biophysical sustainability and approaches to marine ... · aquaculture might develop in a more...

A R E P O R T T O T H E M A R I N E A Q U A C U L T U R E TA S K F O R C E

Biophysical Sustainability andApproAches to Marine Aquaculture

Development Policy in the United States

F E B R U A RY 2 0 0 7

Peter Tyedmers • Nathan Pelletier • Nathan Ayer School for Resource and Environmental Studies, Dalhousie University

Abstract

Over recent decades, aquaculture has been

the single fastest growing food production

sector in the world. As landings from tradi-

tional capture fisheries have effectively

plateaued, growth in aquaculture production

is seen by many as essential to address

increasing demand for seafood in both

developed and developing countries.

Although the United States is currently a

minor contributor to global aquaculture

production, the U.S. Department of

Commerce has adopted a policy to increase

the value of domestic production fivefold by

2025. Against this backdrop, this report was

commissioned by the U.S. Marine

Aquaculture Task Force, convened by the

Woods Hole Oceanographic Institution, to

review existing research regarding the bio-

physical performance of aquaculture produc-

tion systems—essentially their dependence

on flows of matter and energy along with

resulting environmental impacts—and

provide insight regarding how U.S. marine

aquaculture might develop in a more

sustainable manner in this regard.

While the concept of sustainability is

widely debated, this report adopts a “strong”

sustainability approach, which recognizes

that the scale of human activities is ultimate-

ly limited by its impacts on ecosystem

goods and services. Sustainable development

therefore requires that the rate of resource

extraction and waste emissions not exceed

the regenerative capacity of renewable

resources and the ability of ecosystems to

absorb waste and respond to change.

Achieving strong sustainability in the U.S.

aquaculture sector should thus be informed

by comparative analyses of the biophysical

performance of various production scenarios

and technologies.

To date, three broadly related analytical

techniques—energy analysis, ecological foot-

print analysis, and life cycle assessment—

have been used to quantitatively assess bio-

physical performance in aquaculture. This

report introduces each of these methodolo-

gies, describes how they have been applied to

evaluating various aquaculture technologies,

and discusses the implications of existing

research for sustainable aquaculture develop-

ment policy in the United States.

B I O P H Y S I C A L S U S T A I N A B I L I T Y O F A Q U A C U L T U R E1

B I O P H Y S I C A L S U S T A I N A B I L I T Y O F A Q U A C U L T U R E 2


Glossary

Acidification—The process whereby emis-

sions to the atmosphere are converted into

acid substances, which can damage both ter-

restrial and aquatic ecosystems. Acidification

potential is a commonly used environmental

impact category in life cycle assessment.

Aquaculture—The farming of aquatic

organisms. This involves some level of inter-

vention in the rearing process to enhance

production and individual or corporate own-

ership of the stock being cultivated.

Biophysical—Pertaining to the flux of mat-

ter and energy that collectively underpins

ecosystem structure and function, and ulti-

mately all human activities.

Benthic Impacts—Impacts to the benthic

environment, or the surface of contact

between a body of water and its underlying

strata.

Eco-Efficiency—A measure of the total eco-

logical/environmental burden associated with

the provision of goods and services.

According to the World Business Council for

Sustainable Development, eco-efficiency is

improved through the delivery of “...compet-

itively priced goods and services that satisfy

human needs and bring quality of life while

progressively reducing environmental

impacts of goods and resource intensity

throughout the entire life-cycle to a level at

least in line with the Earth's estimated carry-

ing capacity.”

Ecosphere—The sum of atmosphere,

oceans, biosphere (plant and animal life),

and outermost portion of the earth's crust,

the collective relationships of which consti-

tute a self-regulating environment conducive

to life.

Ecosystem Services—The life support ser-

vices provided by natural ecosystems, includ-

ing the purification of air and water, decom-

position of wastes, regulation of climate,

regeneration of soil fertility, and the produc-

tion and maintenance of biodiversity.

Environmental “Hot Spots”—A term used

frequently in life cycle assessment to describe

product or process life cycle stages that con-

tribute disproportionately to overall environ-

mental impact.

Eutrophication—An increase in the rate of

supply of organic matter to an ecosystem.

Eutrophication potential is a commonly used

environmental impact category in life cycle

assessment.

Extensive Aquaculture—Forms of aquacul-

ture characterized by (i) a low degree of

control (e.g., of environment, nutrition,

predators, competitors, and disease agents);

(ii) low initial costs, low-level technology,

and low production efficiency; (iii) high

dependence on local climate and water

quality; and (iv) use of natural waterbodies

(e.g. lagoons, bays, embayments) and of

natural, often-unspecified food organisms.

Feed Conversion Ratio—Formally defined

as the ratio between the dry weight of feed

fed and the weight of harvestable product

gain. In some instances it is also used to

describe the ratio of dry weight of feed fed to

the weight of organisms harvested.

Global Warming Potential—A common

environmental impact category used in life

cycle assessments in which total greenhouse

gas emissions associated with the provision of

a product or service are typically expressed in

terms of CO2 equivalents.

Intensive Aquaculture—Culture systems

characterized by (i) production rates of up to

200 metric tons/ha/yr; (ii) a high degree of

control; (iii) high initial costs; (iv) tendency

towards increased independence of local cli-

mate and water quality; and (v) use of man-

made culture systems.

Mariculture or Marine Aquaculture—The

cultivation of organisms in which their final

rearing takes place in seawater; earlier stages

of the life cycle of these species may be spent

in fresh or brackish waters.

Net Primary Productivity—The net accu-

mulation of carbon in plants. Essentially the

total amount of carbon fixed by plants

through photosynthesis minus that lost

through respiration.

Semi-Intensive Aquaculture—Aquaculture

systems characterized by: (i) production of

0.5-5 metric tons/ha/yr, possibly augmented

through supplementary feeding with low-

grade feeds; (ii) stocking with wild-caught or

hatchery-reared fry; (iii) regular use of organ-

ic or inorganic fertilizers; (iv) rain or tidal

water supply and/or some water exchange;

(v) simple monitoring of water quality; and

(vi) normally occurring in traditional or

improved ponds, also some cage systems.

Trophic Level—The position that a species

occupies in a food web, typically expressed

relative to the energetic foundation of that

web. By convention the foundation of a food

web, for example plants or detritus, is

assigned a trophic level of 1. Consumers of

this material are defined as a trophic level 2

organism while primary carnivores are

defined as a trophic level 3, etc.


Introduction

Introduction

Aquaculture and sustainability: words that

are frequently combined but seldom critical-

ly explored when used in combination. This

report to the U.S. Marine Aquaculture Task

Force attempts such an analysis. Specifically,

this report provides a context within which

the biophysical sustainability of aquaculture,

essentially its dependence on matter and

energy and the implications of this depen-

dence, can be understood, and reviews

insights gained from available research

regarding the biophysical performance of

various aquaculture systems. Reflecting the

marine focus of the Task Force’s work, par-

ticular attention is given to marine-based

culture, or mariculture. However, due to the

relatively limited research undertaken to date

on exclusively marine culture systems, we

review and draw insight from a wide range

of production technologies.

Aquaculture, the farming of aquatic organ-

isms, is a highly diverse activity. More than

220 different species of finfish and shellfish,

along with many species of seaweed, are

farmed globally in freshwater, brackish, and

marine environments. Production systems

are equally varied, ranging from more tradi-

tional, low-intensity subsistence aquaculture

to highly intensive industrial production

facilities. The continuum between these

extremes encompasses a diverse range of

farming technologies. These include mono-

culture and polyculture systems, freshwater

pond and raceway farming, land-based tanks

for both marine and freshwater species, and

open-water culture systems such as cages and

pens for finfish, and poles, rafts, or longlines

for seaweed and mussel culture.

Not surprisingly, the biophysical impacts

of aquaculture range widely, but generally

vary with the intensity of the culture system

(Troell et al. 2004). While not exhaustive or

pervasive across all forms of culture, these

impacts can include localized nutrient

enrichment or depletion (Folke et al. 1992,

Merceron et al. 2002, Homer et al. 2001),

the effects of therapeutants and other

chemicals on receiving waters and associated

organisms (Hastein 1995, Black et al. 1997,

Collier and Pinn 1998, Davies et al. 1998,

Ernst et al. 2001, Haya et al. 2001), the

disturbance or outright replacement of local

ecosystems (Finlay et al. 1995, Pohle et al.

2001, Janowicz and Ross 2001, Alongi

2002), the introduction of exotic species

(Canonico et al. 2005, DeSilva et al. 2006),

gene flow from farmed to wild populations

(Einum and Fleming 1997, Youngson and

Verspoor 1998, Fleming et al. 2000), the

potential amplification and transmission of

disease/parasite loads (Kautsky et al. 2000,

Heusch and Mo 2001, Bjorn et al. 2001,

Bjorn and Finstad 2002, Morton et al. 2005,

Krkosek et al. 2005), high levels of energy

dependence and associated greenhouse gas

emissions (Tyedmers 2000, Troell et al.

2004), and dependence on the products of

capture fisheries (Naylor et al. 2000, Naylor

and Burke 2005).

Sustaining average growth rates exceeding

10 percent per annum in recent decades,

aquaculture has been one of the fastest grow-

ing food production sectors in the world

(FAO 2004) (Figure 1A). As a result, aqua-

culture currently contributes almost 50 per-

cent of seafood consumed globally. Moreover,


Sustaining average growthrates of 10 percent per

annum, aquaculture hasbeen one of the fastest

growing food productionsectors in the world.

the United Nations Food and Agricultural

Organization (FAO) predicts a further 70

percent increase in aquaculture production

from current levels by the year 2030 (FAO

2004). As landings from capture fisheries,

the historic source of most of the world’s

seafood, have effectively plateaued (Botsford

et al. 1997, Pauly et al. 2002, Christensen et

al. 2003, Myers and Worm 2003, FAO

2004), the growth in aquaculture production


F I G U R E 1 A

F I G U R E 1 B

MARINE FISH

DIADROMOUS FISH

CRUSTACEANS

MOLLUSKS

PLANTS

FRESHWATER FISH

CRUSTACEANS

ALL FINFISH

MOLLUSKS

PLANTS

60

50

40

30

20

10

01970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003

Pro

duc

tion

(mill

ion

me

tric

tons

)

30

25

20

15

10

5

01970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003

Pro

duc

tion

(mill

ion

me

tric

tons

)

FIGURE 1 Global Total (A) and Marine-Based (B) Aquaculture Production by Major Species Groups1970 to 2003 (FAO 2005)

is seen by many as essential to address

increasing demand for seafood in both rich

and poor nations (FAO 2004). This is

despite the fact that fisheries and aquaculture

together contribute only about one percent

of total food and six percent of total protein

energy consumed globally. While the majori-

ty of contemporary aquaculture production

originates from extensive and semi-intensive

freshwater carp and tilapia culture in Asia

(and China in particular), intensive produc-

tion of shrimp and finfish species in marine

waters has increased significantly in recent

years in both developed and developing

countries (Figure 1B). This is particularly

true of marine-based shrimp culture.

Although this sector, along with all other

farmed crustaceans, still accounts for a rela-

tively small fraction of global production by

mass, it has expanded at average annual rates

of almost 25 percent per annum over the last

25 years (Figure 1B).

Accounting for only one percent of global

output, the United States is currently a rela-

tively minor player in aquaculture (FAO

2005). Moreover, while production else-

where has experienced double-digit average

rates of increase, the U.S. aquaculture sector

has grown at a modest five percent per

annum over recent decades (Figure 2A).

Nationwide, U.S. production is dominated

by a single species. Channel catfish, grown

intensively in freshwater pond systems,

accounts for over 50 percent of all U.S.

aquaculture production and well over 90

percent of total finfish production (FAO

2005). U.S. mariculture is dominated by the

extensive culture of oysters, clams, and mus-

sels. Interestingly, while production of these

mollusks has varied through time, the overall

scale of this segment of the industry has

barely changed over the last 30 years (Figure

2B). In contrast, both intensive culture of

shrimp and finfish in U.S. marine waters has

increased dramatically in recent decades,

averaging rates of increase of over 10 percent

per annum despite some recent declines in

the latter sector (Figure 2B).

Given the growing popularity of seafood

in the U.S., imports of both wild-caught and

cultured species have increased substantially

in recent years, resulting in a current annual

U.S. seafood trade deficit of $7.5 billion. In

response to this growing gap between domes-

tic production and consumption, the U.S.

