Restricting feed intake of lactating dairy cows: effect on feed efficiency and methane emissions
Résumé
An emerging hypothesis claims that feed inefficiency can be due to overconsumption. The objective was therefore to test feed restriction as a lever to improve feed efficiency. Feed efficiency was estimated with the residual feed intake (RFI), as the difference between observed and expected dry matter intake (DMI). A cohort of 32 lactating Holstein cows was identified among 70 lactating cows as the 25% least efficient cows and the 25% most efficient cows. This identification was done using the last 2.5 months of ad libitum feeding, before feed restriction started. For a given expected DMI during ad libitum period, the offered DMI during restriction was set to the observed DMI of the 10% most efficient cows during the last 2 months of ad libitum feeding. Feed restriction was applied during 2.5 months, starting on average at 194 +/- 16 days in lactation. A single diet was fed during the whole experimentation based on 65.8% of corn silage and 34.2% of concentrates on a DM basis. Individual daily DMI, morning BW, daily milk yield, weekly milk composition, monthly BCS were recorded individually, as well as methane (CH4) emissions with two Greenfeed units during the whole study. RFI was the residual of the linear regression predicting each period’s average DMI with period’s averages of net energy in milk, metabolic BW, BCS, BCS gain and BW loss, the fixed effects of parity and period. RFI was averaged per period. The least efficient group ate 2.7 kg DM/d (p<0.01) and emitted 23 g CH4/d (p=0.23) more than the most efficient group during ad libitum feeding. Feed restriction decreased DMI (p<0.01) and daily CH4 emissions (p<0.01), with a 1.8 kg DMI/d and a 20g CH4/d higher decrease for the least efficient group, but had no effect on CH4 yield per DMI (p=0.81). Feed restriction reduced RFI SD from 1.14 kg DM/d to 0.72 kg DM/d. The 25% least efficient cows remained the least efficient cows in both periods (RFI ad libitum vs RFI restriction, r = 0.59). Feed efficiency was not correlated to CH4 emissions, but was negatively correlated to CH4 yield per DMI (ad libitum r = -0.43, restriction r = -0.25). Overall, feed restriction applied to inefficient lactating dairy cows seems to improve their feed efficiency and to reduce their CH4 emissions.