«International 17 Workshop th Nitrogen The was jointly organised by Teagasc and AFBI Printed by Print Depot Suggested citation Authors, 2012. Title ...»
3. Results & Discussion The product carbon footprint of milk from PS and CS was similar at 1.1 kg CO2eq kg-1 ECM, if C sequestration and land use change were excluded. However, milk from PS had much lower PCF than milk from the CS, if C sequestration of pastures and land use change induced C losses were included (Figure 1). PCFs were dominated by enteric CH4 emissions and field-level N2O emissions
Nitrogen Workshop 2012
in the PS and by enteric fermentation and the resource inputs in the CS (Figure 1). Taking into account C sequestration of grasslands associated with forage production and pasture grazing considerably improved the PCFs of milk, i.e. total GHG emissions were offset by 48% and 9% at PS and CS, respectively (Figure 1). Animal feeding in CS was essentially based on concentrates and it has therefore been considered necessary to include GHGs related to concentrate components. The loss of soil C through changing land use (grassland/forest to arable land) is particularly associated with soybean cultivation in South America, which significantly worsens the PCF of the high-input CS farm by producing an additional 0.34 kg CO2eq kg-1 ECM (Figure 1).
4. Conclusion The investigated dairy farms exhibited large differences in their PCF of milk, not only regarding total amounts of GHGs but also regarding the contribution of GHG sources. Owing to their potential of sequestering atmospheric CO2 in grassland soil C stocks, pasture-based systems hold the potential to improve the PCF of milk. However, estimation of soil C sequestration still lacks accuracy and further knowledge and methodological standardization is required to increase the confidence in estimations and to achieve comparability among systems and studies.
References Biskupek B., Patyk A., Radtke J. 1997. Daten zur Pflanzenölproduktion. In: Nachwachsende Energieträger, Kaltschmitt M., Reinhardt G.A. (Eds). Braunschweig/Wiesbaden. 167-225. (In German) Clemens J., Trimborn M., Weiland P. and Amon B. 2006. Mitigation of greenhouse gas emissions by anaerobic digestion of cattle slurry. Agriculture, Ecosystem & Environment 112, 171-177.
FAO. 2010. Greenhouse Gas Emissions from the Dairy Sector. A Life Cycle Assessment. Food and Agricultural Organization of the United Nations. Rome.
Kirchgessner M., Windisch W., Müller H.L., Kreuzer M. 1991. Release of methane and carbon dioxide by dairy cattle.
Agribiol. Res. 44, 2-3.
Eriksson I.S., Elmquist H., Stern S. and Nybrant T. 2005. Environmental Systems Analysis of Pig Production - The Impact of Feed Choice. Int J LCA 10 (2), 143-154.
Patyk A. and Reinhardt G. 1997. Düngemittel-, Energie- und Stroffstrombilanzen. Braunschweig/Wiesbaden. (In German) VDLUFA (2004) Humusbilanzierung – Methode zur Beurteilung und Bemessung der Humusversorgung von Ackerland. Verband Deutscher Landwirtschaftlicher Untersuchungs- und Forschungsanstalten. Bonn. (In German)
Nitrogen Workshop 2012
The effect of nitrogen fertiliser application rate on nitrous oxide emission intensities of arable crop products Thorman, R.E.a, Pappa, V.A.b, Smith, K.a, Rees, R.M.b, Chauhan, M.a, Bennett, G.c, Denton, P.d, Munro, D.G.a, Sylvester-Bradley, R.a a ADAS UK Ltd., Battlegate Road, Boxworth, Cambridge, CB23 4NN, UK b SAC, West Mains Road, Edinburgh, EH9 3JG, UK c ADAS UK Ltd., Gleadthorpe, Meden Vale, Mansfield, Nottingham, NG20 9PF, UK d ADAS UK Ltd., Bentinck Farm, Rhoon Road, Terrington St Clement, King’s Lynn, Norfolk, PE34 4HZ, UK
1. Background & Objectives The demands of feeding a rapidly expanding global population without exacerbating climate change are a major challenge which urgently needs to be addressed. Arable farms in Northern Europe are likely to play an increasingly vital role in food production, but most crops receive significant amounts of inorganic nitrogen (N) fertilisers, which can be associated with large losses of the greenhouse gas (GHG) nitrous oxide (N2O). In national GHG inventories where direct N2O emissions from soil are calculated using the standard Tier 1 Intergovernmental Panel on Climate Change (IPCC) methodology (IPCC, 2006) and in current commercial GHG accounting procedures, direct N2O emissions from soil are assumed to be linearly related to N inputs. This assumption implies that drastic reductions in N fertiliser use and crop productivity would be required to minimise N2O intensities of crop products (kg N2O-N per kg product). We hypothesise that the response of annual N2O emissions to N supply is, to some extent, related to the surplus of N supply over crop N uptake (Figure 1a). If so, fertiliser N application strategies to minimise N2O emission intensities of crop products may have much less severe implications for crop productivity (Figure 1b).
