Livestock greenhouse gas emissions and mitigation potential in Europe

Abstract

The livestock sector contributes considerably to global greenhouse gas emissions (GHG). Here, for the year 2007 we examined GHG emissions in the EU27 livestock sector and estimated GHG emissions from production and consumption of livestock products; including imports, exports and wastage. We also reviewed available mitigation options and estimated their potential. The focus of this review is on the beef and dairy sector since these contribute 60% of all livestock production emissions. Particular attention is paid to the role of land use and land use change (LULUC) and carbon sequestration in grasslands. GHG emissions of all livestock products amount to between 630 and 863 Mt CO2 e, or 12-17% of total EU27 GHG emissions in 2007. The highest emissions aside from production, originate from LULUC, followed by emissions from wasted food. The total GHG mitigation potential from the livestock sector in Europe is between 101 and 377 Mt CO2 e equivalent to between 12 and 61% of total EU27 livestock sector emissions in 2007. A reduction in food waste and consumption of livestock products linked with reduced production, are the most effective mitigation options, and if encouraged, would also deliver environmental and human health benefits. Production of beef and dairy on grassland, as opposed to intensive grain fed production, can be associated with a reduction in GHG emissions depending on actual LULUC emissions. This could be promoted on rough grazing land where appropriate.

Full text

REVIEW
Livestock greenhouse gas emissions and mitigation
potential in Europe
JESSICA BELLARBY*, REYES TIRADO†,ADRIANLEIP‡,FRANZWEISS‡, JAN PETER
LESSCHEN§and PETE SMI T H *
*Institute of Biological and Environmental Sciences, School of Biological Sciences, University of Aberdeen, Aberdeen, AB24 3UU,
UK, †Greenpeace Research Laboratories, School of Biosciences, University of Exeter, Exeter, EX4 4RN, UK, ‡European
Commission –DG Joint Research Centre, Institute for Environment and Sustainability, Via E. Fermi 2749, Ispra (VA), 21027,
Italy, §Alterra, Wageningen University and Research Centre, P.O. Box 47, Wageningen, 6700 AA, The Netherlands
Abstract
The livestock sector contributes considerably to global greenhouse gas emissions (GHG). Here, for the year 2007 we
examined GHG emissions in the EU27 livestock sector and estimated GHG emissions from production and consump-
tion of livestock products; including imports, exports and wastage. We also reviewed available mitigation options
and estimated their potential. The focus of this review is on the beef and dairy sector since these contribute 60% of all
livestock production emissions. Particular attention is paid to the role of land use and land use change (LULUC) and
carbon sequestration in grasslands. GHG emissions of all livestock products amount to between 630 and 863 Mt
CO
2
e, or 12–17% of total EU27 GHG emissions in 2007. The highest emissions aside from production, originate from
LULUC, followed by emissions from wasted food. The total GHG mitigation potential from the livestock sector in
Europe is between 101 and 377 Mt CO
2
e equivalent to between 12 and 61% of total EU27 livestock sector emissions in
2007. A reduction in food waste and consumption of livestock products linked with reduced production, are the most
effective mitigation options, and if encouraged, would also deliver environmental and human health benefits.
Production of beef and dairy on grassland, as opposed to intensive grain fed production, can be associated with a
reduction in GHG emissions depending on actual LULUC emissions. This could be promoted on rough grazing
land where appropriate.
Keywords: carbon sequestration, consumption, Europe, grassland, greenhouse gas emissions, livestock, mitigation, waste
Received 9 March 2012; revised version received 22 June 2012 and accepted 25 June 2012
Introduction
The publication of FAO’s ‘Livestock’s Long Shadow’ in
2006 (Steinfeld et al., 2006) highlighted the importance
of greenhouse gas (GHG) emissions from animal pro-
duction. Since then, many scientific papers and institu-
tional reports have been published, providing a wealth
of information on GHG emissions of livestock at farm,
regional or global levels (e.g., Garnett, 2007, 2009;
Fairlie, 2010; Nguyen et al., 2010; Foley et al., 2011; Less-
chen et al., 2011; Weiss & Leip, 2012). It is widely
acknowledged that food production and consumption
contribute significantly to anthropogenic greenhouse
gas emissions, with livestock being a major source
(Steinfeld et al., 2006; Smith et al., 2007). Currently,
there is only one whole life cycle estimate of GHG emis-
sions from the global livestock sector including land
use change (LUC), which suggested that global contri-
butions from livestock were 18% of total GHG emis-
sions (Steinfeld et al., 2006). This estimate decreases to
16% if higher total global emission estimates are used
(O’Mara, 2011).
The future security of food and energy supply is of
major concern. Livestock, especially ruminants, occupy
80% of anthropogenic land use (Stehfest et al., 2009)
and consume about 35% of agricultural crops (Foley
et al., 2011), and so are in direct competition with crop
production for human consumption, and with potential
alternative land uses, such as bioenergy crop produc-
tion and nature conservation (Smith et al., 2010). The
future management of the livestock systems is, there-
fore, a key issue in the discussion of land competition
and food security.
There is a vast difference in meat consumption
between developed and developing countries with
developed countries consuming far more animal prod-
ucts per capita than developing countries. In developing
countries, the GHG emissions associated with the same
product can also be much higher due to a low feed
Corresponding: Jessica Bellarby, tel. + 44 1224 273 810,
fax +44 1224 272 703, e-mail: j.bellarby@abdn.ac.uk
©2012 Blackwell Publishing Ltd 1
Global Change Biology (2012), doi: 10.1111/j.1365-2486.2012.02786.x

conversion rate, as a result of e.g., low digestible feed
(Steinfeld et al., 2006). In contrast, developed countries
increase the amount of GHG emissions by overusing
resources and wasting end products.
In this critical review, we examine the GHG emis-
sions in the year 2007 for the EU27 livestock sector
as an example of a group of developed countries.
Besides looking at the livestock sector in general, we
give a special focus to beef and dairy. We examine
the GHG emissions from livestock production, con-
sumption of imported livestock products and wast-
age. We then analyse the potential of different
mitigation strategies in the production and consump-
tion of livestock products.
Greenhouse gas emissions of the EU27 livestock
sector
Life cycle analysis of production in EU27
Generally, the total GHG emissions of a product are
assessed with a life cycle assessment (LCA), which
examines every step and input during the creation of a
product to calculate its total GHG emissions. Therefore,
a comprehensive LCA for livestock would not only
include direct emissions from animals but also emis-
sions of fodder production in and outside the country/
region under consideration, encompassing those from
fertilizer production to fuel emissions and related pro-
cesses. A complete analysis of the impact of livestock
production and consumption would consider all stages
up to its disposal –from cradle to grave.
A review of LCAs within EU27 showed that most
studies set a system boundary at the farm gate and only
a few consider LUC (Table 1).
Of all reviewed studies, Weiss & Leip (2012) is the
most comprehensive to date for the EU27, and one of
the few to attempt to include emissions from LU
(including carbon sequestration of grasslands) and
LUC. The study uses the CAPRI (Common Agricultural
Policy Regionalised Impact) model (Britz & Witzke,
2008), which originally is an economic model to assess
agricultural policies. It now also implements a detailed
GHG module incorporating emissions via emission fac-
tors such as LU (Weiss & Leip, 2010a). Weiss & Leip
(2012) describe different livestock production systems
of the entire range represented in the EU27, which have
been characterized from information drawn from Euro-
pean databases. They look at three scenarios relating to
LUC differing in their assumption on the origin of
required additional land for imported feed products
and reflecting the uncertainty associated with estimates
on LUC emissions. The total emissions provided for the
livestock sector range between 623 and 852 Mt CO
2
e
with the high-emission scenario assuming a high share
of forest being converted (see Weiss & Leip, 2012).
Emissions from LUC contribute between 9 and 33%,
and are mainly attributable to feed imports (Fig. 1).
Beef and dairy make the greatest contribution to total
EU27 GHG emissions, as a result of the high per-kg-
product emissions associated with beef, and the high
EU27 consumption of dairy in comparison to other live-
stock products (Fig. 1). Beef and dairy production is
highly integrated: we estimate that 60 ±5% of beef
originates from the dairy herd in the form of surplus
calves and culled dairy cows with the latter contribut-
ing 17%. The remaining 40 ±5% of beef stems from
suckler cow-calf systems. This estimate is based on fig-
ures of animal numbers from FAOSTAT (2007) and car-
cass weight estimates from Weidema et al. (2008).
Furthermore, it assumes 20% of dairy cows are replaced
each year, which would be equivalent of a dairy cow
being milked for 5 years. A shorter milking period of
3 years (replacement rate of 33%) would increase the
contribution from the dairy herd to between 59% and
68% (Cederberg & Stadig, 2003).
With regard to the different GHGs, the highest emis-
sions are methane (32–42% and 30–39% of total emis-
sions for beef and dairy respectively), followed by
nitrous oxide (21–27% and 17–22%), carbon dioxide
due to energy consumption (13–17% and 14–19%) and
CO
2
fluxes from LULUC (14–34% and 20–38%) (Weiss
& Leip, 2012). Emissions from LULUC form a greater
proportion for dairy cows than for beef, as there is a
greater reliance on imported concentrate feeds, i.e., soy-
bean. These GHG emission estimates include carbon
sequestration rates relative to natural grassland of
0.087 kg CO
2
per m
2
per year for permanent grassland,
and 0.042 kg CO
2
per m
2
per year for cropland culti-
vated with grass and legumes. Croplands are estimated
to be a source of 0.217 kg CO
2
per m
2
per year as a
result of the lost opportunity of carbon sequestration
compared to natural grassland. These rates were calcu-
lated by Weiss & Leip (2010a), based on average num-
bers supplied by Soussana et al. (2007, 2010). These
figures are only slightly higher than an estimate by
Schulze et al. (2009), who give values between 0.167
and 0.175 kg CO
2
per m
2
per year after transformation
from C to CO
2
-eq. to make them comparable. The
application of the above emission factors to the year
2007 results in GHG emissions of between 630 and
863 Mt CO
2
e.
Imports and exports
Roughly 15.4% of the sheep meat, 3.5% of the beef and
3.8% of the chicken meat consumed in EU27 is
imported (Table 2), and thus it is important to include
©2012 Blackwell Publishing Ltd, Global Change Biology, doi: 10.1111/j.1365-2486.2012.02786.x
2J. BELLARBY et al.

