Peter H. Tyedmers, Reg Watson and Daniel Pauly
Fueling Global Fishing Fleets
Over the course of the 20th century, fossil fuels became
the dominant energy input to most of the world’s fisheries.
Although various analyses have quantified fuel inputs to
individual fisheries, to date, no attempt has been made to
quantify the global scale and to map the distribution of
fuel consumed by fisheries. By integrating data repre-
senting more than 250 fisheries from around the world
with spatially resolved catch statistics for 2000, we
calculate that globally, fisheries burned almost 50 billion
L of fuel in the process of landing just over 80 million t of
marine fish and invertebrates for an average rate of 620 L
t?1. Consequently, fisheries account for about 1.2% of
global oil consumption, an amount equivalent to that
burned by the Netherlands, the 18th-ranked oil consum-
ing country globally, and directly emit more than 130
million t of CO2into the atmosphere. From an efficiency
perspective, the energy content of the fuel burned by
global fisheries is 12.5 times greater than the edible-
protein energy content of the resulting catch.
Marine capture fisheries are the most diverse of the major
global food-producing sectors, both in terms of the range of
species harvested (1) and harvesting technologies used (2). One
characteristic, however, common to nearly all contemporary
fisheries, is their dependence on fossil fuels. In a process that
began with the launch of the first coal-fired steam trawler in the
late 1880s and accelerated through the latter half of the 20th
century, fossil fuels have become the dominant energy input to
the world’s fisheries. Consequently, vessels powered by diesel,
and to a lesser degree gasoline and kerosene, now account for
the vast majority of global fisheries landings. Indeed, the extent
of this dependence has prompted one commentator to observe
that ‘‘Fishing continues to be the most energy-intensive food
production method in the world today...’’ (3).
Spurred initially by the oil price shocks of the 1970s, analyses
have been undertaken of a wide range of fisheries, either to
evaluate their energetic performance (4–12) or to assess their
economic vulnerability to potential increases in oil prices (13–
16). This research indicates that direct fuel inputs typically
account for between 75 and 90% of total energy inputs to fishing
activities (7, 17, 18). The scale of direct fuel inputs, however, can
range widely. Purse seine fisheries for small pelagic species, such
as herring and menhaden, that are destined for reduction to fish
meal and oil, typically use under 50 L of fuel per tonne of fish
landed (11, 17). In contrast, fisheries targeting high value
species like shrimp, tuna, or swordfish (Xiphias gladius)
frequently consume in excess of 2000 L per tonne of landings
(8, 11, 12). Where time series data are available, the energy
performance of fishing fleets often has declined over time (7, 9,
11, 12), owing to the need for vessels to search longer and to fish
deeper in offshore waters as coastal stocks decline (19–21).
Despite the importance of fuel inputs to contemporary fishing
activities, to date there has been no comprehensive analysis of
the scale of fuel inputs to global fisheries, nor is there a clear
understanding of the spatial distribution and intensity of where
those fuel inputs are being expended.
MATERIALS AND METHODS
For this analysis, we assembled detailed fuel consumption,
catch, and vessel/gear characteristic data from a wide range of
published and unpublished sources. From these we calculated,
in step-wise fashion, species-specific, globally- and where
possible, regionally-representative average fuel use values.
These values then were integrated with species-specific, spatially
resolved catch data for 2000 to provide both estimates of global
total and average fuel use intensity and the basis upon which
fuel consumption could be mapped.
More specifically, to proceed from individual fuel use case
studies to estimates for each reported commercial taxa from
each of 18 statistical areas (Fig. 1) used by the Food and
Agricultural Organization of the United Nations (FAO)
required a process of progressive refinement, where average
values were replaced at each step by more specific (with regard
to taxa and location) estimates where possible. To provide all
combinations of fished commercial taxa and statistical report-
ing areas with an initial estimate, we started with values based
on the average of all case studies within the same broad
taxonomic group (for example, ‘‘shrimp’’ or ‘‘tuna’’), ignoring
geographic area. We then repeated the process, but separated
the case studies by area and where possible, replaced the more
general estimate from the previous step. In this way, if fisheries
targeting similar taxa did not have an estimate specific to the
area where the landings were reported, then they at least had
one for all areas combined. This process was repeated with
progressively more specific taxonomic limits for the target of the
fishery. Recognizing that in many cases, fisheries land more
than one species, a provision also was made to weight averages
based on the relative contribution that a given species made to
the total landings recorded in a case study. In other words, case
studies in which a species was targeted were weighted more
heavily than those studies in which the species was taken
incidentally. Documentation as to the origin of the estimate was
maintained at each step.
