Readers' Rating:

(See details)  Rate this article

Our thanks to the author and Conference Organisers, a Committee consisting of both University and Industry colleagues.

The full paper will appear in the Conference Proceedings ('Recent Advances in Animal Nutrition - 2007', edited by Phil Garnsworthy and Julian Wiseman) published by Nottingham University Press in the autumn of 2007 www.nup.com

Courtesy of the 41st Annual University of Nottingham Feed Conference www.nottingham.ac.uk/biosciences/ah/research/conferences.php

Introduction and methods

Environmental life cycle assessment (LCA) is a holistic way of calculating the environmental burdens resulting from the production of a commodity. The two general types of burdens on the environment are resource use (e.g. fossil fuels) and emissions of pollutants (e.g. nitrous oxide). LCA traces resource use back to materials in the ground (e.g. crude oil) and allows for the effort of extraction and delivery to the farm. All production occurs within defined system boundaries, which for agriculture is normally a set of fields with a soil depth of 0.3 m and burdens beyond the farm gate were not considered. All inputs are considered, including feed, bedding, housing, grazing, direct energy for management, transport and breeding overheads.

All production burdens are related to the functional unit of output. These were 20 000 eggs (about 1 t), and 1 m³ milk and for meats, this was the edible carcass weight (live weight x killing out percentage). In animal production, most of the burdens relate to the supply of feed as grazing or concentrates. This can be partitioned further into the use of feed for breeding overheads or the actual production stages (egg or milk production or finishing for meat).

There are additional burdens resulting from excreta and enteric emissions. These include methane (especially from ruminants), ammonia and nitrous oxide (all of which may arise from housing and manure management, together with nitrate leaching from manure use.

Manure is an inevitable by-product of animal production. The emissions from manure and the effort of managing manure are debited against animal production, while its fertiliser value is credited back to animal production. The latter is determined from a model of substitution of manure for either synthetic N in non-organic production or for a grass-clover ley in organic production. Excretion during grazing is handled in a separate model of grass production.

Concentrates come from arable crops, which may be produced domestically or imported and may fed whole, after on-farm milling or are by-products of other activities.

Soya and rape meal, for example, arise after extracting oil from whole crops and wheatfeed from wheat milling. The allocation of burdens from by-product feeds, like soya meal, thus includes primary crop production, transport, oil extraction, blending and delivery to farms. The allocation of burdens between the primary and by-products (e.g. oil and meal) is done by mass partition, modified by the economic value of the meal and oil.

Crop production was modelled with reference to long-term yield responses to N while losses through leaching and denitrification were derived from Rothamsted’s SUNDIAL simulation model. Emissions of nitrous oxide were estimated with the IPCC (1997) methodology. Crop inputs were derived from the UK’s national surveys of pesticide and fertiliser use. Cultivation requirements were based on surveys and previous modelling work.

Structural models were used to represent the animal sectors. These included the breeding overheads and transport steps and linkages. The sheep industry is the most complex with purebred flocks in the hills, uplands and lowlands together with hybrid flocks and other features.

Each industry was represented by sets of linear equations defining the inputs and outputs of each stage (including feed, housing needs, replacement stock, manure produced, intermediates for another sub-system, e.g. store lambs). Changing the structure (e.g. more outdoor poultry) causes the model to find a new solution to give the revised structure and attendant material flows to continue producing the functional unit of 1 t poultry meat.

Organic and non-organic systems were both modelled, but the use of stockless rotations for the basis of crop production (in order to identify crop needs precisely) is acknowledged as being contentious.


Results and discussion

Burdens such as energy use and greenhouse gas emissions (expressed as global warming potential [GWP] on 100 year time scale) for meats (1 t carcass weight) are typically an order of magnitude greater than for 1 t of crop (fresh weight basis), reflecting the concentration of plant nutrients into high quality proteins and other materials. The losses of N from undigested feeds, however, cause large emissions of ammonia and secondary nitrate leaching. The potentials for eutrophication and acidification from meats are thus about two orders of magnitude higher than from crop production.






Care is needed in comparing meats as they are not simply interchangeable, but poultry and pigs take less energy and emit less than ruminants. Ruminants of course can utilise forages and hence make use of land that is not suitable for arable crop production, while pigs and poultry effectively live on arable land. The higher energy requirements for ruminants may seem counter-intuitive, but this is partly because the breeding overheads are much higher than for pigs (about 20 offspring per year per sow) or broilers (about 160 per breeding broiler hen).

The proportion of energy used for breeding overheads for suckler beef is 0.7 falling, but is 0.3 including the calves from dairying. Overheads for others are: sheep (0.45), pigs (0.3) and 0.1 for poultry meat.

Changing the intensity of dairy production has relatively small effects, but burdens tend to decrease with higher production. The ratio of burdens at yields of 8 000 and 5 500 l/year are 0.89, 0.95 and 0.99 for energy, eutrophication potential and GWP respectively. There is a biological limit to the conversion of feed to milk and this limits the potential improvement of higher yields.

We also assumed than higher yielding cows are heavier than lower yielders and so have higher maintenance requirements. Organic milk production takes about 0.7 of the energy of non-organic, although the global warming and eutrophication potentials are about 1.1 times those of non-organic and land use is about 1.5 times higher.

Organic grass production uses substantially less energy than non-organic by using clover rather than synthetic N, so that beef and sheep meat also use less energy than in nonorganic production, although some emissions are higher from organic production. In contrast, poultry meat and egg production use about 10% more energy in organic than non-organic production and emit more pollutants. This mainly results from the lower productivity in organic systems.

The results above are essentially those from the original study (Williams et al., 2006), but values will change as both the modelling and science develop and the industry changes.


Reference

Williams, A.G., Audsley, E. and Sandars, D.L. (2006) Determining the environmental burdens and resource use in the production of agricultural and horticultural commodities. Main Report. Defra Research Project IS0205.


Authors: Williams, A.G., Audsley, E. and Sandars, D.L.
Natural Resource Management Centre, Cranfield University, Bedford, UK.

Readers' Rating:

(See details)  Rate this article