The relationship between DM intake (DMI) and enteric methane emission is well established in ruminant animals but may depend on measurement technique (e.g. spot v. continuous gas sampling) and rumen environment (e.g. use of fermentation modifiers). A previous meta-analysis has shown a poor overall (i.e. 24 h) relationship of DMI with enteric methane emission in lactating dairy cows when measured using the GreenFeed system (GF; Symposium review: uncertainties in enteric methane inventories, measurement techniques, and prediction models. Journal of Dairy Science 101, 6655 to 6674). Therefore, we examined this relationship in a 15-week experiment with lactating dairy cows receiving a control diet or a diet containing the investigational product 3-nitrooxypropanol (3-NOP), an enteric methane inhibitor, applied at 60 mg/kg feed DM. Daily methane emission, measured using GF, and DMI were clustered into 12 feed-intake timeslots of 2 h each. Methane emission and DMI were the lowest 2 h before feeding and the highest within 6 h after feed provision. The overall (24 h) relationship between methane emission and DMI was poor (R2 = 0.01). The relationship for the control (but not 3-NOP) cows was improved (R2 = 0.31; P < 0.001) when DMI was allocated to timeslots and was strongest (R2 = 0.51; P < 0.001) 8 to 10 h after feed provision. Analysis of the 3-NOP emission data showed marked differences in the mitigation effect over time. There was a lack of effect in the 2-h timeslot before feeding, the mitigation effect was highest (45%) immediately after feed provision, persisted at around 32% to 39% within 10 h after feed provision, and decreased to 13%, 4 h before feeding. These trends were clearly related to DMI (i.e. 3-NOP intake) by the cows. The current analysis showed that the relationship of enteric methane emission, as measured using GF, and DMI in dairy cows depends on the time of measurement relative to time of feeding. The implication of this finding is that a sufficient number of observations, covering the entire 24-h feeding cycle, have to be collected to have representative emission estimates using the GF system. This analysis also revealed that the methane mitigation effect of 3-NOP is highest immediately after feed provision and lowest before feeding.

Hydrogen is an important intermediate that is produced during carbohydrate fermentation to volatile fatty acid and utilized by methanogens to produce methane in the rumen. Ruminal volatile fatty acid and dissolved methane concentrations are more than 500 times greater than dissolved hydrogen concentration. Therefore, we hypothesized that dissolved hydrogen might have a higher sensitivity in response to dietary changes compared with volatile fatty acid and dissolved methane. Using goats, we investigated the effects of increasing dietary starch content (maize replaced with wheat bran) and supplementing with rhubarb rhizomes and roots on the relationships among dissolved hydrogen, dissolved methane and other fermentation end products. The study was conducted in a replicated 4×4 Latin square with a 2×2 factorial arrangement of four treatments: two starch levels (220 v. 320 g/kg dry matter (DM)), without and with rhubarb supplement (0% v. 2.8% of total mixed ration). Increased dietary starch and rhubarb supplementation did not alter volatile fatty acid concentrations or methane emissions in terms of g/day, g/g DM intake and g/g organic matter digested. However, goats fed the high-starch diet had greater dissolved hydrogen (P=0.005) and relative abundance of Selenomonas ruminantium (P<0.01), and lower (P=0.02) copy number of protozoa than those fed the low-starch diet. Rhubarb increased ruminal dissolved H2 (P=0.03) and total volatile fatty acid concentration (P<0.001), but decreased copies of bacteria (P=0.002). In conclusion, dissolved hydrogen appears to be more sensitive to dietary changes with starch content and rhubarb supplementation, when compared with volatile fatty acid concentrations and methane production.