The majority of case studies used provided fishery-specific
fuel use data for a single year, though some also provided
several annual estimates (Fig. 1). Though our final global fuel
estimate represents the year 2000, this is because the landing
statistics (tonnes reported by taxon and area) used to raise
average fuel intensity estimates to total annual fuel consump-
tion were from 2000. When case studies included time series
data, only values closest to 2000 were used.
When the process of generating representative fuel use values
had been completed, some landings were associated with
specific case studies, whereas some were best represented by
weighted averages calculated for the most specific taxonomic
and spatial aggregation applicable.
Total fuel consumption by the world’s fishing fleets in 2000
was estimated by summing the products of catches, by species,
in each of 18 FAO areas (Fig. 1), and corresponding species-
specific fuel use estimates. In order to contextualize the scale of
fuel inputs to global fisheries, comparisons were then made with
national and global levels of oil consumption in 2000 (22).
Resulting CO2emissions from fishing vessels were quantified
using real-world emission data from vessels (23).
Although for many consumers in industrialized countries,
the nutritional importance of seafood has shifted recently to its
value as a source of essential fatty acids and micronutrients (24),
Ambio Vol. 34, No. 8, December 2005
? Royal Swedish Academy of Sciences 2005
for most of humanity, seafood remains an important source of
animal protein (25). Consequently, the most meaningful basis
upon which to compare fisheries with other food-producing
sectors is in terms of their edible-protein yield (9–12). We
therefore quantified the edible-protein energy efficiency of
global fisheries by dividing the maximum edible-protein energy
that could be derived from global catches in 2000 by the energy
content of the fuel burned (26). The maximum edible-protein
yield from catches in 2000 was estimated by first multiplying
species-specific values for the maximum edible fraction of the
animals landed, assumed to be equivalent to the muscle content
of the animal (4, 27–29), by species-specific protein content of
muscle values (1, 28–30). The resulting species-specific edible-
protein fractions were multiplied then by corresponding catches
in 2000, were summed, and then were converted to an energetic
As the 2000 global catch statistics used to calculate total fuel
consumption were available by ½ degree latitude and longitude
cells, and are therefore mappable using a rule-based algorithm
(19, 31), we also generated a global map of fuel consumption by
fisheries, the first of its kind.
In total, data representing more than 250 distinct fisheries or
fleet subsets, based in 20 countries, were assembled (Fig. 1) (32).
Although it is impossible to know exactly where the vessels
represented in this data set actually fished, most either targeted
species caught over large geographic ranges (e.g. fisheries for
various tuna, billfish, or squid) or produced very large catches
of major species. Hence, we believe our data set to be broadly
representative of world fisheries. Similarly, although a number
of small-scale fisheries employing small outboard engines are
represented, most are larger industrialized fisheries in which
average vessel engine outputs are in the range of many tens to
thousands of kilowatts — a pattern generally reflective of global
In 2000, global fisheries reported landings of 80.4 million t of
fish and invertebrates from marine waters (33). In the process of
catching these, the world’s fleets burned approximately 50
billion L of fuel, yielding a global average fuel use intensity of
620 L per live weight tonnes of fish and shellfish landed.
Applying an average diesel fuel density of 0.85 (26), global
fisheries landed approximately 1.9 t of fish and invertebrates for
each tonne of fuel consumed directly in their capture.
As a consequence of burning almost 42.4 million t of fuel in
2000, representing approximately 1.2% of total global oil
consumption, fishing boats released approximately 134 million
t of CO2into the atmosphere at an average rate of 1.7 t of CO2
per tonne of live-weight landed product.
Note that although our estimates of average fuel consump-
tion and resulting emissions per landed tonne should be reliable,
the above estimates of absolute fuel consumption and CO2
emissions by the world’s fishing fleets are likely serious
underestimates, given that we did not account for freshwater
fisheries nor the tens of millions of tonnes of fish caught by
illegal, unreported, or unregulated (IUU) marine fisheries (20).
Moreover, although direct fuel inputs represent the lion’s share
of industrial energy inputs to fisheries, our analysis does not
account for the indirect or ‘‘embodied’’ energy inputs associated
with the provision of fishing vessels, gear, labor, or the fuel
In terms of their energy efficiency, fisheries globally
dissipated 12.5 times the amount of fuel energy as they provided
in the form of edible-protein energy. Whereas an 8% edible-
protein energy return on fuel energy investment ratio is
disturbingly low, it is higher than many other animal protein
production systems (Table 1), including many of the intensive
Figure 1. Distribution of case studies from which fisheries-specific estimates of fuel use intensity were derived, covering most of the major
fisheries of the world. The straight lines cutting ocean basins demarcate the 18 statistical areas used by the FAO to report on global marine
fisheries catches, and used here for geographic stratification.