Figure. 1. Modelled effects of N supply on (a) crop production (circle), and on N2O emissions if related directly to N supply (diamond; as estimated by IPCC Tier 1 approach) or to the balance between N supply and crop N uptake (triangle; as hypothesised here), and (b) consequent contrasting effects of these scenarios on N2O emission-intensities of crop products.
If N2O emissions are entirely N-balance related, N amounts that minimise N2O intensities would be similar to current use, with little effect on crop productivity. The ongoing research described here is assessing the shapes of the responses in annual N2O emissions to increasing amounts of applied N
Nitrogen Workshop 2012
for the main UK arable crops (cereals, sugar beet and oilseed rape), relating these to economic optimum N amounts for crop production, and suggesting better means of N fertiliser management.
2. Materials & Methods At four UK sites: 1) south east England (clay loam), 2) south east England (silt loam), 3) central England (loamy sand) and 4) central Scotland (sandy loam), N2O emissions (5 static chambers/plot) were monitored from replicated (x3) plots for 12 months, following spring ammonium nitrate fertiliser applications in the range nil to 240% of recommended N. Up to three fertiliser applications were made in order to achieve target rates. In the first 2 weeks after each fertiliser application, 7 measurements were taken, decreasing in frequency to give a yearly total of 40-50 measurements.
The crops studied were winter wheat (ww) & winter oilseed rape (site 1), ww (site 2), spring barley (sb) & sugar beet (site 3) and ww, sb & winter barley (site 4). Yield measurements were also taken.
3. Results & Discussion Cumulative direct N2O emissions from all eight experiments were small and emissions were not generally affected immediately following N fertiliser application due to abnormally dry spring conditions and low soil moisture contents. Significant emissions followed later rainfall events, peak emissions (up to c.70 g N2O-N ha-1 d-1) being measured up to c.4 months after the last fertiliser had been applied. Total annual emissions ranged from 0.3 to 1.1 kg N2O-N ha-1 with nil N applied, and calculated emission factor (EFs) for N lost as N2O at 120% of the recommended N rate (excluding the emission with nil N) were generally 0.30% (all 0.60%) of total N applied, compared to the IPCC Tier 1 (EF) of 1.0% (IPCC, 2006). Despite the apparent delay in N uptake compared to normal conditions, most responses of N2O emission to N application rate were linear. However, in Central Scotland the response for spring barley was clearly best described by a non-linear relationship while those responses for winter wheat and winter barley tended to show slight nonlinearity. The lack of non-linear responses in the majority of experiments is surprising given that the generally dry conditions inhibited emissions soon after N application, and that most peak emissions occurred after the main phase of crop N uptake. It remains to be seen whether more normal dynamics of N application, crop N uptake and N2O emission will lead to non-linear responses, which will enable an appropriate N applications strategy to be devised to minimise N2O emissions.
4. Conclusion The main differences between these results and the assumptions currently used in GHG accounting and the UK GHG inventory are the small EFs in relation to fertiliser N rate. However, the results were affected by a dry spring so cannot yet be considered representative of UK arable conditions. If linear responses prove to be the norm, it may have to be accepted that high crop yields will depend on exceeding minimum direct emission intensities, although if the EF for total N applied is small (say 0.50%), this difference will be slight.
Acknowledgements This research is sponsored by the UK Department for Environment, Food and Rural Affairs and Scottish Government through Sustainable Arable LINK Project LK09128, and we acknowledge contributions of ADAS, Agricultural Industries Confederation, Bayer CropScience, British Sugar, Country Land and Business Association, The Co-operative Group, Frontier Agriculture, GrowHow UK, AHDB-HGCA, Hill Court Farm Research, NFU, North Energy Associates, PGRO, Renewable Energy Association, Rothamsted Research (North Wyke), SAC, Scotch Whisky Research Institute, SoilEssentials, Vivergo fuels, Warburtons and Yara UK.
References IPCC. 2006. 2006 IPCC Guidelines for National Greenhouse Gas Inventories. Eggleston H.S., Buendia L., Miwa K., Ngara T., Tanabe K. (eds.). IGES, Japan.
