Abstract #M207

# M207
Comparison of Holstein and Jersey milk production with a new stochastic animal model.
M. Li*1, V. E. Cabrera1, K. F. Reed2, 1Department of Dairy Sciences, University of Wisconsin, Madison, WI, 2USDA-ARS Dairy Forage Research Center, Madison, WI.

Holsteins and Jerseys are the most popular breeds in the US dairy industry. We built a stochastic, Monte Carlo life events simulation model in Python to test if Jersey cattle’s higher conception rate offsets their lower milk production. The model simulates individual cows and their life events such as the onset of estrus, heat detection, artificial insemination, calving, and culling. Each life event occurs at random according to designated probability distributions. Aggregated data from individual cows represent a herd, which gives the model the ability to simulate management decisions at the farm level. For this study, we assigned different stochastic first service conception rates (CR) for Jerseys and Holsteins based on the literature (McAllister, 2010): 34%, 39%, and 44% for Jerseys and 27%, 32%, and 37% for Holsteins. Subsequent services decreased by 2.6%. (Galvão et al., 2013). We simulated cows in both breeds with voluntary waiting periods (VWP) of 50, 60, and 70 d. All other reproduction-related variables, except CR, VWP were constant in each simulation. We used MilkBot’s lactation curve functions for milk production, which had different parameters for Jerseys and Holsteins and distinguished first, second, and later lactations parameters for each breed (Ehrlich, 2011). These lactation curves estimate Jersey milk production to be 67.6% of Holstein production. Simulations had 1000 cows and ran for 3000 d. From the herd, cumulative milk production from the first day of milking until dry-off or culling were sampled from 50 cows for each parity and each breed. Production samples were then divided by the number of days between calvings and multiplied by 365 to get an average milk production/cow/year. Final results represent an average of 5 simulations and indicate that Jerseys still produce less milk than Holsteins but accounting for conception rate differences decreases the gap in milk production slightly. For example, for cows in their second parity with a 50 d VWP, Holsteins produced on average 11,486.8 kg/cow/year, while Jerseys produced 7,869.3 kg/cow/year: 67.9% of Holsteins’.

Key Words: simulation, Jersey, production