Authors: ATTİLA KAYA, CAN UZMAY, YAVUZ AKBAŞ, İBRAHİM KAYA, SENCER TÜMER
Abstract: This research was carried out to determine the effect of various milk sampling intervals and different methods on the accuracy of estimation of lactation milk yield in dairy cattle. Individual daily milk weights of 599 Holstein, Simmental and Brown Swiss lactations from the Aegean Agricultural Research Institute were used. It was assumed that test-day milk yields were obtained by regular and irregular intervals of 28 d and 56 d, with the irregular intervals having a standard deviation of 5 d. From test-day milk yields, 305 d milk yields were estimated by the Holland (Lactation yield = mean of test day yields x lactation duration (days)) and Test Interval methods. It was found that the bias of estimation increased significantly (P<0.01) as the length of testing intervals increased. The effect of the calculation method on the bias of estimation was also significant (P<0.01). The Test Interval method estimated actual yield more accurately than the Holland method. The difference between the regular and irregular testing intervals with respect to bias of estimation was insignificant. Rank correlations between actual yields and estimated yields were significant (P<0.01) and very high. Therefore, any of the combinations of calculating milk yield in the study can be used to estimate milk yield for the purpose of ranking cows. However, if the aim is to estimate actual yield most accurately, the testing interval should be 28 d. In this case, the Holland or the Test Interval method can be used. If the testing interval is 56 d, the Test Interval method must be used to estimate actual yield.
Keywords: Testing intervals, estimation methods, 305 d milk yield, dairy cattle
Full Text: PDF