![Comparison of supply chain emissions from fresh and frozen cod fillets trucked to Paris (kilograms of greenhouse gas emissions per kilogram edible product [kg GHG/kg edible product] at wholesaler).](https://www.researchgate.net/profile/Friederike-Ziegler-2/publication/256048653/figure/fig1/AS:289157399236650@1445951904700/Comparison-of-supply-chain-emissions-from-fresh-and-frozen-cod-fillets-trucked-to-Paris_Q320.jpg)
Content uploaded by Friederike Ziegler
Author content
All content in this area was uploaded by Friederike Ziegler on Aug 17, 2017
Content may be subject to copyright.
RESEARCH AND ANALYSIS
The Carbon Footprint of Norwegian
Seafood Products on the Global Seafood
Market
Friederike Ziegler,Ulf Winther,Erik Skontorp Hognes,Andreas Emanuelsson,Veronica Sund,
and Harald Ellingsen
Keywords:
aquaculture
fisheries
greenhouse gas (GHG) emissions
industrial ecology
life cycle assessment (LCA)
Norway
Supporting information is available
on the JIE Web site
Summary
Greenhouse gas emissions caused by food production are receiving increased attention
worldwide. A problem with many studies is that they only consider one product; method-
ological differences also make it difficult to compare results across studies. Using a consistent
methodology to ensure comparability, we quantified the carbon footprint of more than 20
Norwegian seafood products, including fresh and frozen, processed and unprocessed cod,
haddock, saithe, herring, mackerel, farmed salmon, and farmed blue mussels. The previous
finding that fuel use in fishing and feed production in aquaculture are key inputs was con-
firmed. Additional key aspects identified were refrigerants used on fishing vessels, product
yield, and by-product use. Results also include that product form (fresh or frozen) only
matters when freezing makes slower transportation possible. Processing before export
was favorable due to the greater potential to use by-products and the reduced need for
transportation. The most efficient seafood product was herring shipped frozen in bulk to
Moscow at 0.7 kilograms CO2equivalents per kilogram (kg CO2-eq/kg) edible product. At
the other end we found fresh gutted salmon airfreighted to Tokyo at 14 kg CO2-eq/kg edible
product. This wide range points to major differences between seafood products and room
for considerable improvement within supply chains and in product choices. In fisheries, we
found considerable variability between fishing methods used to land the same species, which
indicates the importance of fisheries management favoring the most resource-efficient ways
of fishing. Both production and consumption patterns matter, and a range of improvements
could benefit the carbon performance of Norwegian seafood products.
Introduction
Global environmental impacts (like greenhouse gas [GHG]
emissions) caused by seafood production systems are receiv-
ing increased attention worldwide by consumers, retailers, eco-
labeling organizations, and the seafood industry itself. Tradi-
tionally, sustainable seafood sourcing has been concerned with
biological sustainability only. This remains critical, with 80%
of the world’s capture fisheries fully exploited, overexploited,
Address correspondence to: Friederike Ziegler, SIK, PO Box 5401, SE-402 29 G¨
oteborg, Sweden. Email: fz@sik.se
c
2012 by Yale University
DOI: 10.1111/j.1530-9290.2012.00485.x
Volume 00, Number 00
or recovering (FAO 2011), as well as problems of fishing prac-
tice that include large amounts of by-catch and discards and
degradation of marine ecosystems (Hall 1999). Recently, how-
ever, stakeholders in the seafood production chain have become
aware that supply chain GHG emissions (i.e., the carbon foot-
print) is an important additional aspect of sustainability that
needs to be addressed.
Life cycle assessment (LCA) methodology for the environ-
mental assessment of products and services through the entire
supply chain has developed considerably since it was first used in
www.wileyonlinelibrary.com/journal/jie Journal of Industrial Ecology 1
RESEARCH AND ANALYSIS
the 1960s. Today it is standardized by the International Organi-
zation for Standardization (ISO) in ISO 14040 and 14044 (ISO
2006a, 2006b). These standards describe the method and basic
requirements for undertaking an LCA. The carbon footprint of
a product is the result of an LCA that was limited to study-
ing only the environmental impact category “global warming
potential.” Specific standards for carbon footprinting are cur-
rently being developed by, for example, ISO (in ISO 14067)
and revised by British standards (publicly available specifica-
tion 2050), who are about to develop a seafood-specific GHG
emission standard in 2012, as well as a whole range of other
organizations; this is in response to the need for standardized
methodology to assess the environmental impact of products
(Peacock et al. 2011). While it is not clear if or when this goal
will ever be reached, these are much needed efforts.
Seafood LCAs have been performed during the past 15 years,
mainly in developed countries and of products originating in
either industrialized fisheries (Hospido and Tyedmers 2005;
Iribarren et al. 2010b, 2011; Thrane 2004, 2006; V´
azquez-Rowe
et al. 2010, 2012; Ziegler et al. 2003; Ziegler and Valentinsson
2008) or intensive aquaculture (Gr¨
onroos et al. 2006; Pelletier
et al. 2009). For fisheries, common findings include fuel used for
fishing as the main driver of environmental impacts included in
LCA, which explains why energy use and GHG emissions are
typically highly correlated. In aquaculture, feed use and com-
position drives the environmental impacts due to emissions of
both methane and nitrous oxide in agriculture (producing most
of the feed). Fossil fuel use for tractors and more importantly
on fishing boats whose catch is used to produce fishmeal and
oil, also plays a role, but is much less important. Because of the
importance of biogenic (non-fossil) emissions, energy use and
GHG emissions of farmed fish are much more weakly correlated
than for capture fisheries. An overview of the potentials and
challenges of the research field environmental assessment of
fisheries and aquaculture is provided by Ziegler (2010).
Norway is among the world’s top-ten seafood exporters
(FAO 2011). Important species from capture fisheries are
herring (Clupea harengus), mackerel (Scomber scombrus), cod
(Gadus morhua), saithe (Pollachius virens), and haddock
(Melanogrammus aeglefinus), altogether constituting 51% of to-
tal seafood export volume and 37% of export value (NSEC
2008). Farmed Atlantic salmon (Salmo salar) is the single most
important species in Norwegian seafood export, constituting
27% of the export volume and 46% of value (NSEC 2008).
In 2008, half of the Atlantic salmon farmed in the world was
farmed in Norway (FAO 2009). This sector has shown remark-
able growth over the last couple of decades and is currently the
animal-producing sector that grows fastest (FAO 2009). Alto-
gether, the species mentioned in this paragraph constituted 78%
of Norwegian seafood export in volume and 83% in value in
2008. Norwegian seafood exporting companies pay a fee (0.3%
of export value) to the Fisheries and Aquaculture Industry Re-
search Fund (FHF). The fee is used to fund industry-relevant
research. In 2008 this fund, in collaboration with the Norwe-
gian Seafood Federation (FHL) and the Norwegian Fishermen’s
Association (Norges Fiskarlag), initiated the present work to
quantify the carbon footprint and energy use of 22 seafood
products.
Seafood production, as stated previously, leads to a whole
range of environmental impacts besides GHG emissions. More-
over, the impacts caused by aquaculture systems, such as trans-
mission of disease and bred genetic material to wild stocks, are
quite different from those caused by capture fisheries, such as
benthic impacts of fishing and impacts on target and by-catch
stocks. This makes a full environmental comparison between
the two production systems difficult. Due to the low level of
knowledge and awareness about supply chain energy use and
carbon footprinting in the Norwegian seafood sector, while
recognizing that important aspects are left out, the focus in this
study was on learning more about the carbon footprint and how
to reduce it. This focus does not reflect the view that energy use
and the carbon footprint are more important than other envi-
ronmental impacts. For many types of seafood products, how-
ever, GHG emissions and biological impacts are correlated, as
the most energy-intensive fisheries often also have the highest
seafloor impact and by-catch levels. While this may often be
the case, it is not universal (Hornborg et al. 2012; Tyedmers
and Parker 2012). Therefore energy use has been suggested as
an indicator of the overall environmental impact of fisheries
(Jacquet et al. 2009; Schau et al. 2009; Thrane 2006; Ziegler
2006).
