do many of the traditional systems of
aquaculture in Asia with their low environmental
impacts.
More generally, the potential benefi ts of leveraging
cross-sector and cross-country learning deserves
close attention as one of the most effective means
for driving improvement. In view of this international
agencies and regional bodies and government
agencies should:
• Support or develop national and regional learn-
ing networks and innovation platforms for both
policies and technologies that bring together
government, the private sector, NGOs and
research agencies to jointly identify and imple-
ment solutions that will overcome problems,
establish and share best practices, and im-
prove sector wide environmental performance.
• Support the research needed to defi ne and
develop practical measures for implementing
the Ecosystem Approach to Aquaculture that
has recently been developed by the FAO.
5. Policy Implications and
Recommendations

5. Policy
Today Impacts Comparison Appendix Glossary ReferencesSummary Looking Forward
Policy
71Managing the environmental costs of aquaculture
• Support emerging aquaculture sectors to
understand cost drivers as a means to stimu-
late innovation and the uptake of more effi cient
production practices.
• Facilitate private sector investment in improving
environmental performance.
Finding 3. Use of fi shmeal and fi sh oils is
widespread and reducing dependency on
this resource requires a concerted focus on
innovation in the feed sector.
Reducing the fi shmeal and fi sh oil component in
aquaculture feeds is a high priority for intensive and
semi-intensive systems. This is true for traditional
fi shmeal and fi sh users such as salmon, but also
for other emerging industries such as tilapia, catfi sh
and shrimp. A range of largely complementary
strategies based on the following principles and
recommendations is needed to reduce feed
constraints on sector development:
• Use locally sourced feedstuffs, including
agricultural by-products (oil cakes, rice bran),
and develop pre-treatment and processing
methods to increase digestibility and nutrient
availability and reduce anti-nutrients.
• Make better use of scarce and costly fi shmeal
and fi sh oil supplies by restricting their use to
when it is a dietary essential or in fi nishing diets
to improve the nutritional value of the product
for consumers.
• Breed fi sh that have more limited demand for
high quality marine lipids and protein.
• Develop systems of intensifi cation for species
such as carps and tilapia that will not rely on
fi shmeal and fi sh oils.
• Develop high quality protein and lipid sources
from plants and microorganisms.
• Develop feeding technologies and manage-
ment systems to optimize the conversion of
feeds into aquatic animal biomass.
Finding 4. Reducing many impacts requires
responses that are generic.
The above recommendations are specifi c to the
aquaculture sector. There are, however, many
steps that the sector can take that are more
generic in nature. Our analysis shows, for example,
that reducing the sector’s impact on climate
change and acidifi cation is best served by adopting
generic energy effi ciency measures throughout the
value chain. In view of this government agencies
should:
• Facilitate energy and other resource use audits
(e.g., water) across aquaculture value chains
to help identify options for effi ciency gains and
cost savings.
• Where practicable, help make available to
producers energy and other resource use
data for their operations on a daily basis. This
would help drive effi cient practices, especially
if combined with comparative data for other
producers.
• Facilitate cross-sectoral dialogue on industry
best practice in the food and agriculture sector.
Finding 5. Fish farming is an ecologically com-
petitive option for producing animal source
foods.
From an ecological effi ciency and environmental
impact perspective the benefi ts of fi sh farming
relative to several other animal source foods
are clear. For many regions, an increase in the
production of fi sh, poultry and dairy products
relative to meat is likely to make more effi cient
use of available resources. These products are
especially suited to meeting the demand of growing
urban populations (including the urban poor)
through local peri-urban production.
In view of this national planning agencies should:
• Examine thoroughly the relative benefi ts of the
various animal production sectors and consid-
er policy drivers that can shift towards a more
ecologically effi cient production portfolio.
Recommending an aquaculture species choice
based on our analysis is diffi cult because the
picture that emerges is somewhat mixed. Eels
are particularly demanding in relative terms, albeit

5. Policy
TodayImpactsComparisonLooking Forward
Policy
AppendixGlossaryReferences Summary
72 Managing the environmental costs of aquaculture
with very low overall production, and shrimps and
prawns and catfi sh generally have higher impact.
Yet they all perform favorably in terms of resource
demands compared to meat. Bivalve and mollusk
farming is the least ecologically demanding of the
animal source foods and provides an ecological
service by removing nutrients. These groups
are a particularly nutritious and environmentally
sustainable option for consumers.
Finding 6. Aquaculture is likely to be an
increasingly important contributor to food
and nutrition security in developing countries
where there is culture of fi sh consumption.
The contribution of fi sh to food and nutrition
security will become increasingly important in the
developing countries. This is particularly true for
African and Asian countries where there is growing
domestic and regional demand, especially from
the growing urban populations, including the urban
poor. In view of this, governments and industry in
these countries will need to pay particular attention
to:
• Stimulating the private sector to invest in
commercial aquaculture where there is access
to strong demand in domestic and regional
markets.
• Evaluating research and policy development
needs along the entire value chain from inputs
to consumer markets.
• Supporting development of aquaculture pro-
duction that will deliver sustained supplies at
affordable prices for poor consumers.
• Supporting aquaculture both as a household
livelihood and food and nutrition security
support strategy in areas where production is
feasible, but markets are weak.
Finding 7. Climate change cannot be ignored.
Without further and more wide ranging analysis it
is diffi cult to anticipate the degree to which climate
change will affect global aquaculture production. To
more fully assess climate change impacts on the
sector, a value chain approach must be adopted
in which not only production but also essential
upstream and downstream activities (e.g., seed
and feed supply, transport and processing) are
included. To make matters even more complex,
climate change will interact with other factors
such as population growth, changes in markets,
trade barriers and energy prices to impact on
aquaculture and aquaculture-related food security.
Aquaculture also affects climate change; although
it is a relatively small contributor to greenhouse gas
generation. To sustain present and future markets,
especially in developed countries, the sector must
minimize its potential for climate change impact.
Certain key principles should be universally applied:
• Avoid use for aquaculture of sites high in
sequestered carbon (mangroves, seagrass,
forests).
• Organically enriched fi sh pond sediments, a
potentially important source of methanogen-
esis, must be carefully dealt with, preferably for
producing other foods.
• Energy consumption associated with pumping
and post-harvest processing, transport and
marketing must be minimized.
Tools such as Life Cycle Analysis (LCA) can help
identify the most energy-consuming steps in value
chains and evidence from other sectors suggests
that often mitigation may not be that costly. But
fi scal and economic incentives may be needed
to encourage changes, and ultimately it may be
consumers who, through exercising choice in what
they eat, play the most important role in promoting
mitigation.
General conclusion
The trends in many of the drivers of demand for
aquaculture products suggest that the aquaculture
sector will continue to grow to meet increasing
demand for fi sh products. The environmental
impacts of such growth can be managed through
innovation, strengthened policy, capacity building
and monitoring.
Increasing wealth and urbanization will result
in rising demand for farmed fi sh in the coming
decades. At a global scale, there is every indication
that the aquaculture sector will be capable of
meeting this demand. This will occur through
both expansion of areas under cultivation and

5. Policy
Today Impacts Comparison Appendix Glossary ReferencesSummary Looking Forward
Policy
73Managing the environmental costs of aquaculture
intensifi cation of production. But to achieve these
increases in ways that limit environmental impacts
we offer four core recommendations to government
and industry in all producer countries:
1. Continue to support innovation in the
aquaculture sector, especially the development
of productive technologies that make best use
of land and water and feed resources and that
minimize demands on environmental services.
2. Ensure that the regulatory environment
keeps pace with sector development and
support policy analysis and development that
internalizes into aquaculture enterprises the
costs of its environmental impacts.
3. Develop capacity in national agencies for
supporting the development of sector
regulation and for monitoring and compliance.
4. Monitor carefully how supply and demand
for fi sh is evolving to ensure that support
and investment is appropriate to the market
opportunity.
These core recommendations apply globally,
but there are regional differences in their relative
importance for attention over the next three to
fi ve years. Based on the fi ndings of this study,
literature review and our own experience, Figure
5.1 summarizes our view of these differences.
Figure 5.1: Core recommendations for government and industry in all producer countries and their relative importance
for each region.
Focus Core Recommendation
1. Innovation
Continue to support innovation in the aquaculture sector, especially the development of productive
technologies that make best use of land and water and feed resources and that minimize demands on
environmental services.
2. Regulation
Ensure the regulatory environment keeps pace with sector development and support policy analysis and
development that internalizes into aquaculture enterprises costs of environmental impacts.
3. Monitoring and compliance
Develop capacity in national agencies for supporting the development of sector regulation and for
monitoring and compliance.
4. Supply and demand analysis
Monitor carefully how supply and demand for fish is evolving to ensure that support and investment is
appropriate to the market opportunity.
K
E
E
Y
Maintain current emphasis Warrants increased attention and
investment
Requires significant increased
attention and investment
A top priority for attention and
investment