Table 1 Greenhouse gas (GHG) emissions from livestock production in Europe as estimated by different studies. Figures for each livestock sector are shown as descending in
GHG emissions. The shaded values are the estimates used in this report, including LULUC emissions
GHG
emission in kg
CO
2
e kg-product
1
Land
requirement
in m
2
kg
1
Land
emissions
considered System boundary Approach Region Reference
Beef
39–56 NA Yes
†
Retail LCA Germany Reinhardt et al. (2009)
14–32
§
27–49 No Farm gate Review EU countries de Vries & de Boer (2010)
28.7 58.9 No Grave Hybrid LCA EU27 Weidema et al. (2008)
24–28 NA No Retail LCA Germany Reinhardt et al. (2009)
16–27.3 16.5–42.9 No Farm gate LCA Europe Nguyen et al. (2010)
15.8–25.3 22.8–42.1 No Farm gate LCA UK Williams et al. (2006)
22.6 37.3 No Farm gate MITERRA model EU27 Lesschen et al. (2011)
18.2–22.2 NA No–Yes
*
Farm gate CAPRI model EU27 Weiss & Leip (2012)
20 NA No Retail Hybrid LCA Sweden Cederberg et al. (2009a)
19.8 NA No Farm gate Hybrid LCA Sweden Cederberg et al. (2009a)
3.1
‡
–16.8 8.0–36.6 No Farm gate LCA Germany Hirschfeld et al. (2008)
Sheep/goats
16.6–20.3 NA No–Yes
*
Farm gate CAPRI model EU27 Weiss & Leip (2012)
10.1–17.5 11.8–31.2 No Farm gate LCA UK Williams et al. (2006)
Pork
11.2 12.2 No Grave Hybrid LCA EU27 Weidema et al. (2008)
3.9–10
§
8.1–9.9 No Farm gate Review EU countries de Vries & de Boer (2010)
4.5–7.5 NA No–Yes
*
Farm gate CAPRI model EU27 Weiss & Leip (2012))
5.6–6.4 6.9–12.8 No Farm gate LCA UK Williams et al. (2006)
2.3–6.1 NA No Retail Review EU countries Gru
¨nberg et al. (2010)
3.54 NA No Retail Hybrid LCA Sweden Cederberg et al. (2009a)
3.5 11.7 No Farm gate MITERRA model EU27 Lesschen et al. (2011)
3.4 NA No Farm gate Hybrid LCA Sweden Cederberg et al. (2009a)
1.7–3.1 6.4–11.8 No Farm gate LCA Germany Hirschfeld et al. (2008)
Poultry
3.7–6.9
§
8.1–9.9 No Farm gate Review EU countries de Vries & de Boer (2010)
4.6–6.7 6.4–14 No Farm gate LCA UK Williams et al. (2006)
2.5–4.9 NA No–Yes
*
Farm gate CAPRI model EU27 Weiss & Leip (2012)
3.6 9.5 No Grave Hybrid LCA EU27 Weidema et al. (2008)
1.7–3.48 NA No Retail Review EU countries Gru
¨nberg et al. (2010)
2.15 NA No Retail Hybrid LCA Sweden Cederberg et al. (2009a)
1.9 NA No Farm gate Hybrid LCA Sweden Cederberg et al. (2009a)
1.6 9.2 No Farm gate MITERRA model EU27 Lesschen et al. (2011)
Eggs
4.4–6.18 5.04–11.84 No Farm gate LCA UK Williams et al. (2006)
©2012 Blackwell Publishing Ltd, Global Change Biology, doi: 10.1111/j.1365-2486.2012.02786.x
EU27 LIVESTOCK GHG EMISSIONS AND MITIGATION 3

Table 1 (continued)
GHG
emission in kg
CO
2
e kg-product
1
Land
requirement
in m
2
kg
1
Land
emissions
considered System boundary Approach Region Reference
3.9–4.9
§
NA No Farm gate Review EU countries de Vries & de Boer (2010)
1.6–2.9 NA No–Yes
*
Farm gate CAPRI model EU27 Weiss & Leip (2012)
1.7 9.0 No Farm gate MITERRA model EU27 Lesschen et al. (2011)
1.47 NA No Retail Hybrid LCA Sweden Cederberg et al. (2009a)
1.42 NA No Farm gate Hybrid LCA Sweden Cederberg et al. (2009a)
Milk
2.4 2.8 No Grave Hybrid LCA EU27 Weidema et al. (2008)
1.4–1.5 1.3–1.8 No Farm gate LCA Netherlands Thomassen et al. (2008)
0.8–1.5 NA No Farm gate Review EU countries Gru
¨nberg et al. (2010)
1.1–1.4 NA No–Yes
*
Farm gate CAPRI model EU27 Weiss & Leip (2012)
1.3 2.4 No Farm gate MITERRA model EU27 Lesschen et al. (2011)
0.8–1.3
§
1.1–2 No Farm gate Review EU countries de Vries & de Boer (2010)
1.0–1.2 1.1–2.0 No Farm gate LCA UK Williams et al. (2006)
0.8–1.2 1.2–2.4 Yes Farm gate LCA Austria Ho
¨rtenhuber et al. (2010)
1.1 NA No Retail Hybrid LCA Sweden Cederberg et al. (2009a)
0.9–1.0 NA No Farm gate LCA Ireland O’Brien et al. (2010)
1.0 NA No Farm gate Hybrid LCA Sweden Cederberg et al. (2009a)
0.8–1.0
†
NA No Retail LCA Germany Reinhardt et al. (2009)
0.6–0.9 1.2–2.3 No Farm gate LCA Germany Hirschfeld et al. (2008)
The term,‘EU countries’ refers to some countries in Europe. EU27 refers to all the 27 member states of the European Union. Williams et al. (2006) is also considered in de Vries &
de Boer’s, 2010 review, but expressed as meat as eaten. Ranges compiled from Gru
¨nberg et al. (2010) have been selected only according to boundaries given within Europe.
Furthermore, Williams et al. (2006) and Weidema et al. (2008) are not considered as they already have been given separately. The kg-product will be meat as eaten when system
boundary is grave, retail or specifically stated and carcass weight when system boundary is farm gate. Hybrid LCA combines top-down sector input-output data with bottom-
up process data.
*LUC induced by the production of feed (excluding grassland and grazing).
†LUC induced by the production of feed which is 13.73 kg CO
2
m
2
for feed and 10.98 kg CO
2
m
2
for additional pasture abroad.
‡Low value for eco-plus meat from old dairy cows.
§Meat as eaten.
©2012 Blackwell Publishing Ltd, Global Change Biology, doi: 10.1111/j.1365-2486.2012.02786.x
4J. BELLARBY et al.

