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

PARALLEL SESSION 3C: SHEEP AND DAIRY PRODUCTION SYSTEMS 8 th Int. Conference on LCA in the
Agri-Food Sector, 1-4 Oct 2012
3.5. Hotspot assessment
Where hotspots were found, variability among provincial averages was analysed to understand the underlying
trends. For climate change, the largest source of variability was the carbon footprint of energy used at
the farm. This, in turn, was linked to the electrical-grid mix used in each province (0.025-0.17 kg CO2e/kg
FPCM), with provinces supplied by hydroelectricity having the smallest footprint.
Next in range of variability was manure management and feed production. For the former, the percentage
contribution of CH4 and N2O varied furthermore, where solid storage caused higher N2O emissions and liquid
storage was dominated by CH4 emissions. The province with the lowest manure footprint had 55% of
farms using solid storage and tanks with a natural crust for liquid storage. Meanwhile, the province with the
highest manure footprint had many liquid lagoons (37% of farms overall), driving its contribution much
higher.
Table 3. Comparison of manure storage practices and their emissions (FPCM = fat-and-protein-corrected
milk).
Averaged Solid Solid on Liquid with Liquid with Liquid with Liquid
Manag’t Storage Pasture Crust Cover No Cover Lagoon
% of manure Canadian avg: 34% 13% 37% 5% 8% 3%
kg CO2e/kg FCPM 0.32 0.17 0.30 0.31 0.34 0.35 0.96
% CH4 47% 15% 4% 40% 52% 61% 86%
% N2O 53% 85% 96% 60% 48% 39% 14%
A similar evaluation of contributions was done for the feed-production stage (Table 4), especially since it
was a hotspot in all impact categories. Rations in the Eastern provinces are more corn by-product rich while
rations in the West use canola meal.
Table 4. Contributions of the feed production stage to impacts of crops and rations
Hay Corn Dry Corn Small Soybeans Rations Rations
Silage
Grains
East West
Overall weight 46% 20% 11% 10% 8% 6% 3%
Climate change 33% 10% 17% 15% 4% 13% 7%
Ecosystem quality 42% 10% 11% 8% 19% 6% 3%
Human health 39% 19% 14% 15% 8% 4% 2%
There was variability in contribution based on type of crop. Corn grain and small grains had relatively
more impact on climate change, mainly to fertilisation rates, while soybeans had less. The same trend was
observed for impact on human health, as ammonia emissions also depended on fertilisation rates. With potential
impacts on ecosystem quality, there were different factors in play. Corn silage, for example, benefitted
from the highest yield per hectare (approximately 13 t of dry matter (DM)/ha) while soybeans were in the
lower range, with a yield around 2 t DM/ha, each affecting impacts on land use.
In evaluating sensitivity of certain parameters, it was important to consider the variable geopolitical and
socio-economic context that influences practices, while remembering that agriculture is a complex system
with many inter-related cause-effects chains that are difficult to model. With this in mind, and to perform a
meaningful scenario analysis, a few “what-if” scenarios were modelled based on current practices and wellknown
alternatives (as opposed to marginal and emergent practices). These tested animal-replacement practices,
alternative fertiliser types, fat supplements in feed, and manure management practices. While animalreplacement
ratios affect the three main hotspots (feed production, enteric fermentation and manure storage),
most of these options were limited in their overall impact and targeted different hotspots (data not shown).
4. Discussion
Applying LCA to production across Canada required a method that allowed and facilitated representation
of differing provincial contexts, both in terms of practices (inventory) and geophysical conditions (CFs).
Results showed that variability was driven by both aspects, depending on the indicator. By separating the
two, it was easier to understand where reductions are possible and where observance of best practices is even
more important (sensitive areas based on location).
Achieving consistency in data collection and interpretation across Canada, however, was a challenge.
Additionally, while great uncertainty exists in modelling emissions from soils, variability is also great due to
organic (manure) and synthetic fertiliser application dosages and techniques. For the most part of dosages,
only recommendations exist, from which assumptions were derived.
331