To our knowledge, this is the first time an entire sector of
food production including more than 20 products was studied
with regard to GHG emissions based on data inventory of re-
source use. A goal of the work, besides identifying important
contributors to total emissions and improvement options, was
to compare the performance of the products to each other, using
the same methodology throughout.
Methodology, Goal, and Scope
The products studied were identified together with the stake-
holders and are shown in table 1. All but blue mussels are impor-
tant export products. Despite their low economic importance,
blue mussels were included because they represent a different
type of seafood production. We used a combined top-down and
bottom-up approach to data inventory; top-down meaning that,
wherever possible, we used national statistics or published data
for the whole sector and then broke them down into different
components. For example, for fuel use we used the national prof-
itability survey, which, on an annual basis, collects information
that includes fuel use of a subset of the Norwegian fishing fleet.
These data were used to calculate gear-specific fuel factors; that
is, liters of fuel per kilogram landed in round weight for the gear
types used in Norwegian fisheries. The gear-specific fuel factors
were combined with data on how (with which gear types) each
species was caught in 2007 according to sales statistics. This
combination led to species-specific fuel factors. (For details, see
the report by Winther et al. [2009], which has been included as
supporting information to this current article, available on the
Journal’s Web site.)
2Journal of Industrial Ecology
RESEARCH AND ANALYSIS
Ta b l e 1 Seafood products included in the study
Product form Transport mode and destination Chain no.
Salmon
1 Fresh, gutted, head on Truck to Paris 1
2 Fresh, gutted, head on Truck to Oslo 2
3 Fresh, gutted, head on Truck to Moscow 3
4 Fresh, gutted, head on Air to Tokyo 4
5 Frozen, gutted, head on Container freighter to Shanghai 5
6 Fresh fillet Truck to Paris 6
7 Frozen fillet Truck to Paris 7
Mussels
8 Living, fresh sorted Truck to Paris 8
Cod
1 Fresh, gutted, head on Truck to Paris 9
2 Fresh fillet Truck to Oslo 10
3 Fresh fillet Truck to Paris 11
4 Frozen fillet Truck to Paris 12
5 Frozen fillet Truck/container freighter to Paris via
processing in China
13
6 Saltfish Truck to Lisbon 14
7 Clipfish Truck to Lisbon 15
Saithe
1 Frozen fillet Truck to Berlin 16
Haddock
1 Fresh, gutted, head on Truck/RoRo vessel to London 17
2 Frozen, gutted, head on Truck/bulk freight to London 18
Herring
1 Round frozen Bulk freight/train to Moscow 19
2 Frozen, deskinned fillet Truck to Moscow 20
Mackerel
1 Frozen round Container freighter to Tokyo 21
2 Frozen round Bulk freight/train to Moscow 22
Note: Numbering is different from numbering in the work of Winther and colleagues (2009); numbers used in their article are given in the right column,
Chain no. RoRo =roll-on/roll-off vessels.
It is important to note that this means that cod fishing, for
example, was modeled as a mixture of different fishing methods
according to how it was landed in 2007. The same mixture of
different fishing gears is used in all products made from the same
species (i.e., fresh gutted cod, fresh and frozen cod fillets, saltfish,
and clipfish), although this does not actually reflect the truth.
In reality, each product has a certain raw material composition
originating in different fisheries (e.g., the cod for saltfish and
clipfish is to a large degree produced from fish landed by coastal
fisheries, whereas the cod processed in China is almost exclu-
sively trawled). However, the mass flow within the Norwegian
wild caught seafood industry is very complex, and it proved
impossible to find reliable quantitative data on the distribution
of the fisheries behind each product, despite considerable ef-
fort both by ourselves and our industry partners. We ended up
modeling each species’ supply chains as if produced by the same
average Norwegian fishery for that species, which means that
the results apply to theoretical “average” products rather than
real products that can be found at a retailer. The project was
started in 2008, and because the goal was to describe and ana-
lyze current Norwegian fisheries and aquaculture practices, we
aimed at finding data for 2007. Where top-down modeling was
not possible due to lack of published data or official statistics,
we used specific data from individual fisheries or companies to
model sector-wide emissions (bottom-up), sometimes not even
from 2007 (e.g., in the case of processing). In those cases, con-
siderable effort was spent on verification of data from other
independent sources, especially for those data that turned out
to be important to the final results. Bait production for long-
lining was included, assuming that all the bait used (around 0.1
kilograms (kg)/kg fish landed) was herring; data used for this
were collected during this study.1Considerable effort was also
spent on verifying the sector-wide data that proved important
to the results, especially in the case of fuel and refrigerants. For
fuel use, we compared our figures with published and unpub-
lished data for single fisheries or fleets to make sure that the
figures were realistic. For refrigerants, we compared the data
we found on total use in Norwegian fisheries with information
Ziegler et al., Carbon Footprint of Norwegian Seafood on Global Market 3
RESEARCH AND ANALYSIS
Figure 1 System boundaries used in the present study. Construction of fishing vessels and gear was excluded; construction and
maintenance of transports (vehicles and roads) were included, all based on previous findings of relative importance.
from interviews with importers, suppliers, and service providers
of marine refrigeration systems.
ISO-standardized LCA methodology was used throughout
(ISO 2006a, 2006b) and the products were followed from the
production of fuel and other supply materials used in fishing and
aquaculture to fishing and farming, processing, and transporta-
tion to wholesalers in European and Asian countries (figure 1).
Various product forms were included (table 1): round, gutted,
and filleted, as well as fresh, frozen, salted, and dried products
made out of cod, saithe, haddock, herring, mackerel, salmon,
and blue mussels.
Due to the variability in edible yield between the products
included, the functional unit was set to one edible kilogram
delivered to a wholesaler, meaning that when the product was
gutted fish, the amount transported to the wholesaler was in-
creased accordingly so as to correspond to one edible kilogram.
The method chosen for coproduct allocation was mass allo-
cation. This was done after extensive discussion and stepwise
exclusion of alternative approaches. Following the hierarchy
of strategies to deal with coproduct allocation as stated by the
ISO (ISO 2006a, 2006b), we could not find a way to avoid
allocation. System expansion was considered impossible to un-
dertake, as there are no alternative systems landing only one
species or producing only fish fillets. Even if there were, the
assumption of which production would replace the one being
studied is an additional source of uncertainty and influences
results considerably. Therefore physical causality is the method
that is recommended, but we could not find any causality to
make a useful basis for allocation. Economic allocation (as used
in most seafood LCAs so far) was excluded, primarily due to
the temporal variability introduced in results from fluctuations
in prices and value relationships between coproducts. Energy
allocation (as used in Pelletier and Tyedmers 2010; Pelletier
et al. 2009) was not chosen due to the counterintuitive result
that cod by-products would carry a higher burden than cod fil-
lets due to the high fat content of the cod liver and the low
fat content of the muscle. We also found it difficult to find
data on the energy content, especially of the various coprod-
ucts. The full rationale behind the choice of allocation strategy
can be found in Appendix A in the work of Winther and col-
leagues (2009), where mass flows for each supply chain along
with important background data are given. The implications of
the choice of method for coproduct allocation are discussed and
the effect of using an alternative approach, economic alloca-
tion, was tested in the sensitivity analysis, along with nine other
important results. Based on previous findings (Magerholm Fet
et al. 2010; Tyedmers et al. 2007; V´
azquez-Rowe et al. 2010),
we chose to exclude construction of the vessel and fishing gear,
which has often proven to give insignificant contributions to
total GHG emissions of seafood products. In an analysis of all
Galician coastal fisheries, vessel construction was also excluded
for the same reason and gear use was shown to be unimportant
(Iribarren et al. 2010b).
