Tropical Animal Health and Production Animal Science and Zoology

Estimation of lactation curve and model fits using different non-linear equations in Anatolian buffaloes

Şahin, Aziz, Aksoy, Yüksel, Ulutaş, Zafer, Yildirim, Arda
Springer Nature · 2026-11-16 · DOI 10.1007/s11250-026-05473-0

Abstract

This research was conducted to estimate lactation curve parameters for Anatolian buffaloes with the use of seven different models (Mixed log (MLM), Grossman model (GRM), Gaines (GM), Rook (RM), Morant (MM), Dhanoa (DM), and Gompertz (GOM)) and to compare the curve models. The research material consisted of 11 658 test day yield records from 1943 Anatolian buffaloes born in Tokat province between 2019 and 2023. Seven different lactation curve models were compared with the use the adjusted determination coefficient (R^2_adj), Bayesian Information Criteria (BIC), mean square error (MSE), root mean square error (RMSE), and Akaike’s Information Criteria (AIC). The Grossman model yielded persistency (S), maximum milk yield (Y_max), and the time at which this milk yield (T_max) was achieved as 2.40, 6.05, and 78.22, respectively. Present findings revealed that the GRM and DM models with the highest R^2_adj and the smallest MSE, RMSE, AIC, and BIC values were the best models for describing the lactation curve of Anatolian buffaloes for all parities. It was concluded based on present findings that using the individual lactation curves and curve parameters as selection criterion in breeding programs to be conducted in herds of which data were used may contribute to improving selection accuracy.

Read at publisher →