emissions from imports into the GHG estimates of the
livestock sector into the EU27. Monni & Wassenaar
(2010) identify New Zealand as the source of 96% of the
sheep meat imported to the EU27. Beef and chicken
products are imported into EU27 in the largest quanti-
ties, with the majority originating from Brazil: 60% and
43% of net beef and poultry meat imported to EU27
respectively (Table 3). Besides Brazil, considerable
amounts of beef are imported into EU27 from Argen-
tina, representing 24% of the EU27 beef imports. In con-
trast, chicken meat does not seem to have any other
single major importing country into the EU27.
Emissions from Brazilian poultry production are esti-
mated to be 1.2 kg CO
2
e per kg of meat by Monni &
Wassenaar (2010). This is much lower than the emission
factor given by Weiss & Leip (2012) for European
chicken meat without consideration of LULUC [3.1 kg
CO
2
e per kg of meat (assuming 80% of poultry can be
used for meat, as stated in Weiss & Leip, 2010a)] and
(a) (b)
Fig. 1 Total greenhouse gas (GHG) fluxes for scenario II of EU-27 livestock production in 2007 provided in Mt CO
2
e and% of total
livestock emissions for (a) different livestock sectors and (b) different GHG, calculated based on Weiss & Leip (2012) emission factors
and production statistics from FAOSTAT (FAOSTAT, 2007).
Table 2 EU27 livestock statistics from FAOSTAT for the year
2007
Livestock
Production
within
EU27 (Mt)
Net imports
into
EU27 (Mt)
Domestic supply
within EU27 after
imports (Mt)
Sheep and
goats
1.1 0.2 1.3
Beef 8.2 0.3 8.5
Poultry 10.9 0.4 10.5
Pig 22.7 1.4 21.3
Dairy 153 7 144
Eggs 6.6 0.2 6.8
Table 3 Greenhouse gas (GHG) emissions from imported livestock products as from Monni & Wassenaar (2010) for EU27
Kg CO
2
e
kg-product
1
Import into EU27 in kg
GHG emissions in Mt CO
2
e as result
of total EU27 imports
from respective
country
total from all non-EU
countries
*
(% of total
consumption
Sheep (New Zealand) 33 192 200 (15.4) 6.6
Beef (Brazil) 48 (80)
†
180
‡
300 (3.5) 14.4 (24)
†
Chicken meat (Brazil) 1.2
§
170 400 (3.8) 0.5
Total 21.5 (31.1)
†
*Value derived from Table 1 (domestic supply –production =net import), gross import much higher.
†Figure in parenthesis is with LUC.
‡Another 24% comes from Argentina.
§Due to a reduction in chicken meat import into Europe after 2005, LUC is not allocated to chicken meat production as exports did
not result, at least directly, in expansion of crop area (Monni & Wassenaar, 2010).
©2012 Blackwell Publishing Ltd, Global Change Biology, doi: 10.1111/j.1365-2486.2012.02786.x
EU27 LIVESTOCK GHG EMISSIONS AND MITIGATION 5

other values listed in Table 1. This is related to the
purely intensive broiler production system in which
Brazilian exports are produced. Equivalent systems in
the United States and Canada exhibit similar GHG
emissions per kg of meat of 1.8 and 1.3 kg CO
2
e per kg
of meat (Pelletier, 2008; Verge
´et al., 2009).
Sheep and beef both show higher GHG emissions
when produced abroad. The main contributor to these
high non-European sheep emissions is methane from
ruminants’ digestive systems. In New Zealand, the
emission factor for this is set to 11 kg CO
2
e per head,
based on the national GHG inventory of New Zealand
(Ministry for the Environment, 2008), whereas in EU27
it is set at only 8 kg CO
2
e per head, based on IPCC
(2006) default figures for developed countries (Monni
& Wassenaar, 2010).
GHG emissions of Brazilian beef are very high even
before LULUC is taken into consideration (48 kg CO
2
e
per kg of meat, Table 4). This is the result of a mainly
extensive 100% suckler cow-calf production system and
as such only a little higher than the same system in the
EU27 with GHG emissions of up to 27.3 kg CO
2
e per
kg of carcass (or 39 kg CO
2
e per kg of meat) (Nguyen
et al., 2010). GHG emissions without LUC provided by
Cederberg et al. (2009b) are even closer to the EU27
values (Table 4).
LUC contributes a considerable amount to total GHG
emissions and estimates vary even more depending on
the assumptions made. Cederberg et al. (2009b) esti-
mate GHG emissions originating just from LUC to be
about twice as high as estimates from Monni &
Wassenaar (2010) (63 rather than 32 kg CO
2
e per kg of
meat, Table 4) for the lowest emission scenario, which
averages exports over the whole of Brazil. As the net
committed emissions and productivity of both studies
are very similar, this difference in figures seems to be
merely a result of different assumptions of the propor-
tion of LUC occurring on newly deforested land: 6% for
Cederberg et al. (2011), and only 3% for Monni &
Wassenaar (2010). The sensitivity of this proportion is
demonstrated in more detail in Cederberg et al. (2011)
and results are given in Table 4.
GHG emissions from total net imported livestock
products (beef, sheep and chicken) amount to 31.1 Mt
CO
2
e in the overall EU27 livestock emissions (Table 3)
in 2007. Beef imports account for 77% of these emis-
sions from imported livestock goods, including emis-
sions from LUC due to beef production in South
America (Monni & Wassenaar, 2010).
When considering imports into EU27, exports should
be deducted from the EU27 total GHG emissions, since
these should be accounted for in the importing country.
The EU27 exported 6% of pork and 5% of dairy in 2007,
which is equivalent to a total of about 20.3 Mt CO
2
e.
Waste
A side effect of consumption is the accumulation of
unnecessary food waste. The European Commission
(EC, 2010) estimated the amount of waste produced
within the EU27 from EUROSTAT for the year 2006
and compared the results to several national studies, of
which the UK study by WRAP for the year 2008
(WRAP, 2010) is recognized to be the most comprehen-
sive (Table 5).
There is a great difference in values for the UK
households between the studies of the European
Commission and of WRAP; this is not a trend but it
demonstrates the uncertainty associated with this
kind of data. The total food wasted in the UK, how-
ever, is very similar between the two studies
(Table 5) and is equivalent to a waste rate of 20%
when related to the amount of food supplied to the
United Kingdom (61 Mt per year) according to the
food balance sheets of the FAO (FAOSTAT, 2007).
The waste rate determined for the EU27 in this way
is slightly higher, at 22.5%. Generally, UK house-
holds waste a high proportion of purchased food
(22% by weight) (WRAP, 2009) compared to, for
example, the Netherlands, which has a wastage of
between 8 and 11% (Parfitt et al., 2010). However, a
recent study by Williams & Wikstro
¨m (2011)
reviewed several studies and concluded that an
average food loss rate of 20% is a reasonable
estimate for developed countries.
WRAP (2010) is the only study that distinguishes
between waste originating from different food products
(meat, bread etc.) within the retail and household
sectors in the United Kingdom. An extrapolation of the
Table 4 Estimates of greenhouse gas emissions from produc-
tion of Brazilian beef imported to Europe
Beef from -
Without LUC
kg CO
2
e
kg-meat
1
With LUC
kg CO
2
e
kg-meat
1
Reference
- Brazil (average) 48 80 *
- Brazil (average) 41 104
†‡
- legal Amazon
region
298
†‡
- newly deforested
land
1078 ±401
†‡
Values from Cederberg et al. (2011) converted from carcass to
meat using the factor 0.7 from Cederberg et al. (2009b).
*Monni & Wassenaar (2010).
†Cederber et al. (2009b).
‡Cederber et al. (2011).
©2012 Blackwell Publishing Ltd, Global Change Biology, doi: 10.1111/j.1365-2486.2012.02786.x
6J. BELLARBY et al.

animal waste rates derived from these figures should
be reasonable considering that United Kingdom and
EU27 waste levels are very similar.
In the retail and household sector, waste of live-
stock products represent 27% and 14% of the total
food waste respectively. The household sector repre-
sents the majority of the country’s food waste
(WRAP, 2009). Therefore, we apply the animal food
waste rate of 14% to the total food waste (11.9 Mt)
and use the resulting value to determine the
percentage from the total domestic supply (22.4 Mt),
which equates to 8% of all livestock goods produced
being wasted in the United Kingdom. WRAP (2009)
also provides the amount of livestock products pur-
chased by households. Relating the animal products
wasted by households directly to this figure would
result in a considerably higher waste rate of 12.5%.
This is a result of the majority of waste originating
from the household, which is partly due to the
‘recycling’ of waste in the food manufacturing
industry, so that it does not end up as landfill
waste (WRAP, 2010) and the high uncertainty of
data in the retail sector. We take this uncertainty
into account by using a range between 8 and 12.5%.
The total domestic supply of animal products to the
EU27 in 2007 was 192 Mt, which would amount to a
waste of between 15 and 24 Mt, when applying the
waste rates as determined above. Therefore, it not only
includes food wasted in households and other food
retail sectors (restaurants, schools etc.) but also e.g.,
food processing. With each tonne of waste from food
and drink on average 0.45 t CO
2
e are emitted (WRAP,
2009), resulting in additional GHG emissions of
between 7 and 11 Mt CO
2
e from the decomposition of
wasted livestock products alone.
Total EU27 emissions related to livestock products
The total EU27 GHG emissions related to the livestock
sector constitute between 630 and 863 Mt CO
2
e for 2007
(Table 6). This equates to between 12 and 17% of total
EU27 GHG emissions in 2007 (EEA, 2011). This com-
parison, however, is not straightforward as total EU27
emissions are for all goods produced in the country
and as such include exports and exclude imports.
Additional (imported) emissions from imported prod-
ucts are much larger than exported ones. Furthermore,
international air and ship transport are not included
either. The EU27 livestock sector is also assumed not to
cause direct LUC in Europe as most EU27 forest con-
versions happened decades to centuries ago and forest
area is increasing (Houghton, 2003). Another estimate
by Lesschen et al. (2011) placed the contribution of live-
stock farming to be 10% of total EU27 emissions, which
is equivalent to the emissions from production alone
without consideration of LULUC (Table 6).
The global contribution of the livestock sector to total
GHG emissions is estimated to be between 16%
(O’Mara, 2011) and 18% (Steinfeld et al., 2006;). The
EU27 livestock sector contributes between 9% and 12%
to the total global livestock emissions, yet has a 21%
share of the global meat demand, with only about 7.5%
of the human population in the world (FAOSTAT,
2007).
Table 6 Greenhouse gas (GHG) emissions from the livestock
production sector in Mt per year
GHG emissions in Mt CO
2
e
Production
*
461
LU 102
LUC 58–284
Imports 22–31
†
Exports 20 to 26
Waste 7–11
‡
Total 630–863
*Excluding LUC.
†With –without deforestation (low estimate), see Table 3.
‡See section 2.3.
Table 5 Animal and vegetal waste in Mt per year (WRAP, 2009; EC, 2010)
Agriculture,
hunting and
forestry (Mt)
Manufacture
of food products;
beverages and
tobacco (Mt) Households (Mt) Other sectors (Mt) Total (Mt)
EU27
*
(EC, 2010) 32.6 37.3 23.4 16.8 110.1
UK
*
(EC, 2010) 0.02 5.1 3.2 3.5 11.9
UK
†
(WRAP, 2009) 3 8.3 11.3
*2006;
†2008 (The two UK studies have different methodologies, which explains some of the variation in magnitude between the two
sources and years).
©2012 Blackwell Publishing Ltd, Global Change Biology, doi: 10.1111/j.1365-2486.2012.02786.x
EU27 LIVESTOCK GHG EMISSIONS AND MITIGATION 7