An external reviewer followed the project and provided
valuable comments in a three-step review, as did a reference
group representing various parts of the Norwegian seafood
industry and nongovernmental organizations (NGOs). Back-
ground data regarding production of fuel, electricity, and pack-
aging materials from the Ecoinvent database were used (Ecoin-
vent 2009). Electricity production was modeled as the Nordic
interconnected grid Nordel, which leads to GHG emissions
five times higher than the average Norwegian grid (Ecoinvent
2009); this was done to avoid underestimating the importance
of electricity use. For modeling of the salmon feed, our industry
4Journal of Industrial Ecology
RESEARCH AND ANALYSIS
Figure 2 Up-to-the-dock (i.e., at landing/slaughter) greenhouse gas (GHG) emissions for the seven types of seafood studied. Note that
only diesel and refrigerant appear in the first five fisheries.
partners provided average data on the composition of the feed,
including type and origin of input and amount of feed used per
tonne of salmon farmed (see Winther et al. [2009] for data).2
Production of agricultural inputs used in salmon feeds was
taken from the work of Flysj¨
o and colleagues (2008) and ma-
rine inputs (fish meal and oil) mainly from the work of Schau
and colleagues (2009) as well as industry data reported in the
work of Winther and colleagues (2009). We used LCA software
SimaPro v.7.2 (PR´
e Consultants, Amersfoort, The Nether-
lands) and GHG emissions indicators based on the United
Nations Intergovernmental Panel on Climate Change (IPCC
2007). IPCC data were modified to exclude biogenic carbon
dioxide (CO2) emissions and biogenic uptake of CO2from the
calculations (assuming their turnover is so rapid that they do
not contribute to global warming).
Results and Discussion
Up to the Dock
The most important contributors to the carbon footprint of
chains originating in capture fisheries were the onboard use of
fuel and refrigerants (figure 2). Inventory results can be found
in the work of Winther and colleagues (2009). Fuel is used for
propulsion, and in some cases for refrigeration; emissions of re-
frigerants result from leakage of onboard cooling systems. Up to
the dock, fuel use (production and combustion) explained 63%
to 87% and refrigerants, 13% to 37% of total GHG emissions
of the various products. A general difference between pelagic
(dwelling in open seas) and demersal (dwelling near the bottom
of a water body) fisheries was found with regard to both aspects:
demersal fisheries use more fuel and emit more refrigerants per
kilogram of fish landed.
The higher fuel use is explained by the nonschooling nature
of demersal species and the fishing methods used in these fish-
eries. Even though the proportion of passive fishing methods (in
this case, fishing methods other than demersal trawling) is rela-
tively high in Norwegian demersal fisheries, especially for cod,
the fuel use of pelagic fisheries per tonne landed is considerably
less. The fishing methods used to harvest pelagic fish like her-
ring and mackerel include purse seining and pelagic trawling—
among the most energy-efficient fishing methods (Schau et al.
2009; Tyedmers 2001, 2004). Average pelagic vessels are also
newer than average demersal vessels, which explains why they
have lower emissions of refrigerants. Modern vessels have cool-
ing systems that use environmentally harmless refrigerants (e.g.,
ammonia) and therefore only a small proportion of the pelagic
vessels are still using the hydrochlorofluorocarbon (HCFC)
R22. This ozone-depleting and climate-forcing refrigerant is
still very common mainly in demersal fishing fleets around the
world. Modern equipment also gives rise to lower leakage rates
in the pelagic fleet, and the refrigerated seawater systems used
to cool the catch on pelagic vessels do not use as much refriger-
ant per tonne of fish kept compared to the refrigeration systems
used on demersal vessels.
For products from aquaculture, there is one major differential
factor: whether or not the species needs feed input. The entire
system of feed production required by carnivorous species like
salmon, starting with farms and capture fisheries all over the
world producing raw materials that are processed and trans-
ported to a feed producer and then processed into aquafeeds,
completely dominates the up-to-the-dock results for salmon.
Energy inputs at the farm site only provide a minor contribu-
tion. The opposite is true for mussel farms, which are inde-
pendent of manufactured aquafeeds because the mussels feed
on planktonic organisms filtered from water flowing through
Ziegler et al., Carbon Footprint of Norwegian Seafood on Global Market 5
RESEARCH AND ANALYSIS
the farm site. While the on-site emissions are somewhat higher
for mussel than for salmon farms—largely due to a mainte-
nance and harvest boat burning diesel oil and the small volumes
produced—the absence of feed leads to far lower up-to-the-dock
emissions for mussels (see figure 2). Low resource use for mussel
farming is in line with previous findings related to the carbon
footprint of farmed mussels (Hall et al. 2011; Iribarren et al.
2010a).
Transport Distance and Transport Mode
With regard to transport, we found that important factors
influencing the GHG emissions of seafood transport include
the transport mode (i.e., truck, ship, train, or aircraft), the size
of the vessel or vehicle, speed, load capacity (and proportion
of it that is used), transportation time, need for refrigeration,
and distance. Food miles alone is therefore a poor indicator
of the climate impact of food transportation. Generally trans-
port to the wholesaler abroad contributes less than 25% to
the total carbon footprint, with some notable exceptions (see
figures 3a and 3b). The exceptions were due to above average
resource-intensive means of transportation (salmon 4, airfreight
to Tokyo), below average resource-intensive fisheries (mackerel
1 and 2), very low edible yield (mussels), extremely long supply
chain distances (cod 5, processed in China, and mackerel 2,
shipped to Tokyo), or a combination thereof.
In the only case included where airfreight was involved
(fresh gutted salmon taken to Tokyo), transportation was re-
sponsible for 78% of the total GHG emissions; this chain
had the highest carbon footprint of all. Airfreight was, as
indicated, the most resource-intensive mode of transporta-
tion (per tonne∗kilometer [km]) and for which refrigeration—
both the energy used for refrigeration and the leakage of
refrigerants—represented the lowest proportion of transport
emissions (0.1%).3This is explained by the short refrigera-
tion time needed (often less than 12 hours), in combination
with the extremely high energy use of airfreight. Shipping in
bulk on a freight ship was the most efficient transport mode,
closely followed by rail freight. Containerized shipping of cod
for processing in China before transport to Europe was the trans-
port in which refrigeration contributed most to total transport
emissions (more than 50%, explained by the extremely long
distance and slow speed leading to a period of almost 80 days of
refrigeration, in combination with the low resource demand of
containerized shipping per tonne∗km). For these reasons, refrig-
eration during transportation cannot simply be accounted for
by adding a fixed percentage of extra resource use to a transport
as is often done.
Fresh and Frozen
Frozen fish did not generally have a higher carbon footprint
than fresh fish despite the energy required for freezing. Rather,
in cases where this aspect is the only one distinguishing the
products from each other (i.e., transport mode and distance
to market are the same), the total results are almost identical.