Department of Commerce has adopted a

policy to increase the value of aquaculture

production within the U.S. fivefold by 2025

from its current level of about $1 billion per

year. Given the diversity of aquaculture sys-

tems and, in particular, the scale and form of

biophysical impacts associated with different

types of culture, deciding how this expansion

will occur demands careful consideration.

Regardless of whether the current U.S.

aquaculture policy objective is achieved,

however, demand for cultured seafood within

the U.S. and elsewhere will likely increase

substantially into the foreseeable future.

Consequently, in the interests of addressing

the sustainability of aquaculture, it is not suf-

ficient for critics to simply oppose its expan-

sion in U.S. waters. What is needed is a

comprehensive understanding of the bio-

physical implications of various forms of

aquaculture, regardless of where it is under-

taken, to provide a basis upon which

informed decisions can be made.

The balance of this report is composed of

five major sections. We begin by briefly

describing differing perspectives on how the

sustainability of human activities can be

understood and argue in support of adopting

what is generally referred to as a “strong” sus-

tainability approach. For aquaculture, this

implies that any analysis of the sustainability

of its activities must be founded on an

understanding of its biophysical perfor-

mance—essentially its dependence upon

flows of matter and energy through the pro-

duction system along with the

ecological/environmental impacts of those

flows. We then briefly review three leading

techniques (energy analysis, ecological foot-

print analysis, and life cycle assessment) that

have been used to evaluate various forms of


What is needed is a comprehensive

understanding of the biophysical implications

of various forms of aquaculture.

aquaculture, summarize the results of avail-

able research that has employed these tech-

niques, and, where possible, make compar-

isons between culture systems and other

competing animal protein production sys-

tems. Before concluding, we discuss the

growing dependence of aquaculture on

aquatic resources and attempt to address the

seeming confusion regarding the relative effi-

ciency of aquaculture’s use of those resources.

F I G U R E 2 A

F I G U R E 2 B

CRUSTACEANS

MOLLUSKS

FRESHWATER AND

DIADROMOUS FISH

CRUSTACEANS

ALL FINFISH

MOLLUSKS

600

500

400

300

200

100

01970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003

Pro

duc

tion

(tho

usa

nd m

etr

ic to

ns)

180

160

140

120

100

80

60

40

20

01970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003

Pro

duc

tion

(tho

usa

nd m

etr

ic to

ns)


Figure 2 U.S. Total (A) and Marine-Based (B) Aquaculture Production by Major Species Groups 1970 to 2003 (FAO 2005)

Biophysical Sustainability

The human economy can be defined as

the set of activities and relationships whose

cumulative purpose is to satisfy human

material needs and wants. This includes both

our biological and industrial metabolisms,

through which resources are extracted, goods

are produced and consumed, and wastes are

returned to the environment (Rees 1999). In

this light, it is apparent that the ecosphere

provides the supporting context for all eco-

nomic activity.

Since the publication of Our CommonFuture (Bruntland 1987) the idea of “sustain-

able development” has been both widely

invoked and debated. A definition of sustain-

ability articulated by the International Union

for the Conservation of Nature (1991) states

that “sustainable development improves

people’s quality of life within the context of

earth’s carrying capacity.” A complementary

definition of sustainable development

included in the 1992 Rio Declaration onEnvironment and Development—whose signa-

tories included the United States—is “devel-

opment that meets the needs of the present

generation without compromising the ability

of future generations to meet their own

needs” (Reid 1995). Both of these definitions

imply that biophysical carrying capacity rep-

resents the limiting factor for the scale of sus-

tainable human activities.

If sustainability depends, in part, on the

maintenance of the goods and services pro-

vided by nature, then it is important to

determine what share of those goods and ser-

vices must be maintained, the extent to

which human activities reduce their availabil-

ity, and the extent to which they can be

replaced by manufactured equivalents (Ekins

et al. 2003). At present there are two distinct

philosophical approaches to addressing these

issues, commonly referred to as “weak” and

“strong” sustainability (Jansson et al. 1994).

Weak and Strong SustainabilityWeak sustainability rests on the assump-

tion that all forms of natural capital, essen-

tially the foundation of the ecosystems goods

and services upon which we depend (see Box

1), can be substituted with manufactured

equivalents (Victor 1991, Daly 1991, 1994,

Pearce et al. 1994, Victor et al. 1995, Pearce

1998). From this perspective, it is possible to

conceive of a "sustainable" human enterprise

in which all natural capital assets are system-

atically liquidated as long as part of the

resulting stream of benefits is re-invested in

other forms of capital so as to maintain the

economic productivity of the remaining

aggregate stock. As a result, weak sustainabil-

Natural Capital Natural, manufactured, and human capital are thethree forms of productive capital recognized byecological economists. They are roughly analogousto the three factors of production - land, capital,and labor respectively - as traditionally defined with-in neo-classical economics. Natural capital, howev-er, encompasses a far greater range of ecosystem-sourced goods and services than is traditionallyassociated with "land". It includes all the biotic andabiotic goods, such as timber, minerals, and fish,and services—including soil formation, climate regu-lation, and water purification—that either directly orindirectly contribute to the maintenance of humansand our economies.


…the industrial energyinputs to modern food

production systems oftenexceed the caloric returnsin food energy by orders

of magnitude.

ity implies that there are virtually no limits

to the growth potential of the human econo-

my, regardless of increasing scarcity of natur-

al capital.

In contrast, strong sustainability posits

that manufactured capital is generally only

partially substitutable for natural capital,

necessitating the maintenance of “critical”

natural capital stocks in perpetuity (Daly

1991, Costanza and Daly 1992, Rees 1998,

Ekins et al. 2003). In other words, many, if

not most, ecosystem goods and services sim-

ply cannot be replaced with technological

equivalents. Moreover, their inherently finite

nature clearly implies that there are limits to

the scale of impact that human activities can

have if life support functions are to be main-

tained. Sustainable development therefore

requires that the rate of resource extraction

and waste emissions not exceed the regenera-

tive capacity of renewable resources and the

ability of ecosystems to absorb waste and

respond to change (Proops and Faber 1985).

When it comes to informing strong sus-

tainability decision-making, concern arises

regarding the appropriateness of relying on

conventional monetary-based techniques.

This is largely because market prices do not

generally reflect the importance of ecosystem

goods and services to human well-being.

While assigning prices to conventional nat-

ural resources such as lumber, fish, etc., is

relatively straightforward, it is much more

difficult, and often simply impossible, to

adequately price the myriad ecosystem goods

and services upon which human activities are

fundamentally dependant. In recognition of

these limitations, ecological economists have

developed a variety of biophysical accounting

tools. These are used to both better under-

stand the relationship between human activi-

ties and large-scale biophysical limits to

growth, and—at finer scales—to provide

insight into the material, energy, and ecologi-

cal efficiency of our choices.

Strong Sustainability and the Eco-Efficiency of Food Production

The intersection of increasing population,

persistent high levels of malnourishment for

many yet rising consumption levels for some,

and limited resources underscores the impor-

tance of improving the ecological efficiency,

or eco-efficiency of human activities if sus-

tainable development is to be achieved (Box

2). This is particularly pressing within the

context of food production, where rapid

industrialization has precipitated numerous

unintended consequences. Not only do the

industrial energy inputs to modern food pro-

duction systems often exceed the caloric

returns in food energy by orders of magni-

tude (Pimentel 2004, Troell et al. 2004,

Tyedmers et al. 2005), the widespread intro-

duction of intensive production technologies

has led to the fragmentation and outright

conversion of habitats (Kerr and Desguise

2004, Hartemink 2005), species extirpation

or extinction (Kruess and Tscharntke 1994,

Kerr and Desguise 2004), widespread losses

of topsoil (Heffernan and Green 1986, Lal

2000), depletion and contamination of fresh

surface and groundwater (Zebarth et al.

1998, Liess et al. 1999), nutrient enrichment

B o x 2

Improving Eco-Efficiency

Eco-efficiency can be improved by producing goods and services while using less materials andproducing less waste and pollution. Living and non-living resources are used more efficiently byreducing the material and energy intensity of production processes and maximizing the use ofrenewable resources. Natural resources and thefunctioning of ecosystems are also protected byreducing the release of pollutants to the environment.


of soils and receiving waters (Zebarth et al.

1998), the proliferation of pests (Mack et al.

2000), and the general degradation of the

productive capacity of both terrestrial and

aquatic environments (El-Hage Sciallaba and

Hattam 2000). This is further compounded

by international trade in foodstuffs, where

transport of goods between geographically

disparate locales creates additional environ-

mental burdens and allows economically

advantaged regions to run ecological deficits

at the expense of less developed regions

(Hansson and Wackernagel 1999).

The common root of these seemingly

disparate problems is a fundamental lack of

regard for biophysical constraints, and the

attendant necessity of restructuring human

activities to maximize efficiency while simul-

taneously respecting the limited capacity of

natural systems to supply material and ener-

gy resources and absorb wastes. Achieving

strong sustainability will therefore require

analyses of food production systems in order

to establish the biophysical performance of

alternative production scenarios, and to facil-

itate decision making informed by eco-effi-

ciency considerations. This is particularly

important in the aquaculture industry, where

rising demand for seafood products and con-

current declines in capture fisheries have

resulted in rapid proliferation of industrial

aquaculture production globally (FAO

2004).


Techniques for Assessing Eco-Efficiency

To date, three broadly related analytical

techniques—energy analysis, ecological foot-

print analysis, and life cycle assessment—

have been used to quantitatively assess aqua-

culture systems and numerous other human

activities. While differing in methodology

and focus, all three speak to various aspects

of the biophysical performance of various

culture systems and—ultimately—their eco-

efficiency. As such, the information derived

from research using these tools is comple-

mentary, and the results of these analyses

should be interpreted in concert to inform

the broadest possible understanding of the

biophysical sustainability of alternative

aquaculture production systems. Aquaculture

policy makers and managers can therefore

apply these results to strategies for sustain-

able aquaculture development that maximize

efficiency of resource use while minimizing

environmental impacts.

Energy AnalysisEnergy flows have been used to evaluate

human activities for over 100 years

(Martinez-Alier 1987). It was not until the

oil price shocks of the 1970s, however, that

energy analysis rose to prominence as indus-

trialized countries struggled to understand

the scale of their dependence on fossil fuels.

In its more traditional form, energy analysis

entails quantifying the primary direct and

indirect industrial energy inputs (essentially

all fossil fuel and electrical energy inputs)

required to provide a product or service

(Peet 1992, Brown and Herendeen 1996).

In essence, then, energy analysis provides a

measure of the energy cost of production.

The primary rationale underlying its use is

"to quantify the connection between human

activities and the demand for this important

(energy) resource" (Brown and Herendeen

1996). However, as industrial energy use—

and in particular fossil energy use—is directly

related to a number of major environmental

concerns including global climate change,

acid precipitation, eutrophication, and biodi-

versity loss, it also has value as an indicator

of biophysical sustainability (Kåberger 1991,

Brown and Herendeen 1996). To date,

energy analysis has been used to analyze

thousands of goods and services, including

hundreds of fisheries and numerous

aquaculture systems.

As the sun is the ultimate source of most

forms of energy upon which human activities

depend, various techniques have been devel-

oped to specifically account for our depen-

dence on this fundamental resource. Emergy

analysis, arguably the most complete tool for

undertaking this sort of assessment, essential-

ly accounts for the solar energy flux under-

pinning all forms of conventional energy and

material inputs, along with the associated

human and environmental services required

to provide a good or service (Odum 1988,

Odum and Arding 1991, Brown and

Herendeen 1996). However, as emergy

analysis has not been used extensively to

assess aquaculture systems, it is not consid-

ered further in this report. Partial solar ener-

gy inputs to some culture systems, in the

form of net primary productivity appropriat-

ed, have been quantified using another


accounting technique, ecological footprint

analysis.

Ecological Footprint Analysis The ecological concept of carrying capaci-

ty, or the maximum population that can be

sustained indefinitely by a given quantity of

habitat without impairing its long-term pro-

ductivity, has been used for decades to help

contextualize the problem of human over-

consumption of natural resources. It also

forms the basis of a biophysical evaluation

technique known as ecological footprint

analysis (Rees and Wackernagel 1994, Rees

1996, Wackernagel and Rees 1996). Within

ecological footprint analysis, various material

and energy flows required to sustain a

defined human population or activity are

re-expressed in terms of a commensurate

variable – ecosystem support area. As such,

it provides a measure of relative ecological

efficiency.

