Knowledge for better food systems

National Trust report on carbon footprints in various beef production systems – and expert comment

A report published by the National Trust entitled What’s your beef? Compares the cradle-to-farm-gate emissions of ten tenanted National Trust farms, selected as representing a cross section of different beef production systems, including 4 organic, 4 conventional but extensive, and 2 semi intensive farms.

The study follows the methodology set out in the Publicly Available Specification guide PAS2050. Additional scenarios were developed to explore the potentially mitigating effect of carbon sequestration by grassland and as a result of organic conversion.

The study finds that while the carbon footprint of grass-fed and conventional farms were comparable (ie. based on the PAS 2050 methodology), the carbon sequestration contribution of well-managed grass pasture on the less intensive systems reduced net emissions by up to 94 per cent, even resulting in a carbon 'net gain' in upland areas. The farms that had recently converted to organic status showed even greater gains.

In other words, the study finds that more extensive, and especially organic findings, perform better from a GHG perspective once the soil carbon sequestration effects are taken into account – based upon the assumptions that underpin the sequestration scenarios.  I’m copying what it says about the assumptions (and the sources of info for the assumptions in square brackets) and the relevant graphic.

As soil carbon densities were not analysed on the farms, published research was used to develop two scenarios to explore the potential effect of carbon sequestration on the PAS 2050 footprint results:

A Grassland carbon sequestration on all farms. Permanent grassland sequesters carbon at a rate of 0.24tC/ha/year [Janssens, I.A. et al. (2005) The carbon budget of terrestrial ecosystems at a country scale – a European case studyBiogeosciences, 2, pp. 15–26]  This assumption was used in a recent beef carbon footprint study for theCountryside Council for Wales [Taylor, R.C. et al. (2010) Measuring holistic carbon footprints for lamb and beef farms in the Cambrian Mountains Report for Countryside Council of Wales]

B Grassland and cropland carbon sequestration on organic farms. During conv ersion from conventional to organic agriculture, soil carbon levels improve at the following rates over 20 years: grassland: 0.42tC/ha/year; cropland: 0.55tC/ha/year [Azeez, G. (2009) Soil Carbon and Organic Farming Soil Association].

Since the sequestration issue seems to be so critical, I sought comment on the validity of the assumptions from two expert soil scientists.  One of them is Prof Pete Smith at Aberdeen University (and lead author of the chapter on agricultural mitigation in the IPCC’s 2007 Fourth Asssessment Report – see here

Pete writes as follows: “This use of an assumed C sequestration rate for grassland is popping up more and more – most of these traced back to the Soussana et al. review of grassland flux sites from CarboEurope. These are based on flux measurements and we don’t know where the C is going – could be leaching or lost as particulate OC – in any case the flux measurements are uncorroborated by soil C stock change measurements. Rothamsted (and other) permanent long term grass experiments record no such long term increase in SOC, so the use of an assumed C sequestration rate for grasslands is highly uncertain. 

Given that soil C sequestration is so critical to the overall GHG balance or any grass based system, conservative assumptions about SOC gain should be used (in the National Trust report, to be fair, they explore scenarios of soil C sequestration rather than assuming that these rates are achieved in reality).

The SOC accumulation rates supposedly found at the farms in Northumberland (3-5 t C ha-1 yr-1) also seem too large to be true. If we assume that a generous 20% of C remains in the soil, this would mean an annual addition of up to 25 t C ha-1 yr-1 – which is a huge amount at reasonable stocking densities. If the C addition is coming from manure, that could also be a lot of N2O emissions!!

I suspect that recorded changes in SOC on these farms, of this magnitude, are either picking up legacy effects of land use change (i.e. grasslands converted from cropland not too long ago), or are miscalculating changes in concentration to stock changes either by not accounting to bulk density change, or not using an equivalent soil mass approach. An annual SOC accrual rate of 5 t C ha-1 yr-1 would give SOC increases of 250 t C h-1 yr-1 over 50 years. This is way to high to be due to management. Based of a meta-analysis of long term experiments, Smith et al. (2008) suggest that grassland management can increase SOC at a rate of 0.81 (-0.11 to 1.50 95% CI) t CO2 ha-1 yr-1. Divide the CO2 numbers by 3.667 to get to C and you have a median estimate of 0.22 t C ha-1 yr-1 – very much at odds with estimates of 3-5 t C ha-1 yr-1.”

