If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Life-cycle assessment (LCA) is the preferred methodology to assess carbon footprint per unit of milk. The objective of this case study was to apply an LCA method to compare carbon footprints of high-performance confinement and grass-based dairy farms. Physical performance data from research herds were used to quantify carbon footprints of a high-performance Irish grass-based dairy system and a top-performing United Kingdom (UK) confinement dairy system. For the US confinement dairy system, data from the top 5% of herds of a national database were used. Life-cycle assessment was applied using the same dairy farm greenhouse gas (GHG) model for all dairy systems. The model estimated all on- and off-farm GHG sources associated with dairy production until milk is sold from the farm in kilograms of carbon dioxide equivalents (CO2-eq) and allocated emissions between milk and meat. The carbon footprint of milk was calculated by expressing GHG emissions attributed to milk per tonne of energy-corrected milk (ECM). The comparison showed that when GHG emissions were only attributed to milk, the carbon footprint of milk from the Irish grass-based system (837 kg of CO2-eq/t of ECM) was 5% lower than the UK confinement system (884 kg of CO2-eq/t of ECM) and 7% lower than the US confinement system (898 kg of CO2-eq/t of ECM). However, without grassland carbon sequestration, the grass-based and confinement dairy systems had similar carbon footprints per tonne of ECM. Emission algorithms and allocation of GHG emissions between milk and meat also affected the relative difference and order of dairy system carbon footprints. For instance, depending on the method chosen to allocate emissions between milk and meat, the relative difference between the carbon footprints of grass-based and confinement dairy systems varied by 3 to 22%. This indicates that further harmonization of several aspects of the LCA methodology is required to compare carbon footprints of contrasting dairy systems. In comparison to recent reports that assess the carbon footprint of milk from average Irish, UK, and US dairy systems, this case study indicates that top-performing herds of the respective nations have carbon footprints 27 to 32% lower than average dairy systems. Although differences between studies are partly explained by methodological inconsistency, the comparison suggests that potential exists to reduce the carbon footprint of milk in each of the nations by implementing practices that improve productivity.
). To achieve this goal in many parts of the developed world, for instance North America, continental Europe, and increasingly in the United Kingdom (UK), dairy producers aim to increase farm revenue by maximizing milk yield per cow. This is typically accomplished by offering cows nutritionally precise diets in confinement and through improving genetic merit (
). Conversely, in some developed countries, notably Ireland and New Zealand, dairy farmers aim to increase profits by minimizing production costs through maximizing the proportion of grazed grass in the diet of lactating cows (
Optimizing resource use has the potential to maximize the profitability of grass-based and confinement dairy systems, and improves the environmental sustainability of milk production (
). Thus, a link exists between economic performance and environmental sustainability. In recent years, there has been an increasing focus on evaluating the environmental effects of milk production systems, particularly in relation to greenhouse gas (GHG) emissions (
). Dairy production is an important source of the dominant GHG emissions, methane (CH4), nitrous oxide (N2O), and carbon dioxide (CO2). Globally, milk production generates 2.7% of GHG emissions, with a further 1.3% caused by meat produced from the dairy herd (
). Recent studies suggest that annual global GHG emissions will have to be cut by up to 80% (relative to 1990 levels) before 2050 to prevent the worst effects of climate change (
Issues related to mitigation in the long term context.
in: Metz B. Davidson O.R. Bosch P.R. Dave R. Meyer L.A. In Climate Change 2007: Mitigation. Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press,
Cambridge, UK2007
). Thus, reducing GHG emissions (carbon footprint) per unit of milk is becoming a necessity for milk producers.
To assess the carbon footprint of milk from contrasting dairy systems, it is necessary to adopt a life cycle approach. This approach, generally referred to as life-cycle assessment (LCA), entails quantifying GHG emissions generated from all stages associated with a product, from raw-material extraction through production, use, recycling, and disposal within the system boundaries (
ISO (International Organization for Standardization)
Environmental management—Life cycle assessment: Principles and framework. ISO 14040:2006. European Committee for Standardization,
Brussels, Belgium2006
ISO (International Organization for Standardization)
Environmental management—Life cycle assessment: Requirements and guidelines. ISO 14044:2006. European Committee for Standardization,
Brussels, Belgium2006
reported the carbon footprint of milk from a high-performing grass-based dairy system was lower than a confinement dairy system exhibiting moderate performance. Conversely,
showed that the carbon footprint of milk from a commercial confinement dairy system with a noted record of environmental stewardship was lower than a recently established grass-based system. Generally, LCA studies not biased by the farms selected to represent grass-based and confinement dairy systems have reported that grass-based systems produce milk with a lower carbon footprint (
Evaluation of the livestock’s sector contribution to the EU greenhouse gas emissions (GGELS)—Final report. European Commission,
Joint Research Centre, Ispra, Italy2010
). However, such studies have only considered average-performing dairy systems. Thus, a need exists to evaluate the carbon footprint of high-performing dairy systems operated at research and commercial farm levels to determine the direction the industry should take to fulfill production and GHG requirements, and to assess their impact on other aspects of the environment, such as fossil fuel depletion and land occupation.
In this study, the primary objective was to compare the carbon footprints of milk from high-performing confinement and intensive grass-based dairy systems using LCA. To achieve this goal, case study farms located in regions accustomed to grass- and confinement-based milk production were selected, namely the United States and UK for confinement dairy systems and Ireland for grass-based milk production. A secondary goal of this study was to assess the effect different LCA modeling methodologies have on the carbon footprints of these contrasting milk production systems.
Materials and Methods
Description of Dairy Farming Systems
This study used data from existing reports, published studies, and databases and required no approval from an animal care and use committee. Physical data (Table 1) for quantifying carbon footprints of milk from the Irish (IRE) grass-based dairy system and UK confinement dairy system were obtained from research studies (
). The data used for the IRE dairy system was based on a study carried out to analyze the effect of stocking rate and genetic potential of cows on various biological and economic components of grass-based farms from 2002 to 2005. The IRE system fed less concentrate than the average or upper quartile of commercial IRE farms in 2011 (590–850 kg of DM/cow;
) and outperformed the top quartile of farms for key technical measures such as milk yield (5,914 kg/cow per year) and milk composition (4.1% fat and 3.5% protein).
Table 1Technical description of a high-performance Irish grass-based dairy system, a high-performance United Kingdom (UK) confinement system, and a top-performing US confinement dairy system
Forage intakes were estimated with the Moorepark Dairy System Model (Shalloo et al., 2004), using milk production, animal BW, concentrate supplementation, and feed ration composition data.
kg of DM/cow per year
4,099
—
—
Alfalfa hay
kg of DM/cow per year
—
—
2,570
Grass silage
kg of DM/cow per year
849
1,142
—
Maize silage
kg of DM/cow per year
—
1,862
2,155
Whole-crop wheat silage
kg of DM/cow per year
—
825
—
Rape straw
kg of DM/cow per year
—
219
—
Total intake
kg of DM/cow per year
5,270
6,953
8,079
On-farm N fertilizer
kg of N/on-farm ha per year
250
106
53
Manure exported
%
0
33
0
1 Off-farm land area required to produce purchased forage and concentrate feedstuffs.
2 Energy-corrected milk = [0.25 + 0.122 × fat (%) + 0.077 × protein (%)] × kilograms of milk (
Evaluation of the livestock’s sector contribution to the EU greenhouse gas emissions (GGELS)—Final report. European Commission,
Joint Research Centre, Ispra, Italy2010
A survey of fertilizer use in Ireland from 2004–2008 for grassland and arable crops. Teagasc, Johnstown Castle Environment Research Centre,
Wexford, Ireland2010
). The technical performance of the UK system was high compared with the upper quartile of commercial herds in the UK in 2011 for milk yield (8,850 kg/cow per year). However, the UK system fed more concentrate than the average or top quartile of farms (2,666–2,684 kg of DM/cow;
), and represented the top 5% of herds in 2010 to 2011 for key technical indicators (e.g., milk yield/cow per year).
