Milk fatty acids can assist in the prediction of methane from dairy cows fed a range of diets

Abstract

It has been proposed that methane emissions from dairy cows could be predicted from the concentrations of
specific milk fatty acids (Chilliard et al. 2009). The proposed relationships were developed from a corn silage and concentrate diet with an addition of one of whole linseed, extruded linseed or linseed oil. The relationships have been tested by others but only for cows fed total mixed rations incorporating a limited range of feed additives (Mohammed et al., 2011; Dijkstra et al., 2011). The aim of this work was to determine if fatty acid concentrations in milk could assist in the prediction of methane emission from dairy cows fed a range of diets.

Full text

Williams, S.R.O., Moate, P.J., Hannah, M.C., Deighton, M.H., Auldist, M.J. 2013. Milk
fatty acids can assist in the prediction of methane from dairy cows fed a range of diets.
Advances in Animal Biosciences 4, 405.

405
Milk fatty acids can assist in the prediction of methane from dairy cows fed a range of diets
S R O Williams, P J Moate, M C Hannah, M H Deighton, M J Auldist
Department of Environment and Primary Industries - Victoria, Ellinbank, Victoria, Australia
Email:richard.williams@dpi.vic.gov.au
Introduction It has been proposed that methane emissions from dairy cows could be predicted from the concentrations of
specific milk fatty acids (Chilliard et al. 2009). The proposed relationships were developed from a corn silage and
concentrate diet with an addition of one of whole linseed, extruded linseed or linseed oil. The relationships have been tested
by others but only for cows fed total mixed rations incorporating a limited range of feed additives (Mohammed et al., 2011;
Dijkstra et al., 2011). The aim of this work was to determine if fatty acid concentrations in milk could assist in the
prediction of methane emission from dairy cows fed a range of diets.
Material and methods Data (151 individual cow records) were collated from five experiments conducted at Department of
Primary Industries, Ellinbank, Victoria, Australia (38°14'S, 145°56'E). Diets included alfalfa forage supplemented with
docosahexaenoic acid, cottonseed oil, grape marc, corn grain and wheat grain, and perennial ryegrass pasture supplemented
with tannin. Methane emissions (g/day), yield (g/kg DMI) and intensity (g/kg milk) were modelled using a range of
variables. For the dependent variable, a search of all possible regression models was conducted in the independent variables
DMI, milk yield (kg/d), milk fat (kg/d), days in milk, DeNovo fatty acids (total C4 to C15, proportion of total fatty acids in
milk), C16 fatty acid (proportion of total fatty acids in milk), C17 fatty acid (proportion of total fatty acids in milk), pre-
formed fatty acids (total C18 +, proportion of total fatty acids in milk), and odd chain fatty acids (C13, C15, C17,
proportion of total fatty acids in milk), using the RSEARCH procedure in GenStat 15 (VSN International, Hemel
Hempsted, United Kingdom). The model with smallest Mallows Cp statistic (Thompson 1978) was identified and re-fitted,
including random effects for experiment and cow, using ReML in GenStat 15.
Results
Table 1 Best-fit models for predicting methane emission, yield and intensity (coefficient ±standard error)
Predicted Constant DMI
(kg/d)
Milk
(kg/d)
DIM DeNovo
(% total
FA)
C16
(% total
FA)
C17
(% total
FA)
Preform
(% total
FA)
adjusted
R2Residual
standard
deviation
CH4(g/d) 441
±12.0
12.8
±1.98 n.s. n.s. 4.0
±1.43
4.1
±0.94
-137
±13.0 n.s. 78.0 54.0
CH4(g/kg DM) 21.0
±0.55
-0.39
±0.097 n.s. n.s. 0.19
±0.071
0.21
±0.047
-6.4
±0.65 n.s. 68.2 2.64
CH4(g/kg milk) 19.4
±0.24 n.s. -0.50
±0.051
0.03887
±0.0059
1.6
±0.39
1.7
±0.35
-4.8
±0.85
1.44
±0.355 89.9 2.75
DMI = dry matter intake, DIM = days in milk, DeNovo = total C4 to C15 fatty acids in milk, C16 and C17 = fatty acid in
milk, Preform = total C18+ fatty acids in milk, n.s. = not significant in best-fit model P ≤ 0.005.
Conclusions Milk fatty acids, when combined with other production parameters, can assist in the prediction of methane
intensity (g/kg milk) and methane yield (g/kg DMI) when cows are fed a wide range of diets. However, there is still a large
error in the prediction of methane emission (g/d).
Acknowledgements This research was funded by the Department of Primary Industries - Victoria, Dairy Australia, the
Gardiner Foundation, Meat and Livestock Australia, and the Australian Government Department of Agriculture, Fisheries
and Forestry under its Australia’s Farming Future Climate Change Research Program.
References
Chilliard, Y., Martin, C., Rouel, J., and Doreau. M, 2009. Journal of Dairy Science. 92, 5199-5211.
Dijkstra, J., van Zijderveld, S.M., Apajalahti, J.A., Bannink, A., Gerrits, W.J.J., Newbold, J.R., Perdok, H.B., and Berends,
H. 2011. Animal Feed Science and Technology. 166-167, 590-595.
Mohammed, R., McGinn, S.M., Beauchemin, K.A. 2011. Journal of Dairy Science. 94, 6057-6068.
Thompson, M.L. 1978. International Statistical Review, 46, 1-19.