Potential for the mitigation of greenhouse gas emissions
from livestock production and consumption in the EU27
The EU27 livestock sector contributes considerably to
global livestock emissions. Mitigation can be purely
technical at the production level, or can be based on a
reduction of production and consumption. In the light
of further drastic increases of emissions resulting from
the projected levels of consumption of livestock prod-
ucts (FAO, 2006), it is important to identify and
describe reversed trends. At a global level it is usually
suggested that intensification is indispensable to meet
the future demand (Bouwman et al., 2005). However,
further intensification of livestock production systems
within EU27 countries with an already high reliance on
imported high protein feed may not be the best option
with respect to reducing GHG emissions. Furthermore,
ethical and environmental concerns of consumers and
NGOs tend to favour less-intensive production sys-
tems. The impact of this movement has been most
strongly observed in the widespread preference of barn
or free-range eggs as an alternative to battery eggs to
the extent that the EU banned conventional battery cage
production in 2012 (EC, 1999). This example clearly
demonstrates a tendency to reduce the intensity of
livestock production in the EU27.
In regard to production, the beef and dairy sector
have the highest GHG emissions in the EU27 on an
absolute basis (Fig. 1). Generally, monogastric animals
(such as pigs and poultry) have much lower emissions
per unit of meat compared to ruminants. Therefore, it
makes sense to look at the reduction potential in the
beef and dairy sectors in more detail. Here, it should be
noted that monogastric animals consume protein and
feed that could be consumed by humans. Grass-fed
ruminants, on the other hand, consume grass-based for-
age that cannot be utilized for human food. Wilkinson
(2011) calculates conversion efficiencies on the basis of
the quantity of intake of human-edible feed and con-
cludes that, on this basis, ruminants are more efficient
than monogastric animals in converting this human-
inedible material into human-edible livestock products.
However, this would only be true if the land used for
growing human-inedible feed is not actually suitable
for growing edible crops on. Indeed, Gill et al. (2010)
noted that not all land currently used for grazing is
suitable for cropping (including land that is too wet,
too dry, on slopes and so on) and the most efficient use
of this land (from a food security and an emission per-
spective) is to use it to raise livestock. Particularly,
some of these marginal lands may be grazed sustain-
ably for the benefit of biodiversity as well, being known
as ‘High Nature Value’ grassland (IEEP, 2007). Further-
more, grasslands have the additional advantage of
acting as a carbon sink compared to croplands, which
can be exploited for GHG mitigation in beef and dairy
production (Soussana et al., 2010).
In this review, we consider the positive impact of
grassland when comparing different beef and dairy
production systems. To obtain a full picture of the miti-
gation potential in production systems technical
approaches are also briefly summarized.
With respect to consumption, use of livestock
products is directly linked to their production. Conse-
quently, a reduction in consumption would result in a
decrease of livestock numbers, which would have a
direct and indirect impact on reducing GHG emissions.
It is commonly recognized that current levels of meat
consumption in Europe and other developed nations
are too high on the grounds of human health (Friel
et al., 2009; Westhoek et al., 2011). Therefore, a reduc-
tion in the consumption of livestock products would
have environmental as well as health benefits. Further-
more, consumption is directly linked to waste, a part of
the life cycle that should be first targeted, as it unneces-
sarily increases demand and creates additional GHG
emissions through its breakdown. Both aspects will be
addressed in this review.
Mitigation potential from production
Choice of production system
There is a great diversity in beef and dairy production
systems within the EU27. To identify differences in
GHG emissions from various management systems, we
review studies that analysed emissions from a number
of different systems.
At European scale, two different approaches to clas-
sify livestock systems have been applied: (1) A differen-
tiation according to the finishing age and origin of the
calf (dairy herd or cow suckler herd) (Williams et al.,
2006; Hirschfeld et al.,2008; Nguyen et al., 2010), and
(2) a definition of clusters according to five variables
(economic importance of livestock production in a
region, level of intensification, housing system, market
dependence and main feedstuff used) (Loudjani et al.,
2010) (Table 7).
The advantage of the approach from Loudjani et al.
(2010) is that the seven clusters are described by feed-
ing strategy (e.g., ‘extensive grassland’). Furthermore,
LULUC emissions and the market share within the
EU27 are provided. However, the GHG emissions for
each cluster are biased by the regions they occur in.
Consequently, this approach is a good description of
the status quo, but not suited for a simple transfer from
one to the other system without the application of the
modelling system. In contrast, Nguyen et al. (2010) give
©2012 Blackwell Publishing Ltd, Global Change Biology, doi: 10.1111/j.1365-2486.2012.02786.x
8J. BELLARBY et al.

an overall average of the respective production system.
The age classification can also be correlated to three of
the systems by Loudjani et al. (2010) based on the
amount of grass or cereal used respectively. This is, of
course, not a perfect match as four of the clusters
defined by Loudjani et al. (2010) are not considered at
all, although 64% of the total EU27 market share are
still represented by the three clusters. Furthermore, the
beef suckler system is represented within all the differ-
ent clusters (Loudjani et al., 2010) (Table 7).
The ‘intensive maize’ cluster shows by far the lowest
GHG emissions when LULUC is not considered, dem-
onstrating the regional bias. The other two clusters are
more similar to Nguyen et al.’s (2010) values, being
slightly lower (‘intensive grass and maize’) and higher
(‘extensive grassland’). The latter may partly be due to
the potential contribution of suckler cow-calf herds,
which exhibit by far the highest GHG emissions com-
pared to the other systems, which stem from the sur-
plus calves of the dairy herd (Nguyen et al., 2010). This
is a result of dairy cow emissions being assigned to
milk production, in contrast to the suckler cow having
to be fully assigned to beef production (Table 7). Gener-
ally, Hirschfeld et al. (2008) state much lower GHG
emissions throughout than any other study; therefore
the figures should be regarded with caution. However,
Hirschfeld et al. (2008) is the only study to look specifi-
cally at the total GHG emissions of old dairy and
suckler cows. Meat from old dairy cows is associated
with the lowest GHG emissions (Table 7). The amount
is highly dependent on the allocation method used,
hence, can even be zero if all is allocated to the produc-
tion of milk (Flysjo
¨et al., 2011).
The cow-calf and the dairy bull calf system with a
high finishing age are not reliant on soy meal and are
mainly grass fed. In addition, the cow-calf system occu-
pies a high proportion of low productive grassland,
which would likely not be suitable for crops. As a
result, both these systems are better when considering
LULUC as well (Nguyen et al., 2010) (Table 8). The
increase of the GHG emission per product is highly
dependent on the amount of deforestation assigned to
soymeal production. Generally, there is a high uncer-
tainty of the mitigation potential of a change of produ-
tion system due to its strong relationship to the amount
of land being deforested for the production of soymeal.
This is demonstrated by the three different scenarios of
Weiss & Leip (2012), in which no deforestation occurs
in scenario I whereas 100% of land for soymeal origi-
nates from forest in scenario III and scenario II applies
a more likely mix (Table 8) (Weiss & Leip, 2010b). It
becomes apparent that only in scenario III it is really
worth to change to an ‘extensive grassland’ system.
However, this is partly the result of cow suckler herds
being represented within the system as well. Conse-
quently, it is better to apply dairy calf figures from
Nguyen et al. (2010) instead. They provide the amount
of land for grass, cereal and soymeal for each produc-
tion system, so that emission factors can be applied for
them. Then we can also adjust the deforestation rate
and the proportion of the cow calf-suckler herd result-
ing in varying mitigation potentials. Hence, at a defor-
estation rate of 30%, a change from a more intensive
(12–16 months) to an extensive (24 months) system for
25–50% of the beef produced, would mitigate between
6 and 15 Mt CO
2
e each year.
Table 7 Beef production systems, not including LUC
Description
GHG emissions in kg CO
2
e kg-meat
1
Months at
slaughter
Nguyen et al.
(2010)
Europe
Williams et al.
(2006) UK
Hirschfeld
et al. (2008)
Germany
Weiss & Leip
(2010b) EU27
Cluster
‘equivalent’
Mix of non-organic
suckler beef (35%)/
dairy cow system
*
15.8
Dairy bull (exclusively
fed on cereal and soymeal)
12 16.0 8.4
*
12.5 ‘Intensive maize’
Dairy bull (high proportion
of cereal and soymeal but
also some grass)
16 17.9 16.2 ‘Intensive grass
and maize’
Dairy steer (mainly fed on
grass with some cereal and
very little soymeal)
24 19.9 21.3 ‘Extensive grass’
Suckler cow-calf
*
27.3 25.3 16.8
Conventional dairy cow
*
6
*Months at slaughter not specified.
©2012 Blackwell Publishing Ltd, Global Change Biology, doi: 10.1111/j.1365-2486.2012.02786.x
EU27 LIVESTOCK GHG EMISSIONS AND MITIGATION 9