We modeled the use of electricity for processing and storage
in Norway using the average electricity production of Nordel,
which is a body for cooperation between the Nordic countries
(Denmark, Sweden, Norway, Finland, and Iceland), as the grids
of these countries are highly interconnected. Even though the
average Nordel electricity production is five times more climate-
intensive than the average Norwegian grid (Ecoinvent 2009),
the additional energy use for freezing is hard to detect (figure 4).
Moreover, the slightly higher emissions for processing and stor-
age of frozen fish are outweighed by more efficient transport due
to the fact that no ice is needed and hence more frozen fish can
be loaded per truck or container.
A situation in which it does matter whether the product
is fresh or frozen is when long-distance, especially interconti-
nental, transport is involved. In these cases the increased shelf
life of frozen compared to fresh fish makes it possible to slow
down transportation when the same mode of transport is used,
or by choosing a more climate-efficient form, such as ship-
ping rather than airfreight or rail freight rather than trucking
(figure 5). The same finding has been reported by Vazqu´
ez-Rowe
and colleagues (2012).
Wild and Farmed
The inevitable question is whether wild or farmed fish is more
climate efficient. It should be kept in mind that climate impact
is but one of the various types of environmental impacts caused
by seafood production. Fishing and aquaculture lead to highly
different types of total environmental impact, and are therefore
hardly comparable. However, when we focused exclusively on
GHG emissions, we were surprised to find close similarity in
total carbon footprints of a salmon and cod fillet delivered to
Paris, given the completely different supply chains behind each
product (figure 6).
Norwegian Seafood in Perspective
The products studied seem to be relatively efficient com-
pared to previously reported results for seafood products. It is
difficult to compare studies directly because they often take dif-
ferent methodological approaches with regard to either system
boundaries, functional unit, or co-product allocation. However,
as many seafood LCAs from capture fisheries have shown, fuel
use is a key input that often determines total GHG emissions
of seafood products. The fuel use in Norwegian fisheries of the
five main species found in this study by combining two datasets
(full details on methodology in Winther et al. 2009) was rel-
atively low, both when compared on a species level (weighted
averages taking into account the various fishing methods used)
and when compared across each fishing method (e.g., com-
pared with data in the work of Thrane [2004], Tyedmers [2001],
and Tyedmers et al. [2005]). The former is due in part to the
fact that a considerable portion of Norwegian whitefish, espe-
cially cod, is landed using coastal fishing gear such as gill nets
and small-scale longlines. On average, these use less fuel to
land a tonne of fish than active fishing gear such as demersal
6Journal of Industrial Ecology
RESEARCH AND ANALYSIS
Figure 3 (a) Carbon footprint results for seafood products from capture fisheries (kg GHG/kg edible product). (b) Overall results for
seafood products from aquaculture (kg CO2-eq./kg edible product). Cod: Fresh gutted, trucked to Paris (1); fresh fillets, trucked to Oslo
(2), Paris (3); frozen fillets, trucked to Paris (4), trucked to Paris, processed in China (5); saltfish, trucked to Lisbon (6); clipfish, trucked to
Lisbon (7). Saithe: Frozen fillets trucked to Berlin. Haddock: Fresh gutted, trucked/shipped to London (1); frozen gutted, trucked/bulk
shipped to London (2). Herring: Round frozen, bulk shipped and rail freight to Moscow (1); frozen skinless fillets, trucked to Moscow (2).
Mackerel: Round frozen, shipped to Tokyo (1), shipped in bulk and rail freight to Moscow (2). Salmon: Fresh gutted, trucked to Paris (1),
Oslo (2), Moscow (3), airfreight to Tokyo (4); frozen gutted, shipped to Shanghai (5); fresh fillets, trucked to Paris (6); frozen fillets, trucked
to Paris (7). Mussels: Fresh sorted, trucked to Paris. kg CO2-eq./kg edible product =kilograms carbon dioxide equivalents per kilogram
edible product.
trawls. The fact that the fuel use was found to be low when
compared with the same fishing method in other studies is
probably related to the management system and the condition
of the fished stocks (Standal 2005). Previous studies (Schau
et al. 2009) used only one of the datasets, in which the coastal
fisheries were underrepresented, probably leading to an overes-
timation of fuel use.
The Role of Fisheries Management and Fisherman
As an example, large-scale long-lining for cod in this study
required 0.31 liters (L) fuel/kg round fish.4In previous stud-
ies of Norwegian fisheries, round cod landed has been shown
to use 0.26 L fuel/kg (Sund 2009), 0.29 L fuel/kg (Svanes
et al. 2011b), and 0.37 L fuel/kg fish landed (Schau et al. 2009).
Ziegler et al., Carbon Footprint of Norwegian Seafood on Global Market 7
RESEARCH AND ANALYSIS
Figure 4 Comparison of supply chain emissions from fresh and frozen cod fillets trucked to Paris (kilograms of greenhouse gas emissions
per kilogram edible product [kg GHG/kg edible product] at wholesaler).
In Icelandic long-lining, 0.36 L fuel/kg gutted cod were used
(Guttormsd´
ottir 2009) and 0.18 L fuel/kg round cod (Fulton
2010). The average of all long-line fisheries targeting cod re-
ported in the work of Tyedmers (2001) was 0.53 L fuel/kg,
probably gutted. Considerable variability within the same fish-
ery between years has also been reported (Ramos et al. 2011;
Schau et al. 2009; Tyedmers et al. 2005). Our study only
concerns one year, and therefore it is important to recognize
that it only represents a snapshot of the situation in 2007.
In the data we found a wide range of resource efficiencies
Figure 5 Comparison of supply chain emissions from fresh salmon taken to Tokyo by airfreight and frozen salmon taken to Shanghai on a
container freighter (kilograms of greenhouse gas emissions per kilogram edible product [kg GHG/kg edible product] at wholesaler). These
destinations are approximately the same transport distance from Norway.
Figure 6 Comparison of supply chain emissions from frozen salmon (aquaculture) and cod fillets (natural) trucked to Paris (salmon 7 and
cod 4).
8Journal of Industrial Ecology
RESEARCH AND ANALYSIS
Figure 7 Variation in fuel use and landings between individual Norwegian factory trawlers.
between individual vessels, a figure influenced by the skippers
themselves (Hassel et al. 2001; Ruttan and Tyedmers 2007).
Thus there is considerable variability in fuel use within one
fishing technique depending on the circumstances under which
it is used. This indicates a major improvement potential by
designing the fisheries management system in such a way that
fisheries can be undertaken in a resource-efficient manner. The
fact that energy efficiency seems to increase with fuel prices
(Schau et al. 2009) and that great efficiency variability exists
between vessels using the same fishing method (figure 7) sug-
gests that there is such a scope for improvement in Norwegian
fisheries.
Several authors have demonstrated that increased fishing
pressure is correlated with increased fuel use per unit landed
despite parallel technological development in gear and
navigation technology over the same time period (Ramos
et al. 2011; Schau et al. 2009; Tyedmers et al. 2005). Ramos
and colleagues (2011) identified an inverse correlation between
GHG emissions of the Basque mackerel fishery and spawning
stock biomass over the study period (2001–2008). The size
of the spawning biomass and the fishing mortality certainly
are important factors that determine both biotic and abiotic
resource use and resulting emissions.
