ArticlePDF Available

Greenhouse gas emissions from life cycle assessment of Norwegian food production systems

Authors:
  • Nordregio, Nordic Centre for Spatial Devleopment

Abstract

This paper presents an assessment of atmospheric emissions of greenhouse gases (GHGs) and associated land use for the production of milk, beef, grain and potatoes. It compares the less intensive (i.e. organic) farming system with the more intensive (i.e. conventional) food production system. The emission sources, trade-offs with land and potential for reduction of GHG emissions were analysed. The sources for farm accounts data on inputs and outputs are representative farm types constructed from data from the Norwegian Farm Accountancy Data Network. The analysis was carried out with life cycle assessment, including processes from manufacturing of inputs to farms, and on-farm production up to the farm gate. The results show that it is worthwhile considering a greater proportion of food energy from vegetable rather than from animal products, analysing grass-based meat production in more detail and reducing mineral fertilizer use.
This article was downloaded by: [NILF]
On: 25 June 2013, At: 06:13
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer
House, 37-41 Mortimer Street, London W1T 3JH, UK
Acta Agriculturae Scandinavica, Section A – Animal
Science
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/saga20
Greenhouse gas emissions from life cycle assessment
of Norwegian food production systems
K. Refsgaard a , H. Bergsdal b , H. Berglann a & J. Pettersen b
a Research Department , Norwegian Agricultural Economics Research Institute (NILF) ,
Storgata 2/4/6, Oslo , Norway
b Environmental Systems Analysis (MiSA) , Trondheim , Norway
Published online: 24 Apr 2013.
To cite this article: K. Refsgaard , H. Bergsdal , H. Berglann & J. Pettersen (2012): Greenhouse gas emissions from life
cycle assessment of Norwegian food production systems, Acta Agriculturae Scandinavica, Section A – Animal Science, 62:4,
336-346
To link to this article: http://dx.doi.org/10.1080/09064702.2013.788204
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to
anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses
should be independently verified with primary sources. The publisher shall not be liable for any loss, actions,
claims, proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or
indirectly in connection with or arising out of the use of this material.
ORIGINAL ARTICLE
Greenhouse gas emissions from life cycle assessment of Norwegian food
production systems
K. REFSGAARD
1
, H. BERGSDAL
2
, H. BERGLANN
1
& J. PETTERSEN
2
1
Research Department, Norwegian Agricultural Economics Research Institute (NILF), Storgata 2/4/6, Oslo, Norway, and
2
Environmental Systems Analysis (MiSA), Trondheim, Norway
Abstract
This paper presents an assessment of atmospheric emissions of greenhouse gases (GHGs) and associated land use for the
production of milk, beef, grain and potatoes. It compares the less intensive (i.e. organic) farming system with the more
intensive (i.e. conventional) food production system. The emission sources, trade-offs with land and potential for reduction
of GHG emissions were analysed. The sources for farm accounts data on inputs and outputs are representative farm types
constructed from data from the Norwegian Farm Accountancy Data Network. The analysis was carried out with life cycle
assessment, including processes from manufacturing of inputs to farms, and on-farm production up to the farm gate. The
results show that it is worthwhile considering a greater proportion of food energy from vegetable rather than from animal
products, analysing grass-based meat production in more detail and reducing mineral fertilizer use.
Keywords: GHG emissions, farming systems, life cycle assessment.
Introduction
Food production in the Western world contributes
heavily to the emission of greenhouse gases (GHGs),
as emphasized by international organizations such
as United Nations Conference on Trade and Devel-
opment (UNCTAD; Hoffmann, 2011) and Interna-
tional Assessment of Agricultural Knowledge, Science
and Technology for Development (IAASTD, 2009)
and by the research community (Tukker et al., 2006;
Tukker & Jansen, 2006). These GHG emissions are
primarily in the form of N
2
O (from fertilizers and
manure application, tillage of moors and nitrogen
emission), CH
4
(from livestock emissions and stored
animal manure) and CO
2
(from fuel for agricultural
machinery). The contribution of agriculture to the
GHG emissions of the EU was 9% in 2007, with
462 Mt of CO
2
-equivalents (CO
2
e) (Eurostat, 2010).
A study on Europe coordinated by the EU Joint
Research Centre (Weidema et al., 2008) shows that
the total annual consumption of meat and dairy
products in EU-27 contributes 14.2% of the GHG
emissions.
According to Bye et al. (2010), Norway had a total
emission of 51 Mt CO
2
e in 2009, of which 4.8 Mt
(9%) came from the agricultural sector (48% CH
4
,
41% N
2
O and 10% CO
2
). Statistics Norway
provides annual data for the direct emission of
GHG from each sector in the Norwegian economy.
However, there are inputs either outside the agri-
cultural sector or factors not accounted for that
contribute to the GHG emission from food produc-
tion. If we include the emissions arising indirectly
from the manufacturing of mineral fertilizer, and
CO
2
emissions from arable land management and
tillage of moor, the total emission from agriculture
has been estimated to increase from 4.8 Mt CO
2
eto
around 7.8 Mt CO
2
e (Trømborg et al., 2007). The
impact of other factors, such as nitrous oxide
emissions from soybean cultivation in tropical coun-
tries, may be as relevant to the assessment of
European livestock production as the GHG losses
from the animals themselves. Therefore, the indirect
emissions related to the use of inputs on a farm
should be included, such as the environmental
impact from producing the imported feeds and
Correspondence: H. Berglann, Norwegian Agricultural Economics Research Institute (NILF), Storgata 2/4/6, Postboks 8024 Dep, Oslo 0030, Norway.
Tel: +4722367236. Fax: +47227299. E-mail: helge.berglann@nilf.no
Acta Agriculturae Scand Section A, 2012
Vol. 62, No. 4, 336346, http://dx.doi.org/10.1080/09064702.2013.788204
(Received 15 November 2012; revised 6 March 2013; accepted 14 March 2013)
#2012 Taylor & Francis
Downloaded by [NILF] at 06:13 25 June 2013
fertilizer (Van der Werf et al., 2009). Furthermore,
the carbon fixation on land (Hoffmann, 2011) is a
major sequestration source to consider as it impacts
the N
2
O emissions from land, and the effect differs
between crops, or in other words between perennials
and annuals. Food processing, transport, storage and
shopping structures after the farm gate may also be
relevant to consider (Coley et al., 2009); although,
literature shows that it is what happens at the
agricultural stage and not throughout the rest of
the supply chain that matters most for GHG emis-
sions. The relevance of investigating the impact of
agricultural production purely from a sectorial point of
view is therefore questionable.
Historically, most industrialized countries have
experienced a shift from animal power to mechanical
power, and hence from biomass ‘‘fuel’’ to fossil fuels,
and increased overall energy use over time. For
example, this has occurred through a change in the
type of fertilizer used, from natural fertilizer such as
manure and night soil to chemical fertilizer, which is
industrially mined and processed (Hall et al., 1992;
Neset & Lohm, 2005; Pimentel et al., 2005).
However ceteris paribus other natural conditions
such as access to water, rainfall, temperature and so
on, and not the least type of crop, also matter for the
effectiveness of food production.
Organic farming comprises production practices
where some fossil fuel-based inputs deliberately are
substituted with land, such as the use of mineral
fertilizer substituted by land used for legumes for
fertilization only. It uses the market to help support
these objectives and compensate for the internaliza-
tion of externalities (Lampkin, 2003). Compared to
conventional farming,
1
organic production often
results in more land being used for the same amount
of food output. Meta-analyses by Mondelaers et al.
(2009) and by Badgley et al. (2007) report an
average organic/conventional yield ratio of 83% and
91%, respectively, for developed countries for all
crops. To compare GHG emissions of these systems,
it is, therefore, of interest to show both the use of
GHG emissions per unit produced and the use of
land for the same unit produced. Any restrictions
imposed in one part of a farming system might be
compensated for in other parts of the system by the
farmer.