The rationale underlying ecological foot-

print analysis is that human activities are

inexorably underpinned by the limited quan-

tity of ecologically productive land and water

that sustains both our biological and indus-

trial metabolisms. To date, it has been used

most frequently to assess the collective and

per capita impacts of societies under specific

technological and cultural conditions on a

variety of scales. It has, however, also proven

useful in assessing the relative eco-efficiency

of alternative technology systems, including

some of those used in fisheries and aquacul-

ture operations (Larsson et al. 1994, Berg et

al. 1996, Folke et al. 1998, Tyedmers 2000).

Life Cycle AssessmentLife cycle assessment (LCA) is an analyti-

cal framework for evaluating the potential

environmental impacts of human activities

from a systems perspective. This means that

it can be used to quantify the wide range of

environmental impacts associated with each

stage in the provision and use of a product

or service, from resource extraction and pro-

cessing through to ultimate disposal or recy-

cling (Consoli et al. 1993), and to pinpoint

opportunities for improving environmental

performance.

Modeled initially on energy analysis, for-

mal development of the LCA methodology

began in the late 1980s. It has since been

refined and improved by the International

Organization for Standardization (ISO), the

U.S. Environmental Protection Agency, and

the Society for Environmental Toxicology

and Chemistry (SETAC), as well as other

national and international organizations.

Now widely accepted by the scientific com-

munity, industry, and policy makers, LCA

methodology is formally standardized under

ISO 14040-14043 (ISO 1997).

While originally developed to evaluate the

life-cycle impacts associated with manufac-

tured products, LCA is increasingly being

applied to food production systems

(Mattsson and Sonesson 2003). Within this

context, it has been used both to compare

the environmental performance of competing

products, processes, or scales of activity and

to identify activities or subsystems that con-

tribute disproportionately to the environ-

mental impacts of specific food production

technologies (Andersson et al. 1998,

Andersson and Ohlsson 1999, Haas et al.

2001, Hospido et al. 2003). A considerable

body of published research has reported the

life-cycle impacts of various agricultural sys-

tems. More recently, LCA has also been used

to evaluate seafood production, including

several forms of aquaculture (Christensen

and Ritter 2000, Seppälä et al. 2001, Ziegler

et al. 2003, Eyjólfsdóttir et al. 2003, Thrane

2004, Hospido and Tyedmers 2005, Trane

2006, Ellingsen and Aanondsen 2006).

The increasing number of life cycle assess-

ments of industrial aquaculture indicates a

growing interest in its use to better under-

stand the environmental performance of

alternative aquaculture production systems.

This may be due to the high degree of reso-

lution that LCA provides with respect to the


relative magnitude of environmental impacts

attributable to specific aspects of different

production scenarios. In contrast to other

techniques such as ecological footprinting,

which allows an estimation of the ecosystem

support required to sustain various forms of

aquaculture production, the LCA framework

is used to evaluate the environmental “costs”

of individual energetic and material inputs

and outputs associated with each stage of a

production system (Figure 3). These costs are

expressed in terms of their relative potential

contributions to a range of global-scale envi-

ronmental problems (e.g. global warming,

eutrophication, biotic and abiotic resource

use, ozone depletion, eco-toxicity, and acidi-

fication) (Table 1). Such analyses facilitate

the identification of environmental “hot

spots” in production systems, providing a

clear basis upon which environmental perfor-

mance improvements can be made.

F i g u r e 3

Simplified life cycle diagram of a salmonnet-cage system

Production of fertilizer,machinery, pesticides

Farming and livestock rearing

Production of agriculturaland livestock inputs

Production of fishing

Reductionfisheries

Production of fishmealand fish oil

Salmonfeed

Salmonsmolts

Grow-out of salmonin net-cage

Processing

Retail and consumption

Transport atall stages

Production and supplyof energy (e.g. diesel,

electricity) at all stages

Wastemanagement

Materialinputs to hatchery

Hachery productionof smolts


T a b l e 1

Impact categoriescommonly employedin LCA research

IMPACT CATEGORY DESCRIPTION OF IMPACTS

GLOBAL WARMING CONTRIBUTES TO ATMOSPHERIC ABSORPTION OF INFRARED

RADIATION

ACIDIFICATION CONTRIBUTES TO ACID DEPOSITION

EUTROPHICATION PROVISION OF NUTRIENTS CONTRIBUTES TO BIOLOGICAL

OXYGEN DEMAND

PHOTOCHEMICAL OXIDANT FORMATION CONTRIBUTES TO PHOTOCHEMICAL SMOG

AQUATIC/TERRESTRIAL ECO-TOXICITY CREATES CONDITIONS TOXIC TO AQUATIC OR TERRESTRIAL

FLORA AND FAUNA

HUMAN TOXICITY CREATES CONDITIONS TOXIC TO HUMANS

ENERGY USE DEPLETES NONRENEWABLE ENERGY RESOURCES

ABIOTIC RESOURCE USE DEPLETES NONRENEWABLE RESOURCES

BIOTIC RESOURCE USE DEPLETES POTENTIAL PRIMARY PRODUCTION

OZONE DEPLETION CONTRIBUTES TO DEPLETION OF STRATOSPHERIC OZONE


Biophysical Performance of Aquaculture

Energy performance of aquaculture

Similar to other food production systems,

aquaculture involves the redirection, concen-

tration, and dissipation of various forms of

energy from the environment to facilitate the

growth of specific organisms (Troell et al.

2004). Given the diversity of farming sys-

tems, it is not surprising that different kinds

of aquaculture consume varied forms and

amounts of energy. In some cases, such as

the extensive culture of photosynthetic sea-

weeds or filter-feeding bivalves, all metabolic

energy is either self-produced or derived

from the immediate environment. Currently,

however, over one third of global aquaculture

output depends on productivity enhance-

ment achieved through the use of feeds

derived from off-farm sources (Tacon 2005).

In the United States, where commercial sea-

weed culture is limited and non-fed shellfish

culture accounts for only 20 percent of pro-

duction, this proportion is much higher

(FAO 2005). In addition to the metabolic

energy content of the feeds themselves are

the direct and indirect industrial energy

inputs associated with the materials, labor,

capital, and technology required to provide

both feed and an appropriate culture envi-

ronment (Figure 4).

Direct Energy Inputs The direct industrial energy dependence of

any particular culture system will vary with

the means of production, the intensity of the

operation, the degree of mechanization, and

the quality and quantity of feed used (Troell

et al. 2004). For intensive systems, this

includes the energetic costs of harvesting,

processing, and transporting feed compo-

nents from often-remote ecosystems. Further

direct energy inputs are typically required for

the hatchery production or wild harvest of

juveniles.

Indirect Energy InputsThe energy required to sustain human

labor inputs and to build and maintain fixed

capital assets, such as farm infrastructure,

processing facilities, harvesting machinery,

and transportation equipment, constitutes

the major indirect energy inputs to aquacul-

ture production systems. Depending on the

nature of the culture system, these inputs will

vary from negligible to considerable.

Extensive Aquaculture Production SystemsExtensive aquaculture typically requires rel-

atively small direct and indirect energy inputs

and consequently supplies a relatively low

yield of edible protein per unit area of pro-

duction when compared to more intensive

systems. Generally, this can be attributed

both to farming practices and to the feeding

requirements of the cultured organisms.

Many species farmed in extensive systems

subsist on locally available primary produc-

tivity (e.g., mussels, seaweed) or supplemen-

tal inputs of low-grade agricultural by-prod-

ucts (e.g., carp and tilapia), and require little,

if any, manufactured feed. Although produc-

tion may be enhanced through the applica-

tion of organic and inorganic fertilizers, these

are typically of relatively low energetic cost.

Depending on the expense and availability of

labor, extensive systems in industrialized


countries often have higher energy consump-

tion than comparable systems in less devel-

oped regions as fossil fuel or electricity inputs

are substituted for human power. The ener-

getic costs of material inputs, processing, and

transport will similarly vary depending on

the location and specific conditions of the

production facility (Troell et al. 2004).

Intensive Aquaculture Production SystemsIntensive aquaculture production systems

are characterized by a high throughput of

material and energy resources, and generate a

significantly higher edible protein yield per

unit area than do extensive systems. The

considerable energy requirements of inten-

sive aquaculture production result from a

combination of factors. These include the

level of mechanization and environmental

intervention required, the intensity of the

production system, the feeding requirements

of the species being grown, and the degree of

dependence on manufactured feeds.

The level of mechanization and environ-

mental intervention required for intensive

land-based systems results in substantially

higher energy consumption when compared

to open water systems. This is largely due to

water quality requirements. To maintain the

chemical and biological parameters necessary

for the health of the cultured organism,

water must be continuously pumped through

the system from an adjacent water body, or

recirculated. The latter option, which

requires aeration and waste removal, is par-

ticularly energy intensive. In open water sys-

tems, these services are provided by the nat-

ural environment—the sustainability of


F i g u r e 4

Simplified energy and matterflow model for various typesof seafood production, fromsolar fixation by plants innature to point of consump-tion by humans. In natural sys-tems, as well as extensivefarming systems, solar energyand the availability of nutri-ents constitute the typical lim-iting factors for production.Due to energy losses betweentrophic levels, the harvest perarea decreases with increas-ing trophic level. Extensiveseaweed and mussel farmingmay require additional mater-ial infrastructure inputs (buoys,ropes, etc.). The extensiveproduction of fish or shrimpusually requires the construc-tion and maintenance ofponds, and inputs of juvenilesand in some cases fertilizerand/or agricultural by-prod-ucts to maintain production.In contrast, intensive culturesystems require larger inputsof industrial energy, mainly toprovide formulated feeds andjuveniles, and maintainappropriate water qualityand stock health. (from Troellet al. 2004)

DECREASED

DIRECT

ECOSYSTEM

DEPENDENCE

INCREASED FOSSIL ENERGY INPUTS

Solarenergy

N, P

Capture fisheries

AUX.ENERGY MATERIALS LABOR CAPITAL

TECHNOLOGY

Seaweed farmingIndustrial fisheries

Mussel farming

Feed millAgriculture

Hatchery

Agriculture

Extensivefish/shrimp

pond fishingIntensive salmon/shrimp

pond/cage fishing

Systems dominated by ecological processes & cycling Increasing dependence on economic systems inputs

…approximately 90 percent of the totalindustrial energy inputs

to farmed salmon production are

associated with the provision of feed.


F I G U R E 5 A

F I G U R E 5 B

Chicken Rearing22%

Direct Fuel Inputsto Fishing22%

Feed Transport1%

Indirect Inputsto Fishing5%

Fish Reduction11%

Crop InputsProduction2%

Crop Input Processing1%

Fodder Corn Production

20%

By-Product Rendering3%

Component Transport7%

Feed Milling6%

FIGURE 5 Breakdown of industrial energy inputs to (A) intensive Atlantic salmon culture and (B) the production of a generic salmon feed in British Columbia, Canada as of the later 1990s (from Tyedmers 2000).

which relates to the productive and assimila-

tory capacity of the host ecosystem.