IRE Grass-Based Dairy System
Milk production in Ireland is based mainly on seasonal-calving grass-based dairy systems. Therefore, the objective of the IRE dairy system was to maximize utilization of grazed grass in the diet of lactating dairy cows. This was accomplished through a combination of extended grazing (early February to late November), tight calving patterns in early spring, and rotational grazing of pasture (
). Grass silage was harvested in the IRE dairy system when grass growth exceeded herd feed demand, and fed during the housing period with supplementary minerals and vitamins. Overall, the IRE system was self-sufficient for farm-produced forage. Concentrate feed was purchased by the farm and offered to cows at the beginning and end of lactation when forage intake was not sufficient to meet nutritional requirements. The total quantity of concentrate offered was 320 kg of DM per cow. Concentrate was given to cows in equal feeds during morning and evening milking. Cows were milked in a 14-unit herringbone milking parlor. The stocking rate of the system was 2.53 livestock units (LU; equivalent to 550 kg of BW) per hectare (
Replacement heifers were raised on farm in the IRE dairy system and produced their first calf, on average, at 24 mo of age. Heifers primarily grazed pasture, but between November and March, heifers were mainly offered grass silage indoors. Bull calves were sold as early as possible (<3 weeks) in the IRE dairy system. Replacement and cull rates were 18%. The genetics of cows in the IRE dairy system were Holstein-Friesian of New Zealand origin, which were selected over many generations from animals grazing pasture. The genetic potential of each New Zealand Holstein-Friesian trait of economic importance has been reported (
). Average calving interval in the IRE dairy system was 368 d and average annual milk yield per cow was 6,262 kg. The on-farm synthetic N fertilizer input in the IRE dairy system was 250 kg of N/on-farm ha. Manure produced on farm was used for on-farm forage production. The majority of manure was deposited by grazing cattle on pasture. Manure was stored as slurry in tanks during the housing period and spread on grassland mainly in spring.
UK and US Confinement Dairy Systems
Dairy systems increasingly in the UK and United States are based on TMR or partial mixed ration (PMR) diets, where Holstein-Friesian cows typically produce milk all year round. Thus, in the UK and US dairy systems, cows calved throughout the year, were housed full time and fed TMR or PMR. In the UK dairy system, cows were milked individually at automatic (robotic) milking stations. The diet offered was based on data from a UK research herd (
) where cows had ad libitum access to PMR and concentrates were given to cows during milking. In the US dairy system, it was assumed that cows were milked in an 18-unit herringbone parlor. The composition of the TMR in the US system was from the survey of
, which identified corn silage, alfalfa hay, dry ground corn grain, and soybean meal as the typical feedstuffs used in US dairy production. Diets fed in the UK and US dairy systems (Table 2) were formulated to fulfill nutrient requirements and maximize production. The chemical composition of the TMR diets offered were similar to previous studies (
). Maize, grass, and whole-crop cereal silages were grown on farm in the UK dairy system. Alfalfa hay and maize silage were assumed to be grown on farm in the US dairy system. The remaining feed in both systems was treated as purchased feed. The origin of purchased feed used in the UK, US, and IRE dairy systems was based on trade flow data from the Food and Agriculture Organization of the United Nations (
Table 2Formulation and composition of diets fed to lactating Holstein-Friesian dairy cows in the United Kingdom (UK) and US confinement dairy systems and concentrate offered to lactating Holstein-Friesian cows at pasture for the Irish dairy system
Based on FAOSTAT (2012), 95% of the soybean meal in the UK dairy system was from South America and 5% from the United States; for the Irish system, 92% of the soybean meal was from South America and 8% from the United States; for the US system, all soybean meal was from the domestic market.
Megalac = calcium salts of palm oil FA distillate (Volac International Ltd., Royston, UK). Palm oil was sourced from sustainable forest plantations in Malaysia.
Contained 18% Ca, 10% P, 5% Mg, 17% salt, 5,000mg of Cu/kg, 5,000mg of Mn/kg, 100mg of Co/kg, 6,000mg of Zn/kg, 500mg of I/kg, 25mg of Se/kg, 400,000 IU of vitamin A/kg, 80,000 IU of vitamin D3/kg, and 1,000mg of vitamin E/kg.
Contained 18% Ca, 10% P, 5% Mg, 17% salt, 5,000mg of Cu/kg, 5,000mg of Mn/kg, 100mg of Co/kg, 6,000mg of Zn/kg, 500mg of I/kg, 25mg of Se/kg, 400,000 IU of vitamin A/kg, 80,000 IU of vitamin D3/kg, and 1,000mg of vitamin E/kg.
Contained 33% calcium carbonate, 23% dicalcium phosphate, 20% sodium bicarbonate, 13% salt, 7% magnesium oxide, 13,350mg of Cu/kg, 23,990mg of Fe/kg, 51,000mg of Mn/kg, 430mg of Co/kg, 62,010mg of Zn/kg, 1,030mg of I/kg, 320mg of Se/kg, 700,000 IU of vitamin A/kg, 222,000 IU of vitamin D/kg, and 17,600,mg of vitamin E/kg.
Contained 60mg of Se/kg, 700mg of I/kg, 4,000mg of Cu/kg, 5,000mg of Zn/kg, 250,000 IU of vitamin A/kg, 50,000 IU of vitamin D/kg, and 2,000 IU of vitamin E/kg.
, 95% of the soybean meal in the UK dairy system was from South America and 5% from the United States; for the Irish system, 92% of the soybean meal was from South America and 8% from the United States; for the US system, all soybean meal was from the domestic market.
2 Megalac = calcium salts of palm oil FA distillate (Volac International Ltd., Royston, UK). Palm oil was sourced from sustainable forest plantations in Malaysia.
3 Contained 18% Ca, 10% P, 5% Mg, 17% salt, 5,000 mg of Cu/kg, 5,000 mg of Mn/kg, 100 mg of Co/kg, 6,000 mg of Zn/kg, 500 mg of I/kg, 25 mg of Se/kg, 400,000 IU of vitamin A/kg, 80,000 IU of vitamin D3/kg, and 1,000 mg of vitamin E/kg.
4 Contained 33% calcium carbonate, 23% dicalcium phosphate, 20% sodium bicarbonate, 13% salt, 7% magnesium oxide, 13,350 mg of Cu/kg, 23,990 mg of Fe/kg, 51,000 mg of Mn/kg, 430 mg of Co/kg, 62,010 mg of Zn/kg, 1,030 mg of I/kg, 320 mg of Se/kg, 700,000 IU of vitamin A/kg, 222,000 IU of vitamin D/kg, and 17,600, mg of vitamin E/kg.
5 Contained 60 mg of Se/kg, 700 mg of I/kg, 4,000 mg of Cu/kg, 5,000 mg of Zn/kg, 250,000 IU of vitamin A/kg, 50,000 IU of vitamin D/kg, and 2,000 IU of vitamin E/kg.
). Heifers were primarily fed TMR diets in both systems and bull calves were sold within 1 wk. The replacement rate in the UK dairy system was 41% and the cull rate 34%. The discrepancy is because the UK herd was expanding. However, to standardize the comparison between dairy systems, the UK herd was assumed to be static (34%). In the US dairy system, the replacement and cull rate was 38%. The genetics of Holstein-Friesian cows in the UK and US dairy systems were of North American origin (
). The on-farm N fertilizer usage in the UK dairy system was 106 kg of N/on-farm ha and in the US dairy system 53 kg of N/on-farm ha. Manure produced on farm was recycled for forage production in the US dairy system. Approximately 33% of manure produced on farm in the UK dairy system was exported and the remainder used for on-farm forage production. Manure from all animals was stored as slurry in the UK dairy system. In the US dairy system, manure from replacements was stored in a dry lot system and manure from cows was stored in a slurry system.
GHG Modeling
To make the IRE, UK, and US dairy systems as comparable as possible, GHG emissions were calculated using the same dairy farm GHG model (
). The GHG model estimates all known GHG emissions from dairy production: CO2, CH4, N2O, and fluorinated gases. The model uses “cradle-to-gate” LCA to quantify all on- and off-farm GHG sources (e.g., fertilizer, pesticide, and fuel manufacture) associated with milk production up to the farm gate. The GHG model operates in combination with the Moorepark Dairy System Model (MDSM;
). The MDSM is a whole-farm simulation model, which provides input data (animal inventory and feed intakes, among other factors) for the GHG model. The MDSM uses the net energy and ME systems to determine feed requirements (
The GHG model calculates on- and off-farm GHG emissions by combining farm input data from the MDSM with literature GHG emission algorithms (Tables 3 and 4). On-farm emission algorithms for CH4, N2O, and CO2 emissions from sources such as manure storage and crop residues were obtained from Intergovernmental Panel on Climate Change (IPCC) guidelines (
IPCC (Intergovernmental Panel on Climate Change) 2006 IPCC Guidelines for National Greenhouse Inventories.
in: Eggleston H.S. Buendia L. Miwa K. Ngara T. Tanabe K. Agriculture, Forestry and Other Land Use. Vol. 4. Institute for Global Environmental Strategies (IGES),
Hayama, Japan2006
UK greenhouse gas inventory, 1990 to 2010: Annual report for submission under the framework convention on climate change. AEA Technology Plc,
Didcot, UK2012
Greenhouse gas emissions 1990–2010 reported to the United Nations framework convention on climate change. Johnstown Castle Estate, Co. Wexford,
Ireland2012
IPCC (Intergovernmental Panel on Climate Change) 2006 IPCC Guidelines for National Greenhouse Inventories.
