Temperature, but not submersion or orientation, influences the rate of sulphur hexafluoride release from permeation tubes used for estimation of ruminant methane emissions

• SF6 release from permeation tubes is unaffected by tube submersion or orientation.
• Rate of SF6 release from permeation tubes changes by 2.5% per °C between 37 and 41 °C.
• SF6 release rate in vivo changes due to fluctuation of reticulo-rumen temperature.
• Species-specific reticulo-rumen temperature effects in vivo rate of SF6 release.

Predictable release of sulphur hexafluoride (SF6) tracer gas from permeation tubes into the reticulo-rumen is necessary to estimate methane emissions from ruminants using the SF6 tracer technique. Any discrepancy between the laboratory determined rate of SF6 release from permeation tubes and the actual rate of release in the reticulo-rumen would bias calculated methane emissions. The purpose of this investigation was to determine the effect of temperature, submersion and orientation on the rate of SF6 release from permeation tubes. Four experiments were undertaken. Experiment 1 determined that release of SF6 increased by 2.5 ± 0.14% per degree Celsius increase in temperature between 37 and 41 °C (P < 0.001). Experiment 2 determined that the Arrhenius equation can be used to describe the temperature dependence of SF6 release rate from permeation tubes between 0 and 70 °C, consistent with a change in release rate of 2.3 ± 0.08% per degree Celsius change in temperature. Experiment 3 determined that submersion of permeation tubes in water did not affect the rate of SF6 release (P = 0.13). Experiment 4 determined that SF6 release rate was not influenced by permeation tube orientation (P = 0.42). In addition we determined the activation energy of permeation, Ep, describing the overall temperature dependence of SF6 permeation flux from permeation tubes, to be 18,424 ± 680 J/mol. This research implies that the short-term release rate of SF6 from permeation tubes within the reticulo-rumen will vary in response to temperature change due to animal, diet and/or environmental factors. A short term decrease in temperature of reticulo-rumen contents, induced by drinking cold water, is expected to have a larger influence on the accuracy of estimated methane emission derived from time-averaged sampling periods less than 24 h. Use of the SF6 technique to detect differences in enteric methane emissions due to diet or between animal species may be confounded by diet or genetic effects on body temperature. Unless the effect of temperature is managed through careful implementation of the technique, substantial errors could be caused as illustrated by the following example: a +2 °C error in calibration temperature (41 °C), and a −2 °C discrepancy between the actual (37 °C) and assumed reticulo-rumen temperature (39 °C), could bias estimated methane emissions by approximately +10%.