The greenhouse emissions footprint of free-range eggs

Transcription

1 The greenhouse emissions footprint of free-range eggs R. C. Taylor,* 1,2 H. omed, and G. Edwards-Jones 3 * British Trust for Ornithology, The Nunnery, Thetford, Norfolk, UK IP24 2PU; and School of the Environment, Natural Resources and Geography, Bangor University, Bangor, Wales, UK LL57 2UW ABSTRACT Eggs are an increasingly significant source of protein for human consumption, and the global poultry industry is the single fastest-growing livestock sector. In the context of international concern for food security and feeding an increasingly affluent human population, the contribution to global greenhouse-gas (GHG) emissions from animal protein production is of critical interest. We calculated the GHG emissions footprint for the fastest-growing sector of the uk egg market: free-range production in small commercial units on mixed farms. Emissions are calculated to current Intergovernmental Panel on Climate Change and uk standards (PAS2050): including direct, indirect, and embodied emissions from cradle to farm gate compatible with a full product life-cycle assessment. We present a methodology for the allocation of emissions between ruminant and poultry enterprises on mixed farms. Greenhouse gas emissions averaged a global warming potential of 2.2 kg of Co 2 e/dozen eggs, or 1.6 kg of Co 2 equivalent (e)/kg (assuming average egg weight of 60 g). one kilogram of protein from free-range eggs produces 0.2 kg of Co 2 e, lower than the emissions from white or red meat (based on both kg of meat and kg of protein). of these emissions, 63% represent embodied carbon in poultry feed. A detailed GHG emissions footprint represents a baseline for comparison with other egg production systems and sources of protein for human consumption. Eggs represent a relatively low-carbon supply of animal protein, but their production is heavily dependent on cereals and soy, with associated high emissions from industrial nitrogen production, land-use change, and transport. Alternative sources of digestible protein for poultry diets are available, may be from waste processing, and would be an effective tool for reducing the industry s GHG emissions and dependence on imported raw materials. Key words: free-range egg, greenhouse-gas emissions, carbon footprint, life cycle assessment 2014 Poultry Science 93 : INTRODUCTION 2014 Poultry Science Association Inc. Received July 15, Accepted october 12, Corresponding author: r.c.taylor@bangor.ac.uk 2 Present address: School of the Environment, Natural Resources and Geography, Bangor university, Bangor, Gwynedd LL57 2uW. 3 Deceased August Since 1996, the proportion of eggs commercially in battery farms in the united Kingdom has decreased from 88 to 49%; barn production (deep-litter housed systems) remains fairly stable at 5% and organic production has averaged 2% since its entry into the uk market in Total uk production continues to increase, associated with a net shift into free-range production from 10% in 1996 to 45% in 2013 (DEFRA, 2013). In common with the wider uk market, recent changes in legislation relating to caged egg-production systems have promoted increases in the Welsh flock (7.6 million birds in 2010); much of this increase in freerange units forming a component of mixed livestock farms (H. Gittins, Farming Connect, Bangor, uk, personal communication, 2010). Such systems therefore represent a significant and increasing proportion of national egg production and protein supply. Information on the sources and magnitudes of greenhouse-gas (GHG) emissions from livestock food production is of considerable interest to policymakers, cf. recent government commitment to reduce greenhouse gas emissions and minimize the projected impacts of climate change while maintaining food security (Climate Change, Agriculture and Food Security, 2009; Ranger et al., 2009; Gerber et al., 2010; Foresight The Future of Food and Farming, 2011). Greenhouse gas emissions from red meat production are relatively well understood (olesen et al., 2006; Edwards-Jones et al., 2009; Vayssieres et al., 2010) and high in comparison with poultry meat production (Williams et al., 2006; Xin et al., 2011; Leinonen et al., 2012). The difference is largely due to the contributions of methane [from enteric fermentation: global warming potential (GWP) 25 Co 2 equivalents (e)] and nitrous oxide (from soils in response to N application in manure/excreta: GWP 298 Co 2 e; Intergovernmental Panel on Climate Change, 2010). Eggs constitute a major alternative source of 231