in: Eggleston H.S. Buendia L. Miwa K. Ngara T. Tanabe K. Agriculture, Forestry and Other Land Use. Vol. 4. Institute for Global Environmental Strategies (IGES),
Hayama, Japan2006
guidelines, gross energy intake (GEI) used to calculate enteric CH4 emissions excluded GEI from rumen-protected fat supplements (e.g., calcium salts) because they are not fermentable. On-farm emissions of CO2 were limited to fossil fuel combustion and urea and lime application. Short-term biogenic sources and sinks of CO2 such as animals, crops, and manure were considered to be neutral with respect to GHG emissions, given that the
IPCC (Intergovernmental Panel on Climate Change) 2006 IPCC Guidelines for National Greenhouse Inventories.
in: Eggleston H.S. Buendia L. Miwa K. Ngara T. Tanabe K. Agriculture, Forestry and Other Land Use. Vol. 4. Institute for Global Environmental Strategies (IGES),
Hayama, Japan2006
and International Dairy Federation (IDF, 2010) guidelines assume all carbon absorbed by animals, crops, and manure to be quickly released back to the atmosphere through respiration, burning, and decomposition.
Table 3Emission factors used in the baseline scenario of the dairy farm greenhouse gas (GHG) model (
ISO (International Organization for Standardization)
Environmental management—Life cycle assessment: Principles and framework. ISO 14040:2006. European Committee for Standardization,
Brussels, Belgium2006
Country-specific emission factors were used to estimate enteric fermentation CH4 emissions for the Irish seasonal grass-based dairy system (IRE), United Kingdom confinement dairy system (UK) and US confinement dairy system (USA). The remaining emission sources were estimated according to the Intergovernmental Panel on Climate Change guidelines (IPCC, 2006).
Manure VS excreted on pasture × 0.24 × 0.67 × 0.01
kg/yr
Ammonia (NH3-N)
Synthetic N fertilizer
0.1 × N fertilizer
kg/kg of N
Slurry storage
0.4 × slurry N stored
kg/kg of N
Solid manure storage
0.3 × solid manure N stored
kg/kg of N
Manure application
0.2 × (N stored − NH3 storage loss)
kg/kg of N
Grazing returns
0.2 × N excreted on pasture
kg/kg of N
Nitrate (NO3−-N)
N leaching
0.3 × N applied
kg/kg of N
Nitrous oxide (N2O-N)
Grazing returns
0.02 × N excreted on pasture
kg/kg of N
Synthetic N fertilizer
0.01 × N fertilizer
kg/kg of N
Manure application
0.01 × (N stored − N storage loss)
kg/kg of N
Crop residues
0.01 × N crop residues
kg/kg of N
Slurry storage
0.005 × slurry N stored
kg/kg of N
Solid manure storage
0.005 × solid manure N stored
kg/kg of N
Dry lot
0.02 × dry lot manure N stored
kg/kg of N
Nitrate leaching
0.0075 × N leached
kg/kg of NO3−-N
Ammonia redeposition
0.01 × sum of NH3 emissions
kg/kg of NH3-N
CO2
Diesel
2.63 × diesel use
kg/L
Gasoline
2.30 × gasoline use
kg/L
Kerosene
2.52 × kerosene use
kg/L
Lime
0.44 × lime application
kg/kg of lime
Urea
0.73 × urea application
kg/kg of urea
1 Country-specific emission factors were used to estimate enteric fermentation CH4 emissions for the Irish seasonal grass-based dairy system (IRE), United Kingdom confinement dairy system (UK) and US confinement dairy system (USA). The remaining emission sources were estimated according to the Intergovernmental Panel on Climate Change guidelines (
IPCC (Intergovernmental Panel on Climate Change) 2006 IPCC Guidelines for National Greenhouse Inventories.
in: Eggleston H.S. Buendia L. Miwa K. Ngara T. Tanabe K. Agriculture, Forestry and Other Land Use. Vol. 4. Institute for Global Environmental Strategies (IGES),
Hayama, Japan2006
ISO (International Organization for Standardization)
Environmental management—Life cycle assessment: Principles and framework. ISO 14040:2006. European Committee for Standardization,
Brussels, Belgium2006
The baseline scenario and scenario 1 used emission algorithms from the current Intergovernmental Panel on Climate Change guidelines (IPCC, 2006) to estimate emissions from agricultural GHG sources related to the production of feedstuffs.
Scenario 2 applied the same emission factors as the baseline scenario to estimate emissions from non-agricultural products (e.g., electricity), but applied emission algorithms from the original IPCC (1997) guidelines and IPCC (2000) good practice guidelines to estimate emissions from agricultural GHG sources related to the production of feedstuffs.
Scenario 3 used country-specific emission factors to estimate emissions from the manufacture of non-agricultural products and used country-specific emission algorithms to estimate emissions from agricultural GHG sources related to the production of feedstuffs.
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Effects of a perennial ryegrass diet or total mixed ration diet offered to spring-calving Holstein-Friesian dairy cows on methane emissions, dry matter intake, and milk production.
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Effects of a perennial ryegrass diet or total mixed ration diet offered to spring-calving Holstein-Friesian dairy cows on methane emissions, dry matter intake, and milk production.
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Effects of a perennial ryegrass diet or total mixed ration diet offered to spring-calving Holstein-Friesian dairy cows on methane emissions, dry matter intake, and milk production.
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Effects of a perennial ryegrass diet or total mixed ration diet offered to spring-calving Holstein-Friesian dairy cows on methane emissions, dry matter intake, and milk production.
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Effects of a perennial ryegrass diet or total mixed ration diet offered to spring-calving Holstein-Friesian dairy cows on methane emissions, dry matter intake, and milk production.
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Effects of a perennial ryegrass diet or total mixed ration diet offered to spring-calving Holstein-Friesian dairy cows on methane emissions, dry matter intake, and milk production.
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Effects of a perennial ryegrass diet or total mixed ration diet offered to spring-calving Holstein-Friesian dairy cows on methane emissions, dry matter intake, and milk production.
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Effects of a perennial ryegrass diet or total mixed ration diet offered to spring-calving Holstein-Friesian dairy cows on methane emissions, dry matter intake, and milk production.
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Effects of a perennial ryegrass diet or total mixed ration diet offered to spring-calving Holstein-Friesian dairy cows on methane emissions, dry matter intake, and milk production.
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Effects of a perennial ryegrass diet or total mixed ration diet offered to spring-calving Holstein-Friesian dairy cows on methane emissions, dry matter intake, and milk production.
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Effects of a perennial ryegrass diet or total mixed ration diet offered to spring-calving Holstein-Friesian dairy cows on methane emissions, dry matter intake, and milk production.
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Effects of a perennial ryegrass diet or total mixed ration diet offered to spring-calving Holstein-Friesian dairy cows on methane emissions, dry matter intake, and milk production.
IPCC (Intergovernmental Panel on Climate Change) Changes in atmospheric constituents and in radiative forcing.
in: Solomon S. Qin D. Manning M. Chen Z. Marquis M. Averyt K.B. Tignor M. Miller H.L. In Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press,
Cambridge, UK2007
IPCC (Intergovernmental Panel on Climate Change) 2006 IPCC Guidelines for National Greenhouse Inventories.
in: Eggleston H.S. Buendia L. Miwa K. Ngara T. Tanabe K. Agriculture, Forestry and Other Land Use. Vol. 4. Institute for Global Environmental Strategies (IGES),
Hayama, Japan2006
) to estimate emissions from agricultural GHG sources related to the production of feedstuffs.
4 Scenario 2 applied the same emission factors as the baseline scenario to estimate emissions from non-agricultural products (e.g., electricity), but applied emission algorithms from the original
IPCC (Intergovernmental Panel on Climate Change). 2000. Good practice guidance and uncertainty management in national greenhouse gas inventories. J. T. Houghton, L. G. Meira Filho, B. Lim, K. Tréanton, I. Mamaty, Y. Bonduki, D. J. Griggs, and B. A. Callander, ed. Organisation for Economic Co-operation and Development (OECD/OCDE), Paris, France.
good practice guidelines to estimate emissions from agricultural GHG sources related to the production of feedstuffs.
5 Scenario 3 used country-specific emission factors to estimate emissions from the manufacture of non-agricultural products and used country-specific emission algorithms to estimate emissions from agricultural GHG sources related to the production of feedstuffs.