The feeding requirements of intensively

cultured organisms often play a major role in

the total energy demands of the production

system. For example, approximately 90 per-

cent of the total industrial energy inputs to

farmed salmon production are associated

with the provision of feed (Folke 1988,

Tyedmers 2000, Troell et al. 2004) (Figure 5,

Table 2). For those species, such as salmon,

that feed in the wild at mid-to-higher troph-

ic levels, formulated feeds often include rela-

tively high levels of animal-derived feedstuffs

such as fishmeal, fish oil and—less frequent-

ly—livestock processing wastes (Tacon

2005). It is important to note, however, that

the animal-derived fraction of a formulated

diet is not inherently fixed. As long as the

basic nutritional requirements of the cultured

species are met, relatively high levels of sub-

stitution of plant- and animal-derived inputs

are possible (Watanabe 2002). In recognition

of the limited nature of fishmeal and oil

resources relative to projected demand, the

salmon aquaculture industry (a major con-

sumer of global fishmeal and oil supplies) is

investing considerable effort to reduce the

proportion of these ingredients in feed for-

mulations and to identify appropriate plant-

derived substitutes (Watanabe 2002, Tacon

2004). Conversely, animal-derived inputs are


T a b l e 2

Direct and indirect energyinputs to arange ofmarine andfreshwateraquaculturesystems (modi-fied from Troellet al. 2004a)

SYSTEMCHARACTERISTICS MARINE FRESHWATER

SHRIMP SALMON MUSSEL CARP CARP

(RECIRCULATING (SEMI-INTENSIVE

POLYCULTURE) POLYCULTURE) CATFISH TILAPIA

Site area (hectares) 10 0.5 2 1 1 1 .005Production/yr (tons) 40 200 100 4 4.3 3.4 0.058Production/yr/ha (tons) 4 400 50 4 4.3 3.4 12

ENERGY INPUTSb (kJ/kg) EI % EI % EI % EI % EI % EI % EI %FIXED CAPITALStructures/equipment 2,500 2 5,940 6 2,700 58 3,556 7 2,114 8 11,691 10 0 0

OPERATING INPUTS–Fertilizer, chemicals 22,750 15 0 0 0 0 9,316 18 633 2 369 0 1,000 3–Seed 18,750 12 2,970 3 0 0 15 0 0 0 11,076 10 0 0–Feed 58,250 37 78,210 79 0 0 15,451 31 287 1 86,389 75 23,280 97–Electricity, fuel 54,250 35 11,880 12 1,900 42 21,928 44 26,176 97 5,415 5 0 0

TOTAL 156,750 99,000 4,600 50,265 27,096 114,940 24,000

ENERGY INTENSITYc

Edible product (MJ/kg) 275 142 12 84 45 192 40Edible protein (MJ/kg) 784 688 116 272 135 575 199

LABOR INPUTSPerson-days/ton n.a. 10 5 6.5 66.7 3.5 172.4Notes:

a) Original data sources Larsson et al. (1994) (shrimp), Stewart (1994) (salmon, mussel and tilapia), Bardach (1980) (recirc. carp), Singh and Pannu (1998) (semi-intensive carp), Waldrop and Dillard (1985) (catfish)

b) Energy inputs expressed in kilojoules per kilogram (wet weight)

c) Energy intensities expressed in megajoules per kilogram

increasingly being used to enhance the pro-

duction of certain herbivorous and detrivo-

rous fish, including carp and tilapia (Tacon

2004, 2005).

Given the central importance of the ener-

gy inputs associated with feed provision, any

steps to reduce the energy costs of feed can

significantly alleviate the environmental bur-

dens of intensive aquaculture as a whole. To

this end, plant-derived inputs are in general

less energy intensive than many animal-

derived alternatives (Tyedmers 2000), while

transport-related energy costs can generally

be reduced by using locally sourced inputs

(Troell et al. 2004). However, when making

specific substitutions, care must be taken

because there can often be hidden energy

costs that may result in counterintuitive out-

comes. For example, where feed inputs are

supplied by local fisheries, energy costs of

transportation will be small. However,

depending on the nature of the fishery, total

energy inputs may be much higher than

those associated with large-scale reduction

fisheries conducted half way around the

world (Tyedmers 2004), thereby negating

any benefits from reduced transport dis-

tances. Similarly, as some production tech-

nologies for feed ingredients derived from

agriculture are highly energy intensive (for

example fertilizer production and some spe-

cialized processing of certain crops), substitu-

tion of animal by-products with these ingre-

dients may not necessarily reduce energy

consumption. Improving energy efficiency in

feed production for intensive aquaculture

therefore requires a comprehensive under-

standing of the relative energy intensity of

producing alternative feed ingredients.

In discussing the relative merits of aqua-

culture, proponents often cite the feed-to-

flesh conversion efficiency of aquaculture

species relative to those obtained in terrestrial

livestock production systems as evidence of

their superior environmental performance

(Hardy 2001). Although comparing the con-

version efficiency of different animals in cul-

ture based on the mass of feed consumed rel-

ative to the mass of animal produced can be

misleading (feeds often have very different

nutritional value) there is no doubt that fish

are generally very efficient converters of the

food energy they ingest. This is due in part

to the fact that cold-blooded aquatic organ-

isms require much less energy to fuel meta-

bolic processes and consequently are able to

utilize a much higher proportion of ingested

food energy for biomass gain. In contrast,

warm-blooded animals metabolize as much

as 90 percent of food energy to maintain

body temperature alone. However, unless

such comparisons include the full range of

energetic costs associated with feed provision,

this argument is somewhat misleading. For

example, the ratio of industrial energy

requirements to edible protein energy output

of intensive net cage culture of salmon is

greater than that associated with production

of milk, eggs, and even broiler chicken, and

is similar to that of feedlot beef production

(Table 3). This is largely due to the substan-

tial energy inputs associated with the nutri-

tionally dense, concentrated feeds used. By

comparison, more traditional low-input cul-

ture of carp and tilapia in extensive systems

requires five to 15 times less industrial ener-

gy per unit of edible protein energy pro-

duced, while semi-intensive tilapia culture

requires less than half as much energy

(Table 3).

Sustainable development requires attention

to a combination of social, economic, eco-

logical, and technological parameters.

However, understanding the comparative

energy dependencies of aquaculture produc-

tion systems and the associated costs to both

the natural environment and society is of

particular importance to ensure the sustain-

able development of the aquaculture indus-

try as a whole. Given growing concerns in

the United States over the impacts of climate

change and dependence on foreign sources of

oil, the relative energy efficiency of aquacul-

ture systems should be considered an impor-


Given growing concernsin the United States over the impacts of

climate change anddependence on foreignsources of oil, the relative

energy efficiency ofaquaculture systems

should be considered an important indicator of

sustainability.

tant indicator of sustainability, and should

figure prominently in any national policies

relating to how the aquaculture industry

develops. This could include financial or

other measures to encourage the develop-

ment of those forms of culture that existing

data and experience tell us are inherently less

energy intensive. In the marine context this

includes all unfed plant and bivalve culture

systems. Where more intensive culture sys-

tems will be pursued, incentives might be

provided to undertake analyses to identify

and adopt opportunities to improve their

energy performance.

Ecological Footprint of Aquaculture

Several studies have used ecological foot-

print analysis to evaluate the productive

ecosystem capacity required to sustain differ-

ent forms of aquaculture production (Folke

et al. 1998). To date, researchers have vari-

ously examined tilapia, shrimp, and salmon

culture systems.

Folke (1988) evaluated the amount of pri-

mary production appropriated by the culture

of Atlantic salmon in the Baltic Sea, and

found that the production of the fish compo-

nent of salmon feed required a supporting

marine production area 40,000 to 50,000

times larger than the surface area of the cul-


T a b l e 3

Ranking offoods (aqua-culture prod-ucts in bold) by ratio of edible proteinenergy outputto industrialenergy inputs

Protein Energy Output/Industrial Energy Input

Food Type (technology, environment, locale) (%)

Carp (extensive freshwater pond culture, various) 100 – 11a

Herring (purse seining, North Atlantic) 50-33b

Seaweed (marine culture, West Indies) 50-25a

Chicken (intensive, U.S.A.) 25c

Salmon (purse seine, gillnet, troll, NE Pacific) 15 - 7b

Tilapia (extensive freshwater pond culture, Indonesia) 13a

Rainbow Trout (intensive net pen culture, Finland, Ireland) 13 - 4.2a

Cod (trawl and longline, North Atlantic) 10 - 8b

Mussel (marine longline culture, Scandinavia) 10 - 5a

Turkey (intensive, U.S.A.) 10c

Carp (unspecified culture system, Israel) 8.4a

Wild-caught seafood (all gears, marine waters, global average) 8.0d

Milk (U.S.A.) 7.1c

Swine (U.S.A.) 7.1c

Tilapia (unspecified freshwater culture system, Israel) 6.6a

Tilapia (freshwater pond culture, Zimbabwe) 6.0a

Shrimp (trawl, North Atlantic and Pacific) 6.0 – 1.9b

Beef (pasture-based, U.S.A.) 5.0c

Catfish (intensive freshwater pond culture, U.S.A.) 3.0a

Eggs (U.S.A.) 2.5c

Beef (feedlot, U.S.A.) 2.5c

Tilapia (intensive freshwater cage culture, Zimbabwe) 2.5a

Atlantic salmon (intensive marine net pen culture, Canada) 2.5a

Shrimp (semi-intensive culture, Colombia) 2.0a

Chinook salmon (intensive marine net pen culture, Canada) 2.0a

Lamb (U.S.A.) 1.8c

Seabass (intensive marine cage culture, Thailand) 1.5a

Shrimp (intensive culture, Thailand) 1.4a

Sources: a. Troell et al. 2004, b. Tyedmers 2004, c. Pimentel 2004, and d. Tyedmers et al. 2005

ture facility. Based on these results, Folke and

Kautsky (1989) calculated that the

Norwegian salmon aquaculture industry of

the late 1980s depended on primary produc-

tion equivalent to 15 percent of that pro-

duced in the North Sea.

Berg and colleagues (1996) compared the

ecological support requirements for semi-

intensive pond farming and intensive cage

farming of tilapia and found that the inten-

sive system appropriated a much greater area

of ecosystem support than did the pond cul-

ture system (Figure 6). The production of 1

kg of fish in the intensive cage farming sys-

tem was found to require 1.5 times more

industrial energy than in the semi-intensive

pond farming system. Intensive cage farming

was also found to require 21,000 m2 of

ecosystem support area per square meter of

farm area for feed production, and 160 m2

and 115 m2 for oxygen production and

nutrient assimilation respectively (Figure 6).

Larsson and colleagues (1994) estimated

the spatial ecosystem support required to

operate semi-intensive shrimp aquaculture in

Caribbean Colombia. The ecological foot-

print for this type of culture system was cal-

culated to be 35 to 190 times larger than the

area of the farm itself. More than 80 percent

of the primary productivity required to

supply feed was external to the farm, and the

ratio of industrial energy inputs to edible

shrimp output was 40:1.

In the only known analysis to directly

compare competing wild capture fisheries

and culture systems, Tyedmers (2000) calcu-

lated the ecological footprint of salmon fish-

eries and aquaculture in British Columbia,

Canada, and found that salmon farming was

less eco-efficient than commercial salmon

fisheries for chinook, coho, sockeye, chum,

and pink salmon. While this study was based

on data that are now almost a decade old,

cultured chinook salmon appropriated a total

of 16 hectares (ha) of marine and terrestrial

ecosystem support area per metric ton pro-

duced while Atlantic salmon culture required

12.7 ha/mt. In contrast, the least eco-

efficient species harvested commercially

(chinook) appropriated 11 ha/mt, while

the most eco-efficient fishery (pink salmon)

required only 5 ha/mt (Figure 7).

The results of the above analyses under-

score the need to consider a broad range of

material and energetic processes linked to

aquaculture production as possible in order

to arrive at representative evaluations of the

relative sustainability of particular produc-

tion systems. While the physical area of an

aquaculture facility may be quite small, the

F i g u r e 6

Ecosystem supportareas required tosustain one squaremeter of intensivecage farming andsemi-intensive pondfarming (modifiedfrom Berg et al.1996)

Aquatic Ecosystem

Oxygen Production 160m2

PhosphorousAssimilation 115m2

Feed Production21,000m2

Feed Production420m2

Agriculture Ecosystem

Phosphorus Assimilation0.9m2

Oxygen Production0.5m2

Enlarged Pond Surface

Semi-Intensive Ponds1m 2

Intensive Cages1m 2


ecosystem support area required to sustain

feed and other inputs and assimilate resulting

wastes can be dramatically larger than the

area of the farm itself. This is particularly

true in the case of intensive production

systems, where the material and energy

throughputs are largely independent of the

farm’s actual location and physical dimen-

sions. In contrast, less intensive systems may

require little, if any, inputs beyond those

which can be supplied by the ecosystem

goods and services present within the farm’s

boundaries.