Pitcher and colleagues (2009) studied the compliance of
fisheries management systems with the Food and Agriculture
Organization of the United Nations (FAO) Code of Conduct
for responsible fishing (CoC) by surveying managing authori-
ties. The overall conclusion was that CoC compliance was low:
compliance among nations was 60% or less. Norway’s fisheries
management system had the highest CoC compliance, and al-
though the study showed considerable scope for improvement
(the 40% not complied with), this finding is likely part of the
explanation of the relatively high energy efficiency in Norwe-
gian fisheries (Iribarren et al. 2011; Tyedmers 2001; Tyedmers
et al. 2005). The fishing sector is a highly regulated and politi-
cized sector (Hersoug 2005) and energy use in fisheries is, among
many factors, determined by the framework set by fisheries man-
agement systems (Driscoll and Tyedmers 2010; Standal 2005).
This framework includes
•total available quotas and quota allocation policy;
•structural policies to cut down unprofitable overcapacity;
•technical regulations and spatial and temporal limitations
of fisheries
•regulations connected to gear adaptations and rules for
minimum size to avoid catches of juvenile fish.
Representatives from the fishing sector throughout the
present project have claimed that energy efficiency in fisheries
could be much higher if only the fisheries were managed dif-
ferently. It is therefore very promising that work is now being
published in which fisheries’ management decisions are eval-
uated using life cycle methodologies (Driscoll and Tyedmers
2010; Hornborg 2012; Hornborg et al. 2012).
Optimally, life cycle thinking should be integrated in the
planning and evaluation process of fisheries management and
Ziegler et al., Carbon Footprint of Norwegian Seafood on Global Market 9
RESEARCH AND ANALYSIS
into long-term management plans, for example, through the
use of performance indicators measured on a regular basis
(Hornborg 2012).
Aquaculture is also heavily influenced by political decisions
in terms of regulation of where farm sites and slaughter plants
are located and legislation regarding feeds and medication. The
carbon footprint of live-weight salmon produced in Norway was
in line with previous findings, despite the fact that a slightly dif-
ferent methodology was used and key data were much more de-
tailed than prior studies (Ellingsen et al. 2009; Hall et al. 2011)
and were obtained from different sources than in the work of
Pelletier and colleagues (2009). Due to the general finding that
crop-based feed inputs are less climate-intensive than animal-
derived inputs (Pelletier and Tyedmers 2007)—a finding that
was confirmed—the proportion of marine inputs in the feed
is a key factor for future improvement. However, the range
within both marine and crop-based inputs is large and there is
some overlap; the most intensive crop-based inputs have higher
GHG emissions per kilogram than the most efficient marine in-
puts. The value of generic studies that use generic data from
other regions based on different methodological approaches is
therefore of limited value for making comparisons or identifying
improvement options (Hall et al. 2011). When the proportion
of marine feed inputs is minimized, it seems that rather inten-
sive inputs are often chosen (e.g., soy, corn gluten meal, and
wheat gluten meal) to replace fish meal and oil; provided the
most climate-efficient marine inputs are being replaced (e.g.,
anchoveta and menhaden meal and oil), this could ultimately
lead to an increase in the feed carbon footprint. Therefore, in
addition to the proportion of marine inputs, the composition
of both the marine and crop-based inputs is highly important.
Life cycle thinking should be incorporated by feed formulators
to avoid suboptimization of feeds.
The aquafeed industry is and has been working on mini-
mizing the use of marine inputs for many years because of the
limited nature of wild fish stocks from which marine feed inputs
are produced. In light of the expected continued growth of the
aquaculture industry, reduced dependence on limited capture
fisheries is required. It is easier to replace fish meal than fish oil
in the feed, as the marine oils give salmon many of its healthy
properties through long-chain omega 3 fatty acids (Kiessling
2009). When marine inputs are replaced by agricultural inputs
such as soy, corn, or rape seed, it must also be borne in mind
that we increase competition for these crops that are used as
both food as well as feed in other livestock production systems.
Moreover, in doing so we start using more agricultural land
to produce seafood, which, in the face of challenges related to
world food supply for a rapidly growing human population, must
be regarded as a highly limited resource.
It should be noted that the picture regarding the climate
impacts from deforestation for soy and palm oil production
(Cederberg et al. 2011), for example, remains incomplete. In
the present calculations, only the direct emissions related to the
resources used in farming are included. No indirect emissions
due to land use change are factored in, meaning the emissions
from these systems are underestimated.
Methodological Discussion Around Processing
By-Products
Currently about a quarter of the fish meal produced comes
from by-products from fish processing for human consumption
(i.e., by-products from fish filleting plants), and many eco-
labeling organizations that label farmed seafood have rules say-
ing that marine inputs in feed should originate either from fish
processing by-products or from sustainably used stocks. Behind
these rules there is an assumption that the fishery producing
the fish for human consumption operates in the same way inde-
pendently of whether the stream of by-products is used or not
used, hence the by-products are “free” of environmental bur-
den. In other words, the fact that the by-products can be used
in organic aquaculture is judged not to influence the fishery,
which is assumed to produce only for the fillets. In practice,
this is probably not the case. Even though the economic value
of the by-products is much lower than the value of the fillets,
the income generated by by-products does increase the overall
profitability of the processing plant compared to a value of zero.
The growth of certified aquaculture using this criteria increases
demand for the by-products from (sometimes) unsustainable
fisheries and improves the market for them. The extreme case
would be the use of processing by-products from a collapsed
stock of bluefin tuna. One of the feed producers involved in
this project understood our rationale for using mass-based al-
location using this example, since they would never engage in
using a fish meal based on bluefin tuna by-products, exactly
for the reasons mentioned above. Therefore it is our view that
by-products should carry their part of the environmental bur-
den caused upstream and was an important reason behind our
choice of mass-based allocation.
Probably the single most debated issue in this study is the
fact that we ascribe environmental impact based on mass to
by-products that are used further. The rationale for doing so is
described above, and it is our opinion that whitefish trimmings
should not be free from environmental burden simply because
they have a low economic value compared to the fillets. They
are still part of an often resource intensive production system.
Another drawback of basing the allocation on the economic
value of a main product over a by-product is the high variability
in seafood prices over time. When economic valuation is used
to compare environmental performance over time, it leads to
results that reflect only differences in prices, not real differences
in resource use.
Mass allocation has its own drawbacks, most notably the fact
that it becomes very important whether a by-product is used or
not, and it can be very difficult to obtain reliable data on this
matter. Another drawback of mass allocation is that it does not
matter how a coproduct is used, whether it is composted for
further use as fertilizer or if it is used for feed—or food. (See Ap-
pendix B in the work of Winther et al. [2009] for an extensive
discussion of this matter.) Despite the large difference of opin-
ions that exists concerning coproduct allocation in seafood pro-
duction systems (Ayer et al. 2007; Pelletier and Tyedmers 2011;
Svanes et al. 2011a), it is at least reassuring that the overall
10 Journal of Industrial Ecology
RESEARCH AND ANALYSIS
results in terms of hot spots and comparisons did not change
even though the absolute numbers change dramatically when
going from mass allocation to economic allocation (Winther et
al. 2009).
Does Carbon Footprinting Take Away the Focus from
the Real Problems?
Some NGOs fear that the increasing focus on the car-
bon footprint is a move from the industry to draw attention
away from the “real problems” of overfishing, destructive fish-
ing methods, high discard rates, and a lack of proper fisheries
management. We do not agree with this view and base our ar-
gument on a general observation: unsustainable fisheries (from
a biological perspective) also tend to have a higher carbon foot-
print than sustainable fisheries (Jacquet et al. 2009; Schau et al
2009; Thrane 2006; Ziegler 2006). This finding is related mainly
to two factors: first, using the same methods, more effort, and
thereby fuel, has to be spent to land a tonne of fish from an
overfished stock than an optimally fished stock. Second, the
most destructive fishing methods in terms of by-catch, discards,
and seafloor impacts are also the most energy intensive (Ziegler
2006). In these cases, the energy use or carbon footprint could
actually serve as an overall indicator of sustainability. More
importantly, however, is that even when the two areas do not
conform and a biologically sustainable fishery has a large carbon
footprint (or vice versa), it is very important to become aware
of these trade-offs and, in the long run, try to incorporate this
aspect into the framework for the assessment of sustainability
(Hornborg et al. 2012).