6 Emissions were allocated between coproducts based on their economic value using national,
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Effects of a perennial ryegrass diet or total mixed ration diet offered to spring-calving Holstein-Friesian dairy cows on methane emissions, dry matter intake, and milk production.
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
In addition to animals, crops, and manure, soils also have the potential to emit or sequester CO2. Agricultural soils typically lose carbon following the conversion of land to cropland, but gain carbon during the conversion of cropland to grassland. The rate of soil carbon loss or increase declines over time and typically ceases after 20 yr, once new soil carbon equilibrium is reached (
). Over the past few decades, the grassland area has declined in the regions analyzed, but this area has not been converted to cropland, which has also declined in area (
UK greenhouse gas inventory, 1990 to 2010: Annual report for submission under the framework convention on climate change. AEA Technology Plc,
Didcot, UK2012
Greenhouse gas emissions 1990–2010 reported to the United Nations framework convention on climate change. Johnstown Castle Estate, Co. Wexford,
Ireland2012
, carbon sequestration rates for permanent IRE, UK, and US grassland soils vary from 0.79 to 1.74 t/CO2 per hectare per year, partly due to management practices. However, to compare dairy systems, we used the average annual value of these studies (1.19 t/CO2 per hectare) to estimate carbon sequestration by grassland soil.
Off-farm GHG emissions associated with production and supply of non-agricultural products (e.g., pesticide manufacture) were estimated using emission factors from the Ecoinvent database (http://www.ecoinvent.org/database/) and data from literature sources (Table 4). Emission factors for on-farm sources and purchased non-agricultural products were used in combination with physical data from national statistics (
A survey of fertilizer use in Ireland from 2004–2008 for grassland and arable crops. Teagasc, Johnstown Castle Environment Research Centre,
Wexford, Ireland2010
Methodology used in FeedPrint: A tool quantifying greenhouse gas emissions of feed production and utilization. Wageningen UR, Livestock Research,
Lelystad, the Netherlands2012
) to quantify emission factors for growing and harvesting purchased feedstuffs. Emissions from processing and transporting feedstuffs were estimated using emission factors from
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Methodology used in FeedPrint: A tool quantifying greenhouse gas emissions of feed production and utilization. Wageningen UR, Livestock Research,
Lelystad, the Netherlands2012
. Emission factors for importing feedstuffs were estimated by summing emission factors for the farm, processing, and transportation stages (Table 4).
Emissions from land use change were estimated for South American soybean and Malaysian palm fruit. The approach used to calculate land use change emissions from these crops was taken from
and involved dividing the total land use change emissions for a crop by the total crop area to estimate the average land use change emissions per crop. This resulted in average land use change emissions per hectare from South American soybeans of 2.6 t of CO2 and from Malaysian palm fruit 5.5 t of CO2. For Megalac, which is a calcium salt of palm FA, land use change emissions were not included. This was because the feedstuff is reported to be produced from existing palm forest plantations that do not cause land use change emissions from deforestation (
Outputs of the dairy farm GHG model were a static account of annual on-farm and total (on- and off-farm) GHG emissions in CO2 equivalents (CO2-eq). The
IPCC (Intergovernmental Panel on Climate Change) Changes in atmospheric constituents and in radiative forcing.
in: Solomon S. Qin D. Manning M. Chen Z. Marquis M. Averyt K.B. Tignor M. Miller H.L. In Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press,
Cambridge, UK2007
global warming potentials were used to convert GHG emissions into kilograms of CO2-eq using a 100-yr time horizon, where the global warming potential of CO2 = 1, CH4 = 25, and N2O = 298. The GHG model expresses total GHG emissions as the carbon footprint of milk in kilograms of CO2-eq per tonne of ECM, which, per kilogram of milk, is equivalent to 4% milk fat and 3.3% milk protein (
A Nordic proposal for an energy corrected milk (ECM) formula.
Performance recording of animals: State of the art 1990. EAAP Publication No. 50:156–157. Centre for Agricultural Publishing and Documentation,
Wageningen, the Netherlands1991
Besides producing milk, dairy farms may also export crops, manure, and produce meat from culled cows, male calves, and surplus female calves. Thus, the carbon footprint of dairy systems should be distributed between these outputs. A multitude of methods are recommended by various LCA and carbon footprint guidelines to allocate GHG emissions among the coproducts of multifunctional systems (
ISO (International Organization for Standardization)
Environmental management—Life cycle assessment: Principles and framework. ISO 14040:2006. European Committee for Standardization,
Brussels, Belgium2006
IDF (International Dairy Federation). A common carbon footprint for dairy.
The IDF guide to standard lifecycle assessment methodology for the dairy industry. Bulletin of the International Dairy Federation 445,
IDF, Brussels, Belgium2010
). The dairy farm GHG model applies different allocation approaches based on the various guidelines and previous LCA studies of milk.
Allocation of GHG emissions to exported crops was avoided by delimiting the dairy farm GHG model to consider only emissions from crops grown for dairy cattle raised on farm. The system expansion method recommended by the IDF (2010) LCA guidelines was followed to attribute emissions to exported manure. The method assumes that exported manure displaces synthetic fertilizer emissions, but allocates no storage emissions to exported manure. Several methods exist to distribute GHG emissions between milk and meat. The following allocation methods were evaluated:
(1)
Milk: no allocation to meat; all GHG emissions were attributed to milk.
(2)
Mass: the GHG emissions of the dairy system were attributed between coproducts according to the mass of milk and meat sold.
(3)
Economic: allocation of GHG emissions between milk and meat was based on revenue received for milk and meat (sales of culled cows and surplus calves). Prices of milk and animal outputs were estimated using the 2006 to 2010 market average (
Protein: edible protein in milk and meat was used to allocate GHG emissions. The protein content of milk was estimated based on Table 1 and the protein content of meat was assumed to be 20% of carcass weight equivalent (
Biological: the GHG emissions of the dairy system was allocated based on feed energy required for producing milk and meat. The IDF (2010) guidelines and the MDSM (
) were used to estimate feed energy required to produce milk and meat.
(6)
Emission: the GHG emissions associated with producing surplus calves, dairy females <24 mo of age, and finishing culled cows were allocated to meat, with the remaining emissions assigned to milk (
Les gaz à effect de serre en élevage bovin: Evaluation et leviers d’action (Greenhouse gases in cattle breeding: Evaluation and mitigation strategies).
System expansion: this approach assumes that meat from culled cows and surplus dairy calves raised for meat replaces meat from alternative meat production systems (
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.
). In general, meat from traditional cow-calf beef systems is considered as the alternative method of producing meat from a dairy system. The first step of the approach uses LCA to estimate GHG emissions from surplus dairy calves raised for meat and was calculated using the emission factors of the GHG model where relevant (Tables 3 and 4) and physical data from
for the US system. The GHG emissions from raised surplus dairy calves were then added to the dairy systems GHG emissions. Subsequently, the meat produced by culled cows and surplus calves raised for meat was summed to estimate the total quantity of meat produced from the dairy system, which was multiplied by the average GHG emissions per kilogram of meat from cow-calf beef systems. This estimates the displaced GHG emissions from traditional cow-calf meat production, which was subtracted from the emissions generated by the dairy system cows, replacements, and surplus dairy calves raised for beef, to estimate GHG emissions per unit of milk. The GHG emissions per kilogram of meat from traditional cow-calf beef systems were calculated according to LCA, using the emission factors of the GHG model where applicable and using physical data and emission factors from
Allocation of GHG emissions was also required for concentrate feeds that are coproducts (e.g., maize gluten feed). The economic allocation procedure described by the IDF (2010) LCA guidelines was used to allocate GHG emissions between concentrate coproducts. National reports,
Methodology used in FeedPrint: A tool quantifying greenhouse gas emissions of feed production and utilization. Wageningen UR, Livestock Research,
Lelystad, the Netherlands2012
) were used to estimate concentrate coproduct yields and average prices.
Scenario Modeling
To assess variability in the emission algorithms of the base dairy farm system described (Tables 3 and 4), the carbon footprint per unit of milk was tested via scenario modeling. The following scenarios were tested relative to the base dairy farm system or baseline scenario:
•
Scenario 1 (S1): enteric CH4 emissions of all dairy systems in S1 were estimated according to the default
IPCC (Intergovernmental Panel on Climate Change) 2006 IPCC Guidelines for National Greenhouse Inventories.
in: Eggleston H.S. Buendia L. Miwa K. Ngara T. Tanabe K. Agriculture, Forestry and Other Land Use. Vol. 4. Institute for Global Environmental Strategies (IGES),
Hayama, Japan2006
guidelines, which estimate enteric CH4 emissions as 6.5% of GEI and include GEI of fat supplements. The remaining emissions sources were estimated using the same algorithms as the baseline scenario.