In analyses comparing GHG emission among food
production causes, the unit of comparison varies
according to dietary needs. Food consumption is
most often measured in kg or Mcal, although other
dietary needs also need to be considered. Studies
comparing GHG emissions per area or per mass
from different production systems have already been
conducted for some common food products in
various countries, using the well-known approach
of life cycle assessments (LCA) (Mondelaers et al.,
2009). LCA is used to assess the environmental
impacts of production systems and services, ac-
counting for emissions and resource uses during
production, distribution, consumption and disposal
of a product (Carlsson-Kanyama et al., 2003;
Brentrup et al., 2004; Hertwich, 2005). Today,
LCA is an internationally standardized tool (ISO,
1997, 2006); although, questions about system
delimitation and general principles remain.
In this work, we aim to represent average Norwe-
gian farm production rather than production in a
more specific and narrowly defined geographical
area and farming practice. Importantly, we also use
data from whole farms rather than disaggregated
data from trials, because simple relations between
individual input factors, on-farm processes, GHG
emissions and land use cannot be expected, because
input factors and processes may substitute for each
other (Refsgaard et al., 1998). By using data from
whole farms, we are able to track substitutions and
‘‘deficient’’ resource use due to integrated produc-
tion processes.
Purpose of the study
This article presents an assessment of GHG emis-
sions and associated land use linked to the produc-
tion of milk, meat, grain and potatoes in both a
conventional and an organic production system, and
comparing plant and meat production. The assess-
ment includes analysis of the following questions:
(1) What are the major GHG-emission sources
and processes up to the farm gate for the
specific products and production systems?
(2) What is the trade-off between GHG emissions
and land use regarding organic versus conven-
tional production systems?
(3) What is the difference in GHG emissions and
land use between different products (meat
versus plants)?
(4) How do the results from our study of some
Norwegian food products compare to those in
other studies? How valid are our results?
(5) Where do the analyses show a potential for
reduction in GHG emissions?
Methodology and data
Production systems, system boundaries, indicators and
scope
We use LCA to analyse the food production systems
of food at the farm level. To conduct a comprehen-
sive assessment of the environmental consequences
Atmospheric emissions of greenhouse gases 337
Downloaded by [NILF] at 06:13 25 June 2013
of a product, the indirect emissions related to the use
of inputs on a farm should be included. We, there-
fore, used a product-oriented assessment quantifying
the environmental impacts per kilogramme product
delivered. The system boundary includes the whole
production chain until farm gate. We followed the
LCA methodology proposed by ISO (1997, 2006).
The environmental impacts were measured with an
indicator for GHG emission measured in CO
2
eand
an indicator for land use measured in decare (daa). To
consider the difference in nutritional values between
different food products, the food production also was
measured in megacalorie (Mcal). This allows for
substitution of different products with each other.
Milk, beef, grain and potatoes were chosen for
analysis. Together these four product groups make
up approximately 50% of the food basket measured
in kilogramme (Statistics Norway, years 2005
2007). The study includes domestic production
that represents the average inventory across different
farming practices and local variations. This gives
results that might be used to draw general conclu-
sions about trends between the product groups and
the identification of the key inputs and sources for
each product group. These results are indicative and
do not release numbers that can be directly com-
pared with the results of studies where specific
cultivation experiments are performed and evalu-
ated. That is, the emission figures are subject to
relatively large uncertainty (due to the overall data)
for each product, but are valuable for comparing
general features.
Direct emissions
The LCA carried out shows the ratio of emissions to
air of GHG emissions. In our analyses, the major
drivers considered included CH
4
from the enteric
fermentation in cows and from the handling of
manure; further N
2
O emissions from handling
manure, either directly from evaporation or through
ammonia; and emissions directly from the fuel
combustion in the manufacturing of mineral fertili-
zer and the use of diesel and other fuels mainly for
tractor use on the farm and transport of inputs and
products outside the farm. The impact from N
2
O
arising from microbiological processes in soil was
discussed, but was not included in the primary
results, as this is uncertain and varies significantly
with natural and agronomic conditions. Direct
emissions from agricultural soils, direct soil emis-
sions from animal production and N
2
O emissions
indirectly induced by agricultural activities were
analysed and included in the results.
Data sources
farm input and farm output
We used data on input quantities and output
quantities from model farms producing milk, meat,
grain and potatoes. Each model farm is a represen-
tative farm for different productions, farms sizes and
areas and is constructed based on calculations from
the Far m Accountancy Data Network (FADN) with
registered numbers from around 900 farms in Nor-
way. The aggregation into model farms is carried out
on an annual basis by the Norwegian Agricultural
Economics Research Institute and includes 27 dif-
ferent farm types (Reference farms), of which 15 are
on a national level and the remaining are subsets
representing different regions (Budsjettnemda for
jordbruket, 2007, 2008a, 2009). Data for model
farms and prices were averaged for 20052007 to
provide more robust inventory data and account for
price fluctuations of inputs and outputs, fluctuations
in yield due to weather, changes in production
composition and crop rotation.
Table I shows the model farms included in the
analysis. Organic cereal production was not covered
by any of the reference farms and had to be
modelled separately. An organic cereal production
Table I. Model farms used for analysis of GHG emissions in Norway.
Acronym Name Production system Size Number of farms as source
R1 Reference farm no. 1 Milk and beef meat 20 annual cows 341
R2 Reference farm no. 2 Grain 333 daa 88
R5 Reference farm no. 5 Pork and grain 46 sows, 357 daa grain 38
R7 Reference farm no. 7 Potatoes and grain 103 daa potatoes, 13
R8 Reference farm no. 8 Beef cattle/suckler cows 25 suckler cows 33
R10 Reference farm no. 10 Chicken and plant products 80,383 chicken slaughtered 15
R11 Reference farm no. 11 Organic milk and beef 20 annual cows 23
R14 Reference farm no. 14 Milk (the 40 largest) 40 annual cows 40
Ow Farm modelled by Gaffke
(2010)
6-year crop rotation with green
manure
Green manure, wheat, oat w/
under culture, oat/peas, rye w/
grass, green manure
Experiences from extension
service
Sources: Adapted from Budsjettnemda for Jordbruket (2007, 2008a, 2009) and Gaffke (2010).
338 K. Refsgaard et al.
Downloaded by [NILF] at 06:13 25 June 2013
model without animals was defined based on
information from the Agricultural Advisory Service
(Gaffke, 2010), which is located at Østlandet, the
area that has the largest area with organically grown
wheat in 2009 (Agricultural Authority, 2010). The
model farms represent the basis for comparisons of
food production in this study. Most Norwegian
farms have co-production and are not specialized
for only one product. This is reflected in the model
farms which are aggregated according to common
production mixes. This also implies that an average
Norwegian production of an individual food product
is composed of outputs from more than one model
farm. Average Norwegian production of individual
products is, therefore, represented by a weighted
average from the relevant model farms. Grain
production is, for example, composed of production
in model farms R2, R5, R7 and R10. The results
presented here are for average Norwegian produc-
tion, i.e. not for individual model farms.
The statistical accounts report only monetary
flows entering and exiting the farm gate, because
internal production and use not represented by a
purchase or a sale cannot be identified. Allocation
of farm operations is based purely on cost of
production. Economic allocation is used for dis-
tributing the inputs among the outputs. For exam-
ple, if a farm output represents 25% of total income
from sales (subsidies excluded), this output is
assigned 25% of each input to the farm and the
corresponding impacts from production of these
inputs. Consequently, some outputs are also as-
signed a share of inputs that are not used in their
production.
GHG-emission factors for direct and indirect
emissions from fertilization and husbandry
Direct and indirect emissions from fertilization and
husbandry were modelled using an adaptation of the
national GHG accounts for agriculture, with tier-2
resolution for most factors, following national re-
porting rules (Sandmo, 2009). Emissions were
normalized per animal, per mass of fertilizer N and
P applied and per volume of manure, with emission
factors reported elsewhere (Pettersen, 2010). Ferti-
lizer use and animal stocks are reported for each
reference farm, and were used directly in the
inventory model without adjustment for organic or
conventional feeds. We assume that organic and
conventional animals have the same direct and
indirect emissions, in other words, there was no
adjustment for different feeding practices or treat-
ment of manure.
Manufacturing of input factors
We used the Ecoinvent database version 2.1 for most
background processes, with some modifications for
the Norwegian context. Most notable adjustments
were for fertilizer production and feed-concentrate
production, which were modified according to Nor-
wegian production and use (see Appendix 1).