2 232 Taylor et al. animal protein in the UK and DEFRA surveys estimate annual production is greater than 11 billion eggs (DEFRA, 2013), but information on the GHG emissions associated with UK egg production is limited to studies of partially comparable US intensive systems (Xin et al., 2011) and UK studies modeling aggregated national data (e.g., Williams et al., 2006; Leinonen et al., 2012). A recent review of the sustainability of egg production highlighted these gaps in our understanding of their environmental impact. In particular, the authors noted that because each type of production system has a characteristic pattern of emissions and impacts, sustainability assessments should not be assumed to be comparable across different systems (Xin et al., 2011; Leinonen et al., 2012). The present study enhances the model-based findings of Leinonen et al. (2012) by presenting detailed farmenterprise life-cycle assessment (LCA) GHG footprint calculations for eggs in free-range units on mixed-enterprise farms. To inform policy discussions on protein supply and national sustainability, we present these data for cooperatively managed egg production units on 3 contrasting farm systems: a) single-enterprise free range egg production; b) eggs as the primary enterprise on a dual-enterprise farm; c) eggs as the secondary enterprise on a multiple-enterprise farm. We also take this opportunity to compare the GHG emissions from eggs (and spent hens) with emissions from red meat production a) on these farms and b) on a comparable mixed farm producing red meat only. MATERIALS AND METHODS Life-cycle assessment has been used to analyze the environmental impacts of many products including food and agricultural products (e.g., Foster et al., 2006; Milà i Canals et al., 2007; Sim et al., 2007; Edwards- Jones et al., 2009). Life-cycle assessment typically considers the global warming potential of GHG as part of its output; however, it is possible to calculate a carbon footprint for a specifically defined part of a product lifecycle without undertaking a full LCA. The methodology of LCA is standardized (International Organization for Standardization, 2006a,b) and tends to follow set methodologies and use standard databases. Although there are many similarities in approaches between LCA and carbon footprinting, they are not the same thing. For this reason, we present the GHG footprint independently of a full LCA according to published international and national guidelines for system boundaries and calculation methodologies (IPCC, 2006; BSI, 2008, 2011). For further information on the system boundaries applied to livestock farming, see Edwards-Jones et al. (2009). The flow of GHG into and out of agricultural plants and soils remains relatively poorly understood, and although their inclusion in footprinting studies is now permitted as an emissions offset under the 2011 revision of PAS 2050 (BSI, 2011), they are excluded from the current paper. Data Collection and Description of Case Study Farms Four farms representative of Welsh production systems were selected for the present study. Detailed farm records on land management, livestock, all energy use, and all indirect inputs were collected for each farm for the 12 mo between October 2008 and September 2009; all input data are real farm records and published relevant GHG emissions values: no input values were assumed or estimated. The calculations used were very detailed Tier 1 (IPCC, 2006, 2010; Taylor et al., 2010) and the system boundary compliant with the relevant British standard for cradle-to-grave GHG emissions footprinting, including embodied GHG emissions for all inputs and complete to the farm gate (PAS-2050 standard: BSI, 2011). Enterprise-specific inputs (e.g., poultry and cattle feeds, N inputs to soils from poultry and cattle manure) were segregated throughout the calculations. Inputs without enterprise-specific records included diesel, electricity, and noncommercial grazing stock (horses); these inputs were allocated between the poultry and ruminant enterprises in proportion to the revenue from each enterprise (i.e., economic allocation). Similarly, total emissions for a single enterprise (e.g., the poultry unit) were economically allocated between coproducts (eggs, spent hens) in accordance with PAS 2050 (BSI, 2008, 2011). Calculated GHG emissions are presented in kilograms of CO 2 e per ha and per unit product (kg of eggs, dozen eggs, kg of live weight of spent hens, lamb, beef; liters of milk). All 3 poultry units were free-range Rhode Island Red layer flocks bought-in as 1.5 kg point-of-lay (POL) pullets in mo 1; sold in mo 14 as 2.2 kg spent hens and the unit cleaned and left empty for 30 d before the production cycle restarts. After discussion with farmers, birds were assumed to range outdoors on pasture for an average of 12 h per day across the 14-mo cycle. All soils were mineral based. Farm 1 supplies egg-boxes to the other 2 units at zero cost because the 3 farms operate as a producer group. Over the whole production cycle, laying rates averaged 90% and mortality rates were 3.6%, although farm-specific values were used in calculations. Embodied C in feed was calculated using milling labels and published embodied C values for each ingredient (Carbon Trust, 2010; Taylor et al., 2010). Farm 1 was a 28-ha single-enterprise egg producer with 3,000 laying hens (at cycle start) and using 20 adult nonlactating sheep for 2 mo of the cycle for sward maintenance in the hen pasture. This grazing service was considered to be an integral part of the poultry enterprise, but the sheep remained the property of another farm: soils and excreta emissions were therefore included in the poultry unit GHG footprint but enteric fermentation emissions remain within the source farm system boundary and are excluded from the farm 1 footprint. This allocation is used because of the economic basis of the arrangement: the sheep are not hired by the poultry unit to provide a service;

