Determination of the best lactation curve model and lactation curve parameters using different nonlinear models for anatolian buffaloes

Abstract

The present investigation was conducted to determine the lactation curve parameters and to choose the best model that defines the lactation curve. Data were composed of 4897 test day milk yield (TDMY) records from Anatolian buffaloes calved during 2014-2017 raised under different farm conditions in Amasya Province, Turkey. Five different lactation curve models such as Wood (WOD), Cobby and Le Du (CLD), Exponential (EXP), Parabolic Exponential (PEXP) and Quadratic (QUA) models were used. The adjusted multiple coefficient of determination (R2 adj) and the residual standard deviation (RSD) were evaluated as parameters to detect the best fitted lactation curve model. The results revealed that WOD model described the highest R2 adj (0.98) and with the lowest RSD (0.065), besides the highest R2 adj (0.97) and the lowest RSD (0.087) were also observed in CLD model. To conclude, WOD and CLD models were detected to be the most suitable models defining the lactation curve of Anatolian buffaloes. As a result, using the parameters detected via WOD and CLD models in breeding studies will contribute significantly to the researches in this direction in these herds. © 2020 Zoological Society of Pakistan.