An additional benefit would also lie in the reduction
of the amount of cropland that would be required for
grass-fed beef raised over 24 months (Nguyen et al.,
2010), potententially sequestering between 6 and 11 Mt
CO
2
e each year, provided that the previous cropland is
converted into a permanent vegetation type. Therefore,
this would result in a total GHG mitigation potential
between 12 and 26 Mt CO
2
e per year at a current defor-
estation rate of 30% and an increase of a more extensive
grass-based system by 25–50%.
Increasing carbon sinks on grazing lands
Intensive production systems rely on high protein feed
imports, which not only increase GHG emissions via
LULUC but also improve the feed conversion. In con-
trast, grass-fed production systems may have the
advantage of serving as a carbon sink, but have lower
feed conversion. The carbon sequestration potential of
grasslands is highly variable (Table 9) as shown in a
review by Soussana et al. (2010). The literature
reviewed suggests that grasslands range from sinks to
sources depending on climate, management and site
characteristics such as soil type. Even though there is
the potential of fully compensating for beef or dairy
emissions at the farm level, there is the risk that accu-
mulated carbon can be lost in events such as an unusu-
ally dry summer (Soussana et al., 2010). Drained
organic soils are certainly a carbon source and restora-
tion of these would compensate for livestock produc-
tion on other areas.
Generally, Soussana et al. (2010) acknowledge that
improved management practices could increase carbon
stock levels, referring to the meta-analysis by Conant
et al. (2001) of 115 sites worldwide (Table 9). However,
increased carbon sequestration by a management prac-
tice may increase other GHG emissions and, as such,
decrease or even negate the sequestered CO
2
in the soil.
The application of synthetic fertilizer, for example, was
considered to result in net GHG emissions when con-
sidering emissions from fertilizer production and
nitrous oxide emissions after application (Powlson
et al., 2011). Soussana et al. (2004) modelled the poten-
tial of European grassland management systems and
obtained carbon sequestration rates of between 0.7 and
1.8 t CO
2
e per ha per year. It should be noted that
carbon will only be accumulated over a finite period
depending on climate and soil type (Powlson et al.,
2011). Weiss & Leip (2012) already consider an average
carbon sequestration rate of managed grassland. How-
ever, the sequestration rates provided by Soussana et al.
(2004) are additional carbon sequestration rates under
improved grassland management options that may not
already be implemented. Therefore, assuming that half
of the intensively managed grassland area of the EU27
can be further optimized, results in a carbon sequestra-
tion potential of between 4 and 10 Mt CO
2
e per year,
which lies in the same range as estimated mitigation
potentials for the EU27 for measures on arable land
(Frelih-Larson et al., 2008).
Technical mitigation options in production
The main emission sources of livestock production,
which have not already been considered, are as follows:
(1) methane from enteric fermentation in the animal
Table 8 Beef production systems, including LUC
Description
GHG emissions in kg CO
2
e kg-meat
1
(contribution of LULUC in%)
Cluster ‘equivalent’
(% market share)
Months at
slaughter
Nguyen et al.
(2010) Europe
Weiss & Leip (2010b) EU27
Scenario I Scenario II Scenario III
Dairy bull (exclusively
fed on cereal and
soymeal)
12 30.2 (47)
†
16.4 (24) 18.0 (31) 26.0 (52) ‘Intesive maize’ (6)
Dairy bull (high
proportion of cereal
and soymeal but also
some grass)
16 29.3 (39)
†
20.3 (20) 21.6 (25) 27.5 (41) ‘Intensive grass
and maize’ (41)
Dairy steer (mainly fed
on grass with some cereal
and very little soymeal)
24 19.4 (3)
†
19.4 (10) 19.8 (8) 21.6 (1) ‘Extensive
grass’ (17)
Suckler cow-calf
*
25.5 (7)
†
*Months at slaughter not specified.
†contribution of LUC derived by using weighted average of 2.8 kg CO
2
m
2
yr
1
, LU calculated by using emission factors from
Weiss & Leip (2010a).
©2012 Blackwell Publishing Ltd, Global Change Biology, doi: 10.1111/j.1365-2486.2012.02786.x
10 J. BELLARBY et al.

digestive system and (2) methane and nitrous oxide
from manure management (Tubiello & Loudjani, 2010).
Methane from enteric fermentation. This can either be mit-
igated by lowering cattle numbers via reduced con-
sumption, increasing feed conversion, or by diet
regimes that reduce the methane emission from enteric
fermentation.
Reducing animal numbers by a reduction in con-
sumption has a large potential for mitigation, as
expressed in section 3.2. An increase of the yield per
animal will also reduce the number of animals
required, but may have other implications. For exam-
ple, increasing the milk yield per cow may trigger an
increase in dedicated beef suckler herds, an effect that
was observed in Sweden between 1990 and 2005
(Cederberg et al., 2009a). Here, the GHG emissions of
beef production increased from 18 to 20 kg CO
2
e per
kg of beef, as a result of a higher share of beef originat-
ing from beef suckler herds as opposed to the dairy sec-
tor. However, overall GHG emissions in the Swedish
beef and dairy sector still showed a decrease compared
to 1990 levels as a result of the increased yield (Ceder-
berg et al., 2009a).
The reduction of enteric fermentation by different
dietary additions has been reviewed by Eckard et al.
(2010), who identified a potential of a 5–10% reduction
in emissions equivalent to between 8 and 15 Mt CO
2
e
per year. However, a careful consideration of potential
environmental and health effects will be necessary
before recommending any widespread use.
Methane and nitrous oxide from animal waste manage-
ment. A variety of approaches that have been identified
by Tubiello & Loudjani (2010) to manage methane and
nitrous oxide from animal waste are anaerobic diges-
tion, composting, temperature of storage tanks, com-
pacting and coverage. Based on a literature review,
they concluded that with the full uptake of the current
technologies (mitigation of emissions from soils, meth-
ane from enteric fermentation, and methane and
nitrous oxide from animal waste management), GHG
emissions from manure would reduce by 30%, pro-
vided that anaerobic digesting and composting are the
key strategies.
Tubiello & Loudjani (2010) estimated a total technical
mitigation (enteric fermentation, waste management
and grassland management) between 15% and 19%,
Table 9 Overview of carbon sequestration rates in European grassland
Observed effect on carbon
sequestration in grasslands soil
Net carbon sequestration
() or emissions (+)inkg
CO
2
em
2
yr
1
Reference
Increased soil carbon storage under grass with liming 2.0 to +0.33 Fornara et al. (2011)
Variable soil carbon storage in a review of a range of studies 1.3 to +0.6 Soussana et al. (2010)
Treatments with and without T. pratense, seed addition and
fertilizer use
1.2 to +0.03 De Deyn et al. (2011)
Carbon input by roots depending on plant diversity 1.5 to 0.21 Steinbeiss et al. (2008)
Predicted range of European grassland carbon balance 0.6 to +0.2
*
Janssens et al. (2005)
Conversion from cropland to pasture 0.37 Conant et al. (2001)
Net soil carbon sequestration increases with
plant diversity
0.3 to +0.05
†
Tilman et al. (2006)
Net root carbon sequestration increases with
plant diversity
0.2 to 0.06 Tilman et al. (2006)
Carbon sequestration after conversion of arable
cropland to grassland
0.18 Soussana et al. (2004)
Improved grazing 0.13 Conant et al. (2001)
Fertilization 0.11
‡
Conant et al. (2001)
Reduction of nitrogen input 0.11
§
Soussana et al. (2004)
Conversion to grass legume mixtures 0.11 to 0.18 Soussana et al. (2004)
Short duration leys to permanent grassland 0.11 to 0.15 Soussana et al. (2004)
Increasing duration of leys 0.07 to 0.18 Soussana et al. (2004)
*Only in Portugal grassland was a net source, potentially due to climate.
†Monoculture.
‡But net GHG emissions when considering emissions from fertilizer production and application.
§in highly intensive grass leys.
©2012 Blackwell Publishing Ltd, Global Change Biology, doi: 10.1111/j.1365-2486.2012.02786.x
EU27 LIVESTOCK GHG EMISSIONS AND MITIGATION 11