3 GREENHOUSE EMISSIONS FOOTPRINT OF FREE-RANGE EGGS instead the source farm pays for the grazing supplied by the poultry unit. Feed use was 132 t of 70% wheat, 12% soy, 4% sunflower + minerals: average consumption of 13 g/bird per day. Farm 2 was a 39-ha mixed farm where the 5,000 laying hens were the primary enterprise: secondary income was from 250 ewes and 6 suckler cows with followers (calves, lambs, youngstock) producing lamb and beef. Poultry feed use was 267 t: average consumption 16 g/bird per day. Farm 3 was a 79-ha mixed enterprise where the primary income was dairy from a 50-head dairy herd (with followers), and the secondary enterprise was 6,000 laying hens. Poultry feed use was 220 t: average consumption 11 g/bird per day. A small proportion of farm income came from an additional suckler beef herd (8 cows plus followers) and the sheep flock (120 ewes plus followers). Enterprise inputs were separated where records permitted (direct and indirect stock emissions, bought feedstuffs, soil emissions, and so on) and inputs that could not be separated (diesel, electricity, noncommercial stock) allocated economically based on revenue. The emissions allocation between ruminant and poultry enterprises is summarized in Table 1. For details of IPCC/PAS2050 GHG emissions calculations used, see Edwards-Jones et al. (2009) and Taylor et al. (2010). Embodied Carbon Value for POL Pullets No LCA or published production data could be found for the GHG emissions associated with the production of POL pullets. These emissions represent a significant input to the egg production system and were assumed a priori to constitute >5% of the enterprise GHG footprint, and therefore to require estimation and inclusion under PAS2050 guidelines (BSI, 2008). The value was estimated using a published GHG footprint for UK broiler production corrected for BW [4.5 kg broiler finishing weight: 1.5 kg POL pullet (Williams et al., 2006)]. It was assumed that the longer but lessintensive production cycle for layers was approximately comparable (in terms of electricity use, feed consumption and indirect emissions) to the very short, highly intensive broiler cycle (H. Omed, Bangor University, Bangor, UK, personal communication, 2010). Management Scenarios for Reduced GHG Emissions 233 Calculating the GHG emissions footprint for each farm involved building a virtual farm business a detailed month-by-month recalculation of stock management, inputs, and exports for the business year (Taylor et al., 2010). It was consequently possible to manipulate each farm model to investigate the effects of management changes on the farm-gate product footprint. Five theoretical changes were investigated. Three of these have the potential to reduce total farm emissions (and therefore product emissions), and the remaining 2 were modeled as improvements in efficiency that would therefore decrease the product (egg) footprint while leaving farm emissions unchanged. These were as follows: a) source egg-boxes from UK suppliers (applies only to farm 1); b) manage manure stores anaerobically (e.g., closed lagoons, pit, or tank storage); c) export all stored manure to an anaerobic digester unit (without retrieving N in digestate); d) reduce production-cycle casualty rate to 2%; and e) increase laying rate by 5%. Options a) to c) are expected to reduce total GHG emissions, whereas d) and e) represent efficiency gains Table 1. Summary of the main economically allocated and enterprise-allocated components of the greenhouse gas (GHG) emissions associated with egg and red meat production on 3 farms producing free-range eggs in Wales, in kilograms of CO 2 e per year Farm % Footprint 1 Item Poultry Ruminants Poultry Ruminants Poultry Ruminants Poultry Ruminants Overall Economic allocation Diesel and petrol 33,045 3, ,956 12, Electricity 6,276 2, ,209 7, Emissions from noneconomic stock 3, ,740 3, Enterprise-separation Embodied C from: Point-of-lay pullets (estimated) 14,464 24,107 24, Concentrate feeds 83, ,744 19, , , Field and veterinary agrochemicals , , Bedding , Packaging 3, Other consumables , Transport emissions for inputs 8, ,197 6,540 2, Direct and indirect emissions from: Manure and manure handling 3,585 4,900 7,638 8,801 7, Soil (from manure N) 14,935 26,262 51,981 17,024 86, Soil (from fertilizer N) , , Stock (enteric fermentation, CH 4 ) 1,036 1,885 99, , Total 169, , , , ,147 1 Percentage contributions to the overall footprint are listed in the last 3 columns.