Land use
The land use included the area used for crop
production on the farm and the area used from
imported and domestically produced inputs.
Average yield data from 2005 to 2007 for reference
farms were used for calculating the land use for
production of grain, potatoes and fodder used as input
in dairy and meat production (Figure 3). The land use
for imported feedstuff was based on Food and
Agriculture Organization of the United Nations
(FAO) data from IAASTD (2009)as an estimate for
10 North American and European countries. For the
organically grown grain, yields were based on informa-
tion from the Agricultural Advisory Service (Gaffke,
2010).
Results
GHG emissions for food production at the farm gate
among production systems
Figure 1 shows the estimated emissions of CO
2
e per
kilogramme of the four food products and per Mcal
of dietary energy. There is a variation in GHG
emissions depending on the type of production
system, also known as type of reference farm.
For grain production, the average emission is 0.58
kg CO
2
e per kg, with a variation from 0.44 to 0.72 kg
CO
2
e per kg. The organic grain production is
calculated to 0.34 kg CO
2
e per kg. The GHG
emission from milk production has an average of
1.88 kg CO
2
e per kg. GHG emissions from meat
production vary heavily with the type of production
system. GHG emissions from beef produced in
combination with milk (R1 and R14) have 15.1 kg
CO
2
e per kg, while there is 34.0 kg CO
2
e per kg beef
when produced from suckler cows.
For all the products where organic production was
analysed, organic production was found to be more
GHG efficient than conventional production. The
average emissions of CO
2
e for conventional products
are 70% higher for beef, while only 14% higher for milk.
When comparing different types of food, however,
it is more relevant to use other functional units. The
energy content of food is considered most relevant,
although other nutritional aspects, variation and
taste are also relevant. Comparing the GHG emis-
Atmospheric emissions of greenhouse gases 339
Downloaded by [NILF] at 06:13 25 June 2013
sions measured per Mcal reduced the differences
between meat and milk (8.9 kg CO
2
e per Mcal for
conventional produced meat versus 4.1 kg CO
2
e per
Mcal for conventionally produced milk). Compared
to these animal products, grain only contributes
marginally (0.24 kg CO
2
e per Mcal for convention-
ally produced grain).
The composition of GHG emissions and land use
In Figure 2, we see the estimated average of GHG
emissions from the total annual Norwegian production
of potatoes, grain, milk and meat grown either con-
ventionally or organically and how it is composed. The
four products contribute 4.6 Mt of CO
2
eatfarmgate
and 3.6 Mt of CO
2
eatfarmgatefromorganic
2
Figure 1. GHG emissions for different food products at the farm gate in Norway (in kg CO
2
e/Mcal and kg CO
2
e/kg product).
Figure 2. The GHG emissions from the production of the national consumption of potatoes, grain, milk and meat in Norway.
340 K. Refsgaard et al.
Downloaded by [NILF] at 06:13 25 June 2013
production. We see that the animal products, milk and
beef, contribute more than 80% of the total GHG
emissions from these four products.
The largest source for GHG emissions from
both milk and beef is the direct emissions, in
other words, CH
4
and N
2
O from enteric digestion
and manure. Comparing organic and conventional
production of milk and meat shows similar results.
The difference arises mainly from the manufactur-
ing of and fertilizing with mineral fertilizer and
some imported feedstuff for conventionally pro-
duced milk and meat. Further imported soya pro-
teins for conventionally produced milk and meat
are substituted with Norwegian grain if organically
produced.
For the systems without animals, the highest
emission is from mineral fertilizer manufacturing
and use for the conventional grain and potato
production and from green manure for the organic
grain production.
The differences between GHG emissions from
different production systems in our analyses, such as
organic versus conventional and between different
systems for producing grain or beef, are related to
the substitution of some inputs such as mineral
fertilizer and chemical crop protection with land.
Within each type of production system, the land use
for animal products is much higher than that of plant
products, with around 1617 m
2
per Mcal meat
compared to around 1 m
2
per Mcal grain.
In Figure 3, we see that for milk, beef and grain
there is a higher land use associated with organic
production systems than with conventional produc-
tion systems and so is also the composition of land,
i.e. organic farming requires more land per produc-
tion output than conventional farming. Organically
produced grain has a 70% higher land use than
conventionally produced grain, on average, while
there is less than a 10% higher land use for beef that
is organically produced. In general, GHG emissions
per kg or Mcal are lower for organic products than
for conventional products, but land use is higher.
For the national consumption of milk protein
from 1,500,000 daa is imported while if organic
produced only protein from 400,000 daa is im-
ported. For conventionally produced meat protein
Figure 3. The land use for the production of the national consumption of potatoes, grain, milk and meat in Norway.
Atmospheric emissions of greenhouse gases 341
Downloaded by [NILF] at 06:13 25 June 2013
from 610,000 daa is imported compared to
140,000 daa if organically produced.
Discussion
LCA of food is a rapidly growing field and thus the
number of analyses of GHG emissions from food is
growing as well. Comparing results across studies
and countries is not straightforward. First, the
functional unit (i.e. kilogramme mass, calories or
kilogramme protein) varies, but more important is
that the studied systems differ in their delimitation of
system boundaries. Therefore, broad measures taken
from reviews based on secondary sources are difficult
to use when comparing GHG emissions from food.
For milk and beef, two of the most recent sources
are of particular interest. FAO (2010) is a compara-
tive analysis of dairy production across regions of the
world, while Weiss and Leip (2012) report a major
study of emissions from all kinds of livestock
production for EU-27. Without the contributions
from carbon sequestration, the emissions from milk
production in the EU-27 total around 1 kg CO
2
e/kg
milk (Weiss & Leip, 2012), while the FAO (2010)
result is around 1.4 kg CO
2
e/kg. Weiss and Leip
(2012) refer to 2128 kg CO
2
e/kg for beef; however,
there is no information on whether this is based on
suckler cow or dairy cow production, which, accord-
ing to our analyses, is of considerable importance. In
our analyses, beef based on dairy production is
around 15 kg CO
2
e/kg and based on suckler around
34 kg CO
2
e/kg. Nguyen et al. (2010) compared four
beef production systems in Europe three from
intensively reared dairy calves and one from suckler
herds showing that the production of 1 kg beef at
farm gate had lower emissions and lower land use
from dairy calves than from suckler herds (16.019.9
vs. 27.3 kg CO
2
e/kg and 16.522.7 vs. 42.9 m
2
per
year, respectively).
GHG emissions from beef meat produced in
combination with milk are only half that produced
from suckler cows, where the emissions are around
34 kg CO
2
e per kg beef meat. However, one can
(and we do) produce beef from suckler cows where
dairy production is difficult and inefficient and
where grass can form a higher proportion of the
diet. In Norway, grazing of large areas of remote and
unutilized grassland is only possible by suckler cows,
not dairy cows. In this way, grasslands in Norway
replace imported protein, decreasing the GHG
emissions from suckler cow beef compared to beef
from dairy cows.
For cereals, the results show the same level as in
many previous studies. In Denmark studies from
LCA Food 2003/2006, there is 1.13 kg CO
2
e per kg
flour, but they include the process downstream until
retail, while our analyses differ 0.420.72 kg CO
2
e
per kg grain, depending on the type of production
system. In Hille et al. (2012), an extensive literature
review of the GHG emissions from different food
products, and production systems can be found.
The CH
4
emissions from manure from animal
production are major emitters, but the amount of
emissions depends on the type of storage, spreading
and treatment. We used a constant for CH
4
losses
from enteric fermentation
3
and from animal manure
in stable and storage as well as for N
2
O from animal
manure for milk and beef production. Also, the loss
of N
2
O (indirectly through NH
2
) and directly from
manure management depends on the type of storage
and spreading. However, these losses have a much
lower contribution to direct emissions than CH
4
and
will not be further investigated here. Trømborg et al.
(2007) estimated the variability in enteric CH
4
emissions as a function of production intensity to
be 145.8 kg CH
4
emissions per 6000 kg milk per
year. Using these estimates, and assuming a linear
relationship between the yield and CH
4
emission, a
decrease in milk yield of 1000 kg per cow would
imply a 10% increase in CH
4
emission per kg milk.