equivalent to between 55 and 70 Mt CO
2
eyr
1
. How-
ever, we have considered grassland management sepa-
rately, so that it needs to be deducted, hence resulting
in a total of between 45 and 66 Mt CO
2
eyr
1
. These
technical solutions need to be seen in conjunction with
changes in the production system: biogas installation,
for example, would favour high animal numbers and
indoor production systems that could have other detri-
mental implications.
Mitigation potential from reduced consumption
There are two approaches by which consumers can
have a direct impact on the reduction of GHG
emissions: a reduction in the consumption of animal
products and a minimization of food waste.
Reduction in consumption
Protein intakes in the EU are 70% higher than the levels
recommended by the World Health Organization
(Westhoek et al., 2011). If reducing livestock consump-
tion further than the current 70% over-consumption
level, human dietary protein should be replaced with
plant protein. Consumption of beef results in the high-
est GHG emissions, especially when imported from
South America or reared from 100% beef suckler herds.
Table 10 lists the impact of a few different actions by
the EU27 consumer and their associated reduction in
GHG emissions. However, it has recently been demon-
strated by Wolf et al. (2011) that changes in consump-
tion of meat within Europe would not necessarily be
followed by equal changes in production, as products
may be traded in the global market. Although exports
to non-European countries would increase, imports
would decrease. There would also be a tendency to
move towards less extensive production systems
according to Wolf et al. (2011). Therefore, a consider-
able decrease in GHG emissions could still be expected,
especially from lower imports. These may well be
imported to other countries instead, but still the compe-
tition for land would be reduced. On the other hand, a
decrease in production only would encourage an
increase of imports with potentially even more GHG
emitted if consumption is not reduced at the same time.
This makes clear that a shift in consumption should go
hand in hand with a shift in production, or encouraged
by a shift in production of less, but higher valued meat.
With a higher price for meat, it will also increase
incentives to increase the quality of meat from culled
dairy cows (Vestergaard et al., 2007; Minchin et al.,
2010).
Minimizing food waste
In section 2.3 of this review it has been estimated that 7
–11 Mt CO
2
e are emitted by the breakdown of wasted
animal-derived food products in the EU27 every year.
Furthermore, another 49–104 Mt CO
2
e are emitted due
to the production of an extra 15–24 Mt of livestock
goods that are later wasted. Therefore, the total actual
food waste impact sums up to 56–115 Mt CO
2
e. This
represents 8% and 17% of total livestock emissions
depending on the actual waste rate of animal products
used and the scenario under consideration respectively
(Table 6). WRAP (2009, 2010) estimated that the major-
ity of this waste could be avoided reducing the rate to
between 2.4 and 3.9%. This corresponds to an overall
reduced waste rate of between 7.1 and 7.8%. In compar-
ison, only about 2% of food was wasted in the United
Table 10 Impact of the reduction in consumption of different livestock products by European consumers
Action Livestock sector Total Mt product
Total savings
of CO
2
e emissions
in Mt per year
Change in
consumption
per person g (%)
per week
Eat no beef from South America
*
Beef import 0.3 22–31 12 (3.5)
Eat no meat from European beef
suckler herd
†
Beef suckler 2.6–3.7 67–94 106 to 149 (32–45)
One less serving of milk or 20 g
less cheese (per week)
‡
Dairy 6.6 9–11 200 (4.3)
A reduction in dairy products
(reducing the dairy herd would also
decrease the amount of meat)
‡
Beef dairy 0.3 6–812 (3.5)
*Range given for beef raised on historic grassland and partly on newly deforested land.
†Assuming that between 32 and 45% of European beef is produced from suckler herds at a replacement rate of 33% and 20%,
respectively (see section 2.1), with a higher carbon footprint of 25.5 kg CO
2
ekg
1
product (Table 8).
‡Range given for scenario I–II according to Weiss & Leip (2012).
©2012 Blackwell Publishing Ltd, Global Change Biology, doi: 10.1111/j.1365-2486.2012.02786.x
12 J. BELLARBY et al.

Kingdom in the 1930s (Parfitt et al., 2010), which dem-
onstrates that these levels should be achievable.
Depending on the actual waste reduction and scenario,
GHG mitigation potential would be between 39 and
79 Mt CO
2
e just for avoiding the production and
decomposition of wasted animal products. An alterna-
tive would be to collect all animal food waste for
energy production. However, the amount of GHG
emissions saved would depend strongly on the mode
of energy generation replaced. The replacement of the
average UK electricity grid emissions in 1996, for exam-
ple, would result in savings of 46–71 Mt CO
2
e per year
(calculated after values given in Sims et al., 2006 and
Bernstad & la Cour Jansen, 2011). This estimate is in the
same range as waste reduction, but this potential will
decrease with an increasing efficiency of power genera-
tion. Anaerobic waste digestion is also a solution for
unavoidable waste, where it could provide savings of
an additional 14–22 Mt CO
2
e per year. Another option
for avoiding waste is to use it as feed for monogastric
animals, some of which is already being used as animal
feed.
Discussion and conclusion
A reduction in livestock emissions can be based on
technical approaches or, by reduced demand. Reduced
consumption would also be beneficial for public health,
but may not necessarily reduce production to the same
extent in a global trade environment. Reducing produc-
tion in one area may simply stimulate production in
another, thereby partly displacing GHG emissions
rather than reducing them. Technical approaches are
often more economical on a larger scale for the farmer.
Carbon sequestration through improved management
of grasslands may reduce net GHG emissions (depend-
ing on soil type) and may have other environmental
benefits, such as improved biodiversity in high nature
value grasslands. Policies will need to take regional
differences into account as well as the needs of both
producer and consumer, while aiming across sectors
such as public health and consumption behaviour,
agricultural production and marketing and trade. The
previous sections have provided a series of mitigation
options, which are summarized in Table 11.
Table 11 Summary of mitigation options
Description
Emission savings
in Mt CO
2
e per year
Emission
reduction in%
*
Reduction in
consumption in%
Production related mitigation options
Choice of production system to grass-fed beef 12–26 2–4X%
Grassland management 4–10 1–20
Consumer-impacted mitigation options
Eat no beef from South America 22–31
†
3–54
Eat no meat from European beef suckler herd 67–94 10–14 32–45
One less serving of milk or
20 g less cheese (per week)
15–19 2–34
Waste reduction (waste rate of 2.4–3.9%)
Waste minimization 56–115 8–17 0
Anaerobic digestion of unavoidable waste 14–22 2–30
Technical approaches
Anaerobic digestion of all food waste 46–71 7–11 0
Combined techno-fixes 51–60 8–90
Totals
‡
No reduction in consumption
§
101–207 15–31 0
Additional reduction in consumption
¶
216–377 33–57 32–45
*From a scenario II (Weiss & Leip, 2012) total adjusted to 2007 of 661 Mt, the total level of emissions from meat produced in Europe
with LUC.
†Range is from without LUC to with LUC.
‡Low estimates do not include waste reduction, but digest all food waste, high estimates are a combination of high levels of waste
minimization and anaerobic digestion of unavoidable waste at a waste rate of 3.9% (reduced from 12.5%); furthermore, other low/
high estimates are utilized where available.
§Only technical approaches are used –also see note ‡in regard to waste; furthermore, only grassland management is included
under ‘production-related emissions’ as a change to a grass-fed system would likely indirectly result in an unknown level of reduc-
tion in consumption.
¶To the total mitigation potential with no reduction in consumption, mitigation options are added that do result in a reduction in
consumption listed under ‘Consumer-impacted mitigation options’ as well as the choice of production system.
©2012 Blackwell Publishing Ltd, Global Change Biology, doi: 10.1111/j.1365-2486.2012.02786.x
EU27 LIVESTOCK GHG EMISSIONS AND MITIGATION 13