5. Policy
TodayImpactsComparisonLooking Forward
Policy
AppendixGlossaryReferences Summary
74 Managing the environmental costs of aquaculture
Table 5.1: Recommendations summarized for key stakeholder groups.
Stakeholder Group Recommendations
Policy makers
• Use audits of energy and other ecological resources across aquaculture value chains
as a guide for management decisions.
• Make information on energy and other ecological resource impacts and effi ciency
measures accessible to producers.
• Review and improve certifi cation standards, Good Aquaculture Practice, Codes of
Practices and other industry management codes and guidance documents to ensure
they refl ect ecologically effi cient approaches to farm management and value chains.
• Facilitate cross-sectoral comparisons and dialogue on best practices in food
production within the livestock, fi sheries and agriculture sectors.
• Examine thoroughly the relative benefi ts of the various animal production sectors and
consider policy drivers that can shift towards a more ecologically effi cient production
portfolio.
• Avoid siting aquaculture farms in those wetland or coastal ecosystems with high values
as sinks for sequestration of carbon, other greenhouse gases or nutrients.
Development and environmental
organizations
• Encourage and support China and other Asian and Latin American countries to better
manage the sector towards improved environmental performance.
• Continue to encourage adoption in practice and policy of the Ecosystem Approach to
Aquaculture.
• Monitor performance of certifi cation in the aquaculture sector, and seek ways to
support and improve systems to deliver environmental improvements at scale.
• Support development of regional knowledge sharing and learning networks for both
policies and technologies.
• Invest now in improvements in aquaculture technologies in Africa that will help set an
ecologically sound foundation for future aquaculture growth.
• Pay particular attention to carps, shrimps and prawns.
• Pay particular attention to pond culture systems and to pen and cage systems in
freshwater; focus on improving inland pond aquaculture.
• Continue to engage and seek to partner with key retail chains to improve the ecological
performance of the sector.
Private sector operators and
investors
• Make better use of scarce and costly fi shmeal and fi sh oil supplies.
• Avoid using areas high in sequestered carbon for aquaculture.
• Use locally sourced feedstuffs and develop pre-treatment and processing methods to
increase digestibility and nutrient availability and reduce anti-nutrients.
• Breed fi sh that have more limited demand for high quality marine lipids and protein.
• Deal carefully with organically enriched fi sh pond sediments.
• Minimize energy consumption on-farm and in the following value chain.

5. Policy
Today Impacts Comparison Appendix Glossary ReferencesSummary Looking Forward
Policy
75Managing the environmental costs of aquaculture
Photo by Mark Prein
BANGLADESH

5. Policy
TodayImpactsComparisonLooking Forward
Policy
AppendixGlossaryReferences Summary
76 Managing the environmental costs of aquaculture
Research needs
Acting on the above recommendations should be
guided by sound science and implementing many
will benefi t considerably from further research. In
this section we summarize the fi ve research foci
that we think are most important.
1. Support the adoption of inter- and intra-
sectoral best practice in environmental
performance by improving the knowledge
base.
The analysis presented here indicates major
differences in environmental resource demands
within and between countries, species and farming
systems. This indicates major opportunities for
improving ecological performance. Research is
needed to identify the better performers, combined
with fi eld verifi cation, to align incentives and
investments that will drive improvement.
Life Cycle Analyses, the methods of Volpe et al.
(2010), certifi cation standards and the Ecosystem
Approach to Aquaculture are being used in various
ways to measure performance and encourage
improvement. Further work is needed, however,
to improve the consistency and comparability
of fi ndings across the aquaculture sector and
to provide practical guidance to farmers and
regulators. The research needed includes:
• Developing a common and comprehensive
analytical framework to facilitate comparisons
of animal source food production systems that
captures impacts on key planetary boundar-
ies, such as the nitrogen cycle, biodiversity and
climate change.
• Developing cost-effective LCA-based indica-
tors for measuring ecological performance
status and improvements that can be applied
across scales, from farm to global levels.
• Developing LCA indicators for use with inte-
grated farming systems and identifying in-
centives (e.g., economic, policy, markets) to
improve the ecological performance of inte-
grated aquaculture and agriculture at farm and
landscape levels.
• Improving the LCA database on systems that
are currently poorly covered by global datas-
ets — focus fi rst on major production systems
in major producing countries (e.g., carps in
China, Bangladesh; products for domestic
markets).
• Determining the environmental benefi ts of cer-
tifi cation using LCA tools, to identify improve-
ments in certifi cation standards.
• Determining how emerging supermarket chains
in Asia and other entry points can be used to
improve the environmental performance of
aquaculture products for domestic or regional
markets.
• Carrying out more in-depth LCA studies on
trends in intensifi cation, choice of farmed spe-
cies, system design and management practic-
es, to understand entry points for improvement
and costs.
• Identifying the present frontiers of environ-
mental performance and what can be done to
support their adoption.
• Identifying which investment strategies, in-
centives and institutional arrangements best
facilitate environmental improvement among
small- and medium-sized enterprises.
2. Improve modeling and understanding of
demand for farmed aquatic foods.
While there is strong evidence that the aquaculture
sector will continue to grow to meet the anticipated
increasing demand for farmed aquatic products,
policy makers, producers and retailers need to
better understand the drivers of fi sh consumption.
This will require improved quantitative models
of fi sh supply and demand. The Fish to 2030
initiative that is currently being supported by the
World Bank, is particularly welcome in this regard.
Research is also needed to ensure that policies
designed to help meet demand for aquaculture
produce are consistent with policy objectives
for other sectors, such as environment, energy,
food and nutritional security, and poverty and that
policies are consistent at national and regional
levels.

5. Policy
Today Impacts Comparison Appendix Glossary ReferencesSummary Looking Forward
Policy
77Managing the environmental costs of aquaculture
3. Provide guidance to help reduce
environmental impact in high production
domains.
Research is needed to help China and other
Asian and Latin American countries better
manage the aquaculture sector towards improved
environmental performance. Because carp and
shrimp and prawn aquaculture have among the
largest overall impacts in absolute terms and
pond and cage production systems dominate
global aquaculture, efforts should focus on these
commodities and systems. Attention should be
paid to both technological and management
interventions, and the incentives (e.g., policies,
legislation, taxation, market) that produce the
greatest environmental benefi ts.
Work in this area should also build on the recent
efforts of Volpe et al. (2010) to further disaggregate
the “other fi nfi sh” category, which has high
aggregate impact, to help identify the species and
systems to focus on.
4. Innovate in the feed sector to reduce
dependency on fi shmeal and fi sh oils.
Feed contributes a high proportion of the
ecological footprint in many aquaculture systems,
including impact on biodiversity. Further nutritional
research is required to reduce dependency on wild
fi sheries as ingredients in aquaculture feeds. At the
same time, replacement by other ingredients (e.g.,
internationally sourced plant ingredients) can lead
to ecological resource demands that could offset
any environmental improvements from fi shmeal or
oil replacement. Further research on aquaculture
feeds using the LCA tool would be useful to identify
feed and feed management strategies leading to
genuine improved environmental performance.
5. Better integrate climate change
considerations into the aquaculture sector.
The specter of climate change demands that we
better understand how it will affect food security,
at national, regional and global scales and whether
this will affect demand and supply of aquaculture
produce. Work is also needed to determine how
the impacts of aquaculture on climate change
can be mitigated and whether emerging funding
mechanisms for climate change mitigation and
adaptation can be used to support environmental
improvements in developing country aquaculture.
The bottom line
Aquaculture is one of the most environmentally
effi cient ways to produce the animal source foods
that a growing and urbanizing world population
needs. It is one of the fastest growing food
production sectors in the world and demand for
aquaculture production will most likely continue to
grow with rapid pace. But increasing production
will have increasing environmental costs unless
developed in a way that minimizes the demand on
the environment.
This study is the fi rst to provide a global picture of
the demands fi sh farming makes on environmental
resources using Life Cycle Analysis. It shows that
there are huge opportunities for improvement in
ecological performance across countries, regions
and species groups. But we will only capture
these opportunities if governments, businesses,
non-government actors and researchers take
steps together to improve production systems
and techniques, invest in innovation, especially
to reduce reliance on fi sh meal and oils, and
strengthen regulation including improving
monitoring and compliance.
If we do these three things we can make
aquaculture a more sustainable endeavor that uses
biophysical resources prudently so that it can play
its role fully in meeting our future needs for fi sh.