IDF (International Dairy Federation). A common carbon footprint for dairy.
The IDF guide to standard lifecycle assessment methodology for the dairy industry. Bulletin of the International Dairy Federation 445,
IDF, Brussels, Belgium2010
IPCC (Intergovernmental Panel on Climate Change). 2000. Good practice guidance and uncertainty management in national greenhouse gas inventories. J. T. Houghton, L. G. Meira Filho, B. Lim, K. Tréanton, I. Mamaty, Y. Bonduki, D. J. Griggs, and B. A. Callander, ed. Organisation for Economic Co-operation and Development (OECD/OCDE), Paris, France.
good practice guidelines were applied to estimate emissions from on- and off-farm agricultural activities (Supplementary Table S1, available online at http://dx.doi.org/10.3168/jds.2013-7174). Emissions from non-agricultural activities (e.g., pesticide manufacture) were estimated using the same emissions factors as the baseline scenario (Table 4).
•
Scenario 3 (S3): country-specific emission algorithms from national GHG inventories (
UK greenhouse gas inventory, 1990 to 2010: Annual report for submission under the framework convention on climate change. AEA Technology Plc,
Didcot, UK2012
Greenhouse gas emissions 1990–2010 reported to the United Nations framework convention on climate change. Johnstown Castle Estate, Co. Wexford,
Ireland2012
On-Farm GHG Emissions and Carbon Footprint of Milk from Dairy Systems
Table 5 shows GHG profiles, on-farm GHG emissions, and carbon footprints (on- and off-farm GHG emissions) per tonne of ECM, with no allocation of GHG emissions to meat, for the IRE, UK, and US dairy systems. On-farm GHG emissions per tonne of ECM were lowest for the UK confinement dairy system, 12% greater for the IRE grass-based dairy system, and 14% greater for the US confinement dairy system. Carbon footprint per tonne of ECM was lowest for the IRE grass-based dairy system, 6% greater for the UK confinement dairy system, and 7% greater for the US confinement dairy system.
All GHG emissions associated with the dairy production system up to the point milk is sold from the farm are expressed in kilograms of CO2-equivalents per tonne of ECM.
with all greenhouse gas (GHG) emissions attributed to milk of a high-performance Irish grass-based dairy system, a high-performance confinement United Kingdom (UK) dairy system, and a top-performing confinement US dairy system calculated using a life-cycle assessment dairy farm GHG model (
ISO (International Organization for Standardization)
Environmental management—Life cycle assessment: Principles and framework. ISO 14040:2006. European Committee for Standardization,
Brussels, Belgium2006
The baseline scenario used fixed emission factors to estimate emissions from non-agricultural inputs (e.g., fossil fuel) and used the Intergovernmental Panel on Climate Change guidelines (IPCC, 2006) to estimate emissions from agricultural GHG sources, except for enteric fermentation, where country-specific approaches were applied.
Scenario 1: fixed emission factors were used to estimate emissions from non-agricultural inputs and emission algorithms from the IPCC (2006) guidelines were applied to estimate emissions from agricultural GHG sources.
Scenario 2: fixed emission factors were used to estimate emissions from non-agricultural inputs, and emission algorithms from the original IPCC (1997) guidelines and IPCC (2000) good practice guidelines were used to estimate emissions from agricultural GHG sources.
Scenario 3: country-specific emission factors were applied to estimate emissions from the manufacture of non-agricultural inputs and from agricultural GHG sources.
No Seq = carbon sequestration by permanent grassland was excluded.
kg of CO2-eq/t of ECM
On farm
771
630
705
0.4
6.2
6.1
14.0
29.2
31.2
−11.3
−7.5
−8.7
CFP no Seq, kg of CO2-eq/t of ECM
Total
914
895
898
0.4
4.4
4.8
12.1
23.2
28.4
−9.4
−4.1
−9.4
1 All GHG emissions associated with the dairy production system up to the point milk is sold from the farm are expressed in kilograms of CO2-equivalents per tonne of ECM.
2 The baseline scenario used fixed emission factors to estimate emissions from non-agricultural inputs (e.g., fossil fuel) and used the Intergovernmental Panel on Climate Change guidelines (
IPCC (Intergovernmental Panel on Climate Change) 2006 IPCC Guidelines for National Greenhouse Inventories.
in: Eggleston H.S. Buendia L. Miwa K. Ngara T. Tanabe K. Agriculture, Forestry and Other Land Use. Vol. 4. Institute for Global Environmental Strategies (IGES),
Hayama, Japan2006
IPCC (Intergovernmental Panel on Climate Change) 2006 IPCC Guidelines for National Greenhouse Inventories.
in: Eggleston H.S. Buendia L. Miwa K. Ngara T. Tanabe K. Agriculture, Forestry and Other Land Use. Vol. 4. Institute for Global Environmental Strategies (IGES),
Hayama, Japan2006
IDF (International Dairy Federation). A common carbon footprint for dairy.
The IDF guide to standard lifecycle assessment methodology for the dairy industry. Bulletin of the International Dairy Federation 445,
IDF, Brussels, Belgium2010
IPCC (Intergovernmental Panel on Climate Change). 2000. Good practice guidance and uncertainty management in national greenhouse gas inventories. J. T. Houghton, L. G. Meira Filho, B. Lim, K. Tréanton, I. Mamaty, Y. Bonduki, D. J. Griggs, and B. A. Callander, ed. Organisation for Economic Co-operation and Development (OECD/OCDE), Paris, France.
good practice guidelines were used to estimate emissions from agricultural GHG sources.
5 Scenario 3: country-specific emission factors were applied to estimate emissions from the manufacture of non-agricultural inputs and from agricultural GHG sources.
6 CO2-eq = CO2 equivalent, where CO2 = 1, CH4 = 25, and N2O = 298 (
IPCC (Intergovernmental Panel on Climate Change) Changes in atmospheric constituents and in radiative forcing.
in: Solomon S. Qin D. Manning M. Chen Z. Marquis M. Averyt K.B. Tignor M. Miller H.L. In Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press,
Cambridge, UK2007
Evaluation of the livestock’s sector contribution to the EU greenhouse gas emissions (GGELS)—Final report. European Commission,
Joint Research Centre, Ispra, Italy2010
The GHG profiles of Table 5 show that the main sources of GHG emissions from the IRE dairy system were enteric CH4 (47%), N2O emissions from manure deposited on pasture by grazing cattle (15%), CO2 and N2O emissions from fertilizer application (12%), GHG emissions from fertilizer production (8%), and CH4 and N2O emissions from manure storage and spreading (8%). The key sources of GHG emissions from the UK dairy system were enteric CH4 (42%), CH4 emissions from manure storage (13%), GHG emissions from imported concentrate feed (12%), N2O emissions from manure storage and spreading (9%), CO2 emissions from electricity generation and fuel combustion (7%), and CO2 emissions from land use change (6%). The main sources of GHG emissions from the US dairy system were enteric CH4 (42%), N2O emissions from manure storage and spreading (17%), CH4 emissions from manure storage (14%), GHG emissions from imported concentrate feed (12%), and CO2 emissions from electricity generation and fuel combustion (8%).
The GHG profiles also show that sequestration by grassland soil had no effect or a minor mitigating effect on GHG emissions of the UK and US dairy systems (0–1%), but had a large effect on the IRE dairy system (9%). Thus, excluding carbon sequestration affected the ranking and relative difference between dairy systems in on-farm GHG emissions and carbon footprint per tonne of ECM. The analysis showed that when carbon sequestration was excluded, on-farm GHG emissions per tonne of ECM were lowest for the UK confinement dairy system, 12% greater for the US confinement dairy system, and 22% greater for the IRE grass-based dairy system. Excluding carbon sequestration resulted in the confinement and grass-based dairy systems emitting a similar carbon footprint per tonne of ECM.
Allocation of GHG Emissions Between Milk and Meat
The effects of using different methods to allocate GHG emissions between milk and meat on the carbon footprint per tonne of ECM for the IRE, UK, and US dairy systems are shown in Figure 1. Within the dairy systems, a difference of up to 41% was detected in the proportion of dairy system GHG emissions that were allocated to milk, depending on the methodology used. Excluding attributing all GHG emissions to milk, mass allocation attributed the most GHG emissions to milk, followed by protein, economic allocation, biological energy, emissions allocation, and system expansion.