4 234 Taylor et al. Table 2. Greenhouse gas (GHG) footprint of products and by-products from free-range eggs and ruminant livestock (sheep, dairy, and beef cattle) on 3 Welsh farms 1 Enterprise separation between ruminant and poultry units on the same farm 2 Farm GHG emissions per ha (all enterprises) Eggs (dozen) Poultry products: economic allocation Eggs (kg) kg Spent hens kg of LW Lamb Ruminant products: economic allocation kg of LW Beef kg of LW Milk Liters Farm 1 6, , ,336 Farm 2 11, , , , ,950 Farm 3 11, , , , , ,500 1 GHG emissions values (bold type) are global warming potential in kilograms of CO 2 equivalent (100 yr). 2 LW = live weight. and therefore product footprint reductions without affecting whole-farm GHG emissions. RESULTS Carbon Footprint of Free-Range Eggs The GHG footprint of these 3 producers averaged 2.2 kg of CO 2 e/dozen eggs, or 1.6 kg of CO 2 e/kg (assuming free-range average egg weight of 60 g; Table 1). The whole-farm GHG footprint of the single-enterprise farm was lowest at 6 t of CO 2 e/ha, whereas on the mixed farms, the additional CH 4 and N 2 O emissions associated with ruminant livestock (greater land area notwithstanding) brought the whole-farm emissions to more than 11 t of CO 2 e/ha (Table 2). The largest component of the GHG footprint of the eggs was the embodied carbon in concentrate feeds, representing 50% in farm 1, 73% in farm 2, and 65% in farm 3 (Figure 1). This contrasted with the red meat and milk footprints from farms 2 and 3, where the largest component was methane from enteric fermentation [46% on farm 2 and 50% on farm 3, comparable with 65% of the GHG footprint for a dedicated red meat producer (Figure 1b); R. C. Taylor and G. Edwards- Jones, unpublished data]. On farms 2 and 3, the next largest components were embodied C in POL pullets bought in at the start of the production cycle (average 11%), and the N 2 O emissions from soil responding to N in manure and excreta (average 8%). The N 2 O emissions per dozen eggs were similar across the 3 farms, at 0.16, 0.18, and 0.17 kg of CO 2 e/dozen, as was the embodied C from POL pullets (0.19, 0.20, and 0.28 kg of CO 2 e/dozen). Farm 1 had a very large contribution from diesel (19.6%) and was the only farm to buy egg boxes, adding 2% in the form of embodied C in the boxes themselves and a further 5% for their transport to the farm gate. Economic Allocation in Singleand Multiple-Enterprise Free-Range Egg Production The GHG contributions from diesel and transport together represented 28.4% of the footprint of farm 1. On farms 2 and 3, these inputs (along with emissions from noneconomic livestock) were much smaller and were economically allocated between the poultry and ruminant enterprises (Table 2 and Figure 1). On farm 2, where 83.8% of revenue is derived from eggs, these economically allocated diesel, transport, and electricity inputs represented only 4% of the egg footprint and 0.9% of the ruminant product footprint. On farm 3, the high income from milk reduced the egg enterprise to 48.9% of revenue, but direct emissions from the ruminant enterprise are much greater because of the high rates of methane emission from dairy cows. The 3 economically allocated inputs (diesel, transport, electricity) together represented 10.2% of the egg footprint and 3.4% of the ruminant product footprint. If the whole-enterprise GHG emissions were allocated between all the farms products entirely by revenue (strict economic allocation: BSI, 2008), the GHG footprints of those products would be very different, and not representative of the individual enterprise emissions. The comparatively high poultry revenue on farm 2 would have drawn-in emissions from the ruminant enterprise and increased the egg footprint to 3 kg of CO 2 e/dozen, reducing the red meat footprints to unrealistic 3.1 (beef) and 4.3 (lamb) kg of CO 2 e/ kg live weight respectively. On farm 3, emissions from ruminant products included large direct emissions from dairy cattle and from imported soy-based concentrates. Strict economic allocation would have made the egg footprint very high at 4.5 kg of CO 2 e/dozen, and would have reduced the red meat footprints to 3.4 (beef) and 8.3 (lamb) kg CO 2 e/kg of live weight, whereas the milk footprint would have been brought lower than industry average, at 0.9 kg of CO 2 e/l. Footprint Reduction The modeled management scenario offering the greatest reduction in enterprise-level emissions was exporting manure to an anaerobic digester, which reduced overall GHG emissions an average of 7% by removing both the direct manure storage emissions (CH 4, NH 3 secondary N deposition, and N 2 O) and the soil emissions (N 2 O) resulting from its disposal on land (Table 3). This represents total reductions in direct and indirect N 2 O emissions of 11 t of CO 2 e per year on farm 1, 13 t of CO 2 e on farm 2, and 23.5 t of CO 2 e on farm 3,

5 GREENHOUSE EMISSIONS FOOTPRINT OF FREE-RANGE EGGS 235 Figure 1. The origins and percentage breakdown of greenhouse gas (GHG) emissions for eggs and red meat production on small farms. Farm 1 (single-enterprise eggs) is compared with a reference example of a single-enterprise red meat producer. Farms 2 and 3 contrast the GHG emissions from the poultry enterprise with the emissions breakdown of red meat production on the same farm. POL = point of lay.

6 236 Taylor et al. Table 3. Modeled effects of packaging origin, manure management handling, and efficiency improvements on the greenhouse gas footprint of free-range egg production on 3 Welsh farms Farm Scenario for emissions reduction (kg CO 2 equivalent per farm ha or dozen eggs) ha 1 dozen 1 ha 1 dozen 1 ha 1 dozen 1 Current management (baseline reference value) 6, , , Reducing the enterprise footprint (%) Source all egg-boxes from United Kingdom 1.5 Manage manure anaerobically All manure sold to AD 1 unit (N exported) Reducing the product footprint (%) Decrease casualty rate to 2% Increase laying rate by 5% AD = anaerobic digester. although on real mixed farms some mineral N or digestate would be likely required (imported) to compensate for the lost stored-manure N application on pastures. Farm 1 could reduce transport emissions (CO 2 from diesel fuel) by only using UK-sourced packaging (overall reduction 1.5% on this farm, a total reduction of 2.5 t of CO 2 e per year). Anaerobic manure storage (pit and anaerobic lagoon systems) could decrease farm emissions by 2.6% (Table 3), representing total reductions in direct and indirect N 2 O emissions of 5.2 t of CO 2 e on farm 1, 4.3 t of CO 2 e on farm 2, and 7.8 t of CO 2 e on farm 3. Although these changes have relatively modest individual impacts, they are compatible with each other and could together reduce total GHG emissions from the 3 farms by over 65 t of CO 2 e/yr. Alternatively, production improvements might reduce the product footprint while enterprise emissions remain constant: in these cases the same total GHG emissions are allocated to greater production volumes (more eggs for the same total emissions). Reducing layer casualties to 2% increases the number of productive birds through the cycle, reducing the egg footprint down by 1%. Increasing laying rates by 5% reduces the egg footprint by 5% (Table 3). DISCUSSION The production of free-range eggs has a lower GHG emissions footprint per kilogram than either red or white meat in the United Kingdom (Welsh lamb 11 kg and beef 13 kg; UK broilers 3.75 kg of CO 2 e/kg, all values quoted live weight at farm gate; Williams et al., 2006; Taylor et al., 2010). In terms of protein supply, 1 kg of protein from free-range eggs produces 0.2 kg of CO 2 e, compared with 5.5 kg of CO 2 e for lamb; 6.5 kg of CO 2 e for beef, and 1.25 kg of CO 2 e for chicken meat. Milk protein has a GHG emissions footprint of 0.04 kg of CO 2 e (assuming 50% carcass weight for lamb and beef: meat protein content 25%; eggs 14%, and milk 3%). Globally, poultry numbers are increasing more rapidly than any other class of domesticated animals, having increased from 7 to over 23 billion between 1961 and 2009; the European poultry flock increased by 37% over this period, whereas in Asia the increase was 544% (Food and Agriculture Organization, 2010). Given the relatively low GHG impact of eggs as a protein source, this might be considered a positive trend, except that chickens are widely in intensive systems