Background and Objective: There is almost no study in the literature about the Dhofari cow breed lactation curve, as it is the main source of dairy products for large population in Sultanate of Oman and the only one of two native cattle breed in Arabia. The objective of this study was to use different mathematical nonlinear model functions to describe and predict the Dhofari cow breed. Materials and Methods: A total data of 1640 test day records of monthly milk production between years 2010 and 2014, from 173 dairy herds located at Salalah Livestock Research Station was analyzed using GLM method (SPSS 2010). Models (Gompertz, Von Bertalanffy, Logistic, Brody, Wood, Wilmink, Cobby and LeDu, Cappio Borlino and Dijkstra) for goodness of fit using adjusted coefficient of determination (R2), root mean square error (RMSE), akaikes information criterion (AIC) and bayesian information criterion (BIC). Results: Gompertz model provided the best lactation curve than the other models due to the lower values of RMSE (4.05), AIC (38.50) and BIC (6.83) but the Dijkstra model did not fit the observed data as well as the rest. The peak yield (9.72 kg), peak time (60 days), 305 days milk (1580 kg) and persistency (90.65%) were best estimated by the Gompertz model. Conclusion: Gompertz model provided the best estimation for the lactation curve for the Dhofari cows.
展开▼
