Evaluation of existing equations to rank methane yields of sheep fed a range of fresh forages

The objective of the current study was to evaluate the performance of published equations to predict CH4 yield (g CH4/kg dry matter intake; DMI) in NZ sheep fed a range of fresh forages. A database with 182 NZ sheep CH4-yield treatment means was used to validate 17 published CH4-yield prediction equations. The CH4-yield predictions were evaluated using the square root of mean-square prediction error (RMSPE) and Lin’s concordance correlation coefficient (CCC). The three prediction equations with the smallest error (<22% of RMSPE) and the only ones with moderate CCC with observed CH4 yield were all based on data from animals fed forage only. These three models predicted CH4 yield based on DMI plus dietary neutral detergent fibre, and animal body weight, feed crude protein, or feed organic matter digestibility; and they performed better than the DMI only equations that are commonly used in national greenhouse gas inventory models and on-farm accounting models.