6. Appendix
Today SummaryImpactsComparisonLooking ForwardPolicy
Appendix
Glossary
References
78 Managing the environmental costs of aquaculture
Systems modelled in this
study
Country Habitat Species Group Production
System
Intensity Feed Regime Production
2008
Bangladesh Inland Carps Ponds Extensive Natural 173521
Intensive Pellet 83547
Semi-Intensive Mash 385602
Canada Coastal Salmonids Cages & Pens Intensive Pellet 73260
Chile Coastal Salmonids Cages & Pens Intensive Pellet 627878
China Coastal Bivalves Bottom culture Extensive Extractor 3348250
Off-Bottom
Culture
Extensive Extractor 5713407
Ponds Extensive Extractor 750112
Crabs and Lobsters Cages & Pens Extensive Trash 197655
Gastropods Off-Bottom
Culture
Extensive Natural 224967
Other fi nfi sh Cages & Pens Intensive Trash 78141
Semi-Intensive Trash 470175
Other Invertebrates Ponds Semi-Intensive Mash 196575
Aquatic Plants Off-Bottom
Culture
Extensive Extractor 9703005
Shrimps and
Prawns
Ponds Intensive Pellet 95275
Semi-Intensive Pellet 539893
Inland Bivalves Ponds Extensive Extractor 89392
Carps Ponds Extensive Natural 3325593
Intensive Pellet 1801363
Semi-Intensive Mash 8729682
Catfi sh Ponds Extensive Natural 337334
Semi-Intensive Mash 337334
Crabs and Lobsters Cages & Pens Semi-Intensive Pellet 518357
Eels Ponds Intensive Paste 417454
Gastropods Off-Bottom
Culture
Extensive Natural 93629
Other fi nfi sh Cages & Pens Semi-Intensive Mash 2225936
Other Vertebrates Ponds Intensive Pellet 286010
Shrimps and
Prawns
Ponds Extensive Natural 124004
Intensive Pellet 62002
Semi-Intensive Pellet 1054041
Tilapias Ponds Intensive Pellet 1110298
Ecuador Coastal Shrimps and
Prawns
Ponds Semi-Intensive Pellet 150000
Egypt Coastal Other fi nfi sh Ponds Semi-Intensive Pellet 58650
Tilapias Ponds Intensive Pellet 43575
Semi-Intensive Mash 283238
Inland Other fi nfi sh Ponds Semi-Intensive Pellet 150663

6. Appendix
TodaySummary Impacts Comparison Looking Forward Policy
Appendix
Glossary
References
79Managing the environmental costs of aquaculture
Country Habitat Species Group Production
System
Intensity Feed Regime Production
2008
India Inland Carps Ponds Extensive Natural 863927
Intensive Pellet 415965
Semi-Intensive Mash 1919839
Indonesia Coastal Other fi nfi sh Ponds Semi-Intensive Pellet 277002
Aquatic Plants Off-Bottom
Culture
Extensive Extractor 1937591
Shrimps and
Prawns
Ponds Extensive Natural 113431
Intensive Pellet 141789
Semi-Intensive Pellet 28357
Inland Catfi sh Ponds Intensive Pellet 86556
Semi-Intensive Mash 129835
Tilapias Ponds Extensive Natural 72358
Intensive Pellet 14471
Semi-Intensive Mash 202603
Japan Coastal Bivalves Off-Bottom
Culture
Extensive Extractor 416000
Other fi nfi sh Cages & Pens Intensive Trash 229300
Aquatic Plants Off-Bottom
Culture
Extensive Extractor 337900
Korea, Dem.
Rep.
Coastal Aquatic Plants Off-Bottom
Culture
Extensive Extractor 444300
Korea, Rep. Coastal Bivalves Off-Bottom
Culture
Extensive Extractor 317418
Aquatic Plants Off-Bottom
Culture
Extensive Extractor 381076
Mexico Coastal Shrimps and
Prawns
Ponds Semi-Intensive Pellet 121601
Norway Coastal Salmonids Cages & Pens Intensive Pellet 818292
Philippines Coastal Other fi nfi sh Ponds Extensive Natural 245117
Intensive Pellet 30639
Semi-Intensive Mash 30639
Aquatic Plants Off-Bottom
Culture
Extensive Extractor 1422691
Inland Tilapias Ponds Extensive Natural 24193
Intensive Pellet 24193
Semi-Intensive Mash 193546
Thailand Coastal Bivalves Bottom culture Extensive Extractor 65439
Off-Bottom
Culture
Extensive Extractor 239946
Shrimps and
Prawns
Ponds Intensive Pellet 485800
Inland Tilapias Ponds Intensive Pellet 27275
Semi-Intensive Mash 182536
UK Coastal Salmonids Cages & Pens Intensive Pellet 128744
USA Inland Catfi sh Ponds Intensive Pellet 233564
Viet Nam Coastal Shrimps and
Prawns
Ponds Extensive Natural 288894
Intensive Pellet 9738
Semi-Intensive Pellet 22722
Inland Catfi sh Ponds Intensive Pellet 1250000

7. Glossary
Today SummaryImpactsComparisonLooking ForwardPolicyAppendix
Glossary
References
80 Managing the environmental costs of aquaculture
Glossary
Acidifi cation
A process that happens when compounds like ammonia, nitrogen oxides and sulphur dioxides are
converted in a chemical reaction into acidic substances. The Acidifi cation Potential (AP) is expressed
relative to the acidifying effect of SO
2.
Algal bloom
A sudden and rapid increase in biomass of the plankton population. Seasonal blooms are essential for the
aquatic system productivity. Sporadic plankton blooms can be toxic.
Alien species
A species occurring in an area to which it is not native.
Aquaculture
The farming of aquatic organisms in inland and coastal areas, involving intervention in the rearing process
to enhance production and the individual or corporate ownership of the stock being cultivated.
Benthic
Of or relating to or happening on the bottom under a body of water.
Biodiversity
The variability among living organisms from all sources including, inter alia, terrestrial, marine and other
aquatic ecosystems and the ecological complexes of which they are a part: this includes diversity within
species, between species and of ecosystems.
Biophysical resources
Resources such as soil, nutrients, water, plants and animals.
Biotic depletion
The volume of wild fi sh required to support observed aquaculture production.
Bivalves
Common name for a class of aquatic mollusks characterized by two calcareous valves joined by a fl exible
ligament along a hinge line. This class includes various edible species, many of which are cultivated (e.g.
mussels, oysters, scallops, clams).
Cage culture
Culture of stocks in cages. Cages are rearing facilities enclosed on the bottom as well as on the sides by
wooden, mesh or net screens. They allows natural water exchange through the lateral sides and in most
cases below the cage.

