Life cycle assessment of milk at farm gate - focus on green house gas emission Troels Kristensen Institute of Agroecology Århus University, Denmark EAAP 2011 Stavanger Norway Session 7
Structure of the presentation 1) Introduction ti focus on GHG Carbon foot print 2) The Life Cycle Assessment - method 3) Farm as part of the total chain 4) Farm emission how to calculate, results, reduction potential 5) Conclusion and perspectives
Carbon footprint (CF) from livestock in EU 27 Lesschen et al. 2011
Emission from different livestock product shown for three different functional units beef chic- ken per kg of product per kg of protein per daily intake pork 0 5 10 15 20 25 Emission relative to milk = 1 de Vries & de boer, 2011
Carbon footprint through the chain in relation to milk output per cow, national data -eq. per kg FPCM 12.00 10.00 8.00 6.00 ad Multipurpose p Extensive kg CO2 4.00 2.00 Intensive 0.00 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000 Output per cow, kg FPCM per year Gerber et al. 2011
Life cycle of milk CF from different stages through the chain 4% 2% 5% 5% 4% Farm gate Transport dairy 3% Processing Packaging Distribution Retail Consumption 77% www.usdairy.com/sustainability
LCA whole farm approach Emissions to air (CH 4, N 2 O, NH 3, CO 2 ) From feeding, manure storage and application, crop production INPUT Materials e.g. fertilizer feed Energy e.g. fuel Chemicals e.g. pesticides Other LIVESTOCK FARM LIVESTOCK FIELD Feed Manure Enteric Changes in fermentation soil carbon OUTPUT Main products Residues or co-product Emissions to soil and water (NO 3- ) EMISSIONS TO AIR Transport Transport Transport Transport Production of inputs Agricultural production Processing Packaging Supermarket EMISSIONS TO SOIL AND WATER Hermansen & Kristensen, 2011
LCA of Danish milk production Production, annual data Production system Conventional Organic No of farms 35 32 Herd size, no of cows 122 115 Milk, kg ECM per cow 8201 7175 Live weight gain, kg per cow 1) 179 174 Feed intake, kg DM per cow 6593 6618 - roughage, % of DMI 55 69 - pasture, % of DMI 8 19 1) Herd live weight gain (Kristensen et al, 2011)
LCA of Danish milk production Production, annual data Production system Conventional Organic Stocking rate, LSU per ha 1.80 1.12 Maize, % of area 17 3 Grassland in rotation, % of area 24 45 Grassland permanent, % of area 6 10 Fertilizer, kg N per ha 68 0 Manure, kg N per ha 168 130 Production, NE (1000 MJ)per ha 50 37 (Kristensen et al, 2011)
LCA of Danish milk production GHG emission from different pollutants, kg CO 2 eq. per kg of ECM Production system Conventional Organic Total 1.20 1.27 Internal (farm level) 1.05 1.24 -Methane 0.62 0.69 -Nitrous oxide 029 0.29 035 0.35 -Fossil energy 0.14 0.20 External (import) 0.15 0.0303 -Feed 0.10 0.01 -Manure 0 0.02 -Fertilizer 0.05 0 (Kristensen et al, 2011)
LCA of Danish milk production Emission in CO 2 eq. from various sources 8% 4% 12% 52% 24% Farm methane Farm N2O Farm fossil Import feed Import manure Import fertiliser
LCA of Danish milk production Methane where do the emission occurs? Ma nure Enteric pasture straw deep litter slurry bulls heifers cow 0 10 20 30 40 50 60 70 % of methane