The numbers in Table 11 only give an indication
of the mitigation potential as feedback mechanisms
have not been considered: for example, lower fertil-
izer use in extensively managed grassland or higher
nitrous oxide emissions from grazing cows. Further-
more, a complete abandonment of beef suckler herds
would be undesirable as they often serve biodiver-
sity function within High Nature Value grassland.
Generally, the EU reduction target of 20% by 2020
is achievable, but requires also a reduction in waste
and consumption. Waste minimization has the great-
est impact on a reduction of GHG emissions, which
is followed by the avoidance of deforestation linked
to livestock production, indirectly considered when
changing the type of production system and the
high estimate of beef imports. A purely technical
approach aimed at improving production will rely
on optimum conditions being met to be sufficient.
The digestion of all food waste may be reasonably
attractive in Table 11, but it has to be kept in mind
that it depends on the GHG emissions of the energy
it replaces and other uses, e.g., for animal feed,
might be more sustainable. A reduction in consump-
tion is a realistic strategy to increase the mitigation
potential.
Based on the findings of this review, it is recom-
mended that EU27 policies be developed to influence
both production and consumption. It would be desir-
able to improve the efficiency of livestock production
and reduce GHG emissions per kg of product. How-
ever, it has been acknowledged that in view of historic
breeding improvements, further increases in the future
are debatable (Wilkinson, 2011). Therefore, it is sug-
gested to address the following:
(1)Food waste represents the greatest single mitigation
potential. A minimization of waste should be encour-
aged, to reduce waste levels to be as low as possible.
(2)LUC emissions are considerable from newly defor-
ested land in South America. Therefore, a continued
support for avoiding deforestation in these countries
should be aimed for.
(3)Consumption patterns should be changed to meet
healthy eating guidelines, and production reduced
accordingly by favouring more sustainable production
systems.
(4)There is some evidence that intensive grain-based
livestock systems need not be better, but could even be
worse in terms of GHG emissions if LULUC emissions
are taken into account. Therefore, current land manage-
ment practices should be optimized and the less-inten-
sive grazing systems favoured, with lower fertilizer
inputs. Reliance on grain feeding of livestock should be
reduced. The exploitation of rough grazing lands
should be encouraged in areas that cannot otherwise be
exploited and where this does not interfere with other
conservation of biodiversity objectives.
(5)Anaerobic digestion of food waste (after waste levels
have been minimized and appropriate food waste has
gone to feed pigs and poultry) and animal manures
should be implemented. This reduces methane emis-
sions and also offsets fossil fuel emissions from the
energy sector. The digested end product can be used as
high quality, low odour fertilizer.
These recommendations are in line with GHG mitiga-
tion options suggested by Steinfeld et al. (2006) tailored
for the EU27. Specific options for the EU27 or devel-
oped countries in general are the reductions in con-
sumption of livestock products and food waste
although the latter has not been addressed in the FAO
report (Steinfeld et al., 2006). The total GHG mitigation
potential from the livestock sector in Europe is between
101 and 377 Mt CO
2
e, equivalent to between 12 and
61% of EU27 emissions. It would also equate to
between 0.5 and 2% of total global projected GHG emis-
sion increases in the baseline scenario for 2050 (Stehfest
et al., 2009). This demonstrates that GHG emission
reductions in the European livestock sector could have
a global impact. A further reduction in the consump-
tion of livestock products, and the inclusion of other
sectors and regions of the world would enable an even
greater impact.
Acknowledgements
We thank Greenpeace International for funding this review. Jan
Peter Lesschen and Adrian Leip acknowledge funding from the
AnimalChange project (EU FP7 Grant Agreement 266018). Pete
Smith is a Royal Society Wolfson Research Merit Award holder
and his input contributes to the EU FP7 project GHG-Europe,
and to ClimateXChange.
References
Bernstad A, la Cour Jansen J (2011) A life cycle approach to the management of
household food waste –a Swedish full-scale case study. Waste Management,31,
1879–1896.
Bouwman AF, Van der Hoek KW, Eickhout B, Soenario I (2005) Exploring changes in
world ruminant production systems. Agricultural Systems,84, 121–153.
Britz W, Witzke P (2008) CAPRI Model Documentation 2008: Version 2. Available at:
www.capri-model.org (accessed 27 July 2012).
Cederberg C, Stadig M (2003) System expansion and allocation in life cycle assess-
ment of milk and beef production. The International Journal of Life Cycle Assessment,
8, 350–356.
Cederberg C, Flysjo
¨A, Sonesson U, Sund V, Davis J (2009a) Greenhouse Gas Emissions
from Swedish Consumption of Meat, Milk and Eggs 1990 and 2005 (SIK Report No794).
The Swedish Institute for Food and Biotechnology, Gothenburg.
Cederberg C, Meyer D, Flysjo
¨A (2009b) Life Cycle Inventory of Greenhouse Gas Emis-
sions and Use of Land and Energy in Brazilian Beef Production (SIK Report No792). The
Swedish Institute for Food and Biotechnology, Gothenburg.
Cederberg C, Persson UM, Neovius K, Molander S, Clift R (2011) Including carbon
emissions from deforestation in the carbon footprint of Brazilian beef. Environmen-
tal Science & Technology,45, 1773–1779.
©2012 Blackwell Publishing Ltd, Global Change Biology, doi: 10.1111/j.1365-2486.2012.02786.x
14 J. BELLARBY et al.

Conant RT, Paustian K, Elliott ET (2001) Grassland management and conversion into
grassland: effects on soil carbon. Ecological Applications,11, 343–355.
De Deyn GB, Shiel RS, Ostle NJ et al. (2011) Additional carbon sequestration benefits
of grassland diversity restoration. Journal of Applied Ecology,48, 600–608.
EC (1999) Council Directive 1999/74/EC of 19 July 1999, laying down minimum
standards for the protection of laying hens (O.J. L 203/53). Available at: http://
eur-lex.europa.eu/JOIndex.do (accessed 27 July 2012).
EC (2010) Preparatory Study on Food Waste Across EU27. European Commission
(DG ENV), Available at: http://ec.europa.eu/environment/eussd/reports.htm
(accessed 27 July 2012).
Eckard RJ, Grainger C, de Klein CAM (2010) Options for the abatement of methane
and nitrous oxide from ruminant production: a review. Livestock Science,130,
47–56.
EEA (2011) Annual European Union Greenhouse Gas Inventory 1990–2009 and Inventory
Report 2011. Submission to the UNFCCC Secretariat. Technical report No. 2/2011.
European Environment Agency, Copenhagen, Denmark.
Fairlie S (2010) Meat: A Benign Extravagance. Permanent Publications, East Meon, UK.
FAO (2006) World Agriculture: Towards 2030, 2050. Prospects for Food, Nutrition, Agricul-
ture and Major Commodity Groups. Food and Agriculture Organization of the Uni-
ted Nations, Global Perspectives Studies Unit, Rome.
FAOSTAT (2007) Food and Agriculture Organization of the United Nations, FAOSTAT
Database. Available at: http://faostat.fao.org/site/291/default.aspx (accessed 24
February 2012).
Flysjo
¨A, Henriksson M, Cederberg C, Ledgard S, Englund J-E (2011) The impact of
various parameters on the carbon footprint of milk production in New Zealand
and Sweden. Agricultural Systems,104, 459–469.
Foley JA, Ramankutty N, Brauman KA et al. (2011) Solutions for a cultivated planet.
Nature,478, 337–342.
Fornara DA, Steinbeiss S, McNamara NP et al. (2011) Increases in soil organic carbon
sequestration can reduce the global warming potential of long-term liming to
permanent grassland. Global Change Biology,17, 1925–1934.
Frelih-Larson A, Leipprand A, Naumann S, Beucher O (2008) Deliverable D11: Climate
change mitigation through agricultural techniques. Policy recommendations. Report pre-
pared for PICCMAT: Policy Incentives for Climate Change Mitigation Agricultural
Techniques within the Sixth Framework Programme for Research. European Com-
mission, Brussels.
Friel S, Dangour AD, Garnett T et al. (2009) Public health benefits of strategies to
reduce greenhouse-gas emissions: food and agriculture. The Lancet,374, 2016–
2025.
Garnett T (2007) Meat and Dairy Production and Consumption. Exploring the Livestock
Sector’s Contribution to the UK’s Greenhouse Gas Emissions and Assessing What Less
Greenhouse Gas Intensive Systems of Production and Consumption Might Look Like
(Working paper). Food Climate Research Network, University of Surrey, UK.
Garnett T (2009) Livestock-related greenhouse gas emissions: impacts and options for
policy makers. Environmental Science & Policy,12, 491–503.
Gill M, Smith P, Wilkinson JM (2010) Mitigating climate change: the role of domestic
livestock. Animal,4, 323–333.
Gru
¨nberg J, Nieberg H, Schmidt TG (2010) Carbon footprints of food: a critical reflec-
tion. Landbauforschung Volkenrode,60,53–72.
Hirschfeld J, Weiss J, Preidl M, Korbun T (2008) Klimawirkung der Landwirtschaft in
Deutschland.Institut fu
¨ro
¨kologische Wirtschaftsforschung GmbH, Berlin, Germany.
Ho
¨rtenhuber S, Lindenthal T, Amon B, Markut T, Kirner L, Zollitsch W (2010) Green-
house gas emissions from selected Austrian dairy production systems –model cal-
culations considering the effects of land use change. Renewable Agriculture and Food
Systems,25, 316–329.
Houghton RA (2003) Revised estimates of the annual net flux of carbon to the atmo-
sphere from changes in land use and land management 1850–2000. Tellus B,55,
378–390.
IEEP (2007) Final Report for the Study on HNV Indicators Forevaluation. Institute for Euro-
pean Environmental Policy. Contract Notice 2006-G4-04. Report prepared by the
Institute for European Environmental Policy for DG Agriculture, Brussels. Avail-
able at: http://www.ieep.eu/publications/2007/12/final-report-for-the-study-on-
hnv-indicators-for-evaluation (accessed 28 July 2012).
IPCC (2006) 2006 IPCC Guidelines for National Greenhouse Gas Inventories. Prepared by
the National Greenhouse Gas Inventories Programme, (eds Eggleston HS, Buendia
L, Miwa K, Ngara T, Tanabe K). IGES, Japan.
Janssens IA, Freibauer A, Schlamadinger B et al. (2005) The carbon budget of terres-
trial ecosystems at country-scale –a European case study. Biogeosciences,2,15–26.
Lesschen JP, van dBM, Westhoek HJ, Witzke HP, Oenema O (2011) Greenhouse gas
emission profiles of European livestock sectors. Animal Feed Science and Technology,
166–167,16–28.
Loudjani P, Wassenaar T, Grandgirard D (2010) Typology of livestock production
system in Europe. In: Evaluation of the Livestock Sector’s Contribution to the EU Green-
house Gas Emissions (GGELS) –Final Report (eds Leip A, Weiss F, Wassenaar T,
Perez I, Fellmann T, Loudjani P, Tubiello F, Grandgirard D, Monni S, Biala K), pp.
61–70, JRC European Commission, Ispra, Italy.
Minchin W, Buckley F, Kenny DA, Monahan FJ, Shalloo L, O’Donovan M (2010) An
evaluation of over-wintering feeding strategies prior to finishing at pasture for cull
dairy cows on live animal performance, carcass and meat quality characteristics.
Meat Science,85, 385–393.
Ministry for the Environment (2008) New Zealand’s Greenhouse Gas Inventory 1990–
2006. Fulfilling reporting requirements under the United Nations Framework Convention
on Climate Change and the Kyoto Protocol. Ministry of the Environment, Wellington,
New Zealand.
Monni S, Wassenaar T (2010) Estimation of emissions of imported animal products.
In:Evaluation of the Livestock Sector’s Contribution to the EU Greenhouse Gas Emissions
(GGELS) –Final Report (eds Leip A, Weiss F, Wassenaar T, Perez I, Fellmann T,
Loudjani P, Tubiello F, Grandgirard D, Monni S, Biala K), pp. 280–300, JRC Euro-
pean Commission, Ispra, Italy.
Nguyen TLT, Hermansen JE, Mogensen L (2010) Environmental consequences of
different beef production systems in the EU. Journal of Cleaner Production,18,
756–766.
O’Brien D, Shalloo L, Grainger C, Buckley F, Horan B, Wallace M (2010) The
influence of strain of Holstein-Friesian cow and feeding system on greenhouse
gas emissions from pastoral dairy farms. Journal of Dairy Science,93, 3390–
3402.
O’Mara FP (2011) The significance of livestock as a contributor to global greenhouse
gas emissions today and in the near future. Animal Feed Science and Technology,166
–167,7–15.
Parfitt J, Barthel M, Macnaughton S (2010) Food waste within food supply chains:
quantification and potential for change to 2050. Philosophical Transactions of the
Royal Society B: Biological Sciences,365, 3065–3081.
Pelletier N (2008) Environmental performance in the US broiler poultry sector: life
cycle energy use and greenhouse gas, ozone depleting, acidifying and eutrophying
emissions. Agricultural Systems,98,67–73.
Powlson DS, Whitmore AP, Goulding KWT (2011) Soil carbon sequestration to miti-
gate climate change: a critical re-examination to identify the true and the false.
European Journal of Soil Science,62,42–55.
Reinhardt G, Ga
¨rtner S, Mu
¨nch J, Ha
¨fele S (2009) O
¨kologische Optimierung regional
erzeugter Lebensmittel: Energie- und Klimagasbilanzen. ifeu–Institut fu
¨r Energie- und
Umweltforschung Heidelbergh GmbH, Hedeilberg.
Schulze ED, Luyssaert S, Ciais P et al. (2009) Importance of methane and
nitrous oxide for Europe’s terrestrial greenhouse-gas balance. Nature Geoscience,
84,2–850.
Sims REH, Hastings A, Schlamadinger B, Taylor G, Smith P (2006) Energy crops: cur-
rent status and future prospects. Global Change Biology,12, 2054–2076.
Smith P, Martino D, Cai Z et al. (2007) Policy and technological constraints to imple-
mentation of greenhouse gas mitigation options in agriculture. Agriculture, Ecosys-
tems and Environment,118,6–28.
Smith P, Gregory PJ, Van Vuuren D et al. (2010) Competition for land. Philosophical
Transactions of the Royal Society B: Biological Sciences,365, 2941–2957.
Soussana JF, Loiseau P, Vuichard N, Ceschia E, Balesdent J, Chevallier T, Arrouays D
(2004) Carbon cycling and sequestration opportunities in temperate grasslands.
Soil Use and Management,20, 219–230.
Soussana J-F, Allard V, Pilegaard K et al. (2007) Full accounting of the greenhouse gas
(CO2, N2O, CH4) budget of nine European grassland sites. Agriculture, Ecosystems
& Environment,121, 121–134. doi: 10.1016/j.agee.2006.12.022.
Soussana JF, Tallec T, Blanfort V (2010) Mitigating the greenhouse gas balance of
ruminant production systems through carbon sequestration in grasslands. Animal,
4, 334–350.
Stehfest E, Bouwman L, Vuuren DP, Elzen MGJ, Eickhout B, Kabat P (2009) Climate
benefits of changing diet. Climatic Change,95,83–102.
Steinbeiss S, Beßler H, Engels C et al. (2008) Plant diversity positively affects short-
term soil carbon storage in experimental grasslands. Global Change Biology,14,
2937–2949.
Steinfeld H, Gerber P, Wassenaar T, Castel V, Rosales M, de Haan C (2006) Livestock’s
Long Shadow. FAO, Rome, Italy.
Thomassen MA, van Calker KJ, Smits MCJ, Iepema GL, de Boer IJM (2008) Life cycle
assessment of conventional and organic milk production in the Netherlands.
Agricultural Systems,96,95–107.
Tilman D, Hill J, Lehman C (2006) Carbon-negative biofuels from low-input high-
diversity grassland biomass. Science,314, 1598–1600.
©2012 Blackwell Publishing Ltd, Global Change Biology, doi: 10.1111/j.1365-2486.2012.02786.x
EU27 LIVESTOCK GHG EMISSIONS AND MITIGATION 15