Figure 1The effect of different methods of allocating greenhouse gas (GHG) emissions between milk and meat on the carbon footprint [kg of CO2 equivalents (CO2e)/t of ECM, with carbon sequestration] in a high-performance Irish (IRE) grass-based dairy system, a high-performance United Kingdom (UK) confinement dairy system, and a top-performing US confinement dairy system (USA). Milk = all GHG emissions were allocated to milk; Mass = mass of milk and meat was used to allocate GHG emissions; Economic = economic value of milk and meat sold was used to allocate GHG emissions; Protein = edible protein in milk and meat was used to allocate GHG emissions; Biological = feed energy required for producing milk and meat was used to allocate GHG emissions; Emission = the GHG emissions associated with surplus calves, dairy females <2 yr of age, and from finishing cows was allocated to meat, with the remainder allocated to milk; System expansion = assumes meat from milk production substitutes emissions generated by meat from traditional cow-calf beef systems. Color version available in the online PDF.
The comparison of allocation methods shows that mass and protein allocation attributed a fixed proportion of GHG emissions to milk for each dairy system: 98 and 94%, respectively. Thus, the ranking and relative difference between dairy systems carbon footprint per tonne of ECM was unchanged compared with attributing no GHG emissions to meat. The proportion of GHG emissions allocated to the carbon footprint of ECM varied between dairy systems for economic, biological, and emission allocation methods. For instance, allocation on an emission basis attributed 85% of GHG emissions to milk for IRE dairy system, 84% for the UK dairy system, and 81% for the US dairy system. This resulted in the UK dairy system, instead of the US dairy system, having the highest carbon footprint per tonne of ECM. Thus, the ranking of dairy systems’ carbon footprint per tonne of ECM was inconsistent between allocation methods.
System expansion did not affect the ranking of dairy systems carbon footprint per tonne of ECM, but the approach led to a significantly greater relative difference between the carbon footprints of grass-based and confinement dairy systems compared with the other allocation methods analyzed. The approach showed that the IRE grass-based system had a carbon footprint per tonne of ECM 19% lower than the UK confinement system and 22% lower than the US confinement dairy system.
Scenario Analysis
The results of S1 (Table 5) showed that applying the general
IPCC (Intergovernmental Panel on Climate Change) 2006 IPCC Guidelines for National Greenhouse Inventories.
in: Eggleston H.S. Buendia L. Miwa K. Ngara T. Tanabe K. Agriculture, Forestry and Other Land Use. Vol. 4. Institute for Global Environmental Strategies (IGES),
Hayama, Japan2006
guidelines to estimate enteric CH4 emissions as 6.5% of GEI (with GEI from fat supplements included) increased carbon footprints per tonne of ECM of the confinement dairy systems by 4 to 5% compared with the baseline scenario. However, using this approach to estimate enteric CH4 emissions had little effect on carbon footprint per tonne of ECM (<1%) of the grass-based dairy system, because enteric CH4 emissions were estimated as 6.45% of GEI in the baseline scenario. Thus, the relative difference between grass-based and confinement dairy systems carbon footprint per tonne of ECM was greater in S1 than the baseline scenario.
IDF (International Dairy Federation). A common carbon footprint for dairy.
The IDF guide to standard lifecycle assessment methodology for the dairy industry. Bulletin of the International Dairy Federation 445,
IDF, Brussels, Belgium2010
IPCC (Intergovernmental Panel on Climate Change). 2000. Good practice guidance and uncertainty management in national greenhouse gas inventories. J. T. Houghton, L. G. Meira Filho, B. Lim, K. Tréanton, I. Mamaty, Y. Bonduki, D. J. Griggs, and B. A. Callander, ed. Organisation for Economic Co-operation and Development (OECD/OCDE), Paris, France.
) emission algorithms for agricultural sources increased estimates of CH4 emissions from manure storage, GHG emissions from concentrate production, and N2O emissions from manure and fertilizer compared with the baseline scenario. The increase in N2O emissions from on-farm fertilizer use was greater for the grass-based dairy system than for the confinement dairy systems in S2. However, the increase in CH4 emissions from manure storage and GHG emissions from concentrate production was greater for the confinement dairy systems than for the grass-based dairy system. In addition, S2 increased enteric CH4 emissions from the confinement dairy systems, but had the opposite effect on the grass-based dairy system. As a result, S2 caused a greater increase in the carbon footprints per tonne of ECM of the confinement dairy systems (24–28%) than the grass-based dairy system (13%) relative to the baseline scenario.
The country-specific emission algorithms of S3 reduced N2O emissions from manure excreted by grazing cattle, and CH4 and N2O emissions from manure storage and spreading relative to the baseline. In addition, S3 estimated lower GHG emissions from concentrate and fertilizer production for the US dairy system. However, the scenario had no effect or increased emissions from concentrate and on-farm fertilizer use for the IRE and UK dairy systems. This resulted in the country-specific emission algorithms of S3 reducing the carbon footprint of the UK dairy system by 4% relative to the baseline, but by 9 to 10% for the IRE and US dairy systems. Consequently, the order of carbon footprints per tonne of ECM of dairy systems in S3 was not consistent with the baseline scenario.
Discussion
Life-cycle assessment studies that directly compare carbon footprints of milk from high-performance grass-based and confinement dairy systems within or across countries are rare. The direct comparison in this study, therefore, provided a unique opportunity to evaluate the effect that contrasting high-performance dairy systems have on the carbon footprint of milk and individual GHG sources. The results implied that high-performance grass-based systems are capable of having a lower carbon footprint per unit of milk compared with top-performing confinement dairy systems. However, this difference was principally due to the inclusion of carbon sequestration, which confers a degree of uncertainty upon the conclusions due to the lack of solid sequestration data available. The ranking of the carbon footprint of milk from high-performance grass-based and confinement dairy systems was also influenced by LCA modeling choices (e.g., allocation methods and emissions algorithms). This agrees with the outcomes of previous research (
) and implies a need to agree to a uniform LCA methodology for milk production. It is also important to emphasize that all physical data used in this study were a snapshot in time and changes in feeding systems and performance could alter the conclusions.
Comparison of GHG Emissions and Carbon Footprint of Milk from High-Performance Grass-Based and Confinement Dairy Systems
Evaluation of the livestock’s sector contribution to the EU greenhouse gas emissions (GGELS)—Final report. European Commission,
Joint Research Centre, Ispra, Italy2010
), the main source of GHG emissions, enteric CH4, was greater per LU from the confinement dairy systems than the grass-based dairy system, but lower per unit of milk. The greater milk yield per cow and higher replacement rate within the confinement systems explained the greater enteric CH4 emissions per LU, because these factors increase DMI per LU, which is a key determinant of enteric CH4 emissions (
Effects of a perennial ryegrass diet or total mixed ration diet offered to spring-calving Holstein-Friesian dairy cows on methane emissions, dry matter intake, and milk production.
). Milk yield per cow was greater in the confinement systems than the grass-based system, given the greater genetic selection for milk yield and increased levels of concentrate feeding. These factors also explained the lower enteric CH4 emissions per unit of milk of the confinement dairy systems, because concentrate-rich diets generally contain less fiber than forage diets and improving genetic merit increases productivity, which facilitates the dilution of maintenance effect whereby the resource cost per unit of milk is reduced (
agreed with our analysis that increasing milk yield reduces enteric CH4 emissions per unit of milk and showed that at similar annual milk yields, improving the fertility of dairy cows decreases enteric CH4 emissions per unit of milk. This was because improving cow fertility reduces the number of replacement heifers required to maintain the herd size for a given milk quota or number of cows, which reduces enteric CH4 emissions. The results of
also partially explain why the lower replacement rate of the UK confinement dairy system resulted in similar enteric CH4 emissions per unit of milk as the US confinement dairy system, even though annual ECM yield per cow was 10% greater in the US dairy system.
Another key reason that explained the similar enteric CH4 emissions per unit of milk of the confinement systems was the different diets fed. Unlike the diet fed in the US system, the formulation of the diet of cows in the UK system included protected lipids, which, compared with forage and most concentrate feeds, reduce enteric CH4 emissions, because protected lipids are not fermentable in the rumen (
). In addition, they slightly increased the feed efficiency (kg of DM/unit of milk) of the UK dairy system relative to the US dairy system, which partly led to the UK and US systems emitting similar enteric CH4 emissions per unit of milk. However, in contrast to the UK system, the diet of cows in the US system was formulated based on a national survey of common feedstuffs (
). Thus, the US diet may not truly reflect high-performance systems, which would also explain in part the difference in feed efficiency between confinement dairy systems.