Estimated enteric methane by different models. Production data from 218 herds Model no. Source Type Kg per cow Kg per 1000 kg ECM Mean SD Mean SD 0 IPCC, 2006 1 117 0 15,0 14 1,4 1 IPCC, 2006 2 130 14 16,6 1,6 2 Kirchgessner et al., 1991 3 106 7 13,6 1,0 3 Mills et al., 2003 3 145 11 18,6 1,6 4 Ellis et al., 2007 3 115 10 14,7 1,3 5 Giger-Reverdin et al., 2003 3 163 11 20,9 1,8 6 Mills et al., 2003 4 147 14 18,8 2,2 7 Yan et al., 2000 4 138 19 17,6 2,3 8 Giger-Reverdin et al., 2003 4 164 23 21,0 1,9 9 Kirchgessner et al., 1995 4 120 12 15,4 3,2 137 12 17,5 1,7 1) TIER 1 IPCC, 2006 2) TIER 2 IPCC, 2006 3) Production models 4) Feed ration models
Reduction of methane emission per kg of milk in intensive systems Mitigations options Feeding Herd structure Breeding Farm management Trade offs Effect on emission of other GHG Pollution swapping Product quality and food security Animal health Social acceptable
LCA of Danish milk production Nitrous oxide where do the emission occurs? leaching NH3 crop residue mineralisation fertiliser pasture manure field stable 0,5 0 0,5 1 1,5 2 Organic Conventional N 2 O, kg per ha
Reduction of N 2 O emission per kg of milk in intensive systems Mitigations options in relation to livestock Reduced N intake Manure management Utilization of legumes
Allocation of emission between multiple products Allocation methods 1) Attributional (average) -mass, kg -protein, kg -biological (NE) -economic value 2) Consequential (marginal) -CO 2 other meat products
LCA of Danish milk production Allocation between milk and meat Product, kg 180000 160000 140000 R² = 0.9054 Milk*0.1 meat 120000 100000 80000 60000 40000 20000 R² = 0.5439 0 0 500000 1000000 1500000 2000000 2500000 CO2 -eq., kg Total CO 2 eq. = 1.03 * kg milk + 4.17 * kg meat ( r 2 = 0.92) Kristensen et al, 2011
CF of milk and meat Effect of different methods for allocation of total emission All emissions allocated to milk Estimated from data (regression) 1.2 1.03 4.17 meat milk Allocation according to economic value 106 1.06 3.41 Allocation according to feed requirement 0.91 6.92 Allocation according to protein in product 0.99 5.05 System expansion 0.94 6.35 0 1 2 3 4 5 6 7 8 CO 2 eq per kg product Kristensen et al, 2011
Variation in CF of milk between farms % of farms 25 20 15 conventional organic 10 5 0 0.85 0.9 0.95 1 1.05 1.1 1.15 1.2 1.25 1.3 1.35 CO 2 -eq. per kg ECM Kristensen et al, 2011
Variation in CF of milk explained by different farming strategies 40% 27% 17% Farming strategy Herd efficiency Farming intensity N yield crop Grassland Combined milk and meat Not accounted for 4% 5% 7% Kristensen et al, 2011
CF of milk related to farming strategy CO 2 eq. per kg milk 14 1.4 1.2 1 On farm Total 0.8 0.6 0.4 0.2 0 Farming strategy top 25% low 25% top 25% low 25% Herd efficiency. Farming intensity Kristensen et al, 2011
Uncertainty 1) Production data and emission factors CF = amount x EF 2) Models 3) Allocation method 4) Other methodological choices
Uncertainty production data Type of production CV, % Concentrate 4 Roughage 19 Fertilizer 5 Manure 16 Milk 3
Effect of uncertainty on EF - Methane 20% - Nitrous oxide - N applied 100% - N pasture 100% - NH 3 emission 20% - Fossil energy 20% Analysed by Monte Carlo simulation Flysjo et al, 2011
Conclusions CF of intensive milk production Uncertainty relative large compared to reported differences between systems Large variation between studies due to methodological choices CF of Danish milk - Farm emission 80-90 % of total emission through the chain - Enteric methane largest source -Fossil energy only 10-20 % of total CF -Variation in herd efficiency the most important factor for CF -Low stocking rate reduces CF from import
Thank you for your attention