highly dependent on imported cereals and soy (c.f. 63% of eggs GHG footprint from feed production: feed cereal content 70% and soy 12% wt/wt). Cereal production globally requires large inputs of industrially nitrogen fertilizers and water, already uses most of the best land, and may be approaching biological and systemic yield limits (Tilman et al., 2002). Increasing cereal and soy production to support increased cerealdependent poultry production is likely to contribute further to land-use change and land degradation, both of which will increase the embodied GHG values for feeds and hence the impact of egg production itself. It is clear that opportunities exist for reducing the GHG emissions associated with egg production, some of which would also reduce the industry s dependence on imported feedstuffs. The first consideration is that excellent system design and husbandry leading to high laying rates and low mortality will produce eggs with the lowest product footprint for any production system. Small improvements may be achievable including from using the most appropriate and productive breeds and genetic stock for a given system type and climate. Second, laying poultry diets typically contain 16 to 19% protein, much of it from soy (around 12 21% by weight). Soy imported into the United Kingdom comes primarily from South America and Brazil, and the raw meal has a very high GHG footprint of 9 to 15 kg of CO 2 e/kg due to significant emissions from land-use change (Carbon Trust, 2010). Alternative protein sources for poultry include worms (Lumbricus sp.) by composting organic wastes (Sogbesan and Ugwumba, 2008), algae (e.g., Spirulina sp.) in biological CO 2 -absorption systems for industrial exhaust emissions (Becker, 2007; Peirettia and Meineri, 2008) and processed animal byproducts from red meat production. All these have the theoretical potential to be incorporated into poultry diets; the 2 proteins de-

7 GREENHOUSE EMISSIONS FOOTPRINT OF FREE-RANGE EGGS rived from waste processing (Spirulina and earthworms) might constitute net carbon sinks under IPCC system boundary guidelines because they absorb CO 2 and create a sales value for industrial wastes. Initial estimates based on feed formulation modeling using IPCC emissions data and replacing soy with these alternative protein sources indicated a potential 60% reduction in the GHG emissions footprint for poultry diets (IPCC, 2006; R. C. Taylor, unpublished data). This change could reduce the egg GHG footprint by 45% and reduce the dependence of the United Kingdom on imported soy. Another potential GHG reduction strategy might be to increase the value of system coproducts to draw emissions away from the primary product. Poultry manures have the potential to be sold raw as fertilizers, or to be sold as feedstock for anaerobic digestion and the production of renewable electricity; digestate retains its nitrogen content and can still be used as fertilizer. Selling raw manure would have the same impact on farm and egg GHG emissions as the modeled export to anaerobic digester, a reduction on average of 7%. Spent hens constitute a second protein source although their market value is very low; improving their carcass value by selective breeding for dual-purpose would further reduce the egg GHG footprint. Finally, all the emissions reductions modeled on these farms are mutually compatible and would have additive effects on the GHG emissions footprint of eggs at the farm gate. REFERENCES Becker, E. W Micro-algae as a source of protein. Biotechnol. Adv. 25: BSI PAS 2050: 2008 Specification for the assessment of the life cycle greenhouse gas emissions of goods and services. British Standards, London, UK. BSI PAS 2050: 2011 Specification for the assessment of the life cycle greenhouse gas emissions of goods and services. British Standards, London, UK. Carbon Trust FPX database version 3.2 using data from: Renewable transport fuels obligation (RTFO). The Carbon Trust London, Omnitech International, DEFRA. Accessed Sep. 21, Climate Change, Agriculture and Food Security (CCAFS) Climate, agriculture and food security A strategy for change. The Alliance of the CGIAR Centers and