Sustainability & efficiency

Alternative approaches for the incorporation of a methane trait into genetic evaluations

---

Abstract

Inclusion of methane in ruminant genetic evaluations is under investigation internationally. In grass-based systems such as New Zealand's, where land area limits livestock production, methane per unit of feed is relevant for reducing national greenhouse-gas emissions. However, feed intake is not measured, and daily methane production is unfavourably correlated with growth rate thereby penalising larger animals when selecting for low methane. The aim of this study was to generate a simulation modelling the New Zealand sheep industry to evaluate alternative methane-trait definitions to quantify their effectiveness and any unintended consequences. True breeding values (TBVs) for live weight, feed intake, methane yield, and carbon-dioxide yield were simulated as correlated traits. A zero genetic correlation between methane yield and carbon-dioxide yield was assumed. The founder population was mated to produce six generations. Results were calculated as averages of 30 replicates of this simulated scenario. Daily methane and carbon dioxide trait phenotypes were created by multiplying methane yield and carbon dioxide yield with feed intake, respectively. These were traits synonymous with portable accumulation chamber (PAC) data currently recorded. Alternative scenarios where daily methane and daily carbon dioxide data were masked for 50% and 70% of the dataset were also evaluated. Estimated breeding values (EBVs) for live weight, daily methane production, and daily carbon dioxide production were computed for all simulated animals using a pedigree based multi-trait animal model. Unadjusted daily methane EBVs as well as three adjusted methane traits derived from daily methane EBV adjusted for: live weight EBVs only, daily carbon dioxide EBVs only, and both live weight and daily carbon dioxide EBVs. Unadjusted and adjusted methane EBVs were compared to simulated methane yield and live weight TBVs. In non-masked datasets, correlations for methane yield TBVs with unadjusted methane and the three-methane adjusted trait EBVs were 0.439 (0.007), 0.580 (0.004), 0.566 (0.005), and 0.568 (0.005), respectively. When data were masked these correlations weakened with increased masking. Unadjusted methane and methane without adjustment for live weight EBVs had moderately unfavourable correlations with live weight. This was exacerbated in animals without phenotypic PAC records. Conversely, methane adjusted for live weight EBVs reduced the unfavourable correlation in animals both with and without PAC records. Adjusting for daily carbon dioxide alone didn't reduce the unfavourable correlation between daily methane and live weight to the same extent as other adjustments. Results from this study highlight the importance of methane-trait definition when selecting low methane emitters as it influences overall methane emissions but can also inadvertently select on other correlated traits. With simulations run with univariate EBV estimation, we found similar or better overall performance of the various predictions, indicating that univariate prediction might be a good option unless multivariate approaches are required to account for preselection.

Keywords: 2026

How to Cite:

Abeykoon, S., Amer, P., McHugh, N., Murphy, C. & Sleator, R., (2026) “Alternative approaches for the incorporation of a methane trait into genetic evaluations”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2285671. doi: https://doi.org/10.31274/wcgalp.23773

Rights: 1