|
MEKARN Workshop 2009: Livestock, Climate Change and the Environment |
The objective of the study was to evaluate environmental efficiency of dairy diet and to predict methane emission in dairy cows based on fecal near infrared reflectance spectroscopy (faecal NIRS). A three-year on-farm survey (2005 - 2008) was carried out in Vietnam and La Réunion Island (France) to collect to animal, feed intake and fecal excretion data. The composition of diet intake was daily calculated for individual cow. Then, methane production was individually estimated according to a previous equation based on diet composition (Moe et Tyrell, 1980). A data for 1322 individual cows was built and four types of regimes were classified. Results showed that total methane emission and methane efficiency varied robustly (439-552 litres CH4/cow/d, 27-61 litres of CH4/litres of fat corrected milk) according to the different types of regimes. When predicting methane production, fecal NIRS equation obtained, at individual level, low RSEC (5% and 7%) and acceptable R² (0.81 and 0.85) for total methane production (litres/day) and methane production per kg DMI. These equations were also investigated for the average data of animals that had the same performance level. In this step, these equation gave excellent accuracy with low RSEP (3% and 4%) and high R² (0.94 and 0.95), respectively. In addition, the high RPD (3.88 to 4.36) illustrated the robustness of the LOCAL technique. Fecal NIRS is thus a rapid and easy support tool for estimating methane production in dairy cows.