Table 1. Edible-protein energy return on investment (EROI) values
for various animal protein production systems.
Production system (locale)
Carp — extensive pond culture (various)
Tilapia — extensive pond culture (Indonesia)
Mussel — longline culture (Scandinavia)
Carp — unspecified culture system (Israel)
Tilapia — unspecific culture system (Israel)
Tilapia — pond culture (Zimbabwe)
Beef — pasture-based (US)
Catfish — intensive pond culture (US)
Beef — feedlot (US)
Tilapia — intensive cage culture (Zimbabwe)
Atlantic salmon — intensive cage culture (Canada)
Shrimp — semi-intensive culture (Ecuador)
Chinook salmon – intensive cage culture (Canada)
Atlantic salmon – intensive cage culture (Sweden)
Sea bass – intensive culture (Thailand)
Shrimp – intensive culture (Thailand)
Sources: aquaculture data (34), livestock data (39).
Note: Range of values for some culture systems reflects differences in production
Ambio Vol. 34, No. 8, December 2005
? Royal Swedish Academy of Sciences 2005
aquaculture systems (34) that compete directly with and have
been proposed as alternatives to capture fisheries (35).
Reflecting, in part, that the world’s oceans are not uniformly
productive (31, 36), the spatial distribution of fuel use by the
world’s fishing fleets is highly variable (Fig. 2). Whereas large
areas of the world’s oceans experience relatively low levels of
fishing effort, as measured in terms of spatially distributed fuel
use, more productive (largely coastal) fishing areas experience
annual effort levels exceeding 1000 Lkm?2. Fishing grounds in
which heavy fuel use is particularly widespread in 2000 include
the western Pacific and adjacent seas, the Bering Sea, and
coastal waters of the northeastern and southwestern Atlantic
and northern Indian Ocean (Fig. 2).
Our results document for the first time the extent to which
global fisheries have become dependent upon large inputs of
nonrenewable, petroleum-derived fuels (37, 38) and illustrate
the scale and intensity of arguably the largest direct industrial
use of the world’s oceans. At a conservative 1.2% of global oil
consumption, this sector turns out to be a far from trivial player
when it comes to the consumption of this important resource,
a detail that should not be overlooked despite the fact that most
of its activities occur far from shore and far from the thoughts
of most of the public. Indeed, the scale of the industry’s fuel
consumption places it on a par with the total amount of oil
consumed annually by the Netherlands, the 18th-ranked oil
consuming country globally.
Interestingly, although the fishing sector consumes a sub-
stantial amount of fuel, its use of energy is far more efficient
than many other contemporary food production systems,
a finding that flies in the face of some widely held perceptions
of capture fisheries in general (3). This seeming incongruity
between perception and reality may, in part, result from the
relatively high proportion of total energy inputs, and resulting
energy-related costs that accrue at the level of the fishing
enterprise itself. In contrast, in the case of many other animal
protein production systems, the majority of energy inputs tend
to occur farther back in the production chain (34, 39).
The spatial distribution of fuel use intensity illustrated in
Figure 2 provides, to our knowledge, the first comprehensive
picture of global fishing effort. In addition, because the scale of
fuel inputs to fisheries are broadly proportional to the value of
the resulting catch (8), the spatial distribution of fuel use
intensity provides a partial indication of the relative extractive-
use value of the world’s oceans. Reflecting the relative
importance of vessel-sourced operational inputs of oil to the
world’s oceans (40), our map of fuel inputs to fishing vessels
also provides a basis upon which future oil pollution monitoring
programs can be tailored.
Given the absolute and relative scales of fuel consumed by
fishing fleets globally, it is essential that it be considered
explicitly in future policy planning (21), both with regard to the
fuel subsidies from which fishing fleets usually benefit and the
needless climate impact of fossil fuels burned by overcapitalized
References and Notes
1. Froese, R. and Pauly, D. (eds.). 2000. FishBase 2000: Concepts, Design and Data
Sources. ICLARM, Los Ban ˜ os, Philippines, 346 pp. (Distributed with four CD-ROMs;
updates available at http://www.fishbase.org)
von Brandt, A. 1984. Fish Catching Methods of the World. Avon Litho, Stratford-upon-
Avon, Warwickshire, 240 pp.
Wilson, J.D.K. 1999. Fuel and Financial Savings for Operators of Small Fishing Vessels.
FAO Fisheries Technical Paper 383, FAO, Rome, pp. 46 at 1.
