Repeatability of enteric methane determinations from cattle using either the SF6 tracer technique or the GreenFeed system

The SF6 tracer technique (SF6) and GreenFeed system (GF) are two methods for measuring enteric methane (CH4) emissions from cattle. Both methods estimate individual daily CH4 emissions from expired gas samples collected either continuously over 24 h in a canister (SF6) or several times a day during short-term periods (3-8 min) when cattle visit an automated head chamber (GF). The objective of this work was to study repeatability (R) of each method according to duration of measurement period as an indicator of their precision. The R of CH4 measurements was evaluated in two different trials using cows. For Experiment 1, the SF6 technique was used for 20 days in six non-lactating dairy cows fed a hay-based diet; for Experiment 2, the GF system was used for 91 days in seven lactating dairy cows fed a maize silage-based diet. The CH4 data were grouped by periods of 1-10 days (SF6) and 1-45 days (GF). The CH4 emissions averaged 23.6 +/- 3.9 g/kg dry matter intake (DMI) for the SF6 and 17.4 +/- 3.3 g/kg DMI for the GF on the measurement period. To achieve an R value of 0.70 for CH4 emissions (g/kg DMI), 3-day periods were necessary for SF6 and 17-day periods for GF. The R did not increase after 4-day periods for SF6 (R = 0.73), but increased for GF until 45-day periods (R = 0.90). In our experimental conditions and R = 0.70, the total number of cows necessary to detect a significant difference in CH4 emissions (g/kg DMI) between two treatments (e.g. diet) was similar for SF6 and GF.