Concerns have been raised by some in

response to ecological footprint evaluations

of aquaculture described above. The method-

ology has been challenged for failing to

account for differences in the efficiency of

alternative production technologies in pro-

ducing similar outputs (Roth et al. 2000). It

is important to note, however, that this cri-

tique is misplaced. For most analysts, a com-

parison of the relative efficiency of the cul-

ture systems considered was never intended

(Kautsky et al. 1997). Roth and colleagues

(2000) have also contended that area-based

calculations of appropriated ecosystem goods

and services may also result in double count-

ing when a given area of productive ecosys-

tem provides more than one service and is

counted repeatedly for each service. While

this is a conceptual possibility given the mul-

tifunctional nature of ecosystems, this cri-

tique is again misguided as all footprinting

analysis of culture systems have been sensitive

to and acknowledge the potential overlap-

ping nature of some areas of ecosystem sup-

port (Larsson et al. 2000). Roth and col-

leagues (2000) also voice concern that

expressing ecosystem appropriation in terms

of water surface area obscures potential dif-

ferences in various water areas – in particular,

the quantity and quality of available renew-

able resources and services. They further

argue that ecological footprinting provides a

static, simplistic interpretation of complex

biological systems because the choices made

regarding which natural resource consump-

tion patterns to track, and by which relation-

ships to express them, are unlikely to capture


F i g u r e 7 .

Area of marine andterrestrial ecosystemsupport appropriatedto sustain the produc-tion of one metric tonof salmon in BritishColumbia, Canada(from Tyedmers 2000).

16

14

12

10

8

6

4

2

0Cultured Cultured Captured Captured Captured Captured Captured Atlantic Chinook Chinook Coho Sockeye Chum Pink

hec

tare

s p

er m

etr

ic to

n

TERRESTRIAL ECOSYSTEM

SUPPORT

MARINE ECOSYSTEM

SUPPORT

the dynamic reality of both the system being

studied and its ecological context. This

apparent “limitation” of ecological footprint-

ing, however, is well understood and com-

municated by practitioners. All models are

simplifications of reality but remain valid

and valuable for specific purposes.

In recognition of these potential limita-

tions, ecological footprint analysis must be

conducted with care and rigor. Avoiding

double counting necessitates understanding

and adjusting for the multiple services pro-

vided by different ecosystems. Moreover, to

account for differences in the quality and

quantity of goods and services provided,

adjustments must be made for the biological

productivity of each area. Finally, because it

is unlikely that ecological footprinting, or

any comparable tool, will ever be able to

account for every aspect of complex ecologi-

cal systems, the results derived from such

analyses should be interpreted as conserva-

tive. Despite these limitations, this technique

can provide valuable insights into the often-

hidden material and energy dependencies of

intensive aquaculture production by making

visible the ecosystem goods and services

required to satisfy these demands.

Life Cycle Assessment ofAquaculture

Published LCA results for aquaculture pro-

duction systems include Norwegian salmon

(Ellingsen and Aanondsen 2006), Thai

shrimp products (Mungkung 2005), French

farmed trout and salmonid feeds

(Papatryphon et al. 2003, 2004), and

Finnish trout production (Seppälä et al.

2001). Additional data are also available for

Danish trout, Greek sea bass, and French

turbot aquaculture. While these studies have

dealt with relatively diverse production sce-

narios (land-based, marine, and fresh water)

and culture organisms, a comparison of life-

cycle impacts indicates a number of striking

similarities between systems.

In almost every production system stud-

ied, the environmental costs associated with

the provision of feeds dominate most, if not

all, impact categories considered. For exam-

ple, Papatryphon and colleagues (2003)

found that feed production for intensive,

freshwater-based rainbow trout culture in

France accounted for 52 percent of the total

energy use, 82 percent of the contributions

to acidification, 83 percent to climate

change, and 100 percent of biotic resource

use. Similarly, Seppälä and colleagues (2001)

reported that the production of raw feed

materials together with the manufacturing of

feed were responsible for most of the atmos-

pheric emissions associated with rainbow

trout aquaculture in Finland. More striking

still, Ellingsen and Aanondsen (2006) found

that feed provision accounted for the majori-

ty of environmental burdens in all impact

categories considered in their analysis of

Atlantic salmon culture, while an LCA of

Danish trout production showed that feed

production and use accounted for the major-

ity of total impacts in six of the ten impact

categories analyzed (Figure 8) (LCA of Food

2006).

Closely related to feed provision, eutrophi-

cation impacts arising from nitrogen and

phosphorous emissions have also been found

to be significant across production systems.

Seppälä and colleagues (2001) reported that

nutrient emissions to water on the farm were

much more significant in terms of environ-

mental impact than atmospheric emissions.

LCA research of Greek sea bass aquaculture

indicated the importance of feed composi-

tion and management in mitigating eutro-

phying emissions.

These results are not surprising consider-

ing the fossil fuel and material consumption

associated with reduction fisheries, agricul-

tural production systems, fish-rendering and

fish feed plants, and the infrastructure

required to supply and process the con-

stituent ingredients and to transport

products used in intensive aquaculture.

Efforts to mitigate the environmental


Efforts to mitigate the environmental impacts

of intensive aquaculturemust pay considerable

attention to improving theeco-efficiency of feedproduction and use.

impacts of intensive aquaculture must there-

fore pay considerable attention to improving

the eco-efficiency of feed production and

use.

In this regard, Papatryphon and colleagues

(2004) evaluated the life cycle impacts of

four hypothetical salmonid feeds in order to

assess the effect of varying feed formulations

on overall environmental performance. The

four feeds compared were a high fishmeal

diet, a high fish by-product meal diet, a

low fishmeal diet, and a no fishmeal diet.

Overall, the results of this study indicated

that substitution of feed ingredients creates

trade-offs in terms of environmental impacts.

Switching from the HF to the NF feed was

shown to decrease the eutrophication poten-

tial and net primary productivity use, but to

increase the overall global warming potential

(Figure 9). Considering these trade-offs, the

perceived costs and benefits of alternative

feed formulations will depend upon which

environmental impacts are thought to be the

most important.

Since many of the ingredients used in con-

centrated aquafeeds are by-products or co-

products from other food production sys-

tems, the manner in which the researcher

decides to allocate environmental burdens

between these products will greatly influence

the apparent environmental performance of

the feed. For example, fisheries by-products

have traditionally been considered a waste

product, which—fortunately—can be cycled

back into the nutrient stream as a feed ingre-

dient in various animal husbandry systems.

For this reason, some researchers have chosen

to allocate the majority of the associated

environmental impacts to the primary

product, in this case fish flesh for human

consumption. However, given the substantial

industrial energy and ecosystem support that

underpins the provision of these by-products

for use in other industries, they are certainly

not “free” in a biophysical sense. Moreover,

the limited nature of fishmeals and oils from

reduction fisheries, and comparable substi-

tutes generally, suggests that using fisheries

by-products will not result in decreased


100%

80%

60%

40%

20%

0

FEED

INORGANIC CHEMICALS

TRANSPORT

ELECTRICITY

F i g u r e 8

Relative contribution of keyinputs to rainbow trout pro-duction in Denmark (from LCA of Food 2006)

demand from dedicated reduction fisheries.

In fact, at current rates of expansion, it is

predicted that the global aquafeed industry

will require 70 percent of the average histori-

cal fishmeal supply and 145 percent of the

fish oil supply by 2015 (New and

Wickstrom 2002, Tacon 2005). Simply

advocating increased reliance on fisheries by-

products over reduction fisheries, which are

generally much less energy intensive than

most fisheries for human consumption

(Tyedmers 2004), does not necessarily

address this problem. Because demand by

aquaculture and other industries for animal

by-products saturates and even exceeds sup-

ply, regardless of origin, it is more appropri-

ate to compare the full costs of producing

alternative ingredients. This applies equally

to choosing by-product or co-product ingre-

dients of plant or livestock origin.

As was the case with respect to energy

inputs, the environmental costs of feed pro-

duction will be relatively high, regardless of

the ingredients chosen, if the feeds contain

substantial fractions of animal by-products

(which is often the case in the culture of

higher trophic level species). Decisions

regarding the use of these limited resources

should therefore be aimed at maximizing

end-use efficiency—for example, by both

developing suitable plant-derived substitutes

and choosing culture organisms that require

less nutrients of animal origin. Moreover,

aquaculture development policy should

explicitly address both the biophysical costs

and the limited nature of animal-derived

feed ingredients, by encouraging the devel-

opment of production capacity and markets

for low trophic level species.

In open-water production systems, such

as net cage salmon aquaculture, the majority

of life-cycle costs are directly attributable to

feed provision. However, LCA research of

land-based aquaculture facilities indicates

that the energy inputs required to maintain

water quality and oxygen levels can also

contribute substantially to the overall

environmental costs of the production

system (Box 3). For example, Papatryphon

and colleagues (2003) found that production

intensity during the dry summer months,

when higher levels of fuel and electricity

were required for water aeration and circula-

tion, was an important indicator of overall


0.5

0.4

0.3

0.2

0.1

0Energy Use Biotic Resource Global Warming Acidification Eutrophication Use Potential

Rela

tive

Co

ntrib

utio

n

HIGH FISHMEAL

HIGH FISH BY-PRODUCT

LOW FISHMEAL

NO FISHMEAL

F i g u r e 9

Relative tradeoffs associated with four hypothetical salmonidfeeds (from Papatryphonet al. 2004)

…the environmental costs of feed production

will be relatively high if thefeeds contain substantial

fractions of animal by-products.

environmental performance. Similarly, in an

LCA of Thai shrimp aquaculture,

Mungkung (2005) found that energy inputs

for aeration contributed heavily to the envi-

ronmental costs of production. An LCA

study of turbot production in a land-based

recirculating system showed that energy use,

global warming, and acidification impacts

were particular environmental hot spots, and

were largely a function of the quantity and

origin of the energy used. Danish LCA

research of trout production similarly report-

ed high global warming and toxicity impacts

associated with on-farm energy inputs for

aeration and recirculation because the

electrical energy used was generated from

natural gas.

These results consistently indicate the

appreciable energy demands of closed-

containment aquaculture. While opponents

of open-water aquaculture have often

championed these technologies as a general

panacea, such a perspective fails to account

for the broader range of environmental

impacts related to the considerable energy

consumption of these systems, and the

implications for overall environmental

performance. Policy decisions regarding best


B o x . 3

Evaluating EmergingAquacultureTechnologies:How Do We DefineSustainability?

Concerns over the sustainability and ecological impacts of some forms of intensive aquaculture (partic-ularly the farming of high-value species such as salmon) have led to considerable research on alterna-tive technologies. Many of these technologies have been designed primarily to reduce or eliminate thepotential impacts of aquaculture on wild fish populations (genetic impacts, spread of disease) and toreduce the impacts of aquaculture wastes on the benthic environment. In particular, research hasrecently focused on the development of land-based tank systems for the culture of relatively high-valuefinfish such as Atlantic salmon, turbot, halibut and Arctic char.

Land-based tank systems make use of concrete tanks to culture fish in a controlled environment. Wateris pumped into the tanks from an adjacent water body or well, and either flows through the systemback into the adjacent water body or is recirculated within the system. Fish are isolated from wild popu-lations, and fish wastes and uneaten feed pellets can be collected and treated.

What is perhaps most interesting about these alternative technologies is that they are often promotedas sustainable fish production systems. In a recent pilot study of a land-based tank system in BritishColumbia, the salmon produced were sold at local retail outlets as “environmentally friendly eco-salmon” (BCMAL, 2004). A recently developed land-based Arctic char farm in Nova Scotia has beenpromoted as a sustainable fish production system that will evolve into a “zero-impact facility”(Summerfelt et al, 2004). These claims of sustainability appear to be based on the fact that impacts onwild fish populations are reduced or eliminated, and waste products from the fish farms can potentiallybe collected and treated or recycled. However, these assertions of sustainability are based on a rathernarrow set of criteria, and do not take into account the environmental impacts of the considerableenergy inputs required to operate these systems.

The advantage of growing fish in an open net cage is that the rearing conditions are largely main-tained by the natural ecosystem services provided by the ocean. These include the generation of cur-rents, the flushing of wastes by tidal action, and the provision of dissolved oxygen. By culturing fish inland-based tank systems, the ecosystem services provided by the ocean are no longer accessible, andtechnology must be used to maintain the appropriate rearing conditions. This can include pumpingwater in and out of the system, the generation of oxygen, the collection and treatment of waste, andthe maintenance of optimal temperature by heaters and chillers. The operation of these technologiesrequires considerable amounts of energy in the form of fuels and electricity. This was evidenced in therecent pilot study of a land-based salmon farm in British Columbia, where it was determined that theenergy costs incurred to continuously pump water and maintain infrastructure had a significant effecton the economic viability of the farm (BCMAL, 2004). The magnitude of these energy inputs, coupledwith the direct and indirect energy inputs required to manufacture the feed that is used, suggest thatenergy use and its associated impacts (e.g. fossil fuel depletion, greenhouse gas emissions) should alsobe considered as significant indicators of the sustainability of land-based tank systems and otheremerging aquaculture technologies.

choices for culture technologies should there-

fore carefully weigh the biophysical costs and

benefits of alternative production scenarios.

