M.M. Schutz*1, J.M. Bewley1, and H.D Norman2
1Department of Animal Sciences, Purdue University, West Lafayette, IN, USA; 2Animal Improvement Programs Laboratory, Agricultural Research Service, USDA, Beltsville, MD, USA
2008 J. Dairy Sci. (?)
© American Dairy Science Association,
2008. All rights reserved.
Individuals may download, store, or print single
copies solely for personal use.
Do not share personal accounts or passwords
for the purposes of disseminating this article.
The objective was to derive factors to predict daily yield when milk weights are recorded once when cows are milked twice (2x) or once or twice when cows are milked thrice (3x) per d. Data were from herds enrolled in Dairy Herd Improvement and automatically recording milking weights and times. Following edits, 83,690 daily records of 1842 first lactation cows and 143,670 records of 2957 later lactation cows in 5 2x herds and 41,657 records of 779 first lactation cows and 84,131 records of 1999 later lactation cows in 4 3x herds remained. Factors currently in use to adjust single milking yields for milking interval (MINT) were applied. Also, 3 methods were compared to estimate factors or equations to predict daily yield. Factors were estimated as the ratio of the sum of daily milk to the sum of partial milk within a parity-MINT class (24 intervals in 2 parities) [Method 1] or as the sum of the ratios of daily milk to partial daily yield for each cow-day divided by the number of cow-days within parity-MINT class [Method 2]. Resulting factors from both methods were smoothed, applied to data, and residuals were regressed on days in milk (DIM). Regression equations (n=168) were also developed within parity-MINT-DIM classes (2x7x12) [Method 3] to simultaneously account for MINT and DIM. Separate factors and equations were derived for each milking in 2x or 3x herds. Method 3 resulted in consistently stronger correlations between estimated and actual yields and smallest root mean squared error (rMSE) for all milkings and parities. For 2x herds, Method 3 resulted in rMSE of 2.3, 3.0, 2.6, and 3.4 kg for AM milkings for Parity 1 and 2 and PM milkings for parity 1 and 2, respectively; compared to rMSE of 2.6, 3.2, 4.1, and 4.8 kg from current factors for the same milkings and parities. Likewise for 3x herds, Method 3 had smallest rMSE, while Method 1 had rMSE similar to those from current factors. Work is ongoing to determine whether equations from Method 3 will allow accurate estimation of daily yield when applied to other herds and traits fat, protein, and somatic cell score.
(Key words: milking interval, adjustment factor, milking frequency)