Tubiello F, Loudjani P (2010) Technological abatement measures for livestock rearing
emissions. In: Evaluation of the Livestock Sector’s Contribution to the EU Greenhouse
Gas Emissions (GGELS) –Final Report (eds Leip A, Weiss F, Wassenaar T, Perez I,
Fellmann T, Loudjani P, Tubiello F, Grandgirard D, Monni S, Biala K), pp. 190–
199. JRC European Commission, Ispra, Italy.
Verge
´XPC, Dyer JA, Desjardins RL, Worth D (2009) Long-term trends in greenhouse
gas emissions from the Canadian poultry industry. Journal of Applied Poultry
Research,18, 210–222.
Vestergaard M, Madsen NT, Bligaard HB, Bredahl L, Rasmussen PT, Andersen HR
(2007) Consequences of two or four months of finishing feeding of culled dry dairy
cows on carcass characteristics and technological and sensory meat quality. Meat
Science,76, 635–643.
de Vries M, de Boer IJM (2010) Comparing environmental impacts for livestock prod-
ucts: a review of life cycle assessments. Livestock Science,128,1–11.
Weidema BP, Wesnæs M, Hermansen J, Kristensen T, Halberg N (2008) Environmental
Improvement Potentials of Meat and Dairy Products (EUR 23491). JRC European
Comission, Seville, Spain.
Weiss F, Leip A (2010a) Methodology for quantification of greenhouse gas and ammo-
nia emissions from the livestock sector the EU-27. In: Evaluation of the Livestock Sec-
tor’s Contribution to the EU Greenhouse Gas Emissions (GGELS) –Final Report (eds
Leip A, Weiss F, Wassenaar T, Perez I, Fellmann T, Loudjani P, Tubiello F, Grand-
girard D, Monni S, Biala K), pp. 71–139. JRC European Commission, Ispra, Italy.
Weiss F, Leip A (2010b) Quantification of the GHG emissions of EU livestock produc-
tion in form of a life cycle assessment (LCA). In: Evaluation of the Livestock Sector’s
Contribution to the EU Greenhouse Gas Emissions (GGELS) –Final Report (eds Leip A,
Weiss F, Wassenaar T, Perez I, Fellmann T, Loudjani P, Tubiello F, Grandgirard D,
Monni S, Biala K), pp. 162–199. JRC European Commission, Ispra, Italy.
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 &
Environment,149, 124–134.
Westhoek H, Rood T, van den Berg M, Janse J, Nijdam D, Reudink M, Stehfest E
(2011) The Protein Puzzle. PBL Netherlands Environmental Assessment Agency,
The Hague.
Wilkinson JM (2011) Re-defining efficiency of feed use by livestock. Animal,5, 1014–
1022.
Williams H, Wikstro
¨m F (2011) Environmental impact of packaging and food losses
in a life cycle perspective: a comparative analysis of five food items. Journal of
Cleaner Production,19,43–48.
Williams AG, Audsley E, Sandars DL (2006) Determining the Environmental Burdens
and Resource Use in the Production of Agricultural and Horticultural Commodities.
Main Report (Defra Research Project No. IS0205). Cranfield University and
Defra, Bedford. Available at: www.silsoe.cranfield.ac.uk, and www.defra.gov.uk
(accessed 25 August 2011).
Wolf O, Pe
´rez-Domı
´nguez I, Rueda-Cantuche JM et al. (2011) Do healthy diets in
Europe matter to the environment? A quantitative analysis. Journal of Policy
Modeling,33,8–28.
WRAP (2009) Household Food and Drink Waste in the UK. Report prepared by WRAP,
Banbury.
WRAP (2010) Waste Arisings in the Supply of Food and Drink to Households in the UK.
Report prepared by WRAP, Banbury.
©2012 Blackwell Publishing Ltd, Global Change Biology, doi: 10.1111/j.1365-2486.2012.02786.x
16 J. BELLARBY et al.