While life cycle assessment research can

contribute to a better understanding of the

relative contribution of specific life-cycle

activities to the overall environmental perfor-

mance of a production system, the degree of

representation of actual environmental costs

that can be achieved will be determined by

the range of impact categories considered.

At present, the standard suite of impact cate-

gories used in most LCA research (Table 1)

focuses attention on core sustainability issues

related to eco-efficiency and broad-scale

environmental impacts that are often over-

looked in public discourse concerning the

environmental interactions of aquaculture

production. However, while these categories

are highly relevant, there are numerous other

environmental burdens unique to aquacul-

ture production systems—such as the

transmission of diseases and parasites

between farmed and wild organisms, impacts

to the benthos from wastes emitted from

open-water culture facilities, and the poten-

tial alteration of trophic dynamics resulting

from large-scale reduction fisheries—that are

currently not quantifiable within the LCA

framework. For this reason, the results

derived from life cycle assessments do not

alone provide sufficient grounds for decision

making. Beyond informing recommenda-

tions for specific product/process improve-

ments, LCA should therefore be treated

as just one tool among many in decision-

making processes.


Aquaculture, Aquatic Ecosystems and Efficiency

As aquaculture has grown in importance,

so too has interest in the role it plays in

relieving or exacerbating pressure on aquatic

ecosystems (Naylor et al. 2000, Pike 2000,

Tidwell and Allan 2001, Tacon 2004, Naylor

and Burke 2005, Pike 2005). Although this

issue represents just one aspect of the bio-

physical sustainability of aquaculture, and is

partially reflected in one of the techniques

described above (ecological footprint analy-

sis), it is of broad concern and is arguably of

importance in and of itself to any discussion

of the eco-efficiency in this sector. In partic-

ular, the scale of dependence of some forms

of aquaculture on commercially harvested

aquatic resources deserves careful considera-

tion with regards to both the environmental

and social objectives of providing food for

people. In this context, the relative efficiency

of aquaculture and other food production

systems can and should be compared.

However, the basis upon which efficiency is

assessed and compared can result in some-

times confusing or contradictory outcomes.

Many forms of aquaculture—including all

plant and mollusk culture systems, polycul-

ture systems that incorporate these species,

and many forms of extensive to semi-inten-

sive freshwater finfish production—make a

clear positive contribution to the global sup-

ply of marine resources potentially available

for human consumption. By relying on little,

if any, input of aquatic resources to produce

generally high quality seafood, these systems

reduce the economic incentive to fish the

world’s oceans more heavily than might

otherwise be the case. This represents a com-

pelling argument for the role of some forms

of aquaculture in reducing the direct

exploitation of aquatic ecosystems.

However, it is also clear that not all forms

of aquaculture currently make such a net

positive contribution. This is largely, but not

exclusively, the result of including fish har-

vested from the wild, or their reduced prod-

ucts, in aquaculture feeds for intensive pro-

duction systems (Table 4). The most obvious

example is the use of largely unprocessed

wild-caught fish, typically comprised of

species available close at hand to the culture

system, as the primary feed input to a num-

ber of emerging marine-based culture sys-

tems. These include bluefin tuna-fattening

operations in Australia, the Mediterranean

and elsewhere, and the culture of various

grouper and other high value marine species

in Southeast Asia (Tacon 2004, Naylor and

Burke 2005). Given the developmental

nature of these culture systems and the

requirements of the species being farmed,

as much as 15 to 20 kilograms of raw fish

inputs may be used for every additional

kilogram of cultured fish produced (Tudela

2002, Weber 2003, Naylor and Burke 2005).

The source of the fish inputs vary but

include dedicated fisheries for a range of

small pelagic species (e.g., herring, anchovy,

and mackerel species), as well as the use of

so-called “trash fish”—typically the low

value by-catch of local fisheries for human

consumption. Despite the mass of aquatic

resource inputs to these systems relative to

the additional production they yield, they

remain economically viable and are expand-


ing. Indeed, bluefin tuna fattening has

experienced average rates of growth of over

15 percent per annum in the last decade

(FAO 2005). Notwithstanding this relatively

rapid expansion, these systems will likely

continue to represent a small fraction of

global aquaculture production.

Much more common, and accounting

for significantly higher levels of total aquatic

resource consumption, is the semi-intensive

and intensive culture of species whose diets

include quantities of fishmeal and oil. These

include salmonids, most crustaceans, eels,

and a growing number of marine finfish

(Table 4) (Tacon 2004).

In most years, four of the five largest ton-

nage capture fisheries globally targeted small

pelagic species for reduction to meal and oil

for use in animal feed production. A sub-

stantial and increasing fraction of reduction

fisheries landings is appropriated for use in

aquaculture production of both carnivorous

species and herbivorous/detrivorous species

that are being raised as functional carnivores

(Tacon 2004). In fact, fully one third of

global fisheries landings are dedicated to the

feed production sector (Naylor et al. 2000,

Tacon 2004).

While aggregate reduction fisheries land-

ings have remained stable over several

decades, the allocation of fishmeal and oil

between competing animal production

sectors has undergone rapid change, with

increasing volumes diverted to the aquacul-

ture sector (Naylor et al. 2000, New and

Wijkstrom 2002, Tacon 2004). Between

1988 and 1997, the amount of fishmeal

used by the aquaculture industry rose from

10 percent to 33 percent of total supplies

(Naylor et al. 2000). Recent estimates suggest

that aquafeeds currently claim between 42

and 46 percent of total available fishmeal and

approximately 80 percent of global fish oil

(Tacon 2004, Pike 2005). Moreover, at cur-

rent rates of expansion, it is predicted that

the global aquafeed industry will require

70 percent of the average historical fishmeal

supply and 145 percent of the fish oil supply

by 2015 (New and Wijkstrom 2002).

What is often overlooked or too quickly

dismissed is that virtually all species used in

reduction can also be, and in many cases are,

used for direct human consumption.

Obvious examples include Atlantic herring

and mackerel (both of which are readily

available in a variety of smoked, pickled, and

canned forms), and capelin. Even parts of the

generally reviled menhaden, the main species

used to produce fishmeal and oil in the

United States, is consumed by some commu-

nities in the United States (Smith and

Ahrenholz 2000). Moreover, it can also be

processed into a modestly suitable surimi

(Hale and Bauersfeld 1991). Jack mackerel

and Peruvian anchoveta, the two species

which in most years account for the largest


T a b l e 4

Ratio of live weight fishused in feed to liveweight of culturedproduct produced

Cultured Organism Naylor and Burke (2005)a Tacon (2005)b

Eels 4.41 3.1-3.9Salmon 2.13 3.1-3.9Marine fish 3.84 2.5-3.2Trout 1.4 2.5-3.1Marine shrimp 2.05 1.6-2.0Freshwater crustaceans n/a 0.9-1.1Milkfish 0.75 0.3-0.37Tilapia 0.47 0.23-0.28Catfish 0.23 0.22-0.28Fed carp 0.47 0.19-0.24Notes: a) data represent 2002, b) data represent 2003.

….at current rates ofexpansion, it is predictedthat the global aquafeed

industry will require 70 percent of the

fishmeal supply and 145 percent of the fish

oil supply by 2015.

tonnages destined for reduction globally, are

also highly consumable species. Indeed, the

FAO is actively promoting the consumption

of anchoveta amongst poor Peruvians (FAO

2001), and canned or smoked jack mackerel

is commercially available in many countries.

It would appear, then, that the only barriers

to further direct human consumption of

these species are related to historic cultural

preferences (FAO 2001), well-established

access rights and processing infrastructure,

or the simple reality that the international

animal feeds industry can pay more for their

rendered products than many potential

consumers can afford.

These issues have, in part, also given rise

to comparisons between the relative efficien-

cy of some forms of aquaculture and other

food production systems. For example, the

website for Salmon of the Americas (SOTA

2006) states that it takes three pounds of for-

age fish to produce one pound of farmed

salmon, which (it is claimed) represents a sig-

nificant ecological advantage when compared

to the apparent 10-15 pounds of wild forage

fish required to produce one pound of wild

salmon. While seemingly a straightforward

and reasonable comparison, it overlooks a

number of critical issues. Not least of these is

the presumption that the species destined for

reduction and ultimately for incorporation in

aquafeeds are ecologically equivalent to the

prey of the wild-caught species. Put another

way, any meaningful comparison of the

aquatic inputs to wild-caught and cultured

fish must first account for the trophic levels

of the species being consumed.

Trophic transfer efficiency, which is the

fraction of energy incorporated by organisms

relative to the total amount ingested, is

generally in the range of 10 percent in most

aquatic ecosystems (Pauly and Christensen

1998). While it is certainly true that farmed

salmon consume a lesser mass of forage fish

on a pound-for-pound basis, the trophic

levels of forage fish reduced for salmon feeds

and that of the prey of wild salmon can be

substantially different. For example, the

trophic level of many of the species incorpo-

rated into the feeds used to grow salmon in

British Columbia in the late 1990s was

similar to or higher that the average trophic

level of most species of salmon in the wild

(Tyedmers 2000). For such a comparison to

be fair, one should calculate the net primary

productivity appropriated by both the cul-

tured fish and the same fish in the wild.

Since the prey of wild salmon typically

occupy lower trophic levels than those of

many of the forage fish targeted by reduction

fisheries, this calculation would yield a

dramatically different result (Table 5).

Moreover, since fishmeal and oil contribute

less than half of the ingredients in concen-

trated salmon feeds, it would also be neces-

sary to include the net primary productivity

appropriated to produce the crop and animal

by-product ingredients that comprise the

balance of the feeds. Finally, when making

comparisons of aquaculture systems and

T a b l e 5

Estimated carbonappropriation andindustrial energy inputsrequired to produceone metric ton ofsalmon in intensive culture and wild capture systems.

Cultured/Captured Carbon Appropriation Industrial Energy Organism (metric tons) (gigajoules)

Farmed Atlantic Salmon 26.4 69.0Wild Pink Salmon 15.4 22.3Wild Chum Salmon 15.8 24.0Wild Sockeye Salmon 16.9 27.2Wild Coho Salmon 31.5 41.2Wild Chinook Salmon 38.7 35.2Note: Data recalculated from Tyedmers (2000) using updated feed formulations and a gross feed conversion ratio of 1.2:1.


aquatic ecosystems generally, it is important

to at least acknowledge (and better to explic-

itly calculate) the substitution of industrial

energy inputs (fossil fuel, electricity, etc.) to

the culture system for the metabolic energy

expended in ecosystems (Table 5). Put

another way, part of the seeming “inefficien-

cy” of aquatic ecosystems simply reflects the

fact that we are substituting exogenous

energy inputs to capture, process, and deliver

feeds, etc., for solar driven processes within

ecosystems.

The SOTA website also advances compar-

isons of the gross feed conversion efficiency

in salmon production (currently in the range

of 1.2:1) relative to beef, pork and chicken

production, for which the gross feed

conversion ratios provided are 10:1, 5:1, and

2:1 respectively. Although there are certainly

efficiency gains inherent in the culture of

cold- versus warm-blooded species, such a

coarse comparison overlooks the very great

differences in nutrient density of the typical

feeds used and the source of those nutrients.

As before, a fuller accounting would address

the relative trophic level and associated

appropriation of net primary productivity of

feed ingredients being used in each culture

system. For example, it seems highly unlikely

that 50 percent of the diet of beef, pork, or

poultry consists of animal-derived inputs, as

is typical for salmon feeds.


Conclusions

Aquaculture represents an important and

growing source of animal protein globally.

As a major seafood consumer with a strong

interest in dramatically expanding domestic

aquaculture production, the United States

has an opportunity to guide the future

sustainable development of this sector both

at home and abroad through the choices it

makes. To this end, we argue that a strong

sustainability objective should be adopted

to guide the development of the industry

domestically. To achieve these ends, efforts

should be made to maximize the eco-effi-

ciency of the sector as a whole and through-

out its various components. This should be

achieved by promoting forms and practices

and undertaking further research where

necessary to reduce resource use and waste

generation intensity. Given the highly diverse

forms of aquaculture currently in use and

the enormous potential for eco-efficiency

improvements to be made to existing systems

as they mature and evolve, it is entirely

possible for the United States to become a

world leader in eco-efficient aquaculture.