The other is Professor Keith Goulding, Head of the Dept of Sustainable Soils and Grassland Systems at Rothamsted Research and Vice President of the British Society of Soil Science.  Keith agrees with Pete’s comments above and adds:

“Looking at our long-term experiments, the rate of sequestration will depend on the age of the grassland. Anything much over 50 years old will sequester very little if any C. Arable does not sequester C unless it is Min- or Zero-Tilled (and then N2O emissions might offset any gains from C) or it receives very large amounts of manure or other bulky organic material, and see Pete’s comment on N2O from manures. Organic rotations will sequester a little C if the ley phase is much longer than the arable phase. There is another issue, that of land requirements. Extensive organic systems require more land, which impacts on other ecosystem services.”

You can download the National Trust report here.  The accompanying press release is here.

You can read related research by browsing the following categories of our research library:

Add comment

Member input

Plain text

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.


FCRN Admin's picture
Submitted by FCRN Admin on

7 August 2012

We were pleased to see that the “What’s Your Beef” report stimulated some interest and comments from Professors Smith and Goulding (FCRN Newsletter, Mailing 31 May 2012). However, on reflection, there are some points arising which we feel need clarification.

We asked Best Foot Forward to construct scenarios to include potential carbon sequestration under grassland use, because we considered the omission of sequestration and land use changes from standard methods was unsatisfactory. Obviously the value used for SOC gain will affect results, which is why a published value, 0.24tC/ha/y, from the Janssens (2005) paper was used. Professor Smith was a co-author of this paper, and he quotes a similar value of 0.22tC/ha/y from his 2008 paper.

Professor Smith mentions “SOC accumulation rates supposedly found at the farms in Northumberland” –there were no Northumberland farms in the beef study. We made reference in passing to a 2010 study of land management for soil carbon at the National Trust Wallington Estate in Northumberland which showed the impact of farm management on soil carbon stock, and we suggested it would benefit land managers to use soil analysis to determine the carbon status of their own soils. In the Wallington report, Madeleine Bell estimated that if all rough pasture on the estate (a total of 1200 hectares) could achieve optimum soil carbon, an annual sink equivalent of up to 21,500 tonnes CO2 might be achieved over a period of 20 years, and that the gains from good grassland management have the potential to outweigh soil carbon losses from limited conversion to arable. If we understand this estimate correctly, this is equivalent to about 5tC/ha/y.

The reference in “What’s Your Beef” to soil carbon stock increases of 3-5tC/ha/y was based on calculations made by individual farmers – in one case the soil carbon status was monitored over a period of seven years. Questions arise about the influence of sampling frequency, arable reversion, manure applications and grazing system on the results, but the calculations from SOM to SOC appear correct and accounted for bulk density. We did point out that such amounts were an order of magnitude greater than amounts typically reported in the literature, that it followed that more information is needed about soil carbon changes under carbon-friendly management, and that we would address this by experiment under real farming conditions. However, it is also worth noting that a range of C sequestration values from 0.11 to 3.04tC/ha/yr was reported by Conant, R.T. et al (Ecological Applications, 11(2), 2001, pp. 343–355. Grassland management and conversion into grassland) who concluded that grasslands can act as a significant carbon sink with the implementation of improved management. This paper is cited several times in IPCC AR4 Ch 8, 2007.

We felt the scenario building provided an insight into what evolved LCA methods might show in the future with regard to the merits of different livestock production systems. The Eblex “Down to Earth” report (2012) indicates the desirability of movement in this direction, and we note also that Defra project SP1113 is calling for research to develop a framework for capturing cropland and grassland management impacts on soil carbon in the UK LULUCF inventory and to calculate mitigation and offset potentials through changing cropland and grassland management in the future.

Our report illustrates that accounting for carbon emissions alone can be highly misleading when comparing different farm systems and that methodologies that include land use change and sequestration give a more rounded picture of how carbon features in farm systems.

Rob Macklin, Food and Farming Specialist
John Kay, Farm Wastes and Resources Consultant
The National Trust, Heelis, Swindon

FCRN Admin's picture
Submitted by FCRN Admin on

Pete Smith and Keith Goulding reply to say that they agree with the authors that both carbon sinks and sources need to be accounted for (and the need for more research in this area).  The disagreement really lies in the assumptions / estimates of soil C sequestration rates under long term grassland – Pete and Keith think they will be rather small, while the study suggests that they could be larger than P or K consider likely.  P and K  also think it necessary to address the issue of equilibrium levels of SOC: one needs to know what this is for each system and how near or far away the system is for those farms that converted to organic.