7. Glossary
TodaySummary Impacts Comparison Looking Forward Policy Appendix
Glossary
References
81Managing the environmental costs of aquaculture
Coastal aquaculture
The cultivation of aquatic organisms where the end product is raised in brackish and marine waters; earlier
stages of the life cycle of these species may be spent in fresh waters or marine waters.
Cumulative energy demand
It represents the direct and indirect use of industrial energy required throughout the production process.
Dissolved oxygen
The amount of oxygen (mg/l O2) in solution in the water under existing atmospheric pressure, temperature
and salinity. Sometimes also expressed as parts per million (ppm) or as percent of saturation level.
Ecological services
Benefi ts arising from the ecological functions of healthy ecosystems. Examples of ecological goods
include clean air, and abundant fresh water. Examples of ecological services include purifi cation of air and
water, maintenance of biodiversity, decomposition of wastes, soil and vegetation generation and renewal,
pollination of crops and natural vegetation, groundwater recharge through wetlands, seed dispersal,
greenhouse gas mitigation, and aesthetically pleasing landscapes.
Ecosystem
A natural entity (or a system) with distinct structures and relationships that liaise biotic communities (of
plants and animals) to each other and to their abiotic environment. The study of an ecosystem provides a
methodological basis for complex synthesis between organisms and their environment.
Ecosystem approach to aquaculture
An ecosystem approach to aquaculture (EAA) strives to balance diverse societal objectives, by taking
account of the knowledge and uncertainties of biotic, abiotic and human components of ecosystems
including their interactions, fl ows and processes and applying an integrated approach to the sector within
ecologically and operationally meaningful boundaries.
Eutrophication
Natural or artifi cial nutrient enrichment in a body of water, associated with extensive plankton blooms and
subsequent reduction of dissolved oxygen. The Nutriphication Potential (NP) is set at 1 for phosphate
(PO
4
). Other emissions also infl uence eutrophication, notably nitrogen oxides and ammonium.
Fatty acid
Organic acid composed of carbon, hydrogen and oxygen that combines with glycerol to form fats.
Feed conversion ratio (FCR)
Ratio between the dry weight of feed fed and the weight of yield gain. Measure of the effi ciency of
conversion of feed to fi sh (e.g. FCR = 2.8 means that 2.8 kg of feed is needed to produce one kilogram of
fi sh live weight).
Feedlot
Type of animal feeding operation, primarily used to fi nish large number of cattle in pens prior to slaughter.
Feedlots are associated with both the provision of high energy feedstuffs and the generation of
considerable amounts of high moisture content wastes.

7. Glossary
Today SummaryImpactsComparisonLooking ForwardPolicyAppendix
Glossary
References
82 Managing the environmental costs of aquaculture
Feedstuff
Any substance suitable for animal feed.
Fish oil
Oil extracted from total fi sh body or from fi sh waste. Fish oils are used in the manufacture of fi sh feeds,
edible fats and industrial products.
Fishmeal
Protein-rich meal derived from processing (boiling, pressing, drying, grinding) whole fi sh (usually small
pelagic fi sh or bycatch) as well as residues and by-products from fi sh processing plants (fi sh offal). Used
mainly as agriculture feeds for domestic livestock (poultry, pigs, cattle, etc.) and as aquaculture feeds for
carnivorous aquatic species. It must contain not more than 10 percent moisture. If it contains more than 3
percent salt (NaCl), the amount of salt must constitute a part of the brand name, provided that in no case
must the salt content of this product exceed 7 percent.
Gastropods
A member of the largest class of phylum Mollusca. Characteristics generally include: a foot upon which the
rest of the body (called the “visceral mass”) sits, a well-developed head, a protective one-piece shell, and
body “torsion” - where most of the visceral mass is normally twisted anticlockwise 180 degrees so that
the back end of the animal is positioned over its head. The class includes the snails, slugs, sea hares, sea
slugs, limpets, conches and abalone.
Inland aquaculture
Aquaculture that takes place in freshwater.
Life cycle analysis
Life Cycle Assessment (LCA) is a method developed to evaluate the mass balance of inputs and outputs
of systems and to organize and convert those inputs and outputs into environmental themes or categories
relative to resource use, human health and ecological areas.
Mollusk
Invertebrate animal belonging to the phylum Mollusca with a soft unsegmented body and covered by a
calcium carbonate shell, of 1 to 8 parts or sections. In some species the shell is lacking or reduced. The
surface is coated with mucus and cilia. Major cultured mollusks are mussels, oysters, scallops, cockles,
clams (bivalves) and abalone (gastropod).
Nitrogen
An odorless, gaseous element that makes up 78 percent of the earth’s atmosphere, and is a constituent of
all living tissue. It is almost inert in its gaseous form.
Pelagic
Relating to living or occurring in open water areas of lakes or oceans.

7. Glossary
TodaySummary Impacts Comparison Looking Forward Policy Appendix
Glossary
References
83Managing the environmental costs of aquaculture
Pen culture
Culture of stocks in pens. Pen is a fenced, netted structure fi xed to the bottom substrate and allowing
free water exchange; in the intertidal zone, it may be solid-walled; the bottom of the structure, however,
is always formed by the natural bottom of the water body where it is built; usually coastal e.g. in shallow
lagoons, but also inland e.g. in lakes, reservoirs. A pen generally encloses a relatively large volume of water.
Poikilothermic
Having a body temperature, which fl uctuates with that of the environment.
Recirculating system
A closed or partially closed system employed in aquaculture production where the effl uent water from the
system is treated to enable its reuse.
Trash fi sh
Small fi sh species, damaged catch and juvenile fi sh are sometimes referred to as ‘trash fi sh’ because of
its low market value. Usually part of a (shrimp) trawler’s bycatch. Often it is discarded at sea although an
increasing proportion is used as human food or as feed in aquaculture and livestock feed.
Zoonotic
Pertaining to a zoonosis: a disease that can be transmitted from animals to people or, more specifi cally, a
disease that normally exists in animals but that can infect humans.

8. References
TodayImpactsComparisonLooking ForwardPolicyAppendix
Glossary
References
Summary
84 Managing the environmental costs of aquaculture
References
ADB (2005). An evaluation of small-scale freswater aquaculture development for poverty reduction. ADB.
Aldhous, P. (2010). Transgenic fi sh swimming towards a plate near you. The New Scientist 2778.
Allison, E. H., Perry, A., Badjeck, M.-C., Adger, W. N., Andrew, N. L., Brown, K., Conway, D., Halls, A.,
Pilling, G. M., Reynolds, J. D. and Dulvy, N. K. (2009). Vulnerability of national economies to potential
impacts of climate change on fi sheries. Fish and Fisheries 10: 173-196.
Aquaculture News (2009). Institute of Aquaculture. 36. University of Stirling.
ARD (2006). Aquaculture: Changing the face of the waters. Meeting the promise and challenge of
sustainable aquaculture. The International Bank for Reconstruction and Development and The World Bank:
148 pp. Washington.
Arthur, J.R.; Bondad-Reantaso, M.G.; Campbell, M.L.; Hewitt, C.L.; Phillips, M.J.; Subasinghe, R.P.
Understanding and applying risk analysis in aquaculture: a manual for policy-makers. FAO Fisheries and
Aquaculture Technical Paper. No. 519/1. Rome, FAO. 113p.
Asche, F. (2008). Farming the sea. Marine Resource Economics 23: 527-547.
Atmomarsono, M. and Nikijulluw, V. (2004). Guide to invest on fi sheries in Indonesia - Milkfi sh. Directorate
of capital and investment system, Directorate of capacity building and marketing and Ministry of marine
affair and fi sheries. Jakarta, Indonesia.
Ayer, N. W. and Tyedmers, P. H. (2009). Assessing alternative aquaculture technologies: life cycle
assessment of salmonid culture systems in Canada. Journal of Cleaner Production 17(3): 362-373.
Barman, B. K. and Karim, M. (2007). Analysis of feeds and fertilizers for sustainable aquaculture
development in Bangladesh. Study and Analysis of Feeds and Fertilizers for Sustainable Aquaculture
Development. FAO fi sheries technical paper No. 497.M.R Hasan, T Hecht, S.S De Silva and A Tacon (Eds).
FAO: 113-140. Rome.
Bartley, D. M., Brugère, C., Soto, D., Gerber, P. and Harvey, B. (Eds) (2007). Comparative assessment
of the environmental costs of aquaculture and other food production sectors: methods for meaningful
comparisons. FAO fi sheries proceedings No. 10. FAO: 241 pp. Rome.
Betru, S. and Kawashima, H. (2009). Patterns and determinants of meat consumption in urban and rural
Ethiopia. Livestock Research for Rural Development 21(9): 143.
Beveridge, M. (2004). Cage aquaculture. Wiley-Blackwell. Third edition: 376 pp.
Beveridge, M. and Phillips, M. (2010). Aquaculture and climate change: what are the challenges for the
WIO region?
Beveridge, O., Humphries, S. and Petchey, O. (2010). The interacting effects of temperature and food
chain length on trophic abundance and ecosystem function. Journal of Animal Ecology 79(3): 693-700.
Biao, X. and Kaijin, Y. (2007). Shrimp farming in China: operating characteristics, environmental impact and
perspectives. Ocean and Coastal Management 50: 538-550.

