Between-cow variation in digestion and rumen fermentation variables associated with methane production.

A meta-analysis based on an individual-cow data set was conducted to investigate the effects of between-cow variation and related animal variables on predicted CH emissions from dairy cows. Data were taken from 40 change-over studies consisting of a total of 637 cow/period observations. Animal production and rumen fermentation characteristics were measured for 154 diets in 40 studies; diet digestibility was measured for 135 diets in 34 studies, and ruminal digestion kinetics was measured for 56 diets in 15 studies. The experimental diets were based on grass silage, with cereal grains or by-products as energy supplements, and soybean or canola meal as protein supplements. Average forage:concentrate ratio across all diets on a dry matter basis was 59:41. Methane production was predicted from apparently fermented substrate using stoichiometric principles. Data were analyzed by mixed-model regression using diet and period within experiment as random effects, thereby allowing the effect of experiment, diet, and period to be excluded. Dry matter intake and milk yield were more repeatable experimental measures than rumen fermentation, nutrient outflow, diet digestibility, or estimated CH yield. Between-cow coefficient of variation (CV) was 0.010 for stoichiometric CH per mol of volatile fatty acids and 0.067 for predicted CH yield (CH/dry matter intake). Organic matter digestibility (OMD) also displayed little between-cow variation (CV = 0.013), indicating that between-cow variation in diet digestibility and rumen fermentation pattern do not markedly contribute to between cow-variation in CH yield. Digesta passage rate was much more variable (CV = 0.08) between cows than OMD or rumen fermentation pattern. Increased digesta passage rate is associated with improved energetic efficiency of microbial N synthesis, which partitions fermented substrate from volatile fatty acids and gases to microbial cells that are more reduced than fermented carbohydrates. Positive relationships were observed between CH per mol of volatile fatty acids versus OMD and rumen ammonia N concentration versus OMD; and negative relationships between the efficiency of microbial N synthesis versus OMD and digesta passage rate versus OMD, suggesting that the effects of these variables on CH yield were additive. It can be concluded that variations in OMD and efficiency in microbial N synthesis resulting from variations in digesta passage contribute more to between-animal variation in CH emissions than rumen fermentation pattern.