Simon Ward's picture
Submitted by Simon Ward (not verified) on

Apart from the considerable variation in the estimate of carbon accumulation following land use change and the dangers of extrapolating equilibrium climax carbon concentrations to the general situation, inclusion of land use change within simplistic life cycle analysis also introduces the issue of stocking density.  The accumulation or loss is largely independent of stocking density so that a low stocking density reduces emissions associated with the product but so too is  the contribution to world food supply.  This is not necessarily addressed by the economic optum although distortion would be greater if value were applied to GHG reduction.

There are other boundary issues such the application of manure emissions to an unrelated output on the farm in a complex farming environment.

I would argue that the product from any farming system is not wheat or beef but food energy and protein and that a standard reporting needs to be adopted stating the GHG per unit of food energy and protein produced from the whole farm activity and the food and protein energy produced per unit area given that land is the economic limitating factor.



Peter Melchett's picture
Submitted by Peter Melchett (not verified) on

The National Trust’s report, ‘What’s your beef?’ was criticised by Professors Goulding and Smith, but the Soil Association supports Rob Macklin of the National Trust’s view that their Trust report ‘illustrates that accounting for carbon emissions alone can be highly misleading when comparing different farm systems and that methodologies that include land use change and sequestration give a more rounded picture of how carbon features in farm systems’ (FRCN 08.08.12).  We would add that such changes, if maintained, can have very long-term impacts.


It is clear that changes in agricultural systems affect rates of soil carbon sequestration or loss (Gregorich et al., 2001; Wu et al., 2003; VandenBygaart et al., 2004; Xie et al., 2007), and that some such changes can continue to make an impact over very long periods of time. For example, in the long-term Hoosfield Barley experiment at Rothamsted, the organic carbon in the soil that has received farmyard manure for 150 years is still increasing, now slowly, after more rapid increases in the early years.  The organic carbon in soil which only received farmyard manure from 1852 to 1871, and none since, is still declining,  Neither of these has yet reached equilibrium (Jenkinson & Johnston, 1977; Johnston et al., 2009). Similarly, in an arable to grassland experiment at Rothamsted, an old arable soil converted to permanent grass in 1949 (and now developing to scrub/woodland) is still gaining soil carbon, while soil carbon in an old grassland soil, ploughed in 1948 and used since then for arable crops, is still decreasing. Again, neither soil has yet reached a new equilibrium (Johnston et al., 2009).


In the National Trust research, the highest sequestration of carbon to soils occurred on their farms managed under organic systems.  However, net greenhouse gas emissions were slightly higher than conventional, grass-based, non-intensive beef production, which the National Trust attributed to their mix of pasture and better quality feeds.  As Rob Macklin notes, the National Trust farms performed similarly to farms monitored by EBLEX.


The National Trust report was right to stress the importance of incorporating soil carbon sequestration data into carbon footprinting calculations.  The rates of carbon accumulation that they used under grassland relied on published data (of 0.24 t C ha-1 y-1).   Conversion of grassland to organic pasture results in higher rates of sequestration and a figure over 20 years of 0.42 t C ha-1 y-1 was also derived from published data, whilst the conversion of cropland to organic pasture has a higher accumulation rate of 0.55 t C ha-1 y-1(these are all similar to the figure for conversion to organic farming suggested some time ago in a Soil Association report on soil carbon, of 0.56 t C ha-1 y-1).  However, the National Trust report noted in its discussion that some of their particularly motivated farmers have begun to monitor carbon levels in soils, and some have reported increases of 3-5 t C ha-1y-1


In their comments, Prof Keith Goulding and Prof Pete Smith seemed to focus on the farmer-derived figure of increases in soil carbon although this was not the basis of the National Trust report, and as Rob Macklin says, the figures for carbon accumulation they proposed in their critique of the National Trust report were in fact of the same order as those used in the report itself.  Of course there are still uncertainties, and a clear need for more research, in which, as the National Trust’s work over several years shows, individual farmers could play an important role.  There are also boundary issues, as Simon Ward points out, which could include not only levels of production, but also potential changes in patterns of consumption.  However, overall, the National Trust’s conclusions remain valid.  The Soil Association welcomes the steps being taken by the National Trust to highlight the significance of soil carbon sequestration, and the positive aspects of grass-based meat production, and to consider differentiation of product based on production systems that account for ecosystem services.  The conclusion that extensive and organic grass-fed systems may be carbon neutral when carbon sequestration to soils is included reaffirms that this is a climate-friendly way to produce meat.  This should further serve to discourage irresponsible, undifferentiated attacks on all red meat.  The four countries of the UK can produce climate-friendly, grass based beef, lamb and mutton, and we should be encouraging consumption of these products in place of intensively produced red or white meat.


Peter Melchett

Policy Director

Soil Association