Development of proxy indicators of methane output by sheep using rapid-throughput laboratory technologies (PhD)



Methane production by ruminants is a significant contributor to agricultural greenhouse gas emissions (Webb et al., 2013). However, current values used to estimate methane output by sheep are default values and do not take into account animal and dietary factors that may affect methane output (Bernstein et al., 2007). Strategies to reduce ruminant methane output are the focus of a large body of research (Iqbal et al., 2008) and, in order to implement these strategies fully, a greater understanding of factors that influence ruminant methane emissions is necessary.
The "gold standard" method for measuring methane output by sheep is the use of respiratory chambers (Blaxter and Clapperton, 1965). However, this method is expensive, time-consuming and labour intensive, making it unsuitable for use in an on-farm situation. The work presented in this thesis explores the potential of three proxies to estimate methane output by sheep, which could be used or adapted to be used as a practical means of estimating methane emissions from sheep on a large scale.

The proxies investigated here are a Laser Methane Detector (LMD), used to take measurements of methane concentration from air expired by sheep, in vitro gas production analysis of feeds offered to sheep, and Fourier-transform infrared spectroscopy (FTIR) analysis of feeds offered to sheep. Predictions of methane output obtained from each of the proxies are validated using respiratory chamber measurements taken from sheep offered a variety of feeds during different experiments.

With further development and validation, all three proxies presented in this thesis demonstrate potential to be used to successfully estimate or predict methane output by sheep as measured in respiratory chambers. A novel and very successful approach to the method for use of the LMD and calculation of daily methane emissions from LMD data is presented in this thesis. However, the methods used were relatively labour intensive and time-consuming. Further work should, therefore, focus on simplifying these methods as much as possible. To my knowledge, the results presented for in vitro gas production and FTIR spectroscopy are also novel, although these are established methods. Both of these methods are rapid-throughput techniques and, therefore, have real potential to be used on a large scale. Further work using larger data sets may provide a more comprehensive idea of the aspects of feeds that affect their methane potentials.

Industry Message:

One of the challenges faced in the reduction of methane emissions by ruminants is the successful implementation of any practices that are shown to be beneficial. Hegarty et al. (2010) argue that there is a lack of policy to motivate farmers to reduce their emissions, with emphasis being entirely on productivity and profit. Without a means of accurately estimating methane output by sheep, or the effects of any measures taken to reduce methane emissions by sheep at a large on-farm scale, it is difficult to introduce incentives for farmers to take measures to reduce methane emissions. Pinares-Patiño et al. (2013) also highlighted the need for shorter and alternative methane emissions measurements in order to facilitate the establishment of selection lines of low-methane-producing animals, as methane output has been shown to be a heritable trait.
The aim of this project was to develop and validate proxies that could potentially be used to obtain accurate estimates of methane output by sheep. Proxy indicators for methane output by sheep could provide a useful tool for accurately estimating methane output at an on-farm level, as well as measuring the impact of introducing different management systems or diets on methane emissions. A brief discussion of the implications for industry of the results for each potential proxy investigated during this project is provided in the following sections.