This implies that the marginal increase in GHG
emissions from milk, if produced organically with a
lower production of 830 kg milk less per year than if
produced conventionally, would be about 0.1 Mt
CO
2
e for the total production of 1.5 Mt of milk.
4
This implies that even with a lower GHG efficiency
for the enteric fermentation our analyses still show a
lower emission from the organic production system.
The other large contributor to GHG emissions
from conventional production of all farm products,
which also accounts for much of the difference with
organic production, is the manufacturing and appli-
cation of mineral fertilizer. These two processes
contribute to roughly 20% of total emissions for
milk and about 70% of total emissions for grain. In
our calculations, the coefficients for GHG emissions
from fertilizer processing are based on the current
technology from Yara,
5
being nearly the most
energy-efficient technology available. This implies
that using technologies with less efficiency will
increase GHG emissions of conventional systems.
The fertilization coefficients are rough estimates,
however, and therefore more uncertain.
6
Changes in the soil-carbon pool have not been
considered in this study either for Norwegian or for
overseas crop production, for example we do not
consider the effects of the forest cut down to increase
soya production in Brazil. Riley (pers. comm. in
Trømborg et al., 2007) calculated CO
2
emissions
from arable land in eastern Norway to be about 2
tonnes CO
2
per hectare. The total loss from
arable crops can then be estimated for all cropland
342 K. Refsgaard et al.
Downloaded by [NILF] at 06:13 25 June 2013
(domestic and imported) used for the national
consumption of the four analysed products to be
about 1.0 Mt of CO
2
e for conventional production
and 1.1 Mt of CO
2
e for organic production, but then
the area with grass-fixing carbon is not taken into
consideration. Uhlen (1991) refer to Norwegian
trials showing that if grass is at least one-third of
the crop rotation this will avoid an annual emission
of around 300 kg C per ha. Therefore as Hoffmann
(2011) claims: ‘‘Giving higher importance to carbon
uptake by grassland through integrated crop and
livestock management (while biomass of forests
expands by 10% annually, that of grassland expands
by 150%)’’ may have considerable potential for
reduction of GHG emissions. Flysjo
¨et al. (2012)
estimated the GHG emissions in combined dairy
production considering the carbon emissions depen-
dent on different land use. They (Flysjo
¨et al. 2012)
argue that increasing per-cow milk production does
not necessarily reduce the GHG emissions per
kilogramme milk, when also considering the alter-
native production of the by-product beef because
often there is higher beef production per kilogramme
milk in a more extensive (organic) dairy system.
As Hospido et al. (2010) argue, the environmental
impacts from animal and vegetable products often
differ by a factor of 10. Our analyses confirm this
finding and show that plant products with high-
energy densities, such as cereals, have the lowest
GHG emission per Mcal while beef remains very
GHG intensive. In total for the four products
analysed in our study, milk and beef contribute
with more than 80% of the GHG emissions while
only providing around 30% of the energy.
Conclusion
LCA has been used to analyse the differences
between production systems and between different
types of food in terms of GHG emissions up to the
farm gate and including relevant upstream activities.
Results show that although direct emissions from
farm activities are the major emission sources for
GHG, upstream activities have a significant contri-
bution to for life cycle emissions from food produc-
tion. Of particular importance is fertilizer and animal
feed production. Conventional farming shows gen-
erally higher GHG emissions than organic farming,
at the expense of higher land use. GHG results
furthermore indicate that emissions intensities for
food products vary significantly depending on farm-
ing practice and farm type. Both beef and grain
production show significant variations, but average
national production intensities are comparable with
other studies.
Direct emissions from farm activities make up the
major part of the GHG emissions for both animal
and plant products. Improving manure management
and fertilization use has potential for achieving
substantial reductions in emission levels, and in
particular for animal products where GHG emis-
sions are comparatively the largest. More efficient
handling and spreading can give direct lower emis-
sions, as well as substitute and reduce the amount of
chemical fertilizers and thereby the indirect emis-
sions from its production. Impacts from upstream
production processes are important for overall im-
pacts from food production systems, and the fertilizer
production and feed concentrate production are key
contributors within the upstream production chain.
Our results are based on LCA in order to include
indirect emissions from upstream production. The
results cover products that make up about 50% of
Norwegian food consumption. Applying the emis-
sion intensities from this study results in 4.6 Mt of
CO
2
e, while Intergovernmental Panel on Climate
Change (IPCC) figures estimate a total of 5.5 million
tons of CO
2
e for all food production. This indicates
that including indirect emissions upstream from the
farm is important, although other factors such as
sector aggregation in statistics and uncertainties in
emissions intensities are also part of the explanation
for this finding.
The life cycle approach to assess GHG emissions
from food production provides insight into poten-
tial trade-offs and a wider perspective for compar-
ing products and farming practices. Combining
LCA with FADN data allows for comparing
and understanding differences between production
systems and associated effects on environmental
indicators.
Acknowledgements
This work was performed in the research project
‘‘Socio-economic and environmental impacts of
organic farming’’ (20082011), funded mainly by
The Research Council Norway and with additional
funding from Oikos, the County Governor Sør-
Trøndelag and KSL Matmerk. We thank Prof.
John M. Bryden for valuable comments to the
manuscript and John Hille for providing us with a
compilation of relevant references.
Notes
1. As in Nemecek et al. (2011), we define conventional farming in
this study as a mode of production not respecting the organic
farming rules. The agricultural system’s main goal is to achieve
high yields and a high economic output.
2. Although the production of potatoes is conventional.
Atmospheric emissions of greenhouse gases 343
Downloaded by [NILF] at 06:13 25 June 2013
3. One hundred and forty-three kilogramme CH
4
emissions/dairy
cow/year (Sundstøl & Mroz, 1988; Volden & Nes, 2006;
Sandmo, 2009; Pettersen, 2010).
4. The average annual milk production from conventional dairy
production is 830 kg higher than the organic dair y production
(see Figure 3).
5. Of about 1.5 kg CO
2
e (0.005 kg N
2
O) per kg N together with
about 3 kg CO
2
from use of gas in the hydrogen production; in
total, 4.5 kg CO
2
e per kg N (personal information from Yara).
6. We use 0.0269 kg N
2
O per kg mineral fertilizer N, which
includes the indirect contribution through evaporation to
ammoniac and nitrogen run-off and direct contribution from
N
2
ON (Pettersen, 2010). Trømborg et al. (2007) use 0.019
kg N
2
O per kg N.
References
Agricultural Authority (2010). Produksjon og omsetning av økolo-
giske landbruksvarer. Rapport for 2009. Rappor t 7/2010
[Production and marketing of organic agricultural products.
Report for 2009. Rapport 7/2010]. Oslo, Norway: Statens
landbruksforvaltning (SLF).
Badgley, C., Moghtader, J., Quintero, E., Zakem, E., Chappell,
M. J. & Aviles-Vazquez, K. (2007). Organic agriculture and
the global food supply. Renewable Agriculture and Food
Systems, 22, 86108.
Brentrup, F., Ku
¨sters, J., Kuhlmann, H. & Lammel, J. (2004).
Environmental impact assessment of agricultural production
systems using the life cycle assessment methodology I.
Theoretical concept of a LCA method tailored to crop
production. European Journal of Agronomy, 20, 247264.
Budsjettnemda for jordbruket (2007). Referansebruksberegninger
2005. Beregninger basert pa˚driftsgranskninga, avgitt juni
2007 [Model farm calculations. Results based on the survey
of account statistics for agriculture and forestry]. Norway:
NILF.
Budsjettnemda for jordbruket (2008a). Referansebruksberegnin-
ger 2006. Beregninger basert pa˚driftsgranskninga, avgitt juni
2008 [Model farm calculations. Results based on the survey
of account statistics for agriculture and forestry]. Norway:
NILF.
Budsjettnemda for jordbruket (2008b). Totalkalkylen for jord-
bruket. Jordbrukets totalregnskap 2006 og 2007 og budsjett
2008 [Total account for the agricultural sector in Norway,
2006 and 2007, and budget 2008]. Norway: NILF.
Budsjettnemda for jordbruket (2009). Referansebruksberegninger
2007. Beregninger basert pa˚driftsgranskninga, avgitt juni
2009 [Model farm calculations. Results based on the survey
of account statistics for agriculture and forestry]. Norway:
NILF.