8. References
Today Impacts Comparison Looking Forward Policy Appendix Glossary
References
Summary
85Managing the environmental costs of aquaculture
Bosma, R. H., Hanh, C. T. T. and Potting, J. (2009). Environmental impact assessment of the Pangasius
sector in the Mekong Delta. Wageningen University: 58 pp.
Bostock, J. C., McAndrew, B., Richards, R., Jauncey, K., Telfer, T. C., Lorenzen, K., Little, D. C., Ross, L.
G., Handisyde, N., Gatward, I. and Corner, R. (2010). Aquaculture: global status and trends. Philosophical
transactions of the royal society B 365: 2897-2912.
Brierley, A. and Kingsford, J. (2009). Impacts of climate change on marine organisms and ecosystems.
Current biology 19(14): 602-614.
Brooks, K. M. (2007). Assessing the environmental costs of Atlantic salmon cage culture in the Northeast
Pacifi c in perspective with the costs associated with other forms of food production. In Comparative
assessment of the environmental costs of aquaculture and other food production sectors: methods
for meaningful comparisons. FAO/WFT Expert Workshop. 24-28 April 2006, Vancouver, Canada. FAO
Fisheries Proceedings. No. 10. D.M Bartley, C Brugère, D Soto, P Gerber and B Harvey (Eds). FAO: 137-
182. Rome.
Brown, L. (2000). “Fish farming may soon overtake cattle ranching as a food source”. Worldwatch Issue
Alerts. Washington, D.C.
Brugère, C. and Ridler, N. (2004). Global aquaculture outlook in the next decades: an analysis of national
aquaculture production forecast to 2030. FAO Fisheries Circular No. 1001. FAO: 47 pp.
Brummett, R. E. (2007). Comparative analysis of the environmental costs of fi sh farming and crop
production in arid areas. In Comparative assessment of the environmental costs of aquaculture and other
food production sectors: methods for meaningful comparisons. FAO/WTF expert workshop. D M Bartley, C
Brugère, D Soto, P Gerber and B Harvey (Eds). FAO: 221-228. Rome.
Brummett, R. E., Angoni, D. E. and Pouomogne, V. (2004). On-farm and on-station comparison of wild and
domesticated Cameroonian populations of Oreochromis Niloticus. Aquaculture 242(1-4): 157-164.
Bunting, S. and Pretty, J. (2007). Global carbon budgets and aquaculture - emissions, sequestration and
management options. Occasional Paper 2007-1. Centre for Environment and Society.
Cao, L., Wang, W., Yang, Y., Yang, C., Yuan, Z., Xiong, S. and Diana, J. (2007). Environmental impact of
aquaculure and contermeasures to aquaculture pollution in China. Environmental Science and Pollution
Resource 14(7): 452-462.
Chen, J. (2003). Overview of sea cucumber farming and sea ranching practices in China. SPC Beche-de-
mer Information Bulletin 18: 18-23.
Cheung, W., Lam, V., Sarmineto, J., Kearney, K., Watson, R. and Pauly, D. (2009). Projecting global marine
biodiversity impacts under climate change scenarios. Fish and Fisheries 10: 235-251.
CIFA (accessed in 2010). http://www.cifa.in/
Costa-Pierce, B. A., Desbonnet, A., Edwards, P. and Baker, D. (Eds) (2005). Urban aquaculture. CABI: 304
pp.
Craig, S. and Helfrich, L.A. (2009). Understanding fi sh nutrition, feeds, and feeding. Virginia Cooperative
Extension no. 420-256: 4 pp.
Cressey, D. (2009). Future fi sh. Nature 458: 398-400.

8. References
TodayImpactsComparisonLooking ForwardPolicyAppendix
Glossary
References
Summary
86 Managing the environmental costs of aquaculture
Cruz-lacierda, E. R., Corre, V. L., Yanmamoto, A., Koyama, J. and Matsuoka, T. (2008). Current status on
the use of chemicals and biological products and health management practices in aquaculture farms in the
Philippines. Mem.Fac.Fish.Kagoshima Univ. 57: 37-45.
Cruz, P. S. (1997). Aquaculture feed and fertilizer resources atlas of the Philippines. FAO Fisheries Technical
Paper No. 366. FAO: 259 pp. Rome.
De Silva, S. S. and Hasan, M. R. (2007). Feeds and fertilizers: the key to long-term sustainability of Asian
aquaculture. Study and analysis of feeds and fertilizers for sustainable aquaculture development. FAO
fi sheries technical paper 497. M.R Hasan, T Hecht, S.S De Silva and A.G.J Tacon (Eds). FAO 19-47.
Rome.
de Vries, M. and de Boer, I. J. M. (2010). Comparing environmental impacts for livestock products: A
review of life cycle assessments. Livestock Science 128(1): 1-11.
Delgado, C. L., Crosson, P. and Courbois, C. (1997). The impact of livestock and fi sheries on food
availability and demand in 2020. MTID discussion paper No. 19. IFPRI.
Delgado, C.L., Wada, N., Rosegrant, M.W., Meijer, S., Ahmed, M. (2003). Fish to 2020. Supply and
demand in changing global markets. IFPRI: 226 pp. Washington, D.C.
Denman, K.L., Brasseur, G., Chidthaisong, A., Ciais, P., Cox, P.M., Dickinson, R.E., Hauglustaine, D.,
Heinze, C., Holland, E., Jacob, D., Lohmann, U., Ramachandran, S., da Silva Dias, P.L., Wofsy, S.C.,
Zhang, X. (2007). Couplings between changes in the climate system and biogeochemistry. In Climate
Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment
Report of the Intergovernmental Panel on Climate Change. Solomon, S., Qin, D., Manning, M., Chen, Z.,
Marquis, M., Averyt, K.B., Tignor, M., Miller, H.L (eds.). Cambridge University Press, Cambridge, United
Kingdom and New York, USA.
Duarte, C., Holmer, M., Olsen, Y., Soto, D., Marbà, N., Guiu, J., Black, K. and Karakassis, I. (2009). Will the
oceans help feed humanity? Bioscience 59: 967-976.
Dugan, P., Sugunan, V.V., Welcomme, R.L., Béné, C., Brummett, R.E., and Beveridge, M. (2007). Inland
fi sheries and aquaculture. In: D. Molden, (ed.) Water for Food, Water for Life. Colombo and London: IWMI
and Earthscan, 459–483.
Durham, S. (2010). Finding alternative fi sh feeds for aquaculture. Agricultural Research: 8-10.
El-Sayed, A.-F.M. (2006). Tilapia culture. 277 pp. CABI, London.
El-Sayed, A.-F. M. (2007). Analysis of feeds and fertilizers for sustainable aquaculture development in
Egypt. Study and analysis of feeds and fertilizers for sustainable aquaculture development. FAO fi sheries
technical paper 497. M.R Hasan, T Hecht, S.S De Silva and A. Tacon (Eds). FAO: 401-422. Rome.
Ellingsen, H. and Aanondsen, S. A. (2006). Environmental impacts of wild caught cod and farmed salmon –
A comparison with chicken. International Journal of LCA 1: 60-65.
Ellingsen, H., Olaussen, J. and Utne, I. (2009). Environmental analysis of the Norwegian fi shery and
aquaculture industry: A preliminary study focusing on farmed salmon. Marine Policy 33(3): 479-488.
Emerson, C. (1999). Aquaculture impacts on the environment. Discovery Guides. CSA.
Fang, X. (2007). Export and industry policy of aquaculture products in China. Global Trade Conference on
Aquaculture. R Arthur and J Nierentz (eds), FAO.
FAO (1995). Code of conduct for responsible fi sheries. FAO. 41 pp. Rome.

