LCA Food 2012 in Saint Malo, France! - Manifestations et colloques ...

PARALLEL SESSION 2C: QUANTIFICATION AND REDUCTION OF UNCERTAINTY 8 th Int. Conference on LCA in the
Agri-Food Sector, 1-4 Oct 2012
instead of synthetic N fertiliser, with rock P and K used instead of triple superphosphate and potassium chloride.
In all cases, long term emissions and yield gains were accounted for to ensure a mass balance of N.
2.3. Environmental impacts
Emissions to the environment were aggregated into environmentally functional groups as follows. Global
Warming Potential (GWP) was calculated using a timescale of 100 years. The main sources of GWP in poultry
industry are carbon dioxide (CO2) from fossil fuel, nitrous oxide (N2O) and methane (CH4). GWP was
quantified as CO2 equivalent: with a 100 year timescale 1 kg CH4 and N2O are equivalent to 25 and 298 kg
CO2 respectively (Foster et al., 2007).
Eutrophication Potential (EP) was calculated using the method of the Institute of Environmental Sciences
(CML) at Leiden University (http://www.leidenuniv.nl/interfac/cml/ssp/index.html). The main sources are
nitrate (NO3 - ) and phosphate (PO4 3- ) leaching to water and ammonia (NH3) emissions to air. EP was quantified
in terms of phosphate equivalents: 1 kg NO3-N and NH3-N are equivalent to 0.44 and 0.43 kg PO4 3- ,
respectively.
Acidification Potential (AP) was also calculated using the method of the Institute of Environmental Sciences
(CML) at Leiden University. The main source in poultry industry is ammonia emissions, together with
sulphur dioxide (SO2) from fossil fuel combustion. Ammonia contributes to AP despite being alkaline; when
emitted into the atmosphere, it is oxidized to nitric acid. AP was quantified in terms of SO2 equivalents: 1 kg
NH3-N is equivalent to 2.3 kg SO2.
Primary Energy Use included all the energy needed for extraction and supply of energy carriers.
2.4. Uncertainty analysis
A Monte Carlo approach was applied to quantify the uncertainties of the modelled systems (Wiltshire et
al., 2009, Leinonen et al., 2012a;b). The systems model, together with the animal production sub-model was
run 5000 times, and during each run a value of each input variable was randomly selected from a predetermined
distribution for this variable. The outcome of the analysis was the Coefficient of Variation, which was
used to evaluate the statistical significance of the differences between the systems at 5% probability level.
The uncertainties in the input variables were divided into two groups, namely “alpha” and “beta” errors.
Alpha errors were considered to vary between systems, and therefore were taken into account in statistical
analyses of the differences between the systems. For example, variation between farms in production, feed
intake and energy use figures were all considered to represent alpha errors. In contrast, beta errors were considered
to be similar between the systems, and had no effect in the statistical comparison between the systems,
e.g. the emission factor for N2O from manure or conversion factor from electricity to primary energy.
The errors in the emission factors were associated with errors in the models used to generate them, and therefore
considered as beta errors (Wiltshire et al., 2009). Also, farms from which activity data were obtained
were not restricted to particular climatic zones. However, it should be noted that the emissions themselves
were affected for example by variation related to bird performance (e.g. N excretion), and therefore contained
also alpha errors.
The uncertainties of the input variables were quantified and their distribution functions specified on the
basis of the data from the industry, and they also included potential errors of the mechanistic models. The
error distributions of the emission factors followed the IPCC (2006) guidelines.
3. Results
3.1. Broilers
The number of broiler birds required to produce the expected edible carcass weight of 1000 kg was higher
in the standard indoor system than in the free range and organic systems because the finishing weight was
lowest in the standard indoor system. The length of the production cycle was much higher in free range and
organic systems than in the standard indoor system, thus the feed consumption per bird was also higher in
these systems. This had a major effect on the trends in environmental burdens (Table 3).
Table 3. Global Warming Potential (GWP), Eutrophication Potential (EP), Acidification Potential (AP) and
Primary Energy Use per 1000 kg of expected edible carcass weight in the main broiler production systems in
the UK. The Coefficient of Variation based on the alpha errors is given in the parentheses.
200