Wiviott, D.J. and Mathews, S.B. 1975. Energy efficiency comparison between the
Washington and Japanese otter trawl fisheries of the northeast Pacific. Mar. Fish. Rev.
37, (4), 21–24.
Rawitscher, M. and Mayer, J. 1977. Nutritional outputs and energy inputs in seafood.
Science 198, 261–264.
Lorentzen, G. 1978. Energy account of the Norwegian fishing sector. Meldingen SSF
M2, 5–9. (In Norwegian).
Watanabe, H. and Uchida, J. 1984. An estimation of direct and indirect energy input in
catching fish for fish paste products. Bull. Jpn. Soc. Sci. Fish./Nippon Suisan Gakkaishi
Watanabe, H. and Okubo, M. 1989. Energy Input in Marine Fisheries of Japan. Bull.
Jpn. Soc. Sci. Fish./Nippon Suisan Gakkaishi 53, 1525–1531.
Mitchell, C. and Cleveland, C.J. 1993. Resource scarcity, energy use and environmental
impact: a case study of the New Bedford, Massachusetts, USA Fisheries. Environ.
Manage. 17, 305–317.
Pimentel, D., Shanks, R.E. and Rylander, J.C. 1996. Bioethics of fish production: energy
and the environment. J. Agric. Environ. Ethics 9, 144–164.
Tyedmers, P. 2001. Energy consumed by North Atlantic fisheries. In: Fisheries’ Impacts
on North Atlantic Ecosystems: Catch, Effort and National/Regional Datasets. Zeller, D.,
Watson, R. and Pauly, D. (eds.). Fisheries Centre, University of British Columbia,
Vancouver, pp. 12–34. http://www.seaaroundus.org/report/method/tyedmers10.pdf
Tyedmers, P. 2004. Fisheries and energy use. In: Encyclopaedia of Energy. Cleveland, C.
(ed.). Elsevier, San Diego, vol. 2. pp. 683–693.
Scott, W.G. 1981. Energy and Fish Harvesting: Conference Proceedings. Fisheries and
Oceans Canada, Ottawa, 166 pp.
Samples, K.C. 1983. An economic appraisal of sail-assisted commercial fishing vessels in
Hawaiian waters. Mar. Fish. Rev. 45, (7–9), 50–55.
George, V.C., Vijayan, V., Varghese, M.D., Radhalakshmi, K., Thomas, S.N. and
Joseph, J. (eds). 1993. Proceedings of the National Workshop on Low Energy Fishing, 8–9
August, 1991. Society of Fisheries Technologists (India), Cochin, 286 pp.
Senthilathiban, R., Venkataramanujam, K., Selvaraj, P. and Sanjeeviraj, G. 1997. Costs
and earnings of trawlers operating at Tuticorin, Tamilnadu. Fish. Technol. 34, (2), 31–34.
Rawitscher, M. A. 1978. Energy Cost of Nutrients in the American Diet. PhD Thesis,
University of Connecticut, Storrs, Connecticutt.
Figure 2. Distribution and intensity of fuel consumption by marine fisheries in 2000. Total fuel inputs amount to 50 billion L, with most of this
being expended in nearshore fishing grounds of the Northern Hemisphere.
Ambio Vol. 34, No. 8, December 2005
? Royal Swedish Academy of Sciences 2005
18.Tyedmers, P. 2000. Salmon and Sustainability: The Biophysical Cost of Producing
Salmon Through the Commercial Salmon Fishery and the Intensive Salmon Culture
Industry. PhD Thesis, University of British Columbia, Vancouver, Canada.
Watson, R. and Pauly, D. 2001. Systematic distortions in world fisheries catch trends.
Nature 414, 534–536.
Pauly, D., Christensen, V., Gue ´ nette, S., Pitcher, T.J., Sumaila, U.R., Walters, C.J.,
Watson, R. and Zeller, D. 2002. Toward sustainability in world fisheries. Nature 418,
Pauly, D., Alder, J., Bennett, E., Christensen, V., Tyedmers, P. and Watson, R. 2003.
The future for fisheries. Science 302, 1359–1361.
British Petroleum (BP). 2003. BP Statistical Review of World Energy 2003. BP, London,
Engineering Services Group. 1995. Marine Exhaust Emissions Research Programme.
Lloyd’s Register of Shipping, London, 63 pp.
Valdimarsson, G. and James, D. 2001. World fisheries — utilization of catches. Ocean
Coast. Manage. 44, 619–633.
FAO. 2002. The State of World Fisheries and Aquaculture. FAO, Rome, 150 pp.
Rose, J.W. and Cooper, J.R. (eds). 1977. Technical Data on Fuel (7th ed.). Scottish
Academic Press, Edinburgh. 343 pp.