To achieve this objective, the following

insights from existing research should be

borne in mind. Although extensive culture

systems typically deliver lower yields per unit

area occupied by the farm site, they are

generally much less material and energy

intensive, and consequently result in smaller

environmental burdens per unit of protein

produced than intensive systems. This is

particularly evident with respect to the use of

industrial energy inputs. While all forms of

industrialized food production are highly

dependent on substantial energy inputs,

typically requiring orders of magnitude more

industrial energy input than they return in

edible energy outputs, extensive aquaculture

systems are amongst the most energy efficient

producers of animal protein currently in

operation. In contrast, existing data suggests

that many forms of intensive aquaculture are

amongst the least energy efficient protein

producers even when compared to many

of the capture fisheries with which they

compete (Table 3).

Many of the most highly desired species,

including most marine and anadromous

finfish and shrimp, are difficult if not

impossible to culture extensively. Where

these species are to be produced, efforts

should be made to dramatically improve

the eco-efficiency of current practices. In this

regard, a great deal of attention must be paid

to reducing the biophysical costs of providing

feeds. Not only does the provision of feed

account for the large majority of industrial

energy inputs to most intensive culture

systems currently, it also drives most of the

life cycle environmental impacts and

accounts for the majority of net primary

productivity appropriated by the culture

system. While it may be impossible to reduce

the impacts of feed provision to trivial levels,

substantial improvements are possible. This

is in part because to date, most feed formula-

tion decision making has been guided

without much attention to the associated

biophysical impacts. When looking to

improve the eco-efficiency of feeds, however,

care must be taken to not simply adopt


seemingly obvious solutions: informed

decisions are required. In many cases, sourc-

ing inputs locally may be more eco-efficient

than importing feedstuffs from afar.

However, in some instances distantly sourced

inputs may be obtained at very much lower

material and energy costs, easily offsetting

the impacts of greater transport distances.

In general, however, less processed inputs are

likely to result in lower impacts than highly

refined alternatives, all other things being

equal, while plant-derived inputs typically,

but not always, have much lower associated

biophysical costs than do their animal-

derived substitutes.

Looking beyond the impact of feeds, in

those intensive culture systems that rely on

wild-sourced farm stock, the acquisition of

that stock can result in nontrivial biophysical

costs both in terms of energy expenditures

and impacts on associated ecosystems.

Although impacts associated with supplying

juveniles cannot be eliminated, they may be

substantially reduced when the production

cycle is effectively closed through the use of

hatcheries or related techniques.

Even though very little hard data exist

regarding the biophysical performance of

some of the newest forms of intensive culture

being adopted, including a range of land-

based systems that are promoted as “green”

alternatives to more conventional systems,

caution should be exercised before they are

more widely deployed. In particular, the

apparently high levels of energy required

to supply and maintain water quality, and

process resulting wastes, suggest that their

eco-efficiency may be much lower, even if

they address a number of proximate environ-

mental concerns associated with the more

conventional open-water systems currently

in use.

In closing, aquaculture must have a bright

future if we are to maintain, let alone

increase, access to highly nutritious seafood

globally. The challenge we all face in a world

in which critical ecosystem functions are

threatened by the current levels of human

activity is to identify and promote those

forms of culture that will allow us to increase

supply while minimizing the associated

biophysical costs. This is a challenge that

must be faced and that can be overcome

with attention and dedication to the

common objective.


References

Alongi, D. (2002) Present state and future

of the world’s mangrove forests.

Environmental Conservation 29(3):331-

349.

Andersson, K., Ohlsson, T., Olsson, P.

(1998) Screening life cycle assessment

(LCA) of tomato ketchup: a case. J. Clean.

Prod. 6:277-288.

Andersson, K., Ohlsson, T., (1999) Life cycle

assessment of bread produced on different

scales. Int. J. LCA. 4(1):25-40.

Bardach, J. (1980) Aquaculture. Pages 431-

437 in D. Pimentel editor. Handbook ofenergy utilisation in agriculture. CRC Press

Boca Raton, Florida.

Berg, H., P. Michelsen, M. Troell, and N.

Kautsky (1996) Managing aquaculture for

sustainability in tropical Lake Kariba,

Zimbabwe. Ecological Economics 18:141-

159.

Bjorn, P., B. Finstad, and R. Kristoffersen

(2001) Salmon lice infection of wild sea

trout and Artic char in marine and fresh-

waters: The effects of salmon farms.

Aquaculture Research 32:947-962.

Bjorn, P. and B. Finstad (2002) Salmon lice,

Lepeophtheirus salmonis (Kroyer) infesta-

tion in sympatric populations of Artic

char, Salvelinus alpinus (L.), and sea trout,

Salmo trutta (L.), in areas near and distant

from salmon farms. ICES Journal ofMarine Science 59:131-139.

Black, K., S. Fleming, T. Nickell, and P.

Pereira (1997) The effects of ivermectin,

used to control sea lice on cage farmed

salmonids, on infaunal polychaetes. ICES

Journal of Marine Science 54:276-279.

Botsford, L.W., J.C. Castilla and C.H.

Peterson (1997) The management of fish-

eries and marine ecosystems. Science.277:509-515.

Brown, M. T., and R. A. Herendeen. (1996).

Embodied energy analysis and EMERGY

analysis: A comparative view. EcologicalEconomics 19:219-235.

Bruntland, G. (ed). 1987. Our CommonFuture. World commission on environ-

ment and development. Oxford University

Press, Oxford (UK).

Canonico, G., A. Arthington, J. Mccrary,

and M. Thieme (2005) The effects of

introduced tilapias on native biodiversity.

Aquatic Conservation – Marine andFreshwater Ecosystems 15(5):463-483.

Collier, L. and E. Pinn (1998) An assessment

of the acute impact of the sea lice treat-

ment ivermectin on a benthic community.

Journal of Experimental Marine Biology andEcology 230:131-147.

Christensen, P., and Ritter, E. (2000) Lifecycle screening of pickled herring in jars.Masters Thesis, Ålborg University, Ålborg

(Denmark).

Christensen, V., S. Guenette, J.J. Heymans,

C.J. Walters, R. Watson, D. Zeller, D.

Pauly (2003) Hundred-year decline of

North Atlantic predatory fishes. Fish andFisheries 4(1):1-24.

Consoli F, D. Allen, I. Boustead, J. Fava, W.

Franklin, A. Jensen, N. de Oude, R.

Parrish, R. Perriman, D. Postlethwaite, B.

Quay, J. Sequin, and B. Vignon (1993)

Guidelines for Life-cycle assessment: A "Codeof Practice". Society for Environmental

Toxicology and Chemistry, Brussels and

Pensacola.


Costanza, R. and H. Daly (1992) Natural

capital and sustainable development.

Conservation Biology 6(1):37-46.

Daly, H. (1991) Steady-state economics. Island

Press, Washington, D.C.

Daly, H. (1994) Operationalizing

Sustainable Development by Investing in

Natural Capital. Pages 22-37 in A. M.

Jansson, M. Hammer, C. Folke and R.

Costanza, editors. Investing in NaturalCapital: The Ecological Economics Approachto Sustainability. Island Press, Washington

D.C.

Davies, I., P. Gilibrand, J. McHenery, and

G. Rae. (1998) Environmental risk of

ivermectin to sediment dwelling organ-

isms. Aquaculture 163:29-46.

De Silva, S., T. Nguyen. and N. Abery

(2006) An evalution of the role and

impacts of alien finfish in Asian inland

aquaculture. Aquaculture Research 37(1):

1-17.

Einum, S. and I. Fleming (1997) Genetic

divergence and interactions in the wild

among native, farmed and hybrid Atlantic

salmon. Journal of Fish Biology 50:634-

651.

Ekins, P., S. Simon, L. Deutch, C. Folke,

and R. de Groot (2003) A framework for

the application of concepts of critical nat-

ural capital and sustainability. EcologicalEconomics 44:165-185.

El-Hage Scialabba, N. and C. Hattam

(2002) Organic agriculture, environmentand food security. Environment and

Natural Resources Service, Sustainable

Development Department, Food and

Agriculture Organization of the United

Nations, Rome.

Ellingsen, H. and S. Aanondsen (2006)

Environmental impacts of wild caught cod

and farmed salmon – a comparison with

chicken. Int. J. of LCA 1(1):60-65.

Ernst, W., P. Jackman, K. Doe, F. Page,

G. Julien, K. Mackay, and T. Sutherland

(2001) Dispersion and toxicity to non-tar-

get aquatic organisms of pesticides used to

treat sea lice on salmon in net pen enclo-

sures. Marine Pollution Bulletin 42(6):433-

444.

Eyjólfsdóttir, H.R., Jónsdóttir, H.,

Yngvadóttir, E., and Skuladóttir, B. (2003)

Environmental Effects of Fish on theConsumers Dish: Life Cycle Assessment ofIcelandic Frozen Cod Products. IFL report

06-03, IceTec report 0305/HDT05.

Finlay, R., L. Watling, and L. Mayer (1995)

Environmental impact of salmon net-pen

culture on marine benthic communities in

Maine – a case-study. Estuaries18(1A):145-179.

Fleming, I., K. Hindar, I. Mjolnerod, B.

Jonsson, T. Balstad, and A. Lamberg

(2000) Lifetime success and interactions of

farm salmon invading a native population.

Proceedings of the Royal Society of London B267:1517-1523.

Folke, C. (1988) Energy economy of salmon

aquaculture in the Baltic Sea.

Environmental Management 12(4):525-

537.

Folke, C. and N. Kautsky (1989) The role of

ecosystems for a sustainable development

of aquaculture. Ambio 18:234-243.

Folke, C., N. Kautsky, and M. Troell (1992)

The cost of eutrophication from salmon

farming: Implications for policy. Journal ofEnvironmental Management 40:173-182.

Folke, C., N. Kautsky, H. Berg, A. Jansson,

and M. Troell (1998) The ecological foot-

print concept for sustainable seafood pro-

duction: A review. Ecological Applications8(1):S63-S71.

Food and Agriculture Organization (2001)

Tipping the scales towards a better diet in

Peru available at: http://www.fao.org/

news/2001/011203-e.htm (accessed April

23, 2006).



The state of world fisheries and aquaculture2004. FAO Fisheries Department, Food

and Agricultural Organization of the

United Nations, Rome.


Fishstat Plus statistical database. Available

for download at: http://www.fao.org/fi/

statist/FISOFT/FISHPLUS.asp

Glass, C. (2000) Conservation of fish stocks

through bycatch reduction: A review.

Northeast Naturalist 7(4):395-410.

Hale, M. and P. Bauersfeld et al. (1991) New

products and markets for menhaden,

Brevoortia spp. Marine Fisheries Review53(4):42-48.

Hansson, C. and M. Wackernagel (1999)

Rediscovering place and accounting space:

How to re-embed the human economy.

Ecological Economics 29:203-213.

Hardy, R. (2001) Urban legends and fish

nutrition 2. Aquaculture Magazine 27:57-

60.

Hartemink, A. (2005) Plantation agriculture

in the tropics – environmental issues.

Outlook on Agriculture 34(1):11-21.

Haas, G., Wetterich, F., Köpke, U. (2001)

Comparing intensive, extensified and

organic grassland farming in southern

Germany by process life cycle assessment.

Agric. Eco. Env. 83, 43-53.

Hastein, T. (1995) Disease Problems, Use of

Drugs, Resistance Problems and Preventive

Measures in Fish Farming World Wide.

Pages 183-194 in H. Reinertsen and H.

Haaland, editors. Sustainable FishFarming: Proceedings of the FirstInternational Symposium on SustainableFish Farming, Oslo, Norway, 28-31

August 1994. A.A.Balkema, Rotterdam

(Holland).

Haya, K., L. Burridge, and B. Chang (2001)

Environmental impact of chemical wastes

produced by the salmon aquaculture

industry. ICES Journal of Marine Science58:492-496.

Heffernan, W. and G. Green (1986) Farm

size and soil loss – prospects for a sustain-

able agriculture. Rural Sociology 51(1):31-

42.

Hereendeen, R. (1999) Should sustainability

analyses include biophysical assessments?

Ecological Economics 29:17-18.

Heuch, P. and T. Mo (2001) A model of

salmon louse production in Norway:

Effects of increasing salmon production

and public management measures. Diseasesof aquatic organisms 45:145-152.

