Comparison of strategies for spot sampling enteric methane and hydrogen emissions to evaluate methane-reducing feed additives in dairy and beef cattle

Abstract

Short-term or “spot” gas emissions measurement systems, such as the GreenFeed system, offer a less intensive alternative for measuring methane (CH4) and hydrogen (H2) emissions from cattle. However, the optimal sampling interval for obtaining representative spot measurements is unclear. Furthermore, it is uncertain whether data derived exclusively from voluntary use of the GreenFeed system by cows accurately represent daily emissions. This study compared spot sampling strategies in 2 different experiments. In the beef experiment, 2 temporal spot sampling schemes were evaluated during a study on a bromoform-based antimethanogenic feed additive (AMFA) in finishing steers. The Q8 scheme involved 8 spot samples from each steer over a 3-d period, covering 24 h. The Q2 scheme sampled steers every 2 h from immediately before morning feeding until evening feeding, over 1 or 2 d. Gas emissions and feed intake data were analyzed using linear mixed-effects models. Methane production and yield did not differ significantly between schemes. Hydrogen production, however, was significantly greater by up to 1.3 g/d under the Q8 scheme, likely reflecting the effect of more frequent spot sampling on feeding behavior. In the dairy experiment, 2 sampling strategies were compared during a study evaluating a linseed-based feed additive in mid-lactation dairy cows. Once in each experimental period, cows were sampled intensively using the Q8 scheme. During the remaining period, cows voluntarily used the GreenFeed system. Data from days with varying minimum voluntary visits (3–8 visits/d, designated as V3–V8) were analyzed. Methane and H2 estimates in the control group did not differ from those under intensive sampling when cows had more than 7 and 5 voluntary samples per day, respectively. The greatest differences in CH4 and H2 production between any voluntary sampling strategy and intensive sampling for the control group were 34.4 g/d and 1.03 g/d, respectively. These findings suggest that differences in CH4 estimates due to sampling strategies are unlikely to affect AMFA efficacy assessments. However, differences in H2 estimates may have implications for energy flow accounting under CH4-inhibition.