Is Carbon Labeling the Way to Go?
Eco-labels of type III according to the ISO are based on quan-
titative and qualitative results of LCA and represent the most
advanced and least subjective type of eco-labels (ISO 14025
[2006c]). LCA-based knowledge and new aspects, like carbon
footprinting, can be incorporated into existing eco-labels of type
I (ISO 14024 [ISO 1999]), as shown by the Swedish eco-labeling
organization, KRAV (Peacock et al. 2011; Thrane et al. 2009).
In 2010 KRAV incorporated climate criteria into several groups
of food products, seafood being one of them, and this work is
being continued by developing and launching climate criteria
for other groups of food products (www.klimatmarkningen.se).
As mentioned previously, new standards for carbon footprint-
ing are currently underway for example in ISO 14067 and the
seafood-specific PAS 2050 standard mentioned earlier. Carbon
labeling can take on many forms, but if the goal is to decrease
GHG emissions from production, then we propose it is more
efficient to apply life cycle-based criteria in the way KRAV has
done (e.g., not allow the use of refrigerants that contribute to
global warming in certified fisheries) than to label products with
an exact number of grams of GHG emissions emitted per pack-
age, especially while data are so sparse and the methodology is
still under development.
Action Items for the Norwegian Fishing Sector
A relatively simple improvement—simple in the sense that
it involves only technological changes—is the replacement of
the HCFC R22 with environmentally harmless refrigerants like
ammonia (Svanes et al. 2011b; Ziegler et al. 2010). Today these
systems have been developed so that they do not represent any
increased risk for accidents onboard, as they previously did. The
importance of this change cannot be overstated, as it can reduce
the carbon footprint of the whitefish products included in this
study by up to 30% if the right substitutes are chosen (Svanes
et al. 2011b; Winther et al. 2009). R22 is a substance that is
being phased out under the Montreal Protocol due to its ozone
depleting properties and, since 2010, refilling cooling systems
using new R22 is prohibited; R22 will be completely phased out
within the European Union by 2015 (Sander Poulsen 2011).
Existing systems can therefore only be refilled using old R22
that is recovered when scrapping old cooling systems. If the
phasing out of R22 is done in the easiest way, however, which
is to let hydrofluorocarbons (HFCs) like R404a or R507 re-
place it, then the carbon footprint will increase rather than
decrease; the IPCC 2007 indicators for these substances are
about double those of R22. This is the easiest method because
these refrigerants are so-called drop-in refrigerants for R22: no
equipment needs to be replaced, only the refrigerant. Switching
to ammonia requires replacement of the entire cooling system,
which means higher initial costs, but the investment pays back
in short time because ammonia is cheaper than synthetic re-
frigerants and is more efficient (NMR 2000; Sander Poulsen
2011; Ziegler et al. 2010). As a result, the Norwegian fish-
ing sector has started working on this issue and, in parts, R22
has already been replaced by ammonia systems (Svanes et al.
2011b).
As already indicated, considerable potential also exists for
the improvement of energy efficiency of fisheries, by both in-
corporating this aspect into the design of fisheries management
systems and operating vessels in a fuel-efficient way within the
framework set by the fisheries management system. Consumers,
industry, and regulating authorities may need to rethink the
costs and benefits of frozen versus fresh fish when interconti-
nental transportation and the measurement of GHGs are in-
volved. It is currently an official policy to increase the propor-
tion of fresh fish in Norwegian seafood export. This could lead
to increased climate impacts if it involves more airfreight than
current operations. Policy making on various levels would ben-
efit from being informed by studies showing quantified impacts
across a sector based on scientifically accepted and standardized
methods, as well as from high-quality data.
Conclusions
•For products originating in capture fisheries, use of fuel and
refrigerants on the fishing vessel were the most important
contributors to the total carbon footprint.
Ziegler et al., Carbon Footprint of Norwegian Seafood on Global Market 11
RESEARCH AND ANALYSIS
•The carbon footprint of demersal fish such as cod, had-
dock, and saithe was higher than that of pelagic fish such
as herring and mackerel.
•For products originating in aquaculture, the amount and
type of feed used determined the carbon footprint.
•Whether products are fresh or frozen only affects the car-
bon footprint if it allows a change in transport mode (e.g.,
from air to sea).
•Food miles are a poor metric, even an irrelevant metric,
even to assess the climate impact of food transportation,
not to mention the entire environmental impact of a
product.
•GHG emissions and energy use can be correlated, but this
is not always the case (e.g., when emissions of biogenic or
synthetic GHGs are significant).
•Processing and transportation are generally of low impor-
tance, with the notable exception of airfreight.
•The carbon footprints of a salmon and cod fillet were
almost the same, despite completely different production
systems.
•Edible yield and by-product use were important factors.
•The variability found on several scales—between fishing
vessels, fisheries, and products—shows a large scope for
improvements, both in production and consumption pat-
terns.
•The carbon footprint is often linked to other impacts such
as stock status, levels of discard, and seafloor impact.
•Policy making for fisheries, aquaculture, and marketing
would benefit from being informed by this type of quanti-
tative analysis.
Acknowledgements
We would like to thank all those in the Norwegian fisheries
and aquaculture sector who have patiently helped us find and
verify data, especially the representatives in the project’s steer-
ing and reference groups. Of course, we are also most grateful
to the Fisheries and Aquaculture Industry Research Fund, for
funding the work. This JIE article summarizes a more extensive
report, available as supporting information on the journal Web
site, presenting the project findings in a scientific framework and
placing those findings in the context of the research literature
and related issues. Three anonymous reviewers commissioned
by the Journal and one external reviewer commissioned by the
project helped us improve the publication.
Notes
1. One kilogram (kg, SI) ≈2.204 pounds (lb).
2. One tonne (t) =103kilograms (kg, SI) ≈1.102 short tons.
3. One kilometer (km, SI) ≈0.621 miles (mi).
4. One liter (L) =0.001 cubic meters (m3,SI)≈0.264 gallons (gal).
References
Ayer, N., P. H. Tyedmers, N. L. Pelletier, U. Sonesson, and A. Scholz.
2007. Co-product allocation in life cycle assessments of seafood
production systems: Review of problems and strategies. Interna-
tional Journal of Life Cycle Assessment 12(7): 480–487.
Cederberg, C., U.M. Persson, K. Neovius, S. Molander, and R. Clift.
2011. Including carbon emissions from deforestation in the carbon
footprint of Brazilian beef. Environmental Science and Technology
45(5):1773–1779.
Driscoll, J. and P. H. Tyedmers. 2010. Fuel use and greenhouse gas
emissions implications of fisheries management: The case of the
new England Atlantic herring fishery. Marine Policy 34(3): 353–
359.
Ecoinvent v. 2.0. 2009. LCA background database. Access can be
purchased at www.ecoinvent.ch.
Ellingsen, H., J. O. Olaussen, and I. B. Utne. 2009. Environmental
analysis of the Norwegian fishery and aquaculture industry – A
preliminary study focusing on farmed salmon. Marine Policy 33(3):
479–488.
FAO (Food and Agriculture Organization of the United Nations).
2009. The State of World Fisheries and Aquaculture (SOFIA) 2008.
Rome, Italy: FAO Fisheries and Aquaculture Department.
FAO (Food and Agriculture Organization of the United Nations).
2011. The State of World Fisheries and Aquaculture (SOFIA) 2010.
Rome, Italy: FAO Fisheries and Aquaculture Department.
Flysj¨
o, A., C. Cederberg, and I. Strid. 2008. LCA-databas f¨
or kon-
ventionella Fodermedel [LCA database for conventional feeds].
Version 1 SIK report no. 772.
Fulton, S. 2010. Fish and fuel: Life cycle greenhouse gas emissions
associated with Icelandic cod, Alaskan pollock and Alaskan pink
salmon fillets delivered to the United Kingdom. M.Sc. thesis,
Dalhousie University, Halifax, Nova Scotia, Canada.
Gr¨
onroos, J., J. Sepp¨
ala, F. Silvenius, and T. M¨
akinen. 2006. Life cycle
assessment of Finnish cultivated rainbow trout. Boreal Environ-
mental Research 11: 410–414.
Guttormsd´
ottir, A. B. 2009. Life cycle assessment on two Icelandic cod
products based on two different fishing methods. Masters thesis,
Mat´
ıs/H´
ask´
oli ´
Islands, Reykjavik, Iceland.
Hall, S. J., A. Delaporte, M. J. Phillips, M. Beveridge, and M. O’Keefe.
2011. Blue frontiers. Managing the environmental costs of aquaculture.
Penang, Malaysia: WorldFish Center.
Hall, S. 1999. The effects of fishing on marine ecosystems and communities.
Oxford, UK: Blackwell Science.
Hassel, K., A. Farstad, and D. Standal. 2001. Fiskerinaeringens mu-
ligheter til ˚
a bidra til oppfyllelse av Kyoto-avtalen [The fishing
sectors potential to contribute to fulfillment of the Kyoto agree-
ment] [in Norwegian]. SINTEF Fisheries and Aquaculture Re-
port No. 830011.00. Trondheim, Norway: SINTEF FIsheries and
Aquaculture.
Hersoug, B. 2005. Closing the commons. Norwegian fisheries from open
access to private property. Delft, The Netherlands: Ebouron Aca-
demic Publishers.
Hornborg, S. 2012. Evaluating fisheries management from a life cycle
perspective with new approaches to by-catch impacts. Licentiate
thesis. G¨
oteborg University, Dept. of Biological and Environ-
mental Sciences and SIK, The Swedish Institute for Food and
Biotechnology, G¨
oteborg Sweden.
12 Journal of Industrial Ecology
RESEARCH AND ANALYSIS
Hornborg, S., P. Nilsson, D. Valentinsson, and F. Ziegler. 2012.
Integrated environmental assessment of fisheries management:
Swedish Nephrops trawl fisheries evaluated using a life cycle ap-
proach. Marine Policy 36(6): 1193–1201.
Hospido, A. and P. H. Tyedmers. 2005. Life cycle environmen-
tal impacts of Spanish tuna fisheries. Fisheries Research 76(2):
174–186.
Iribarren, D., A. Hospido, M. T. Moreira, and G. Feijoo. 2010a. Car-
bon footprint of canned mussels from a business-to-consumer
approach. A starting point for mussel processors and policy mak-
ers. Environmental Science & Policy 13(6): 509–521.
Iribarren, D., I. V´
azquez-Rowe, A. Hospido, M. T. Moreira, and G.
Feijoo. 2010b. Estimation of the carbon footprint of the Gali-
cian fishing activity (NW Spain). Science of the Total Environment
408(22): 5284–5294.
Iribarren, D., I. V´
azquez-Rowe, A. Hospido, M. T. Moreira, and G.
Feijoo. 2011. Updating the carbon footprint of the Galician fish-
ing activity (NW Spain). Science of the Total Environment 409(8):
1609–1611.
IPCC (Intergovernmental Panel on Climate Change). 2007. Contri-
bution of Working Group I to the Fourth Assessment Report of
the Intergovernmental Panel on Climate Change, 2007, edited
by S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K.B.
Averyt, M. Tignor and H.L. Miller. Cambridge, UK: Cambridge
University Press.
ISO (International Organization for Standardization). 1999. ISO
14024:1999. Environmental labels and declarations - Type I envi-
ronmental labelling - Principles and procedures. Geneva, Switzer-
land: ISO.
ISO (International Organization for Standardization). 2006a. ISO
14040:2006(E). Environmental management – Life cycle assessment
– Principles and framework. Geneva, Switzerland: ISO.
ISO (International Organization for Standardization). 2006b. ISO
14044:2006(E). Environmental management – Life cycle assessment
– Requirements and guidelines. Geneva, Switzerland: ISO.
ISO (International Organization for Standardization). 2006c. ISO
14025:2006. Environmental labels and declarations Type III
environmental declarations–Principles and procedures. Geneva,
Switzerland: ISO.
Jacquet, J., J. Hocevar, S. Lai, P. Majluf, N. Pelletier, T. Pitcher, E.
Sala, R. Sumaila, and D. Pauly. 2009. Conserving wild fish in a
sea of market-based efforts. Oryx 44(1): 45–56.
Kiessling, A. 2009. Feed—The key to sustainable fish farming. In Fish-
eries, sustainability and development,editedbyP.Wramner,H.
Ackefors, and M. Cullberg, 303–322. Halmstad Tryckeri, Sweden:
Royal Swedish Academy of Agriculture and Forestry (KSLA).
Magerholm Fet, A., E. M. Schau, and C. Haskins. 2010. A framework
for environmental analyses of fish food production systems based
on systems engineering principles. Systems Engineering 13(2): 109–
118.
NMR (Nordic Council of Ministers). 2000. Alternatives to HCFC as
refrigerant in shipping vessels. TemaNord 2000: 573. Copenhagen,
Denmark: Nordic Council of Ministers.
NSEC (Norwegian Seafood Export Council). 2008. Årsstatistikk 2008
(Annual statistics 2008) [in Norwegian]. www.seafood.no.
Peacock, N., C. de Camillis, D. Pennington, H. Aichinger, A. Parenti,
J.-P. Rennaud, A. Raggi, F. Brentrup, B. S´
ara, U. Schenker, N.
Unger and F. Ziegler. 2011. Towards a harmonised framework
methodology for the environmental assessment of food and drink
products. International Journal of Life Cycle Assessment 16(3): 189–
197.
Pelletier, N. and P. H. Tyedmers. 2007. Feeding farmed salmon: Is
organic better? Aquaculture 272(1–4): 399–416.
Pelletier, N. and P. H. Tyedmers. 2010. Life cycle assessment of frozen
tilapia fillets from Indonesian lake-based and pond-based inten-
sive aquaculture systems. Journal of Industrial Ecology 14(3): 467–
481.
Pelletier, N. and P. H. Tyedmers. 2011. An ecological economic cri-
tique of the use of market information in life cycle assessment
research. Journal of Industrial Ecology 15(3): 342–354.
Pelletier, N., P. H. Tyedmers, U. Sonesson, A. Scholz, F. Ziegler, A.
Flysj¨
o, S. Kruse, B. Cancino, and H. Silverman. 2009. Not all
salmon are created equal: Life cycle assessment (LCA) of global
salmon farming systems. Environmental Science and Technology
43(23): 8730–8736.
Pitcher, T., D. Kalikoski, G. Pramod, and K. Short. 2009. Not honour-
ing the code. Nature 457: 658–659.