8. References
Today Impacts Comparison Looking Forward Policy Appendix Glossary
References
Summary
87Managing the environmental costs of aquaculture
FAO (2004) Carassius carassius cultured aquatic species information programme. Text by Weimin M. In:
FAO Fisheries and Aquaculture Department [online], Rome. (http://wwwfaoorg/fi shery/culturedspecies/
Carassius_carassius/en)
FAO (2009a). The state of food and agriculture. Livestock in balance. FAO.
FAO (2009b). The state of world fi sheries and aquaculture (SOFIA) - 2008. FAO: 196 pp. Rome.
FAO (accessed in 2010). Cultured aquatic species information programme. Mugil cephalus. http://www.
fao.org/fi shery/culturedspecies/Mugil_cephalus/en
FAO (2010). Fact Sheet: the international fi sh trade and world fi sheries. www.fao.org/fi shery. FAO.
FAOStat (2010). http://faostat.fao.org.
FIN (accessed in 2010). http://www.gafta.com/fi n/index.php
FishStat (2010). www.fao.org/fi shery/statistics/software/fi shstat/en. FishStat Plus.
Flachowsky, G. (2002). Effi ciency of energy and nutrient use in the production of edible protein of animal
origin. Institute of Animal Nutrition, Federal Agricultural Research Center (FAL). Braunschweig, Germany.
Flachowsky, G. (2007). Protein – population – politics: how protein can be supplied sustainably in the 21
st
century? Lohmann Information 42(1): 2-9.
Flegel, T. W. (1997). Major viral diseases of the black tiger prawn (Penaeus monodon) in Thailand. World
Journal of Microbiology and Biotechnology 13: 433-442.
Flores-Nava, A. (2007). Feeds and fertilizers for sustainable aquaculture development: a regional review for
Latin America. Study and analysis of feeds and fertilizers for sustainable aquaculture development. FAO
fi sheries technical paper 497 M.R Hasan, T Hecht, S.S De Silva and A. Tacon (Eds). FAO: 49-75. Rome.
Gabriel, U. U., Akinrotimi, O. A., Bekibele, D. O., Anyanwu, P. E. and Onunkwo, D. N. (2007). Economic
benefi t and ecological effi ciency of integrated fi sh farming in Nigeria. Scientifi c Research and Essay 2(8):
302-308.
Gerber, P., Wassenaar, M. R., Castel, V. and Steinfeld, H. (2007). Environmental impacts of changing
livestock production: overview and discussion for a comparative assessment with other food production
sectors. Comparative assessment of the environmental costs of aquaculture and other food production
sectors: methods for meaningful comparisons. FAO/WTF expert workshop. D M Bartley, C Brugère, D
Soto, P Gerber and B Harvey (Eds). Rome, FAO: 37-54.
Goedkoop, M., de Schryver, A. and Oele, M. (2008). Introduction to LCA with SimaPro 7.
Gowing, J. W., Tuong, T. P. and Hoanh, C. T. (2006). Land and water management in coastal zones:
dealing with agriculture-aquaculture-fi shery confl icts. Environment and livelihoods in tropical coastal zones.
Managing agriculture-fi shery-aquaculture confl icts. C.T Hoanh, T.P Tuong, J.W Gowing and B Hardy (Eds),
CABI.
Graslund, S. and Bengtsson, B.-E. (2001). Chemicals and biological products used in south-east Asian
shrimp farming, and their potential impact on the environment - a review. The Science of the Total
Environment 280: 93-131.
Grøttum, J. A. and Beveridge, M. (2007). A review of cage aquaculture: northern Europe. Cage
Aquaculture. Regional Reviews and Global Overviews. M Halwart, D Soto and J.R Arthur (Eds). FAO
Fisheries Technical Paper 498: 126-154. Rome.

8. References
TodayImpactsComparisonLooking ForwardPolicyAppendix
Glossary
References
Summary
88 Managing the environmental costs of aquaculture
Guinée, J. B., Gorree, M., Heijungs, R., Huppes, G., Kleijn, R., de Koning, A., van Oers, L., Wegener
Sleeswijk, A., Suh, S., Udo de Haes, H. A., de Brujin, H., van Duin, R. and Huijbregts, M. A. J. (2002).
Handbook on Life Cycle Assessment: Operational Guide to the ISO Standards, Kluwer Academic
Publishers.
Gupta, M. and Acosta, B. (2004). A review of global tilapia farming practices. Aquaculture Asia 9 (1): 7-16.
Handisyde, N. T., Ross, L. G., Badjeck, M.-C. and Allison, E. H. (2006). The effects of climate change
on world aquaculture: a global perspective. Final Technical Report, DFID Aquaculture and Fish Genetics
Research Programme. Stirling Institute of Aquaculture, Stirling, U.K. : 151 pp.
Hasan, M. R. and Halwart, M. (Eds) (2009). Fish as feed inputs for aquaculture: practices, sustainability and
implications. FAO Fisheries and Aquaculture Technical Paper. No. 518. FAO: 407 pp. Rome.
Hempel, E. (1993). Constraints and possibilities for developing aquaculture. Aquaculture International 1(1):
2-19.
Henriksson, P. (2009). Energy intensity in tropical aquaculture - Cumulative energy demand of milkfi sh,
pangasius and oyster production in South East Asia. Department of System Ecology. Stockholm University.
Master thesis: 40 pp.
Hewitt, C. L., Campbell, M. L. and Gollasch, S. (2006). Alien species in aquaculture. Considerations for
responsible use. IUCN: 46 pp. Gland, Switzerland.
Hung, L. T. and Huy, H. P. V. (2007). Analysis of feeds and fertilizers for sustainable aquaculture
development in Viet Nam. Study and analysis of feeds and fertilizers for sustainable aquaculture
development. FAO fi sheries technical paper 497. M.R Hasan, T Hecht, S.S De Silva and A. Tacon (Eds).
FAO: 331-361. Rome.
Huntington, T. and Hasan, M. R. (2010). Fish as feed inputs for aquaculture - practices, sustainability and
implications: a global synthesis. FAO Fisheries and Aquaculture Technical Paper 518. M.R Hasan and M
Halwart (Eds). FAO: 209-268. Rome.
ICES (2005). Code of Practice on the Introductions and Transfers of Marine Organisms 2004. 29 pp.
IEA (accessed in April 2010). www.iea.org.
Islam, M. S. (2005). Nitrogen and phosphorus budget in coastal and marine cage aquaculture and impacts
of effl uent loading on ecosystem: review and analysis towards model development. Marine Pollution
Bulletin 50(1): 48-61.
Kapuscinski, A.R., K. Hayes, S. Li, and G. Dana, eds. 2007. Environmental Risk Assessment of
Genetically Modifi ed Organisms, Vol. 3: Methodologies for Transgenic Fish, CABI Publishing, UK. 304 pp.
Karaanagiotidis, I. T., Bell, M. V., Little, D. C., Yakupitiyage, A. and Rakshit, S. K. (2006). Polyunsaturated
fatty acid content of wild and farmed tilapias in Thailand: effect of aquaculture practices and implications for
human nutrition. Journal of Agricultural and Food Chemistry 54(12): 4304-4310.
Karapanagiotidis, I., Yakupitiyage, A., Little, D.C., Bell, M.V., Mente, E. (2010). The nutritional value of lipids
in various tropical aquatic animals from rice-fi sh farming systems in northeast Thailand. Journal of Food
Composition and Analysis 23(1): 1-8.
Kaushik, S. and Troell, M. (2010). Taking the fi sh-in fi sh-out ratio a step further... Aquaculture Europe 35(1):
15-17.