Nitrogen Workshop 2012
Carbon footprint of Irish milk production: can white clover make a difference?
Yan, M.-J.a, Humphreys, J.b and Holden, N. M.a UCD School of Biosystem Engineering, University College Dublin. Belfield, Dublin 4, Dublin, Rep. of Ireland Animal and Grassland Research and Innovation Centre, Teagasc, Moorepark, Fermoy, Co Cork, Rep. of Ireland
1. Background and objectives The greenhouse gas (GHG) emissions of dairy production are of considerable concern to Ireland.
Carbon footprint (CF) calculated by life cycle assessment (LCA) is a useful tool to trace the global effect of GHG emissions due to local production. The goal of the study was to model the CF of milk, produced from pasture reliant on fertilizer N (FN) and white clover (WC), at the research farmlet scale, and to evaluate the uncertainty of CF when up-scaling. The four stages of LCA methodology were implemented as follows according to ISO 14040.
2. Materials & Methods The foreground data was based on the comparative experiment at Teagasc Solohead Research Farm for the years 2001-2006 (Table 1; Humphreys et al., 2008; 2009). All cows were Holstein-Friesian breed. The soil on-farm has clay-loam texture and ten-year average rainfall was 1005 mm. The functional unit (FU) was defined as 1 kg energy corrected milk (ECM). The system boundary was set at the farm gate and included relevant pre-farm processes (production and transportation of fertilizer and concentrate feed) and the on-farm production. Soil carbon sequestration, pesticides, medicine, plastic sheets etc. were not included. Economic allocation between milk and meat from surplus calves and culled cows (both 91% for FN and WC) was used applied using average market prices between 2000 and 2006. Emission factors (EFs) were taken from relevant literature while the EF for enteric CH4 was determined by estimating net energy for maintenance, lactation and pregnancy. Simapro 7.3 was used for the LCA modelling and all background processes (such as fertilizer production) were chosen from Ecoinvent 2.2 database. The “IPCC 2007GWP 100a” was selected to assess the GHGs per FU, which defined that the GWP of CO2 (time span of 100 years) as 1, of CH4 as 25, and of N2O as 298. Ratio sensitivity was performed to assess the impact of the uncertainty of EFs on the comparison between the systems.
3. Results & Discussion The CF of WC was 11 to 24% lower than FN across the range of fertilizer N input. With physical allocation (85% to milk), the average CF for WC was 0.81 kg CO2 eq kg ECM-1 and significantly
lower than FN, which was 0.98 kg CO2 eq kg ECM-1 (P 0.001). With economic allocation (91% to milk) the difference was also significant, with 0.87 and 1.05 kg CO2 eq kg ECM-1 respectively. The majority of GHGs were within Ireland and contributed more to WC (85%) than to FN (80%). The contributors that accumulated c.
95% of GHGs were enteric CH4 (WC: 51% and FN: 43%), excreta deposition (WC: 13%, FN: 11%), fertilizer spreading (WC: 6%, FN: 12%), fertilizer production (FN: 10%), electricity production (WC: 8%, FN: 6%), indirect N2O (both 6%), slurry storage (WC:
4%, FN: 3%), concentrate production (WC: 4%, FN: 3%), and slurry spreading (WC: 3%).
Significant correlation was found between surplus N per kg ECM and CF (Figure 1, R2 = 0.66, P 0.001), which indicated that a 1 g reduction of on-farm surplus N could reduce CF by 26 g CO2 eq.
A similar relationship (29 g CO2 eq) was reported by Schils et al. (2006). The ratio sensitivity analysis revealed that to reverse the priority of WC and FN, changes to emission factors (EF) and assumptions had to be much greater than the uncertainty range found in the literature. For example, EF of enteric CH4 in WC needed to be increased by 24 to 59%, but enteric CH4 from cows fed white clover was found to be similar (van Dorland et al., 2007) to those fed grass, and neither higher gross energy content or a larger methane conversion factor of clover is suggested to belikely (Andrews et al., 2007).
Figure 1. Relationship between on-farm surplus N and CF (economic allocation).
Dots: WC; triangles: FN
4. Conclusions The carbon footprint (CF) of milk production from WC was 11 to 24% lower than from compared with FN swards. Sensitivity analysis showed the model on research farmlets was robust and white clover could reduce greenhouse gas emissions. CF model requires further research on system efficiency, productivity and profitability to translate effects to the national scale.