Bye, A. S., Aarstad, P. A., Løvberget A. I., Berge, G, & Hoem, B.
(2010). Jordbruk og miljø: Tilstand og utvikling [Agriculture
and the environment. Condition and development] Rappor-
ter, Oslo-Kongsvinger: Statistics Norway.
Carlsson-Kanyama, A., Ekstrom, M. P. & Shanathan, H. (2003).
Food and life cycle energy inputs: Consequences of diet and
ways to increase efciency. Ecological Economics, 44, 293
307.
Coley, D., Howard, M. & Winter, M. (2009). Local food, food
miles and carbon emissions: A comparison of farm shop and
mass distribution approaches. Food Policy, 34, 150155.
EFMA (2000). Production of NPK Fertilizers by the Nitrophosphate
Route. Booklet no. 7 (Brussels: European Fertilizers Manu-
facturers Association).
Eurostat (2010). Using Official Statistics to Calculate Greenhouse
Gas Emissions. A Statistical Guide. European Union. (Luxem-
bourg: Publications Ofce of the European Union). ISBN
978-92-79-14487-5. doi:10.2785/3842. Cat. No. KS-31-09-
272-EN-C 2010 edition.
FAO (2010). Food and agriculture statistics global outlook.
Accessed 27 October 2011, available at: http://faostat.fao.
org/Portals/_Faostat/documents/pdf/world.pdf.
Flysjo
¨, A., Cederberg, C., Henriksson, M. & Ledgard, S. (2012).
The interaction between milk and beef production and
emissions from land use change critical considerations in
life cycle assessment and carbon footprint studies of milk.
Journal of Cleaner Production, 28, 134142.
Gaffke, H. (2010). Organic Crop Rotation and Yields (Heggenes,
Norway: Løken forskingsstasjon).
Hall, C., Cleveland, A. S., Cutler, J. & Kaufmann, R. K. (1992).
Energy and Resource Quality: The Ecology of the Economic
Process (Colorado: University Press of Colorado).
Hertwich, E. (2005). Life cycle approaches to sustainable con-
sumption: A critical review. Environmental Science and
Technology, 13, 46734684.
Hille, J., Solli, C., Refsgaard, K., Krokann, K. & Berglann, H.
(2012). Environmental and climate analysis for the Norwegian
agriculture and food sector and assessment of actions, Working
Paper 2012-1. Norwegian Agricultural Economics Research
Institute (NILF).
Hoffmann, U. (2011). Sustainable Farming Can Cost-effectively
Cool Down the Earth. United Nations Conference on Trade and
Development: FAO/OECD Expert Meeting on Greening the
Economy with Agriculture (Paris: UNCTAD Secretariat).
Hospido, A., Berlin, J., Davis, J. & Sonesson, U. (2010). A review
of methodological issues affecting LCA of novel food
products. The International Jour nal of Life Cycle Assessment,
15, 4452.
IAASTD (2009). Agriculture at a crossroads: Global report. In B.
D. McIntyre, H. R. Herren, J. Wakhungu & R. T. Watson
(eds.) International Assessment of Agricultural Knowledge,
Science and Technology for Development. Global Report.
IAASTD. Washington, DC: Island Press.
ISO (1997). Environmental Management Life Cycle Assessment
Principles and Framework (Geneva: International Organiza-
tion for Standardization [ISO]).
ISO (2006). Environmental Management Life Cycle Assessment
Principles and Framework (Geneva: International Organiza-
tion for Standardization [ISO]).
Klif (2010). Klif [Norwegian Climate and Pollution Agency],
Norske utslipp [Norwegian emissions inventory]. Accessed
2010, available at: http://www.norskeutslipp.no/Templates/
NorskeUtslipp/Pages/sektor.aspx?id=1106&epslanguage=en
Lampkin, N. H. (2003). From conversion payments to integrated
action plans in the European Union. In OECD, (ed.) Organic
Agriculture: Sustainability, Markets and Policies (Wallingford:
CABI Publishing), pp. 313328.
Mondelaers, K., Aertsens, J. & van Huylenbroeck, G. (2009). A
meta-analysis of the differences in environmental impacts
between organic and conventional farming. British Food
Journal, 111, 10981119.
Nemecek, T., Dubois, D., Hugenin-Elie, O. & Gaillard, G.
(2011). Life cycle assessment of Swiss farming systems: I.
Integrated and organic farming. Agr icultural Systems, 104,
217232.
Neset, T.-S. S. & Lohm, U. (2005). Spatial imprint of food
consumption: A historical analysis for Sweden, 18702000.
Human Ecology, 33, 565580.
Nguyen, T. L. T., Hermansen, J. E. & Mogensen, L. (2010).
Environmental consequences of different beef production
systems in the EU. Journal of Cleaner Production, 18, 756
766.
344 K. Refsgaard et al.
Downloaded by [NILF] at 06:13 25 June 2013
Pettersen, J. (2010). Direkteutslipp fra landbruket i LCA [Direct
emissions from agriculture in LCA].MiSA Report No. 10.
Trondheim, Norway [MiSA report 10/2010. Project report.
Trondheim, Norway].
Pimentel, D., Hepperly, P. & Seidel, R. (2005). Environmental,
energetic, and economic comparisons of organic and con-
ventional farming systems. Bioscience, 55, 573583.
Refsgaard, K., Halberg, N. & Kristensen, E. S. (1998). Energy
utilization in crop and dairy production in organic and
conventional livestock production systems. Agricultural
Systems, 57, 599630.
Sandmo, T. (2009). The Norwegian Emission Inventory 2009.
Documentation of Methodologies for Estimating Emissions of
Greenhouse Gases and Long-range Transboundary Air Pollutants
(Oslo: Statistics Norway).
Steinshamn, H. (2010). Feed Concentrate for Organic Farming.
Bioforsk, Tingvoll, Norway: Norwegian Institute for Agri-
cultural and Environmental Research.
Sundstøl, F. & Mroz, Z. (1988). Utskillelse av nitrogen og fosfor i
gjødsel og urin fra husdyr i Norge. SEFO Rapport nr. 4
[Nitrogen and phosphorus in manure and urine from
domestic animals in Norway], Landbrukspolitikk og miljø-
forvaltning [Agricultural policy and environmental manage-
ment]. A
˚s: Agricultural University of Norway.
Svoldal, B. T. (2005). Bærekraftig matproduksjon med miner-
algjødsel [Sustainable food production with mineral fertili-
zer]. YARA Norge AS. Accessed 2010, available at: http://
www.bioforsk.no/ikbViewer/Content/48086/Yara-PPT-H-B-
rekraftig-matprodu.pdf.
Trømborg, E., Nielsen, A. & Hoen, H. F. (2007). Klimagasser og
bioenergi fra landbruket kunnskapsstatus og forskningsbe-
hov [Climate gases and bioenergy from agriculture status
of knowledge and need for research]. INA fagrapport 11. A
˚s,
Norway: University of Life Sciences, INA.
Tukker, A., Eder, P. & Suh, S. (2006). Environmental impacts of
products: Policy relevant information and data challenges.
Journal of Industrial Ecology, 10, 183198.
Tukker, A. & Jansen, B. (2006). Environmental impacts of
products: A detailed review of studies. Journal of Industrial
Ecology, 10, 159182.
Uhlen, G. (1991). Long-term effects of fer tilizers, manure, straw
and crop rotation on Total-N and Total-C in soil. Acta
Agricultural Scandinavian, 41, 119 127.
Van der Werf, H. M. G., Kanyarushoki, C. & Corson, M. S.
(2009). An operational method for the evaluation of resource
use and environmental impacts of dairy farms by life cycle
assessment. Journal of Environmental Management, 90, 3643
3652.
Volden, H. & Nes S. K. (2006). Methane emissions from enteric
fermentation in Norwegians cattle and sheep population.
Method description. Appendix H. pp. 181189. Vol.
Reports 2006/30, in The Norwegian Emission Inventory
2006. Documentation of methodologies for estimating
emissions of greenhouse gases and long-range transboundary
air pollutants, by B. Hoem. Statistisk sentralbyra˚[Statistic
Norway], 2006.
Weidema, B., Wesnæs, M., Hermansen, J., Kristensen, T.,
Halberg, N., Eder, P. & Delgado, L. (2008). Environmental
improvement potentials of meat and dairy products. JRC
Scientic and Technical Reports. EUR 23491 EN. Luxem-
bourg: Ofce for Ofcial Publications of the European
Communities.
Weiss, F. & Leip, A. (2012). Greenhouse gas emissions from the
EU livestock sector: A life cycle assessment carried out with
the CAPRI model. Agriculture, Ecosystems and the
Environment, 149, 124134.
Appendix 1: Data sources and their assumptions.
Modelling of inputs to production
This section describes the key assumptions on how various
inputs from the model farm statistics are modelled in the
inventory.
Seed grain
Input of seed grain to model farms is reported as one group,
without a split between grain types. Input of seed grain is
distributed on the three cereal types that are reported: barley, grain
and oat, and a category termed ‘‘other grain’’ for the remaining
fractions. Distribution is performed according to production
volumes. Production volumes are calculated from reported income
from output of the individual cereal types in the model farms and
converted to physical units through price information.
Fertilizer production and use
The major share of fertilizer use in Norwegian agriculture is
produced domestically by Yara (Svoldal, 2005). Personal com-
munication confirms that its main production facility in Pors-
grunn, Norway, produces fertilizer in accordance with the Best
Available Techniques (BAT)-document issued by the European
Fertilizer Manufacturers Association (EFMA) for production of
NPK fertilizers by the nitrophosphate route (EFMA, 2000).
Ecoinvent processes for production of N, P and K are modified
to represent the specific mix for the fossil part of the energy use for
the plant in Porsgrunn, averaged for the years 20052007 (Klif ,
2010). Fossil energy use is reported as 70% natural gas, 20% light
fuel oil and 10% heavy fuel oil. The composition, however,
changes considerably between years. The modification to specific
fossil-fuel use has a small effect ( B5%) relative to the generic
database-production process.
The use of fertilizers varies according to farm type and
products, and access to manure. Cereal production requires a
different mix of nutrients than, for example, potatoes or cattle
products. NPK fertilizer mixes are used for all fertilizer use, but
with varying composition of the three elements, according to farm
type and production.
All fertilizer use on cereal farms is modelled with NPK 21-4-10,
with numbers indicating the shares of the N, P and K. Model
farms producing mainly cattle products are modelled using a mix
of NPK 18-3-15, NPK 21-4-10 and NPK 22-2-12. This
composition is based on annual statistics about the trade within
the agricultural sector (Budsjettnemda for jordbruket, 2008b,
2009), and is 21%, 57% and 22%, respectively, for the three
fertilizer mixes. Fertilizer use of potato production is modelled
with NPK 11-5-17. Model farms producing a mix of products
with different fertilizer profiles are modelled with fertilizer mixes
according to the product output composition.
Feed concentrates
Feed concentrate for cattle is composed of various nutrient
sources, and there are several types of feed concentrate mixes
being sold. Feed concentrate for conventional farm production is
modelled in this study, and is based on information from the
manufacturer of the two most widely sold formulas in Norway.
However, while the formulas relate to a certain level of nutrients,
which products go into the production of a specific formula to
provide the given nutrient levels varies both between and within
years. Variations are largely due to price fluctuations and avail-
ability of the commodities that are used in the formulas. Feed
concentrate compositions for the two selected types are collected
and averaged for three points in time in 2007/2008. The
composition is not very different with regard to which ingredients
make up the main part of the mix, but there are considerable
variations for some of the less important ingredients.
Atmospheric emissions of greenhouse gases 345
Downloaded by [NILF] at 06:13 25 June 2013
Feed concentrate for organic farming is modelled according to
formulas from Felleskjøpet and Steinshamn (2010), and personal
information from Bioforsk økologisk. Imported ingredients to the
feed concentrate mixes are modelled with Ecoinvent processes,
while input of feed grain from barley, oat and grain is assumed to
be produced domestically and is modelled as an input from the
cereal model farms. The results are based on a feeding intensity,
with around 4.6 kg protein per kg meat from dairy cows and 3.2
kg protein per kg meat from beef cows.
Energy use
Expenses on fuel use for the model farms are all assumed to
be diesel. Fuel production is modelled with Ecoinvent data.
Combustion emissions from farm machinery are modelled based
on figures from Statistics Norway for combustion of auto diesel
from stationary and mobile machinery.
Energy use from fuel and electricity is provided in the accounts for
the model farms.
Electricity use is modelled as a Norwegian consumption
mix, including imports. As described earlier, organic cereal
production is not included in the model farm accounts, but is
modelled separately for this study. Energy use for organic cereal
production is estimated based on a previous study (Refsgaard
et al., 1998). Fuel use per production output is similar for
conventional and organic, while electricity use is approximately
40% lower.
346 K. Refsgaard et al.
Downloaded by [NILF] at 06:13 25 June 2013
Technical Report
Full-text available
De nya kostråd som Livsmedelsverket publicerade i april 2015 tog även hänsyn till miljöaspekter, förutom närings- och hälsoaspekter. Den här rapporten sammanställer skillnaderna i miljöpåverkan mellan ekologiskt och konventionellt producerade livsmedel och är ett kunskapsunderlag som Livsmedelsverket kan använda för att belysa frågan om det finns några livsmedelsgrupper där ekologiskt alternativt konventionellt bör lyftas fram. Arbetet har fokuserat på ett antal större livsmedelsgrupper och prioriterat studier som har använt livscykelanalys, LCA. Den senare är en miljösystemanalysmetod som kartlägger den potentiella miljöpåverkan en produkt ger upphov till under sin livscykel, från ”vaggan till graven” (eller en väl avgränsad del av livscykeln, till exempel primärproduktionen). Projektet har genomförts i två delar. Del 1 består av en sammanställning av svenska rapporter från bland andra SIK och SLU, där skillnader i miljöpåverkan mellan ekologiskt och konventionellt producerade livsmedel har studerats. Del 2 består av en sammanställning av skillnader i miljöpåverkan (klimatpåverkan, övergödning, försurning, ekotoxicitet, energianvändning och markanvändning) mellan ekologiskt och konventionellt producerade livsmedel och baseras på kvantitativa data från 57 vetenskapligt publicerade studier. Samtliga resultat uttrycks i relation till kilo produkt (tabell 3 i rapporten). För miljöaspekter som är relevanta från ett lokalt eller regionalt miljöperspektiv (övergödning, försurning och ekotoxicitet), uttrycks resultaten även i relation till brukad areal (per hektar) (tabell 4 i rapporten). Några miljöaspekter som det i dagsläget saknas LCA-metodik för diskuteras kortfattat, nämligen biologisk mångfald, ekosystemtjänster och användning av fosfor. Dessutom ges en kort översikt över de viktigaste resultaten från tre tidigare publicerade kunskapssammanställningar på området. Resultatet från del 1 visar att det är svårt att göra rättvisa jämförelser på produktnivå mellan ekologisk och konventionell produktion eftersom växtföljder och produktionssystem oftast är olika och de båda systemen ofta inte producerar samma livsmedelsprodukter. Variationen mellan gårdar inom respektive system är ofta mycket stor och ibland större än mellan produktionssystemen. Resultaten beror också på om jämförelsen görs per hektar eller per kilo produkt. Om skörden är lägre kan en lägre miljöpåverkan per hektar innebära en högre påverkan per kilo produkt. Rapporterna som har studerats i del 1 har delats upp i två kategorier. Den första kategorin är rapporter som behandlar hur olika odlingssystem inverkar på odlingslandskapet och på möjligheten att uppfylla de nationella miljökvalitetsmålen. Den faktor som oftast framkommer som positiv för den ekologiska odlingen är att kemiska växtskyddsmedel i princip inte används, vilket minskar risken för spridning av naturfrämmande ämnen i miljön. Strukturen på det ekologiska jordbruket är också positiv ur flera aspekter, till exempel för bevarande av biologisk mångfald och för det nationella miljökvalitetsmålet ”Ett rikt odlingslandskap”. Effekten av detta är störst i slättbygderna, i storskaliga landskap. När det gäller klimatpåverkan och övergödning framhålls inte ekologisk odling som bättre än konventionell. Den andra kategorin rapporter i del 1 har jämfört livsmedel på produktnivå med hjälp av LCA och redovisar resultatet som miljöpåverkan per kilo eller liter produkt. Olika typer av kött, mejeriprodukter och vegetabilier har studerats. De flesta har producerats i Sverige men några är importerade, som banan och kaffe. Eftersom klimatfrågan har varit i fokus de senaste årtiondena finns det flest jämförelser med avseende på klimatavtryck, medan andra miljöaspekter som går att beräkna med LCA, till exempel övergödning, försurning, mark-, energi- och annan resursanvändning, endast ingår i några rapporter. För klimatavtrycket är de ekologiskt odlade produkterna i de flesta fall i nivå med de konventionellt odlade, eller bättre, medan de ekologiska produkterna är sämre än de konventionella när det gäller övergödande ämnen, med undantag för mejeriprodukter där de är bättre eller lika bra. Ett vanligt resultat från livscykelanalyser är att markanvändningen är högre för ekologisk produktion eftersom skördenivån oftast är lägre. I ekologisk odling används inte mineralgödsel, vilket innebär att förbrukningen av fosfor, som är en ändlig resurs, är lägre. Resultaten från del 2 visar att med avseende på klimatpåverkan och energianvändning per kilo produkt är resultatet i huvudsak otydligt. I sex av nio livsmedelsgrupper går det inte att urskilja vilket produktionssystem som presterar bäst. Med avseende på övergödningspotential i relation till kilo produkt är resultatet blandat. Med avseende på försurningspotential i relation till kilo produkt har konventionellt producerade livsmedel generellt sett en lägre påverkan. I relation till brukad areal har däremot ekologiska produkter i huvudsak både lägre övergödnings- och försurningspotential. Dessa resultat ska tolkas med stor försiktighet och inte generaliseras på grund av att: 1) olika studier har använt olika analysmetoder med avseende på exempelvis systemgränser och allokering, 2) antalet studier är för få för att det ska gå att dra några säkra slutsatser, 3) orsakerna till de observerade skillnaderna är inte kända och 4) endast ett begränsat antal miljöaspekter har studerats. Sammanställningen i del 2 visar även att ekologiskt producerade livsmedel generellt sett ger upphov till en lägre ekotoxisk miljöpåverkan, men i regel kräver mer mark än konventionellt producerade livsmedel. Dessa resultat betraktas däremot som tämligen säkra eftersom de stämmer överens med resultat från andra publicerade kunskapssammanställningar. Enligt de senare har ekologisk livsmedelsproduktion i genomsnitt cirka 20–25 procent lägre skördar än konventionell livsmedelsproduktion, främjar i de flesta fall den biologiska mångfalden mer, eller har en mindre skadlig effekt på den biologiska mångfalden än konventionellt jordbruk. Baserat på sammanställningarna i del 1 och del 2 går det sammanfattningsvis inte att säga att det ena produktionssystemet är överlägset det andra från miljösynpunkt för någon livsmedelsgrupp, då miljöpåverkan jämförs i relation till kilo produkt. Det beror på att resultaten från olika studier pekar åt olika håll för enskilda miljöaspekter, eller att resultatet för olika miljöaspekter pekar åt olika håll för samma livsmedelsgrupp. Det kan också bero på att de båda systemen presterar likvärdigt. I relation till brukad areal är resultatet däremot i huvudsak till det ekologiska jordbrukets fördel. Ekologiskt jordbruk presterar generellt sett bättre än konventionellt jordbruk från ett lokalt och regionalt miljöperspektiv, men det har inte nödvändigtvis en lägre miljöpåverkan per producerad enhet. Båda perspektiven är relevanta och ger kompletterande information. Vilket perspektiv som är mest relevant i en given situation beror på vilken eller vilka miljöaspekter som prioriteras, samt på de lokala och regionala förhållanden där livsmedelsproduktionen sker, exempelvis tillgången på mark och den bakomliggande miljöbelastningen. För att kunna ta ställning till om en ekologisk produkt är fördelaktig från miljösynpunkt måste målbilden vara bestämd samt vilka miljömål som ska prioriteras. Till detta kommer alla övriga aspekter som kan ha betydelse, som till exempel djurvälfärd, sociala aspekter, arbetstillfällen, risker vid hantering av växtskyddsmedel och risken för resthalter av dessa i livsmedel.
Technical Report
Full-text available
European-funded food waste prevention project FUSIONS has developed the criteria and a baseline assessment of environmental and socio-economic impacts of food waste. Although this research has shown that there are still some major data gaps for a more comprehensive assessment, these findings can serve as documentation for the existing knowledge base and provide new information on how to proceed towards socio-economic and environmental assessment of the impacts of food waste.
Article
Full-text available
Purpose: This paper aims at comparing the environmental impacts of organic and conventional farming and linking these to differences in management practises. The studied environmental impacts are related to land use efficiency, organic matter content in the soil, nitrate and phosphate leaching to the water system, greenhouse gas emissions and biodiversity. Design / Methodology / Approach: The theoretic framework uses the driver-state-response framework and literature data were analysed using meta-analysis methodology. Meta-analysis is the statistical analysis of multiple study results. Data were obtained by screening peer reviewed literature. Findings: From our meta-analysis we can conclude that soils in organic farming systems have on average a higher content of organic matter. We can also conclude that organic farming contributes positively to agro-biodiversity (breeds used by the farmers) and natural biodiversity (wild life). Concerning the impact of the organic farming system on nitrate and phosphorous leaching and greenhouse gas emissions the result of our analysis is not that straightforward. When expressed per production area organic farming scores better than conventional farming for these items. However, given the lower land use efficiency of organic farming in developed countries, this positive effect expressed per unit product is less pronounced or not present at all. Original value: Given the recent growth of organic farming and the general perception that organic farming is more environment friendly than its conventional counterpart, it is interesting to explore whether it meets the alleged benefits. By combining several studies in one analysis, the technique of meta-analysis is powerful and may allow to generate more nuanced findings and to generalise findings to a wider scope.
Article
Full-text available
Various organic technologies have been utilized for about 6000 years to make agriculture sustainable while conserving soil, water, energy, and biological resources. Among the benefits of organic technologies are higher soil organic matter and nitrogen, lower fossil energy inputs, yields similar to those of conventional systems, and conservation of soil moisture and water resources (especially advantageous under drought conditions). Conventional agriculture can be made more sustainable and ecologically sound by adopting some traditional organic farming technologies.
Book
Full-text available
This report first presents a systematic overview of the life cycle of meat and dairy products and their environmental impacts, covering the full food chain. It goes on to provide a comprehensive analysis of the improvement options that allow reducing the environmental impacts throughout the life cycle. Finally, the report assesses the different options regarding their feasibility as well as their potential environmental and socioeconomic benefits and costs. The report focuses on improvement options in three main areas: • Household improvements, mainly to reduce food losses (wastage) and to reduce car use for shopping; • Agricultural improvements, mainly to reduce water and air emissions (in particular nitrate, ammonia and methane) and land requirements; • Power savings in farming, food industry, retail, catering, and for household appliances. The study presents the consequences that the adoption of these options might have on a broad range of different environmental issues, including global warming, eutrophication, respiratory health impacts, etc. It shows that when all environmental improvement potentials are taken together, the aggregated environmental impacts (external costs) of meat and dairy products may be reduced by about 20 %. The study has also quantified the economic costs and benefits of implementing the different options.
Article
Full-text available
This study presents detailed product-based net emissions of main livestock products (meat, milk and eggs) at national level for the whole EU-27 according to a cradle-to-gate life-cycle assessment, including emissions from land use and land use change (LULUC). Calculations were done with the CAPRI model and the covered gases are CH4, N2O and CO2. Total GHG fluxes of European livestock production amount to 623–852 Mt CO2-equiv., 182–238 Mt CO2-equiv. (28–29%) are from beef production, 184–240 Mt CO2-equiv. (28–30%) from cow milk production and 153–226 Mt CO2-equiv. (25–27%) from pork production. According to IPCC classifications, 38–52% of total net emissions are created in the agricultural sector, 17–24% in the energy and industrial sectors. 12–16% Mt CO2-equiv. are related to land use (CO2 fluxes from cultivation of organic soils and reduced carbon sequestration compared to natural grassland) and 9–33% to land use change, mainly due to feed imports. The results suggest that for effective reduction of GHG emissions from livestock production, fluxes occurring outside the agricultural sector need to be taken into account. Reduction targets should address both the production side as defined by IPCC sectors and the consumption side. An LCA assessment as presented here could be a basis for such efforts.
Article
Full-text available
Purpose: This paper aims at comparing the environmental impacts of organic and conventional farming and linking these to differences in management practises. The studied environmental impacts are related to land use efficiency, organic matter content in the soil, nitrate and phosphate leaching to the water system, greenhouse gas emissions and biodiversity. Design / Methodology / Approach: The theoretic framework uses the driver-state-response framework and literature data were analysed using meta-analysis methodology. Meta-analysis is the statistical analysis of multiple study results. Data were obtained by screening peer reviewed literature. Findings: From our meta-analysis we can conclude that soils in organic farming systems have on average a higher content of organic matter. We can also conclude that organic farming contributes positively to agro-biodiversity (breeds used by the farmers) and natural biodiversity (wild life). Concerning the impact of the organic farming system on nitrate and phosphorous leaching and greenhouse gas emissions the result of our analysis is not that straightforward. When expressed per production area organic farming scores better than conventional farming for these items. However, given the lower land use efficiency of organic farming in developed countries, this positive effect expressed per unit product is less pronounced or not present at all. Original value: Given the recent growth of organic farming and the general perception that organic farming is more environment friendly than its conventional counterpart, it is interesting to explore whether it meets the alleged benefits. By combining several studies in one analysis, the technique of meta-analysis is powerful and may allow to generate more nuanced findings and to generalise findings to a wider scope.
Article
Full-text available
Various organic technologies have been utilized for about 6000 years to make agriculture sustainable while conserving soil, water, energy, and biological resources. Among the benefits of organic technologies are higher soil organic matter and nitrogen, lower fossil energy inputs, yields similar to those of conventional systems, and conservation of soil moisture and water resources (especially advantageous under drought conditions). Conventional agriculture can be made more sustainable and ecologically sound by adopting some traditional organic farming technologies.
Article
Full-text available
The principal objections to the proposition that organic agriculture can contribute significantly to the global food supply are low yields and insufficient quantities of organically acceptable fertilizers. We evaluated the universality of both claims. For the first claim, we compared yields of organic versus conventional or low-intensive food production for a global dataset of 293 examples and estimated the average yield ratio (organic:non-organic) of different food categories for the developed and the developing world. For most food categories, the average yield ratio was slightly <1.0 for studies in the developed world and >1.0 for studies in the developing world. With the average yield ratios, we modeled the global food supply that could be grown organically on the current agricultural land base. Model estimates indicate that organic methods could produce enough food on a global per capita basis to sustain the current human population, and potentially an even larger population, without increasing the agricultural land base. We also evaluated the amount of nitrogen potentially available from fixation by leguminous cover crops used as fertilizer. Data from temperate and tropical agroecosystems suggest that leguminous cover crops could fix enough nitrogen to replace the amount of synthetic fertilizer currently in use. These results indicate that organic agriculture has the potential to contribute quite substantially to the global food supply, while reducing the detrimental environmental impacts of conventional agriculture. Evaluation and review of this paper have raised important issues about crop rotations under organic versus conventional agriculture and the reliability of grey-literature sources. An ongoing dialogue on these subjects can be found in the Forum editorial of this issue.
Article
Full-text available
Environmental effects of economic activities are ultimately driven by consumption, via impacts of the production, use, and waste management phases of products and services ultimately consumed. Integrated product policy (IPP) addressing the life-cycle impacts of products forms an innovative new generation of environmental policy. Yet this policy requires insight into the final consumption expenditures and related products that have the greatest life-cycle environmental impacts. This review article brings together the conclusions of 11 studies that analyze the life-cycle impacts of total societal consumption and the relative importance of different final consumption categories. This review addresses in general studies that were included in the project Environmental Impacts of Products (EIPRO) of the European Union (EU), which form the basis of this special issue. Unlike most studies done in the past 25 years on similar topics, the studies reviewed here covered a broad set of environmental impacts beyond just energy use or carbon dioxide (CO2) emissions. The studies differed greatly in basic approach (extrapolating LCA data to impacts of consumption categories versus approaches based on environmentally extended input-output (EEIO) tables), geographical region, disaggregation of final demand, data inventory used, and method of impact assessment. Nevertheless, across all studies a limited number of priorities emerged. The three main priorities, housing, transport, and food, are responsible for 70% of the environmental impacts in most categories, although covering only 55% of the final expenditure in the 25 countries that currently make up the EU. At a more detailed level, priorities are car and most probably air travel within transport, meat and dairy within food, and building structures, heating, and (electrical) energy-using products within housing. Expenditures on clothing, communication, health care, and education are considerably less important. Given the very different approaches followed in each of the sources reviewed, this result hence must be regarded as extremely robust. Recommendations are given to harmonize and improve the methodological approaches of such analyses, for instance, with regard to modeling of imports, inclusion of capital goods, and making an explicit distinction between household and government expenditure.
Article
Two most critical factors to address in environmental system analysis of future milk production are 1) the link between milk and beef production, and 2) the competition for land, possibly leading to land use change (LUC) with greenhouse gas (GHG) emissions and loss of biodiversity as important implications. Different methodological approaches concerning these factors, in studies on environmental impacts of dairy production, sometimes lead to contradictory results.Increasing milk yield per cow is often one of the solutions discussed in order to reduce GHG emissions from milk production. However, when also accounting for other systems affected (e.g. beef production) it is not certain that an increase in milk yield per cow leads to a reduction in total GHG emissions per kg milk. In the present study the correlation between carbon footprint (CF) of milk and the amount of milk delivered per cow is investigated for 23 dairy farms (both organic and conventional) in Sweden. Use of a fixed allocation factor of 90% (based on economic value) indicates a reduction in CF with increased milk yield, while no correlation can be noted when system expansion is applied. The average CF for two groups of farms, organic and high yielding conventional, is also calculated. When conducting system expansion the CF is somewhat lower for the organic farms (which have a lower milk yield per cow, but more meat per kg milk), but when a 90% allocation factor is used, the CF is somewhat higher for the organic farms compared to the high yielding conventional farms. In analysis of future strategies for milk production, it is suggested that system expansion should be applied, in order to also account for environmental impacts from affected systems. Thus, scenarios for milk and meat production should be analysed in an integrated approach in order to reduce total emissions from the livestock sector.How to account for emissions from LUC is highly debated and there is no current shared consensus. Different LUC methods result in significantly different results. In this study, four different LUC methods are applied, using data for organic milk production and high yielding conventional milk production systems in Sweden. Depending on which LUC method was applied, the organic system showed about 50% higher or 40% lower CF compared to the conventional high yielding system. Thus, when reporting CF numbers, it is important to report LUC-factors separately and clearly explain the underlying assumptions, since the method of accounting for LUC can drastically change the results.
Article
In long-term experiments it was found that different crop rotations, applications of farm manure and fertilizer-N, and annually ploughed-in cereal straw have an effect on the total-N and total-C contents of the topsoil. C:N ratios remained unaffected by different crop rotations in a 31-year experiment on loam soil, but a long-term increase in N in relation to C seems to have occurred. The residual carbon in the soil amounted to approximately 7% of the C-additions in straw applied annually for 21, 23 and 31 years in experiments on loam and clay loam. Increase in total-N in the soil was roughly 30–40% of the nitrogen added in straw in two of the three field experiments. In an 8-year field lysimeter experiment the mineralization rate of the soil nitrogen corresponded to a half-life breakdown of 55 and 43 years for unfertilized cereal and fallowed plots, respectively. In five experiments with 8–50 years of N-additions, a half-life period of 15 years of nitrogen not utilized or lost in the year of application is calculated from the residual N found in the soil. The nitrogen source was mostly calcium nitrate. Somewhat higher residual N percentages in the soil with N added in farm manure in the same experiments, may largely be due to the much lower crop utilization percentages for farm manure nitrogen in the first year.