Holmer, M., P. Lassus, J. Stewart, and D.

Wildish (2001) ICES symposium on envi-

ronmental effects of Mariculture:

Introduction. ICES Journal of MarineScience 58:363-368.

Hospido, A, Moreira, MT, Feijoo, G. (2003)

Simplified Life cycle assessment of

Galician milk production. Int. Dairy J.13:783–796.

Hospido, A., and P. Tyedmers, (2005) Life

cycle environmental impacts of Spanish

tuna fisheries. Fisheries Research 76(2):174-

186.

International Union for the Conservation of

Nature (1991) World conservation strate-

gy. Living resource conservation for sus-

tainable development. International Union

for the Conservation of Nature, United

Nations Environment Program and World

Wildlife Fund, Gland, Switzerland.

Janowicz, M. and J. Ross (2001) Monitoring

for benthic impacts in the southwest New

Brunswick salmon aquaculture industry.

ICES Journal of Marine Science 58:453-

459.

Kautsky, N., H. Berg, et al. (1997).

Ecological footprint for assessment of

resource use and development limitations

in shrimp and tilapia aquaculture."

Aquaculture Research 28: 753-766.

Kautsky, N., P. Ronnback, M. Tedengren,

and M. Troell (2000) Ecosystem perspec-

tives on management of disease in shrimp

pond farming. Aquaculture 191(1-3):145-

161.


Kerr, J. and I. Desguise (2004) Habitat loss

and the limits to endangered species recov-

ery. Ecology Letters 7(12):1163-1169.

Krkosek, M., M. Lewis, J. Volpe, and A.

Morton (2006) Fish farms and sea lice

infestations of wild juvenile salmon in the

Broughton Archipelago – A rebuttal to

Brooks (2005). Reviews in Fisheries Science14(1-1):1-11.

Kruess, A. and T. Tscharntke (1994) Habitat

fragmentation, species loss, and biological-

control. Science 264(5156):1581-1584.

Larsson, J., C. Folke, and N. Kautsky (1994)

Ecological limitations and appropriation

of ecosystem support by shrimp farming

in Columbia. Environmental Management18(5):663-676.

LCA of Food (2006) http://www.lcafood.dk/

(accessed February, 2006).

Lal, R. (2000) Soil management in the

developing countries. Soil Science165(1):57-72.

Liess, M., R. Schulz, and M. Leiss (1999)

Determiniation of insecticide contamina-

tion in agricultural headwater streams.

Water Research 33(1):239-247.

Mack, R., D. Simberloff, W. Lonsdale, H.

Evans, M. Clout, and F. Bazzaz (2000)

Biotic invasions: causes, epidemiology,

global consequences, and control.

Ecological Applications 10(3):689-710.

Martinez-Alier. J. (1987) EcologicalEconomics. Basil Blackwell, Oxford (UK).

Mattsson, B., and U. Sonesson (2003)

Environmentally-friendly food processing.Woodhead Publishing Limited,

Cambridge (UK).

Merceron, M., M. Kempf, D. Bentley, J-D.

Gaffet, J. Le Grand, and L. Lamort-Datin

(2002) Environmental impact of a

salmonid farm on a well flushed marine

site: I. Current and water quality. Journalof Applied Ichthyology 18:40-50.

Morton A., R. Routledge, and R. Williams

(2005) Temporal patterns of sea louse

infestation on wild Pacific salmon in rela-

tion to the fallowing of Atlantic salmon

farms. North American Journal of FisheriesManagement 25(3):811-821.

Mungkung, R. (2005) Shrimp aquaculture inThailand: Application of life cycle assessmentto support sustainable development. PhD.thesis. Center for Environmental Strategy,

School of Engineering, University of

Surrey, England.

Myers, R.A., and B. Worm (2003) Rapid

worldwide depletion of predatory fish

communities. Nature 423:280-283.

Naylor, R. L., R. J. Goldburg, et al. (2000)

Effect of aquaculture on world fish sup-

plies. Nature 405:1017-1024.

Naylor R. and M. Burke (2005) Aquaculture

and ocean resources: Raising tigers of the

sea. Annual Review of Environment andResources. 30:185-218.

New, M. and U. Wijkstrom (2002) Use of

fishmeal and fish oil in aquafeeds: Further

thoughts on the fishmeal trap. FAO

Fisheries Circular 975, United Nations

Food and Agricultural Organization,

Rome, Italy.

Odum, H. T. (1988) Self-organization, trans-

formity, and information. Science242:1132-1139.

Odum, H. T., and J. E. Arding (1991)

EMERGY Analysis of Shrimp Mariculture

in Ecuador. A working paper prepared for

the Coastal Resources Center, University

of Rhode Island.

Papatryphon, E., J. Petit, S. Kaushik, and H.

Van der Werf (2004) Environmental

impact assessment of salmonid feeds using

Life Cycle Assessment (LCA). Ambio33(6):316-323.

Papatryphon, E., J. Petit, H. Van der Werf,

and S. Kaushik (2003) Life Cycle

Assessment of trout farming in France: A

farm level approach. Life Cycle Assessmentin the agrifood sector. Proceedings from the4th International Conference Dias Report61: 71-77.


Pauly, D., V. Christensen, et al. (1998).

Fishing down marine food webs. Science279:860-863.

Pauly, D., V. Christensen, S. Guénette, T.J.

Pitcher, R.U. Sumaila, C.J. Walters, R.

Watson, and D. Zeller (2002) Towards

sustainability in world fisheries. Nature418:689-695.

Pearce, D.G. (1998) Sustainable

Development. Pages 69-90 in D. Pearce,

editor. Economics and the Environment:Essays on Ecological Economics andSustainable Development. Edward Elgar,

Cheltenham.

Pearce, D., G. Atkinson, and W. Dubourg

(1994) The economics of sustainable

development. Annual Review of Energyand the Environment 19:457-474.

Peet, J. (1992) Energy and the EcologicalEconomics of Sustainability. Island Press,

Washington, D.C.

Pike, I (2000) The real effect of aquaculture

on world fish supplies. Feed International.Sept. 2000. 34-40.

Pike, I. (2005) Eco-efficiency in aquaculture:

Global catch of wild fish used in aquacul-

ture. International Aquafeed 8(1):38-40.

Pimentel, D. (2004) Livestock Production

and Energy Use. Pages 671-676 in C.

Cleveland, editor. Encyclopaedia of Energy.Elsevier, San Diego, CA, vol. 3.

Pohle, G., B. Frost, and R. Findlay (2001)

Assessment of regional benthic impact of

salmon mariculture within Letang Inlet,

Bay of Fundy. ICES Journal of MarineScience 58:417-426.

Proops, J. and F. Faber (1985)

Interdisciplinary research between econo-

mists and physical scientists: Retrospect

and prospect. Kyklos 38:599-616.

Rees, W., E, (1996) Revisiting carrying

capacity: Area-based indicators of sustain-

ability. Population and Environment: AJournal of Interdisciplinary Studies17(3):195-215.

Rees, W. (1998) How should a parasite value

its host? Ecological Economics 25(1):49-52.

Rees, W. (1999) Consuming the earth: The

biophysics of sustainability. EcologicalEconomics 29:23-27.

Rees, W., E, and M. Wackernagel. (1994)

Ecological Footprints and Appropriated

Carrying Capacity: Measuring the Natural

Capital Requirements of the Human

Economy. Pages 362-390 in A.M. Jansson,

M. Hammer, C. Folke and R. Costanza,

editors. Investing in Natural Capital: TheEcological Economics Approach toSustainability. Island Press, Washington,

DC.

Reid, D. (1995), Sustainable Development:An Introductory Guide, Earthscan

Publications, London. (UK).

Roth, E., H. Rosenthal, and P. Burbridge

(2000) A discussion of the use of the sus-

tainability index: ‘ecological footprint’ for

aquaculture production. Aquatic LivingResources 13:461-469.

Salmon of the Americas (2006)

http://www.salmonoftheamericas.com/env

_food.html (accessed April, 2006).

Seppälä, J., F. Silvenius, J. Grönroos, T.

Mäkinen, K. Silvo, and E. Storhammar

(2001) Rainbow trout production and theenvironment. The Finnish Environment (in

Finnish). Suomen ymparisto 529.

Technical Report, Helsinki.

Singh, S. and C. J. S. Pannu (1998) Energy

requirements in fish production in the

state of Punjab. Energy Convers. Mgmt.39:911-914.

Smith, J. and D. Ahrenholz (2000) The

ancillary harvest of Atlantic menhaden,

Brevoortia tryannus, roe on the North

Carolina coast. Marine Fisheries Review

62(1):1-7.

Stewart, J. A. (1994) Assessing sustainabilityof aquaculture development. Ph.D. Thesis,

University of Stirling, Stirling, United

Kingdom.

Tacon, A. (2004) Use of fishmeal and fish oil

in aquaculture: A global perspective.

Aquatic Resources, Culture andDevelopment. 1(1): 3-14.


Tacon, A. (2005) State of information on

salmon aquaculture and the environment.

http://www.worldwildlife.org/cci/dia-

logues/salmon.cfm (accessed October 15,

2005).

Thrane, M. (2004) Energy consumption in

the Danish fishery. Journal of IndustrialEcology. 8(1-2):223-239.

Thrane, M. (2006) LCA of Danish fish

products: New methods and insights. Int.J. LCA 11(1):67-74.

Tidwell, J. H. and G. L. Allan (2001) Fish as

food: aquaculture’s contribution. EMBOReports 2(11): 958-963.

Troell, M, P. Tyedmers, N. Kautsky, and P.

Rönnbäck (2004) Aquaculture and Energy

Use. Pages 97-108 in C. Cleveland editor.

Encyclopedia of Energy. Elsevier, San Diego.

vol. 1.

Tudela S. (2002) Grab, cage, fatten, sell.

Samudra. 32:9-17.

Tyedmers, P. (2000) Salmon and sustainabili-ty: The biophysical cost of producing salmonthrough the commercial salmon fishery andthe intensive salmon culture industry. Ph.D.

Thesis, University of British Columbia,

Vancouver, Canada.

Tyedmers, P. (2004) Fisheries and Energy

Use. Pages 683-693 in C. Cleveland edi-

tor. Encyclopedia of Energy. Elsevier, San

Diego. vol. 2.

Tyedmers, P., R. Watson, and D. Pauly

(2005) Fueling global fishing fleets.

Ambio. 34(8): 635-638.

Victor, P. A. (1991) Indicators of sustainable

development: Some lessons from capital

theory. Ecological Economics 4:191-213.

Victor, P. A., J. E. Hanna, and A. Kubursi

(1995) How strong is weak sustainability?

Economie Appliquee XLVIII:75-94.

Wackernagel, M. (1999) Why sustainability

analyses must include biophysical assess-

ments. Ecological Economics 29:13-15.

Wackernagel, M., and W. E. Rees (1996)

Our Ecological Footprint. New Society

Publishers, Gabriola Island, BC.

Waldrop, J.E. and J.G. Dillard (1985)

Channel Catfish Culture. Elsevier, New

York, NY.

Watanabe, T. (2002) Strategies for further

development of aquatic feeds. FisheriesScience 68: 242-252.

Weber, M. (2003) What Price Farmed Fish: AReview of the Environmental and SocialCosts of Farming Carnivorous Fish. SeaWeb

Aquaculture Clearinghouse, Washington,

DC.

Youngson, A. and E. Verspoor (1998)

Interactions between wild and introduced

Atlantic salmon (Salmo salar). AmericanJournal of Fisheries and Aquatic Sciences55:153-160.

Zebarth, G., B. Hii, H. Liebscher, K.

Chipperfield, J. Paul, G. Grove, and S.

Szeto (1998) Agricultural land use prac-

tices and nitrate contamination in the

Abbotsford Aquifer, British Columbia,

Canada. Agriculture Ecosystems andEnvironment 69(2):99-112.

Ziegler, F., P. Nilsson, B. Mattsson, and Y.

Walther (2003) Life Cycle Assessment of

frozen cod fillets including fishery-specific

environmental impacts. Int. J. LCA8(1):39-47.


Marine Aquaculture Task ForceP.O. Box 5687Takoma, Park, MD 20913

www.whoi.edu/sites/marineaquataskforce

biophysical sustainability and approaches to marine ... · aquaculture might develop in a more...

Documents