Variation in methane emission from dairy cows and beef heifers measured in open-circuit chambers with an infrared laser spectrometer

Introduction Methane emitted in ruminant husbandry accounts for about one quarter of all anthropogenic CH4 emissions. This makes CH4 from ruminants a target for abatement measures, which have been rather unsuccessful so far (Beauchemin et al., 2008). The controlled conditions in open-circuit chambers provide an ideal platform for testing mitigation strategies which show lasting effects. During four trial periods the past 18 months, we have collected methane emission data of 16 dairy cows (Holstein Friesian) who were fed typical Flemish rations and 18 beef heifers (Belgian Blue) all fed with the same finishing diet. The aim of the present work was to analyse the between animal variation in methane emission.

Material and methods The monitoring system with six individual open-circuit chambers was designed to accurately measure methane, carbon dioxide, nitrous oxide and ammonia emissions, and to allow feeding and milking, and collect faeces and urine separately (Peiren and De Campeneere, 2013). The system was successfully used for the long term trials in the SMEthane project (Yanez-Ruiz et al., 2013). Each trial started with an adaptation period of one month for adaptation to the standard ration. This was followed by a control period of two weeks with a restricted feed intake at 95% of ad libitum feed intake to avoid leftovers. At the end of the second week cows entered the chambers to measure the methane production individually from Tuesday morning till Friday morning. The gas concentrations were measured with an infrared laser optical-feedback cavity-enhanced absorption spectrometer (OFCEAS). The air from all chambers (at each of the six outlets) and the ambient air (two, near the air inlets) were continuously sampled and a gas switching device delivered a sample stream to the gas analyser at intervals of 180 seconds. The last 60 seconds were used to calculate the emissions. The chambers operated at an airflow between 300 and 500 m3 /h. The dairy cows (n = 16) were fed and milked twice daily, with concurrent removal of faeces and urine. Three rations were used which differed in the proportions of maize silage, grass silage, pressed beet pulp, concentrate, rapeseed and soybean meal. The beef heifers were fed in the morning, with concurrent removal of faeces and urine. Their ration was a finishing diet with maize silage and concentrate (50:50 on DM). During the trial we recorded other parameters such as feed intake, milk production, milk composition, body weight, etc. Statistical analysis was performed with SAS.

Results Although the measurement of the other gases are necessary for durable long-term strategies only the methane emissions of the control measurements are presented here (Table 1). There was no significant difference between the dairy cow rations (P = 0.58) . The correlation (P < 0.01) between the mean methane emission and mean DMI was r = 0.75. No significant correlation was found between methane emission and milk production or body weight. The methane emission between animals with the same ration and DMI was significantly different (P < 0.05), with coefficients of variations of more than 10%. Variations between three consecutive days for the same animal ranged from 1 to 20%. Because of these large animal variations, a sufficient number of animals per treatment should be used. The within-day variation indicated that most methane was emitted in the first two hours after feeding, while minimal concentrations were measured before feeding. The correlation between the mean methane emission and mean DMI for beef heifers was r = 0.71. The variation of methane emissions between the heifers with the same ration and DMI ranged from 12 to 16%. Variations between three consecutive days in the same animal ranged from 1 to 23%.

Table 1 Cow performance and methane emissions (mean, standard deviation, minimum, maximum and coefficient of variance)
Dairy cows (n = 16) Beef heifers (n = 18)
Mean s.d. min. max. c.v. Mean s.d. min. max. c.v.
Dry matter intake (kg/d) 17.4 2.5 14.8 21.4 13.5 8.5 0.6 6.8 9.3 6.7
Body weight (kg) 620 65 508 700 8.9 538 86 412 696 16.0
Milk production (kg/d) 24.0 4.3 19.1 34.9 18.0 - - - - -
Methane emission (g/d) 301 43 206 382 14.2 144 26 95 188 18.0

Conclusion Our system is adequate to monitor methane emissions. Because methane emissions showed a large between-animal variation, we recommend including sufficient animals per treatment.

Acknowledgements This research was partly funded by the European Commission (FP7-SME-262270)

References
Beauchemin, K. A., Kreuzer, M., O'Mara, F., and. McAllister, T. A. 2008. Australian Journal of Experimental Agriculture 48(1-2), 21-27.
Peiren, N. and De Campeneere, S. 2012. In Technical Manual on Respiration Chamber design (eds C Pinares and G Waghorn), pp. 43–57. Ministry of Agriculture and Forestry, Wellington, NZ.
Yanez-Ruiz, D.R., Losa, R., Nuñez, C., Shearer, A., Tessier, N., Media, B., De Campeneere, S., Morgavi, D., Fievez, V., and Newbold, J., 2013. Advances in Animal Biosciences (GGAA2013)