Laser Methane detector
The results obtained during this project suggest the LMD has potential to be used as a means to estimate methane output by sheep. Although the methods used to collect LMD data were simple, they were relatively time consuming, required close contact with animals on a daily basis, and required expensive equipment (i.e. the LMD). However, the work presented in this thesis demonstrates that LMD measurements can be used to estimate daily methane output by sheep that is in the expected magnitude and correlates with daily methane output as measured in methane chambers. Furthermore, the LMD could potentially be used to rank animals in terms of their methane production, thereby facilitating the introduction of breeding programmes for animals that are low methane producers (Hegarty et al., 2007; Pinares-Patiño et al., 2013).
Further work would be required in order to establish methods for taking LMD measurements that could be used at a large on-farm scale. The LMD can be used at a range of up to 150m, which could allow for collection of LMD measurements from grazing animals, without causing any disruption to animals. However, there is a paucity of published data regarding LMD measurements taken from ruminants at long distances. Similarly, there is a lack of published data regarding the use of the LMD the estimation of methane output from groups of sheep. However, the use of open-path lasers for this purpose has been investigated with some success (Tomkins et al., 2011). The results of this initial experimentation presented in 4.1.4 suggest that further work regarding group LMD measurements should be pursued as a means of quickly and simply estimating methane output from animals, making the LMD potentially useable at a large on-farm scale.
In vitro gas production
The in vitro gas production technique could provide a quick and simple means of estimating methane emissions by sheep based on the methane production potential of the feeds offered and intake measurements or estimates of feed intake. The technique requires small amounts of feed material, which is freeze-dried, allowing analyses to be conducted several weeks or months after collection. Multiple samples can be analysed at the same time, making the technique practical for use on a reasonably large scale.
The results presented in Section 4.2 demonstrate that there were significant positive correlations between methane production per gram of apparently digested DM and per gram of DM using the in vitro method and those obtained using the in vivo methane chamber experiments. The gas production technique, therefore, has the potential to be used as a proxy to predict methane emissions from different animals given different diets, provided a reasonable estimate of the DM intake of animals can be made. The challenge posed by measuring DM intake may limit the usefulness of the in vitro gas production technique, particularly as DM intake does not correlate particularly well with body weight (Lassey et al., 1997). However, in cases where animal intake is already monitored, taking samples to be used for in vitro gas production analysis would be easily manageable. The technique requires collection of rumen fluid, which may present a problem as it requires either fistulated animals or stomach tubing, both of which are subject to Home Office regulation and require careful management. However, if the technique were to be used as a means of predicting methane potentials of diets from farms, the samples would be collected on-farm and sent to a research facility for analysis; this would also reduce any variation in results due to laboratory conditions and staff performing analyses.
FTIR spectroscopy
Fourier-transform infrared (FTIR) spectroscopy is a rapid-throughput laboratory technique that requires a very small amount of plant or feed material (Allison et al., 2009), making it ideal as a proxy indicator for methane output by sheep provided that it can be used to successfully predict methane emissions using feed samples or faecal matter. The data presented in Section 4.3 demonstrate that it is possible to distinguish between different feed samples on the basis of their FTIR spectra. Furthermore, there is potential for FTIR spectra of feed samples to successfully predict daily methane emissions from sheep as measured in methane chambers.
The main limitation of the data presented in this thesis is that the data set is relatively small compared with data sets presented in the literature: for example, Belanche et al. (2013) used 663 samples, representing 80 feed types. Conclusions are therefore limited from this small data set regarding the aspects of the FTIR spectra that enable the prediction of methane emissions. Further work should, therefore, involve using larger data sets, with contrasting feed samples, to perform a similar analysis. This may provide a more comprehensive idea of the properties of feeds that cause animals to produce more or less methane, whether this is related to feed composition, or perhaps whether certain properties of feeds affect feed intake by animals.
While further work is required to optimise the methods used to estimate or predict methane output by sheep, the data collected during this project provide evidence for the potential of three proxy indicators for this purpos
Beef & Lamb
Project code:
01 October 2010 - 30 September 2013
Hybu Cig Cymru - Meat Promotion Wales (HCC)
AHDB sector cost:
Total project value:
Project leader:
Aberystwyth University


7779 Final Report Dec 2014

About this project

The Challenge

Methane is produced by microbial fermentation in the rumen. It is a potent greenhouse gas and its production requires energy, resulting in production losses from ruminant livestock. Accurately measuring methane production by ruminants presents difficulties due to considerable variation between animals, depending on factors such as diet and body weight. Although the “gold standard” method for measuring ruminant methane output is the use of methane chambers, this is too time consuming and labour intensive to be used in large-scale on-farm situations.

The Project

This project aims to develop proxies, which can be used to estimate methane output by sheep at an on-farm scale.

The potential proxies to be investigated during this project include the methane potentials of various feeds, obtained using an in vitro gas production technique and a selection of plants is currently being collected on a monthly basis for this purpose. The use of a laser methane detector (LMD), which uses infrared spectroscopy to give measurements for methane concentration in the column of gas, is also being explored. Finally, the detection of archaeol, a membrane lipid of the methanogenic archaea, in faeces is another potential proxy.

Methane emissions from sheep are being measured using the chamber technique, and these are being used to calibrate methane production measured using the proxy techniques under investigation.  Results to date indicate that the LMD is sensitive enough to detect eructation peaks of methane from the animal’s mouth as well as normal breath concentrations, and investigations are currently focussed on the length of time required to achieve  representative data from individual animals, and whether environmental measurements (i.e. from the sheep’s surroundings) provide useful data.


Sophie Doran