Abstract

The autumn-calving herd at the Dexcel Automatic milking farm was used to test a system of predicting next oestrous date for individual cows and automatically drafting them for focused observation of behavioural oestrus. During the first 21 days of artificial inseminations (AI), the automatic milking system was programmed to draft cows into a holding paddock on each of three days either side of the due date for next oestrus, which was predicted from oestrous dates and milk progesterone concentrations collected prior to the planned start of mating. Cows drafted into the holding paddock were visually checked in the morning, released if not in oestrus or retained if in oestrus for AI later in the morning. The remainder of the herd was also visually checked for oestrus in the morning and afternoon. During the first 21 days of mating, 29 of the 37 cows (78%) were inseminated, and 22 (76%) of these were inseminated within the predicted ‘six-day separation window’. Of these 22 cows, 15 (68%) were detected as being in oestrus after being drafted into the holding paddock. A system for drafting cows due to be in oestrus for focused observation is achievable in practice, although improved performance will be necessary to reach the desired level of reliability. KEYWORDS: oestrous detection; automation; dairy cattle.

CR, Burke, KL Davis, and JG Jago

Proceedings of the New Zealand Society of Animal Production, Volume 65, Christchurch, 231-235, 2005