Crapo, C., Paust, B. and Babbitt, J. 1993. Recoveries and Yields from Pacific Fish and
Shellfish. Alaska Sea Grant College Program, University of Alaska Fairbanks,
Fairbanks, 36 pp.
Bykov, V.P. 1986. Marine Fishes: Chemical Composition and Processing Properties.
A.A. Balkema, Rotterdam, 322 pp.
Torry Research Station. 1989. Yield and Nutritional Value of the Commercially More
Important Fish Species. FAO Fisheries Technical Paper No. 309. FAO, Rome, 187 pp.
Wheaton, F.W. and Lawson, T.B. 1985. Processing Aquatic Food Products. John Wiley,
New York, 518 pp.
Watson, R., Kitchingman, A., Gelchu, A. and Pauly, D. 2004. Mapping global fisheries:
sharpening our focus. Fish Fish. 5, 168–177.
A summary of the fisheries represented inthis datasetand theirsourcesisavailable from P.T.
33.Detailed annual species-specific catches based in large measure on data in FAO 2003
FishStat Plus statistical database. (http://www.fao.org/fi/statist/FISOFT/FISHPLUS.asp)
Troell, M., Tyedmers, P., Kautsky, N. and Ro ¨ nnba ¨ ck, P. 2004. Aquaculture and energy
use. In: Encyclopaedia of Energy. Cleveland, C. (ed.). Elsevier, San Diego, vol. 1, pp. 97–
Naylor, R.L., Goldburg, R.J., Primavera, J.H., Kautsky, N., Beveridge, M.C.M., Clay,
J., Folke, C., Lubchenco, J., et al. 2000. Effect of aquaculture on world fish supplies.
Nature 405, 1017–1024.
Pauly, D. and Christensen, V. 1995. Primary production required to sustain global
fisheries. Nature 374, 255–257.
Deffeyes, K.S. 2001. Hubbert’s Peak: The Impending World Oil Shortage. Princeton
University Press, Princeton, 208 pp.
Heinberg, R. 2003. The Party’s Over: Oil, War and the Fate of Industrial Societies. New
Society, Gabriola Island, BC, Canada, 275 pp.
Pimentel, D. 2004. Livestock production and energy use. In: Encyclopaedia of Energy.
Cleveland, C. (ed.). Elsevier, San Diego, vol. 3, pp. 671–676.
Committee on Oil in the Sea. 2003. Oil in the Sea III: Inputs, Fates, and Effects. National
Academies Press, Washington, DC, 280 pp.
41. D.P, and R.W. are members of the Sea Around Us Project, initiated and funded by the
Pew Charitable Trusts, Philadelphia, PA. D.P. acknowledges support from the Natural
Sciences and Engineering Research Council, Canada. We thank Scott Wallace and
Martin Willison for their feedback on early drafts, the input from two anonymous
referees, and Adrian Kitchingman for his assistance in preparing the figures.
42.First submitted 9 July 2004. Accepted for publication 19 Jan. 2005.
Peter H. Tyedmers is an assistant professor of ecological
economics at Dalhousie University, Canada. His research
interests revolve around the application of tools for improved
natural resource and environmental decision-making for
sustainability. His current research activities focus on the
biophysical performance of food production systems including
fisheries, aquaculture, and livestock production systems. His
address: School for Resource and Environmental Studies,
Dalhousie University, 6100 University Avenue, Suite 5010,
Halifax, NS, Canada, B3H 3J5.
Reg Watson is a senior research associate at the Sea Around
Us Project (SAUP) based at the Fisheries Centre, University of
British Columbia. His expertise includes penaeid biology, trawl
fisheries, computer modeling, stock assessment, limnology,
and ecological modeling. His current research focuses on
examining the global impacts of fishing with novel database
and mapping methods. His address: Fisheries Centre, 2202
Main Mall, University of British Columbia, Vancouver, BC,
Canada V6T 1Z4.
Daniel Pauly is the director of the Fisheries Centre, University
of British Columbia, Vancouver, Canada, and principal in-
vestigator of the Sea Around Us Project, devoted to studying
the impact of fisheries on the world’s marine ecosystems
(www.seaaroundus.org). The concepts, methods, and soft-
ware he helped develop are used throughout the world —
especially the Ecopath modeling approach (www.ecopath.org)
and FishBase, the online encyclopedia of fishes (www.fishba-
se.org). His address: Fisheries Centre, 2202 Main Mall,
University of British Columbia, Vancouver, BC, Canada V6T
Ambio Vol. 34, No. 8, December 2005
? Royal Swedish Academy of Sciences 2005