8. References
Today Impacts Comparison Looking Forward Policy Appendix Glossary
References
Summary
89Managing the environmental costs of aquaculture
Kautsky, N., Ronnback, P., Tedengren, M. and Troell, M. (2000). Ecosystem perspectives on management
of disease in shrimp pond farming. Aquaculture 191: 145-161.
Losinger, W., Dasgupta, S., Engle, C., Wagner, B. (2000). Economic interactions between feeding rates
and stocking densities in intensive catfi sh Ictalurus punctatus production. Journal of the World Aquaculture
Society 31(4): 491-502.
Marris, E. (2010). Transgenic fi sh go large. Nature 467 (259 (published online)).
Mills, D. J., Westlund, L., de Graaf, G., Kura, Y., Willman, R. and Kelleher, K. (2010). Underreported and
undervalued: small-scale fi sheries in the developing world. Small-scale fi sheries management frameworks
and approaches for the developing world. R.S Pomeroy and N Andrew (Eds). Penang, Malaysia.
Moffi tt, C.M. (2006). Environmental, economic and social aspects of animal protein production and the
opportunities for aquaculture. Fisheries 30(9): 36-38.
Molden, D. H., Murray-Rust, R., Sakthivadivel, R. and Makin, I. (2003). A water productivity framework for
understanding and action. In: J. W. Kijne, R. Barker and D. Molden (Eds.). Water Productivity in Agriculture:
Limits and Opportunities for Improvement. Wallingford and Colombo: CABI Publishing and International
Water Management Institute.
Muir, J., Little, D. C., Young, J. A. and Bostock, J. C. (2009). Growing the wealth of aquaculture:
perspectives and potential. OECD: 46 pp.
Mungkung, R., Udo de Haes, H. and Clift, R. (2006). Potentials and limitations of life cycle assessment. In
Setting Ecolabelling Criteria: A Case Study of Thai Shrimp Aquaculture Product. International Journal of
LCA 11(1): 55-59.
Nakada, M. (2002). Yellowtail culture development and solutions for the future. Reviews in Fisheries
Science 10(3&4): 559-575.
Nellemann, C., Corcoran, E., Duarte, C. M., Valdes, L., DeYoung, C., Fonseca, L. and Grimsditch, G.
(2009). Blue carbon. A rapid response assessment. UNEP, GRID-Arendal.
Neori, A., Chopin, T., Troell, M., Buschmann, A. H., Kraemer, G. P., Halling, C., Shpigel, M. and Yarish, C.
(2004). Integrated aquaculture: rationale, evolution and state of the art, emphasizing seaweed biofi ltration in
modern mariculture. Aquaculture 231: 361-391.
Nur, A. (2007). Analysis of feeds and fertilizers for sustainable aquaculture development in Indonesia. Study
and analysis of feeds and fertilizers for sustainable aquaculture development. FAO fi sheries technical paper
497. M.R Hasan, T Hecht, S.S De Silva and A. Tacon (Eds). FAO: 245-267. Rome.
Obayelu, A. E. (2007). Socio-economic analysis of the impacts of avian infl uenza epidemic on households
poultry consumption and poultry industry in Nigeria: empirical investigation of Kwara State. Livestock
Research for Rural Development 19.
OECD (September 2008). Fisheries: improving policy coherence for development. Policy brief.
OECD (2010). Globalization in fi sheries and aquaculture: opportunities and challenges. OECD: 157 pp.
Paris.
Oken, E., Kleinman, K., Berland, W., Simon, S., Rich-Edwards, J. and Gillman, M. (2003). Decline in fi sh
consumption among pregnant women after a national mercury advisory. Obstetric Gynecology 102(2):
346-351.

8. References
TodayImpactsComparisonLooking ForwardPolicyAppendix
Glossary
References
Summary
90 Managing the environmental costs of aquaculture
Olah, J. and Sinha, V. R. P. (1986). Energy cost in carp farming systems. Aquacultura Hungarica: 219-226.
Pearson and Black, K. (2001). The environmental impact of marine fi sh cage culture. Environmnetal
Impacts of Aquaculture. K Black (Ed).
Pelletier, N. and Tyedmers, P. (2007). Feeding farmed salmon: is organic better? Aquaculture 272: 399-
416.
Pelletier, N. and Tyedmers, P. (2010). Life cycle assessment of frozen tilapia fi llets from Indonesian lake-
based and pond-based intensive aquaculture systems. Journal of Industrial Ecology 14(3): 467-481.
Pelletier, N., Tyedmers, P., Sonesson, A., Scholz, A., Ziegler, F., Flysjo, A., Kruse, S., Cancino, B. and
Silverman, H. (2009). Not all salmon are created equal: Life Cycle Assessment (LCA) of global salmon
farming systems. Environmental Science and Technology 43(23): 8730-8736.
Péron, G., François Mittaine, J. and Le Gallic, B. (2010). Where do fi shmeal and fi sh oil products come
from? An analysis of the conversion ratios in the global fi shmeal industry. Marine Policy 34(4): 815-820.
Phan, L.T., Bui, T.M., Nguyen, T.T.T., Gooley, G.J., Ingram, B.A., Nguyen, H,V., Nguyen, P.T., De Silva,
S.S. (2009). Current status of farming practices of striped catfi sh, Pangasianodon hypophthalmus in the
Mekong Delta, Vietnam. 51 pp.
Phuong, N.T. (2010). Aquaculture in Viet Nam: a focus on key farmed species. SEAT project, Bangkok,
Thailand. 14-17 January 2010. http://seatglobal.eu/wp-content/uploads/2010/02/3.-Vietnam-introduction-
Nguyen-Phoung.pdf
Pimentel, D., Berger, B., Filiberto, D., Newton, M., Wolfe, B., Karabinakis, E., Clark, S., Poon, E., Abbett,
E. and Nandagopal, S. (2004). Water resources: agricultural and environmental issues. BioScience 54(10):
909-918.
Ponzoni, R. W., Nguyen, N. H. and Khaw, H. L. (2009). Genetic improvement programs for aquaculture
species in developing countries: prospects and challenges. Matching genetics and environment: a new
look at an old topic. Proceedings of the 18th Conference of the Association for the Advancement of Animal
Breeding and Genetics. 342-349. Barossa Valley, South Australia.
Poštrk, V. (2003). The livestock revolution. Dietary transition: global rise in consumption of animal food
products. EnvironmentalScience. Lund. Master: 50 pp. Lund, Sweden.
Prein, M. (2007). Comparative analysis of material fl ows in low input carp and poultry farming: an overview
of concepts and methodology. Comparative assessment of the environmental costs of aquaculture and
other food production sectors: methods for meaningful comparisons. FAO/WTF expert workshop. D M
Bartley, C Brugère, D Soto, P Gerber and B Harvey (Eds). Rome, FAO: 183-199.
Primavera, J. (2006). Overcoming the impacts of aquaculture on the coastal zone. Ocean and Coastal
Management 49: 531-545.
Ramseyer, L.J. (2002). Predicting whole-fi sh nitrogen content from fi sh wet weight using regression
analysis. North American Journal of Aquaculture 64(3): 195-204. Reardon, T., Timmer, C. P. and Minten,
B. (2010). Supermarket revolution in Asia and emerging development strategies to include small farmers.
Proceedings of the National Academy of Sciences.
Rockström, J., Steffen, W., Noone, K., Persson, A., Chapin, F. S., Lambin, E., Lenton, T. M., Scheffer, M.,
Folke, C., Schellnhuber, H., Nykvist, B., De Wit, C. A., Hughes, T., van der Leeuw, S., Rodhe, H., Sörlin, S.,
Snyder, P. K., Costanza, R., Svedin, U., Falkenmark, M., Karlberg, L., Corell, R. W., Fabry, V. J., Hansen, J.,
Walker, B., Liverman, D., Richardson, K., Crutzen, P. and Foley, J. (2009). Planetary boundaries: exploring
the safe operating space for humanity. Ecology and Society 14(2): 32 pp.
90