Ramos, S., I. V´
azquez-Rowe, I. Artetxe, M. T. Moreira, G. Feijoo, and J.
Zuf´
ıa. 2011. Environmental assessment of the Atlantic mackerel
(Scomber scombrus) season in the Basque Country. Increasing the
timeline delimitation in fishery LCA studies. International Journal
of Life Cycle Assessment 16: 599–610.
Ruttan, L. M. and P. H. Tyedmers. 2007. Skippers, spotters and
seiners: Analysis of the “skipper effect” in US menhaden
(Brevoortia spp.) purse-seine fisheries. Fisheries Research 83(1):
73–80.
Sander Poulsen, T. 2011. Nordiske indsatser for R-22 i k¨
oleanlg p˚
a
skibe [Nordic efforts for R-22 in refrigeration systems on ships].
TemaNord report 2011: 503 37. Copenhagen: Nordic Council of
Ministers. www.norden.org/sv/publikationer/publikationer/2011-
503. Accessed July 2012.
Schau, E. M., H. Ellingsen, A. Endal, and S. A. Aanondsen. 2009.
Energy consumption in Norwegian fisheries. Journal of Cleaner
Production 17(3): 325–334.
Standal, D. 2005. Nuts and bolts in fisheries management–a tech-
nological approach to sustainable fisheries? Marine Policy 29(3):
255–263.
Sund, V. 2009. Environmental assessment of northeast Arctic cod
caught by long-lines and Alaskan Pollock caught by pelagic trawls.
Bachelor of Science thesis, G¨
oteborg University, G¨
oteborg, Swe-
den, and SIK, The Swedish Institute for Food and Biotechnology,
G¨
oteborg, Sweden. SIK report 799.
Svanes, E., M. Vold, and O. J. Hanssen. 2011a. Effect of different
allocation methods on LCA results of products from wild-caught
fish and on the use of such results. International Journal of Life Cycle
Assessment 16(6): 512–521.
Svanes, E., M. Vold, and O. J. Hanssen. 2011b. Environmental assess-
ment of cod (Gadus morhua) from autoline fisheries. International
Journal of Life Cycle Assessment 16(7): 611–624.
Thrane, M. 2004. Energy consumption in the Danish fishery. Identifi-
cation of key factors. Journal of Industrial Ecology 8(1–2): 223–239.
Thrane, M. 2006. LCA of Danish fish products. New methods and
insights. International Journal of Life Cycle Assessment 11(1): 66–
74.
Thrane, M., F. Ziegler, and U. Sonesson. 2009. Eco-labelling of wild-
caught seafood products. Journal of Cleaner Production 17:416–423.
Tyedmers, P. H. 2001. Energy consumed by north Atlantic fisheries.
In Fisheries impacts on North Atlantic ecosystems: Catch, effort and
national/regional data sets, edited by D. Zeller, R. Watson, and D.
Pauley. Fisheries Centre Research Reports, vol. 9, no. 3, 12–34.
Vancouver, British Columbia, Canada: Fisheries Center, Univer-
sity of British Columbia.
Ziegler et al., Carbon Footprint of Norwegian Seafood on Global Market 13
RESEARCH AND ANALYSIS
Tyedmers, P. H. 2004. Fisheries and energy use. In Encyclopedia of
energy, vol. 2, edited by C. Cleveland, 683–693. Amsterdam:
Elsevier.
Tyedmers, P. and R. Parker. 2012. Fuel use and greenhouse gas emis-
sions from global tuna fisheries: A preliminary assessment. ISSF
Technical Report 2012-03. International Seafood Sustainability
Foundation. McLean, VI, USA.
Tyedmers, P. H., N. Pelletier, and N. Ayer. 2007. Biophysical sustain-
ability and approaches to marine aquaculture development policy in the
United States. Tacoma Park, MD, USA: Marine Aquaculture Task
Force.
Tyedmers, P. H., R. Watson, and D. Pauly. 2005. Fueling global fishing
fleets. Ambio 34(8): 619–622.
V´
azquez-Rowe, I., M. T. Moreira, and G. Feijoo. 2010. Life cycle assess-
ment of horse mackerel fisheries in Galicia (NW Spain): Com-
parative analysis of two major fishing methods. Fisheries Research
106(3): 517–527.
V´
azquez-Rowe, I., M. T. Moreira, and G. Feijoo. 2012. Environmental
assessment of frozen common octopus (Octopus vulgaris) captured
by Spanish fishing vessels in the Mauritanian EEZ. Marine Policy
36(1): 180–188.
Winther, U., F. Ziegler, E. Skontorp Hognes, A. Emanuelsson, V.
Sund, and H. Ellingsen. 2009. Carbon footprint and energy use of
Norwegian seafood products. SINTEF Fisheries and Aquaculture
report: SFH80 A096068. Trondheim, Norway: SINTEF Fisheries
and Aquaculture.
Ziegler, F., P. Nilsson, B. Mattsson, and Y. Walther. 2003. Life cycle
assessment of frozen cod fillets including fishery-specific environ-
mental impacts. International Journal of Life Cycle Assessment 8(1):
39–47.
Ziegler, F. and D. Valentinsson. 2008. Environmental life cy-
cle assessment of Norway lobster (Nephrops norvegicus)
caught along the Swedish west coast by creels and con-
ventional trawls. LCA methodology with case study. In-
ternational Journal of Life Cycle Assessment 13(6): 487–
497.
Ziegler, F. 2006. Environmental life cycle assessment of seafood prod-
ucts from capture fisheries. Ph.D. thesis, G¨
oteborg University,
Department of Marine Ecology and SIK, The Swedish Institute
for Food and Biotechnology, G¨
oteborg, Sweden. SIK report 754
(ISBN 91-89677-27-7).
Ziegler, F. 2010. Challenges in assessing the environmental impacts of
aquaculture and fisheries. In Environmental assessment and man-
agement in the food industry: Life cycle assessment and related ap-
proaches, edited by U. Sonesson, J. Berlin and F. Ziegler, chap. 8.
Cambridge, UK: Woodhead Publishing.
Ziegler, F., E. Skontorp Hognes, A. Emanuelsson, U. Winther, H.
Ellingsen, and V. Sund. 2010. There is more to greenhouse gas
emissions of fisheries than carbon dioxide from fuel combustion.
Proceedings from Energy Use in Fisheries, Seattle, WA, USA,
14–17 November 2010.
About the Authors
Friederike Ziegler is a senior researcher in the field of life
cycle assessment (LCA) and eco-labeling of seafood products,
Andreas Emanuelsson is a PhD student, and Veronica Sund
is a consultant, all in the Department of Sustainable Food Pro-
duction at the Swedish Institute for Food and Biotechnology,
G¨
oteborg, Sweden. Ulf Winther is research director in the De-
partment of International Projects and Consulting and Erik
Skontorp Hognes is a consultant to the Department of Fish-
eries Technology both at SINTEF Fisheries and Aquaculture,
Trondheim, Norway. Harald Ellingsen is a professor in the De-
partment of Marine Technology at the Norwegian University
of Science and Technology, Trondheim, Norway, and is also
affiliated with SINTEF Fisheries and Aquaculture.
Supporting Information
Supporting information may be found in the online version of this article.
Supporting Information S1: The supporting information presents the full 2009 SINTEF report titled Carbon footprint and
energy use of Norwegian seafood products (Winther et al. 2009).
Please note: Wiley-Blackwell is not responsible for the content or functionality of any supporting information supplied by
the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.
14 Journal of Industrial Ecology