ESSP. Accessed Oct. 10, ber2009.pdf. Department for Environment, Food and Rural Affairs (DEFRA) Quarterly United Kingdom Egg Statistics, Quarter 3, Accessed Nov. 12, Edwards-Jones, G., K. Plassmann, and I. M. Harris Carbon footprinting of lamb and beef production systems: Insights from an empirical analysis of farms in Wales, UK. J. Agric. Sci. 147:1 13. Food and Agriculture Organization (FAO) Online datasets. Accessed Sep. 5, aspx. Foresight The Future of Food and Farming Final project report. The Government Office for Science, London, UK. Foster, C., K. Green, M. Bleda, P. Dewick, B. Evans, A. Flynn, and J. Mylan Environmental impacts of food production 237 and consumption A report to the Department for Environment, Food and Rural Affairs. Manchester Business School, London, UK. Gerber, P., N. Key, F. Portet, and H. Steinfeld Policy options in addressing livestock s contribution to climate change. Animal 4: Intergovernmental Panel on Climate Change (IPCC) guidelines for national greenhouse gas inventories. Pages in volume 4, Agriculture, Forestry and Other Land Use. H. S. Eggleston, L. Buendia, K. Miwa, T. Ngara, and K. Tanabe, ed. Institute for Global Environmental Strategies (IGES), Hayama, Japan. Intergovernmental Panel on Climate Change (IPCC) Emissions databases. Accessed Aug igquarterly United Kingdom Egg Statistics, Quarter 3, Accessed November 12, publications/egg-statisticses.or.jp/efdb/main.php. International Organization for Standardization (ISO). 2006a. ISO14040 Environmental management, life cycle assessment, principles and framework. Geneva, Switzerland. International Organization for Standardization (ISO). 2006b. ISO Environmental management, life cycle assessment, principles and framework. Geneva, Switzerland. Leinonen, I., A. G. Williams, J. Wiseman, J. Guy, and I. Kyriazakis Predicting the environmental impacts of chicken systems in the United Kingdom through a life cycle assessment: Egg production systems. Poult. Sci. 91: Milà i Canals, L., S. J. Cowell, S. Sim, and L. Basson Comparing domestic versus imported apples: A focus on energy use. Environ. Sci. Pollut. Res. 14: Olesen, J. E., K. Schelde, A. Weiske, M. R. Weisbjerg, W. A. H. Asman, and J. Djurhuus Modelling greenhouse gas emissions from European conventional and organic dairy farms. Agric. Ecosyst. Environ. 112: Peirettia, P. G., and G. Meineri Effects of diets with increasing levels of Spirulina platensis on the performance and apparent digestibility in growing rabbits. Livest. Sci. 118: Ranger, N., A. Bowen, J. Lowe, and L. Gohar Mitigating climate change through reductions in greenhouse gas emissions: The science and economics of future paths for global annual emissions. Policy brief December Publications/Policy/Policy-docs/bowen-Ranger_MitigatingClim atechange_dec09.pdf. Sim, S., M. Barry, R. Clift, and S. Cowell The relative importance of transport in determining an appropriate sustainability strategy for food sourcing A case study of fresh produce supply chains. Int. J. Life Cycle Assess. 12: Sogbesan, A. O., and A. A. A. Ugwumba Nutritional values of some non-conventional animal protein feedstuffs used as fishmeal supplement in aquaculture practices in Nigeria. Turkish J. Fish. Aqua. Sci. 8: Taylor, R., A. Jones, and G. Edwards-Jones Measuring holistic carbon footprints for lamb and beef farms in the Cambrian Mountains Initiative. CCW Policy Research Report No. 10/8 September CCW and Welsh Assembly Government, Cardiff, UK. Tilman, D., K. G. Cassman, P. A. Matson, R. Naylor, and S. Polasky Agricultural sustainability and intensive production practices. Nature 418: Vayssieres, J., A. Thevenot, M. Vigne, E. Tillard, and P. Lecomte Comparing energy use efficiency and greenhouse gas emissions for livestock products. Adv. Anim. Biosci. 1: Williams, A. G., E. Audsley, and D. L. Sandars Determining the environmental burdens and resource use in the production of agricultural and horticultural commodities DEFRA Research Project ISO205. Cranfield University and DEFRA, Bedford, UK. Xin, H., R. S. Gates, A. R. Green, F. M. Mitloehner, P. A. Moore Jr., and C. M. Wathes Environmental impacts and sustainability of egg production systems. Poult. Sci. 90: