EVALUATION AND COMPARISON OF TWO TECHNIQUES FOR ESTIMATING ENTERIC METHANE EMISSION IN YOUNG BULLS

DOREAU, M., ARBRE, M., ROCHETTE, Y., LASCOUX, C., MARTIN, C.

INRA, UMR 1213 Herbivores, 63122 Saint-Genès Champanelle, France

ABSTRACT Enteric methane emission by ruminants is mainly measured using the SF6 or the Greenfeed technique when they are not tied. The objective of this trial is to compare these two methods in fattening bulls, and to provide additional information on the reliability of the Greenfeed technique. Sixteen young bulls weighing 348 kg at the beginning of the experiment were used for 3 months. Reliability of the Greenfeed technique was determined from continuous measurement for 90 days; comparison between methods was made from measurements during the last 15 days and the last 4 days of the experiment for Greenfeed and SF6, respectively. A repeatability higher than 0.70 was achieved for methane emission and yield with a 15-day period of measurement; a total of 16 animals is necessary to test the mitigating effect of a treatment when two groups of animals are compared. Greenfeed resulted in higher methane emission than SF6 (192 vs 170 g/day, respectively) but inter-animal correlation between methods is low. These results show that Greenfeed method can be used for comparing the effect of feeding treatments on methane emission, but additional work is necessary to better understand differences in emission and yield between Greenfeed and SF6.

Keywords: enteric methane, ruminants, measuring method, Greenfeed, repeatability

INTRODUCTION Enteric methane (CH4) represents about 37% of greenhouse gases emission due to livestock activities, when expressed as carbon dioxide (CO2)-equivalents (Gerber et al., 2013).

Among the methods available to quantify the emission of enteric CH4 emission in ruminants housed in free-stalls or on pasture, the sulphur hexafluoride (SF6) tracer method, used since two decades, and the Greenfeed technique (GF) (C-lock, Rapid City, SD, USA) which appeared recently are the most popular. The SF6 technique is based on the release of a tracer gas (SF6) from a bolus placed in the rumen and continuous sampling of gases produced by the animal, by eructation and exhalation; CH4 and CO2 emission are determined from SF6 release rate and from SF6, CH4 and CO2 concentration in air samples (Johnson et al., 2007). The GF technique is based on the spot sampling of gases produced by the animal when eating concentrate at an automatic

Emission of Gas and Dust from Livestock – Saint-Malo, France – May 21-24, 2017 174

composition offered ad libitum made of 67% baled haylage of permanent grassland and 33%

concentrate. Haylage was offered at 0900 h. A fraction of concentrate (ca. 2.1 kg) was introduced in the GF system and the rest was top-dressed on haylage at trough in equal amounts at 0900 and 1600 h. Methane estimation using the SF6 technique was performed at the end of the experiment on 4 successive days according to Martin et al. (2016). Methane estimation using the GF system (C-lock, Rapid City, SD, USA) was made according to Arbre et al. (2016a). The GF was calibrated for a maximum of 6 daily visits per animal. Methane estimation for the analysis of repeatability of GF technique was achieved for 90 successive days; for the comparison with SF6

technique, only data of the last 15 days were taken. Animals were accustomed to the diet and to the GF system for 1 month before the beginning of the experiment. Outliers were eliminated before statistical analysis by calculating the distance of each daily data to axis from the joint distribution of CO₂ vs CH₄ emission according to Arbre et al. (2016a). In order to evaluate repeatability, average data per cow were calculated for periods defined as sequences of consecutive days. From means per cow and per period, variances of animal (varA), of period, and residual error (varR) were calculated, and repeatability (R) was defined as R = varA/(varA+varR). The number of animals to be used for comparing two treatments was calculated as specified in Arbre et al. (2016a) with a level of significance of difference taken as 5% and a statistical power taken as 80%. Comparison between SF6 and GF techniques was made from average data for each animal (mean of 15 days for GF and of 4 days for SF₆) by analysis of variance using a mixed model (SAS 9.1 release, SAS Inst. Inc., Cary, NC, USA) with method as fixed effect and animal as random effect.

RESULTS AND DISCUSSION Dry matter intake regularly increased from 7.48 ± 0.93 to 9.81 ± 0.69 kg/day between the beginning and the end of the experiment. Liveweight gain during the experiment was 1.67 ± 0.17 kg/day. For GF individual daily data, 11.0% were missing due to not enough daily visits of animals to the system and 3.9% were outliers. Due to increase in dry matter intake (DMI) throughout the experiment, gas emission (g/day) increased with time, but gas yield (g/kg DMI) did not. Repeatability (Table 1) was higher than 0.70 from 15-day period for CH4 emission, from 5-day period for CH4 yield, and from 10-day period for CO2 emission and yield and for CO2/CH4 ratio. Repeatability was close or higher than 80% for any variable from a 30-day period.

Table 1. Repeatability of GF technique according to duration of measurement period in bulls.

Duration of the measurement period

Posters

The number of animals to be used in experiments is specified in Table 2. When the objective is to detect a 10%-difference between experimental treatments, the required number of animals is very high, but for a 20%-difference the total number of animals is equal to 16 or 20, i.e. 8 or 10 per group, according to the objective of the study. Comparison of techniques is presented in Table 3. Methane emission and yield was significantly lower for SF6 than for GF (P=0.01 and 0.04, respectively) although numerical differences were moderate (-11 and -8% for SF6

compared to GF). Neither CO2 emission and yield nor the CO2/CH4 ratio varied between methods (-7, -2 and +5%, respectively, for SF6 compared to GF).

Table 2. Total number of animals required for determining CH4 emission with GF according to the significant difference (∆,%) to be detected between treatments and duration of measurement.

Days of measurement

Mean ± standard deviation

∆ = 10% ∆ = 20%

1-sided test 2-sided test 1-sided test 2-sided test

15 263 ± 40 60 76 16 20

30 262 ± 40 58 74 16 20

45 253 ± 39 58 74 16 20

90 235 ± 39 70 88 18 22

1-sided test is used when the mitigating effect of treatment on CH4 emission is known; 2-sided test is used when the direction of the difference between treatments is not known.

Table 3. Comparison between SF6 and GF techniques in bulls.

Technique

yield. Difference in correlation coefficients for CO2 emission and yield is of low extent.

Emission of Gas and Dust from Livestock – Saint-Malo, France – May 21-24, 2017 176

conclude that the average of 7 to 14 days is enough for repeatable data when a minimum of 20 spot samples is reached. In practical conditions, GF measurements are easy to implement in a 30-day period; a longer period of measurement improves measurement accuracy to a low extent. Experiments with GF require 8 or 10 animals per group according the type of comparison between treatments. This number of animals is compatible with most facilities in experimental stations or in commercial farms when a difference of 20% between treatments is expected, whereas a difference of 10% between treatments can be evidenced with a very large number of animals, as previously found in dairy cows (Arbre et al., 2016a). Increasing the number of days of measurement from 15 to 90 days does not result in a decrease in the number of animals required for a comparison. This allows moving from a study to another when animals are available for several months. The higher mean emission for GF is surprising because both methods measure emissions from mouth and do not account for flatulence. Arbre et al. (2016b) using the same techniques found the similar CH4 emissions for GF and SF6 whereas measurements in open chambers gave higher emissions. Hammond et al. (2016) who summarized the available literature found 4 comparisons between SF6 and GF in dairy cows or heifers, with contrasted results, SF6 being higher, similar or lower than GF according to the study. However in all studies the difference in CH4 emissions between methods is lower than 15%. The low correlations between methods for individual CH4 emissions show that at least one of the two methods may not give an accurate estimate of emissions. In dairy cows Arbre et al.

(2016b) found a good correlation between open chambers and SF6 and a lower correlation between GF and the other two techniques; in heifers Hammond et al. (2015) found a significant correlation between SF6 and GF. In the present experiment, the low correlations may be due to limited differences in CH4 emissions between animals.

3. CONCLUSION: This study confirmed that the GF technique, which appeared a few years ago, is reliable for CH4 estimation of enteric emissions. There may be a small difference in emissions measured by SF6 and GF techniques which remained to be confirmed, but the main issue is the low correlation between techniques for individual emissions; additional progress in the implementation of these methods is necessary..

Acknowledgments This study was granted by a consortium of R&D institutes and of private companies: Adisseo, Agrial, Apis Gene, Deltavit, DSM, Institut de l’Elevage, Lallemand, Moy Park Orléans, Neovia, Techna, Valorex.

REFERENCES

Arbre, A., Rochette, Y., Guyader, J., Lascoux, C., Gomez, L.M., Eugène, M., Morgavi, D.P., Renand, G., Doreau, M., Martin, C., 2016a. Repeatability of enteric methane determinations from cattle using either the SF6 tracer technique or the GreenFeed system. Anim. Prod. Sci. 56, 238-243.

Arbre, A., Martin, C., Rochette, Y., Lascoux, C., Eugène, M., Doreau, M., 2016b. Comparison of three techniques for measuring enteric methane emissions by ruminants. Proc. EAAP 67th Annual Meeting, p. 487.

Gerber, P.J., Steinfeld, H., Henderson, B., Mottet, A., Opio, C., Dijkman, J., Falcucci, A., Tempio, G.,

Posters

2013. Tackling climate change through livestock – A global assessment of emissions and mitigation opportunities. FAO, Rome, Italy.

Hammond, K.J., Humphries, D.J., Crompton, L.A., Green, C., Reynolds, C.K., 2015. Methane emissions from cattle: Estimates from short-term measurements using a GreenFeed system compared with measurements obtained using respiration chambers or sulphur hexafluoride tracer. Anim. Feed Sci. Technol., 203, 41-52.

Hammond, K.J., Crompton, L.A., Bannink, A., Dijkstra, J., Yanez-Ruiz, D.R., O’Kiely, P., Kebreab, E., Eugène M.A., Yu, Z., Shingfield, K.J., Schwarm, A., Hristov, A.N., Reynolds, C.K., 2016. Review of current in vivo measurement techniques for quantifying enteric methane emission from ruminants. Anim. Feed Sci. Technol., 219, 13-30.

Huhtanen, P., Cabezas-Garcia, E.H., Utsumi, S., Zimmerman, S., 2015. Comparison of methods to determine methane emissions from dairy cows in farm conditions. J. Dairy Sci., 98, 1-16.

Johnson, K.A., Westberg, H.H., Michal, J.J., Cossalman, M.W., 2007. In Makkar, H.P.S., Vercoe, P.E.

(eds.), Measuring methane production from ruminants. Springer, Dordrecht, the Netherlands, pp. 33-67.

Manafiazar, G., Zimmerman, S., Basarab, J.A., 2017. Repeatability and variability of short-term spot measurement of methane and carbon dioxide emissions from beef cattle using GreenFeed emissions monitoring system. Can. J. Anim. Sci., 97, 118-126.

Martin, C., Ferlay, A., Mosoni, P., Rochette, Y., Chilliard, Y., Doreau, M., 2016. Increasing linseed supply in dairy cow diets based on hay or corn silage: effect on enteric methane emissions, digestion and rumen microbial fermentation. J. Dairy Sci., 99, 3445-3456.

Renand, G., Maupetit, D., 2016. Assessing individual differences in enteric methane emission among beef heifers using the GreenFeed emission monitoring system : effect of length of testing period on precision. Anim. Prod. Sci, 56, 218-223.

Emissions of Gas and Dust from Livestock – Saint-Malo, France – May 21-24, 2017 178

ENTERIC METHANE EMISSIONS FROM RUMINANTS FED FORAGES: A META-ANALYSIS