The greater feed efficiency of the UK confinement system also, in part, reduced GHG emissions from manure storage and on-farm feed production, which resulted in lower on-farm GHG emissions per unit of milk relative to the US confinement system. This was because feed intake is a key determinant of GHG emissions from these sources (
IPCC (Intergovernmental Panel on Climate Change) 2006 IPCC Guidelines for National Greenhouse Inventories.
in: Eggleston H.S. Buendia L. Miwa K. Ngara T. Tanabe K. Agriculture, Forestry and Other Land Use. Vol. 4. Institute for Global Environmental Strategies (IGES),
Hayama, Japan2006
). Manure from all animals was managed in a liquid system for the UK confinement system, but for the US confinement system, manure from replacements was managed in a dry lot. This caused the US system to emit greater N2O emissions and, therefore, greater GHG emissions per unit of milk from manure storage. On-farm GHG emissions per unit of milk were also greater from the US system relative to the UK system, because the US system recycled all manure on-farm to produce forage for ruminants, but the UK system exported one-third of the manure produced to stay within European regulations for slurry application in a nitrate-vulnerable zone. Furthermore, the manure exported from the UK system was assumed to displace synthetic fertilizer (
IDF (International Dairy Federation). A common carbon footprint for dairy.
The IDF guide to standard lifecycle assessment methodology for the dairy industry. Bulletin of the International Dairy Federation 445,
IDF, Brussels, Belgium2010
Compared with the IRE grass-based dairy system, the UK and US confinement dairy systems were more feed and N efficient, but also fed more conserved forages. Thus, the confinement dairy systems harvested more feed mechanically and, albeit based on inconsistent research (
), sequestered less carbon compared with the IRE grass-based dairy system, because the majority of forage was grown on arable land. As a result, on-farm GHG emissions per unit of milk of the IRE grass-based dairy system were lower than the US confinement dairy system. However, the feed efficiency and carbon sequestration of the UK confinement system was greater than the US confinement system. This led to the UK confinement dairy system emitting the lowest on-farm GHG emissions per unit of milk.
), GHG emissions from production and transport of purchased concentrate feed, manufacture of fertilizer for on-farm feed production, and from electricity generation were the main contributors to dairy systems off-farm GHG emissions. The IRE grass-based system emitted the lowest off-farm GHG emissions per unit of milk, which can be explained by the low reliance of the grass-based system on purchased concentrate (
). Off-farm GHG emissions per unit of milk were greater from the UK confinement system than the US confinement system, given the greater feeding of concentrate feeds associated with high GHG emission (e.g., South American soybeans) in the UK system. This is similar to the finding of
, who reported that production of South American soybeans used in European dairy systems emits significant CO2 emissions from deforestation.
The greater off-farm GHG emissions per unit of milk of the UK confinement dairy system led to the UK system emitting a greater carbon footprint than the IRE grass-based dairy system. However, the carbon footprint of the UK confinement dairy system was lower than the US confinement dairy system, because as discussed, on-farm GHG emissions per unit of milk were greater from the US system. The lower carbon footprint of milk from the grass-based dairy system compared with the confinement dairy systems agrees with some reports (
Evaluation of the livestock’s sector contribution to the EU greenhouse gas emissions (GGELS)—Final report. European Commission,
Joint Research Centre, Ispra, Italy2010
). This can be explained by the performance of dairy systems compared, but also by the variation in the application of the LCA methodology.
Influence of LCA Methodology on the Carbon Footprint of Milk from Dairy Systems
Major methodological decisions of LCA include the selection of GHG emission algorithms and the approach to allocate environmental impacts such as GHG emissions between coproducts (e.g., milk and meat) of multifunctional systems. Although international standards (
ISO (International Organization for Standardization)
Environmental management—Life cycle assessment: Principles and framework. ISO 14040:2006. European Committee for Standardization,
Brussels, Belgium2006
IDF (International Dairy Federation). A common carbon footprint for dairy.
The IDF guide to standard lifecycle assessment methodology for the dairy industry. Bulletin of the International Dairy Federation 445,
IDF, Brussels, Belgium2010
) have been developed for LCA methodology, the standards are not consistent particularly regarding allocation methodologies. Several criteria can be used to allocate GHG emissions between milk and meat (e.g., economic value or mass basis). Choosing different methodologies to allocate GHG emissions between milk and meat affects the carbon footprint of milk and can change the ranking of the carbon footprints of milk from dairy systems (
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.
). For instance, choosing to allocate dairy system GHG emissions between milk and meat on a mass basis for the UK confinement dairy system, but on an economic basis for the US confinement dairy system, resulted in the UK system having a greater carbon footprint per tonne of ECM than the US system. However, when mass or economic allocation was used for both dairy systems, the UK system had a slightly lower carbon footprint per tonne of ECM. Thus, to facilitate a valid comparison of the carbon footprints of milk from different dairy systems, the same method must be used to allocate GHG emissions between milk and meat.
), allocation according to physical relationships such as mass, protein content, or economic value resulted in a greater carbon footprint per unit of milk relative to allocation based on physical causal relationships (e.g., biological energy required to produce milk and meat from dairy cows and surplus calves). The differences between these allocation methods was explained by the relatively high energy requirements of producing meat from dairy systems compared with the mass or economic value of meat produced. The assessment of allocation methods showed, similar to
, that even when the same allocation method was applied, the percentage of GHG emissions allocated between milk and meat varied, depending on dairy system. As a result, the ranking of carbon footprints of milk from dairy systems was not consistent between allocation methods. Thus, for a given dairy system, advantages and disadvantages exist in choosing a particular allocation procedure.
Another method evaluated to handle allocation of GHG emissions between coproducts was system expansion. Similar to previous studies, the methodology was applied to assume meat from dairy production (including meat from surplus dairy calves raised for finishing) as a substitute for meat from traditional cow-calf beef systems (
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.
). This assumption resulted in a large deduction in GHG emissions of dairy systems, because meat production from cow-calf beef systems generates a substantially larger GHG emissions per unit of meat (30–40%) than an equal quantity of meat produced from dairy systems (
). Thus, applying this approach resulted in a significantly lower carbon footprint per unit of milk, compared with the other allocation methods. Furthermore, system expansion caused the greatest relative difference between the grass-based and confinement system carbon footprints per tonne of ECM. This was because for a fixed farm, increasing milk yield per cow generally reduces meat production from a dairy system (
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.
). Thus, the confinement systems displaced less meat per unit of milk from traditional cow-calf beef systems, compared with the grass-based system. Consequently, the deduction in confinement systems GHG emissions per unit of milk was lower than the grass-based system.
In addition to the quantity of meat a dairy system produces, the demand for meat and the type of meat a dairy system substitutes can significantly affect the carbon footprint of milk using system expansion. For instance,
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.
reported that conventional dairy systems had a greater carbon footprint per unit of milk than organic dairy systems when meat from dairy systems was assumed to replace meat from traditional beef systems, but conventional systems had the opposite effect when meat from dairy systems was assumed to substitute pork. Thus, this demonstrates that system expansion increases the uncertainty of the carbon footprint of milk from dairy systems compared with allocation based on causal or noncausal relationships. Furthermore, the methodology can create an unfair bias against meat by attributing the production of dairy animals entirely to meat (
). Conversely, some noncausal allocation methods were biased against milk because they attributed little (2%) or no GHG emissions to meat. Thus, this suggests that more moderate options (e.g., economic or biological allocation) are the most suitable methods to distribute GHG emissions between milk and meat.
Aside from allocation methods, LCA choices regarding emission algorithms affect the carbon footprint of milk. For instance, scenario modeling showed that computing GHG emissions with country-specific emission algorithms for each nation ranked carbon footprints of milk from dairy systems differently than calculating emissions with the same emission algorithms for all countries. Thus, this suggests that where nations differ in their efforts to measure emissions, it is more appropriate, albeit less precise, to use the same computation approach for each region (
) relatively few emission algorithms influence the carbon footprints of milk from dairy systems. The algorithms that affected both the grass and confinement systems were enteric CH4 emission algorithms, N2O emission factors for manure spreading, and emission factors related to fertilizer. Similar to results reported by
, the carbon footprint of milk from the grass-based system was also affected by the N2O emission factor for manure deposited during grazing, given the short housing period (80 d). The N2O emission factor for manure excreted by grazing cattle, however, had no effect on the carbon footprint of milk from the confinement systems, which were instead influenced by the CH4 and N2O emission algorithms for manure storage.