8. References
Today Impacts Comparison Looking Forward Policy Appendix Glossary
References
Summary
91Managing the environmental costs of aquaculture
Rosenberry, B. (1999). World shrimp farming 1999, No. 12. Shrimp News International. San Diego.
Schroeter, C. and Foster, K. (2004). The impact of health information and demographic changes on
aggregate meat demand. AAEA annual meeting. Denver, CO.
Smil, V. (2001). Nitrogen and food production: proteins for human diets. Ambio 31(2): 126-131.
Speedy, A. W. (2003). Global production and consumption of animal source foods. Journal of Nutrition
133: 4048S-4053S.
Stage, J. and McGranahan, G. (2010). Is urbanization contributing to higher food prices? Environment and
Urbanization 22(1): 199-215.
Steinfeld, H., Gerber, P., Wassennar, T., Castel, V., Rosales, M. and de Haan, C. (2006). Livestock’s long
shadow. Environmental Issues and Options. FAO.
Sturrock, H., Newton, R., Paffrath, S., Bostock, J., Muir, J., Young, J., Immink, A. & Dickson, M. (2008).
Prospective analysis of the aquaculture sector in the EU. Part 2: characterization of emerging aquaculture
systems. Spain: European Commission Joint Research Centre.
Subasinghe, R. P., Curry, D., McGladdery, S. E. and Bartley, D. (2003). Recent technological innovations in
aquaculture. Review of the State of World Aquaculture, FAO Fisheries Circular. 59-74.
Sumagaysay-Chavoso, N. S. (2007). Analysis of feeds and fertilizers for sustainable aquaculture
development in the Philippines. Study and analysis of feeds and fertilizers for sustainable aquaculture
development. FAO fi sheries technical paper 497. M.R Hasan, T Hecht, S.S De Silva and A. Tacon (Eds).
FAO: 269-308. Rome.
Tacon, A. and Metian, M. (2008). Global overview on the use of fi sh meal and fi sh oil in industrially
compounded aquafeeds: trends and future prospects. Aquaculture 285: 146-158.
Tan, Z., Komar, C. and Enright, W. J. (2006). Health management practices for cage aquaculture in Asia - a
key component. Intervet.
Tanner, D. K., Brazner, J. C. and Brady, V. J. (2000). Factors infl uencing carbon, nitrogen, and phosphorus
content of fi sh from a Lake Superior coastal wetland. Canadian Journal of Fisheries and Aquatic Science
57: 1243-1251.
Tilman, D., Cassman, K. G., Matson, P. A., Naylor, R. and Polasky, S. (2002). Agricultural sustainability and
intensive production practices. Nature 418: 671-677.
Tlusty, M. F. and Lagueux, K. (2009). Isolines as a new tool to assess the energy costs of the production
and distribution of multiple sources of seafood. Journal of Cleaner Production 17(3): 408-415.
Troell, M., Kautsky, N. and Folke, C. (1999). Applicability of integrated coastal aquaculture systems. Ocean
and Coastal Management 42: 63-69.
Troell, M., Tyedmers, P., Kautsky, N. and Ronnback, P. (2004). Aquaculture and energy use. Encyclopedia
of Energy 1: 97-108.
Tveterås, R. (2002). Norwegian salmon aquaculture and sustainability: the relationship between
environmental quality and industry growth. Marine Resource Economics 17(2): 121-132.
UN (2010). World Urbanization prospects. The 2009 revision. Department of Economic and Social Affairs.
Population Division. 56 pp. New York.

8. References
TodayImpactsComparisonLooking ForwardPolicyAppendix
Glossary
References
Summary
92 Managing the environmental costs of aquaculture
Unknown. www.yk-snail.com/introen.asp. Retrieved July, 2010.
UNSTAT (accessed december 2010). UNSD environmental indicators http://unstats.un.org/unsd/
environment/qindicators.htm.
US-EPA (2006). Global anthropogenic non-CO2 greenhouse gases. United States Environmental protection
Agency, EPA 430-R-06-003, June 2006. Washington, D.C.
VDI (1997). Cumulative energy demand-terms, defi nitions, methods of calculation. Ingenieure VD.
Verdegem, M. and Bosma, R. (2009). Water withdrawal for brackish and inland aquaculture, and options to
produce more fi sh in ponds with present water use. Water Policy 11(supplement 1): 52-68.
Verdegem, M., Bosma, R. and Verreth, J. (2006). Reducing water for animal production through
aquaculture. International Journal of Water Resources Development 22: 101-113.
Volpe, J. P., Beck, M., Ethier, V., Gee, J. and Wilson, A. (2010). Global Aquaculture Performance Index.
University of Victoria, Victoria, British Columbia, Canada.
Wackernagel, M., Monfreda, C., Moran, D., Wermer, P., Goldfi nger, S., Deumling, D. and Murray, M. (2005).
National footprint and biocapacity accounts 2005: The underlying calculation method. Global Footprint
Network. Oakland, California.
World Bank (2006). Aquaculture: changing the face of the waters. World Bank report No. 36622-GLB.
World Bank. Washington, DC.
World Bank (2010). World Development Report. Development and Climate Change. 417 pp. Washington,
DC.
Weimin, M. and Mengqing, L. (2007). Analysis of feeds and fertilizers for sustainable aquaculture
development in China. Study and analysis of feeds and fertilizers for sustainable aquaculture development.
FAO fi sheries technical paper 497. M.R Hasan, T Hecht, S.S De Silva and A. Tacon (Eds). FAO: 141-190.
Rome.
White, T. (2000). Diet and the distribution of environmental impact. Ecological Economics 34(1): 145-153.
WHO (2010). Global and regional food consumption patterns and trends. www.who.int/nutrition/topics/3_
foodconsumption/en/index.html.
Wijkstróm, U. N. (2003). Short and long-term prospects for consumption of fi sh. Veterinary Research
Communications 27(suppl. 1): 461-468.
Wiwchar, D. (February 2005). Fish farms causing problems in Muchalet Inlet. Ha-Shilth Newspaper. www.
westcoastaquatic.ca/article_fi shfarms_problems_muchalat0205.htm
Worldfi shing and Aquaculture (2010) US ramps up import tax on pangasius. http://www.worldfi shing.net/
news101/us-ramps-up-import-tax-on-pangasius
WRI (1998). World resources 1998-99: environmental change and human health, WRI.
Ye, Y. (1999). Historical consumption and future demand for fi sh and fi shery products: exploratory
calculations for the years 2015/2030. FAO Fisheries Circular 946. FAO.
92

Partners
OUR VISION
We imagine a healthy, prosperous world in which societies are
forever committed to caring for and valuing nature, our global
biodiversity, for the
long-term bene t of people and all life on Earth.
OUR MISSION
Building upon a strong foundation of science, partnership and
 eld demonstration, CI empowers societies to responsibly and
sustainably care for nature, our global biodiversity, for the well-
being of humanity.
www.conservation.org
The WorldFish Center is an international, nonpro t,
nongovernmental research organization dedicated to reducing
poverty and hunger by improving  sheries and aquaculture.
WorldFish is one of 15 members of the Consortium of International
Agricultural Research Centers supported by the Consultative Group
on International Agricultural Research (CGIAR). The CGIAR is a
global partnership that unites organizations engaged in research
for sustainable development with the funders of this work. The
funders include developing and industrialized country governments,
foundations, and international and regional organizations.
The WorldFish Center is committed to meeting two key
development challenges: 1) improving the livelihoods of those who
are especially poor and vulnerable in places where  sheries and
aquaculture can make a di erence, and 2) achieving large scale,
environmentally sustainable increases in supply and access to  sh
at a ordable prices for poor consumers in developing countries.
www.world shcenter.org
WorldFish Publication Number:2011-33
ISBN: 978-983-2346-78-4