Carbon Sequestration and Land Use Change Emissions
Evaluations of the carbon footprint of milk from dairy systems are affected by LCA methodological decisions regarding carbon sequestration and land use change emissions from tropical deforestation and increased cropping. For instance, when carbon sequestration was included, the grass-based dairy system had the lowest carbon footprint per tonne of ECM, but omitting sequestration resulted in the grass-based and confinement dairy systems having similar carbon footprints per tonne of ECM. On the one hand, LCA standards recommend excluding carbon sequestration, because the
IPCC (Intergovernmental Panel on Climate Change) 2006 IPCC Guidelines for National Greenhouse Inventories.
in: Eggleston H.S. Buendia L. Miwa K. Ngara T. Tanabe K. Agriculture, Forestry and Other Land Use. Vol. 4. Institute for Global Environmental Strategies (IGES),
Hayama, Japan2006
Evaluation of the livestock’s sector contribution to the EU greenhouse gas emissions (GGELS)—Final report. European Commission,
Joint Research Centre, Ispra, Italy2010
) that managed grassland soils can permanently sequester carbon. However, given the uncertainty associated with carbon sequestration by managed permanent grassland, more research and data are required to accurately include sequestration and determine if it causes differences between the carbon footprints of milk from grass-based and confinement dairy systems.
A lack of consensus also exists on how to assess land use change emissions. For instance,
Evaluation of the livestock’s sector contribution to the EU greenhouse gas emissions (GGELS)—Final report. European Commission,
Joint Research Centre, Ispra, Italy2010
assume that the expansion of certain crops in particular regions (e.g., soybeans in South America) causes land use change emissions from deforestation. However, others (e.g.,
Audsley, E., M. Brander, J. Chatterton, D. Murphy-Bokern, C. Webster, and A. Williams. 2009. How low can we go? An assessment of greenhouse gas emissions from the UK food system and the scope for reduction by 2050. Report for the World Wildlife Fund (WWF) and Food Climate Research Network WWF-UK, Woking, UK.
) assume that all land occupation either directly or indirectly causes emissions from land use change. Thus, instead of applying an emission factor for land use change to a particular crop (e.g., Brazilian soybean), the approach applies a general emission factor for land use change to all occupation of land. The method suggested by
Evaluation of the livestock’s sector contribution to the EU greenhouse gas emissions (GGELS)—Final report. European Commission,
Joint Research Centre, Ispra, Italy2010
was followed in this study, but using a different approach, such as a general emission factor for land use change, can alter the order of dairy system carbon footprints per unit of milk (
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.
). Thus, a need exists to develop a harmonized approach to assess land use change emissions.
Comparison with Carbon Footprint Studies of Milk
Results of LCA and carbon footprint studies of milk are difficult and rarely completely valid to compare, because of potentially large differences in the application of the LCA methodology, as outlined previously. Nevertheless, differences can be partly explained by inherent differences between dairy systems. In general, the carbon footprint estimates of the high-performance IRE grass-based dairy system and top-performing UK and US confinement dairy systems were at the lower end of the range of recent carbon footprint reviews and studies of milk (
Evaluation of the livestock’s sector contribution to the EU greenhouse gas emissions (GGELS)—Final report. European Commission,
Joint Research Centre, Ispra, Italy2010
), our findings suggest that high-performance dairy systems of these countries reduce the carbon footprint of milk by 27 to 32%; however, this comparison is partially affected by methodological differences.
Excluding methodology differences, the lower carbon footprint of milk from high-performance dairy systems can be explained by their greater productive efficiency, which potentially reduces resource use per unit of milk, thereby reducing the carbon footprint (
). Furthermore, comparison of carbon footprints of milk from high-performance dairy systems in the current study relative to recent reports of carbon footprints of average IRE, UK, and US dairy systems indicates that the relative difference between average and high-performance dairy systems was likely to be greater than the relative difference between top-performing grass and confinement dairy systems. This is similar to the results of
and suggests that improving productivity of dairy systems has a greater effect on the carbon footprint of milk than converting from a confinement dairy system to an intensive grass-based system or vice versa.
Conclusions
Comparisons of the carbon footprints per unit of milk from high-performing dairy systems showed that the UK and US confinement dairy systems had a similar carbon footprint, but the IRE grass-based dairy system had a lower carbon footprint per unit of milk when carbon sequestration and direct allocation of land use change emissions were included in the calculations. However, the relative differences and ranking of dairy systems carbon footprints per unit of milk were not consistent in this study when different LCA methodologies regarding GHG emission algorithms, carbon sequestration, and allocation decisions between milk and meat were used. In particular, choosing to exclude carbon sequestration resulted in the grass-based and confinement dairy systems having similar carbon footprints per unit of milk. Therefore, this implies that further harmonization of several aspects of the LCA methodology is required to compare carbon footprints of milk from contrasting dairy systems. This study also indicates that significant potential exists to reduce the carbon footprint of milk in each of the countries by adopting farm practices currently implemented at a research level and by top-performing commercial milk producers.
Acknowledgments
The authors express their gratitude for financial support provided by the MultiSward project (FP7-244983) of the European Union Seventh Framework Programme. We also thank the anonymous reviewers for their comments and suggestions.
Audsley, E., M. Brander, J. Chatterton, D. Murphy-Bokern, C. Webster, and A. Williams. 2009. How low can we go? An assessment of greenhouse gas emissions from the UK food system and the scope for reduction by 2050. Report for the World Wildlife Fund (WWF) and Food Climate Research Network WWF-UK, Woking, UK.
UK greenhouse gas inventory, 1990 to 2010: Annual report for submission under the framework convention on climate change. AEA Technology Plc,
Didcot, UK2012
Les gaz à effect de serre en élevage bovin: Evaluation et leviers d’action (Greenhouse gases in cattle breeding: Evaluation and mitigation strategies).
Greenhouse gas emissions 1990–2010 reported to the United Nations framework convention on climate change. Johnstown Castle Estate, Co. Wexford,
Ireland2012 (Environmental Protection Agency)
Ecoinvent. 2010. Ecoinvent Centre. Ecoinvent 2.0 database. Swiss Centre for Life Cycle Inventories, Dübendorf, Switzerland. Accessed Nov. 30, 2012. http://www.ecoinvent.ch
Issues related to mitigation in the long term context.
in: Metz B. Davidson O.R. Bosch P.R. Dave R. Meyer L.A. In Climate Change 2007: Mitigation. Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press,
Cambridge, UK2007
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.
The IDF guide to standard lifecycle assessment methodology for the dairy industry. Bulletin of the International Dairy Federation 445,
IDF, Brussels, Belgium2010
IPCC (Intergovernmental Panel on Climate Change). 2000. Good practice guidance and uncertainty management in national greenhouse gas inventories. J. T. Houghton, L. G. Meira Filho, B. Lim, K. Tréanton, I. Mamaty, Y. Bonduki, D. J. Griggs, and B. A. Callander, ed. Organisation for Economic Co-operation and Development (OECD/OCDE), Paris, France.
2006 IPCC Guidelines for National Greenhouse Inventories.
in: Eggleston H.S. Buendia L. Miwa K. Ngara T. Tanabe K. Agriculture, Forestry and Other Land Use. Vol. 4. Institute for Global Environmental Strategies (IGES),
Hayama, Japan2006
Changes in atmospheric constituents and in radiative forcing.
in: Solomon S. Qin D. Manning M. Chen Z. Marquis M. Averyt K.B. Tignor M. Miller H.L. In Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press,
Cambridge, UK2007
ISO (International Organization for Standardization)
Environmental management—Life cycle assessment: Principles and framework. ISO 14040:2006. European Committee for Standardization,
Brussels, Belgium2006
ISO (International Organization for Standardization)
Environmental management—Life cycle assessment: Requirements and guidelines. ISO 14044:2006. European Committee for Standardization,
Brussels, Belgium2006
A survey of fertilizer use in Ireland from 2004–2008 for grassland and arable crops. Teagasc, Johnstown Castle Environment Research Centre,
Wexford, Ireland2010
Evaluation of the livestock’s sector contribution to the EU greenhouse gas emissions (GGELS)—Final report. European Commission,
Joint Research Centre, Ispra, Italy2010 ()
Effects of a perennial ryegrass diet or total mixed ration diet offered to spring-calving Holstein-Friesian dairy cows on methane emissions, dry matter intake, and milk production.
A Nordic proposal for an energy corrected milk (ECM) formula.
Performance recording of animals: State of the art 1990. EAAP Publication No. 50:156–157. Centre for Agricultural Publishing and Documentation,
Wageningen, the Netherlands1991
Methodology used in FeedPrint: A tool quantifying greenhouse gas emissions of feed production and utilization. Wageningen UR, Livestock Research,
Lelystad, the Netherlands2012
00031-9&id=gr1.jpg){kind=link}