Individual lactation curves in Siboney dairy cows of Cuba

The objective was to study and to model the individual lactation curves in Siboney cows comparing four mathematical models. In total, 31,631 test day milk production records (TD) of 3,697 lactations (1 to 5) from 2,632 Siboney cows of Cuba (5/8 Holstein, 3/8 Cebu Cuban) monthly recorded between 1994 and 2003 were adjusted using the functions of Wood, Wilmink, Ali-Schaeffer and Legendre polynomials. The parameters were estimated using nonlinear regressions and the goodness of fit was measured by the adjusted coefficient of determination (R2A). Values of R2A> 0.75 in 23, 24, 28, and 36 % of the lactations for the models of Wood, Wilmink, Ali-Schaeffer and Legendre polynomials, respectively, were obtained. The models of Wood and Wilmink described four types of curves and the models of Ali-Schaeffer and the Legendre polynomials 17 and 20 for 32 possible theoretical groups. The correlations among the parameters of Ali-Schaeffer function were higher than that estimated for the Legendre polynomials. The proposed functions represented the different shapes among lactation curves and in particular the five parameters models detected higher diversity related to the rest of the function. This pointed that even if additional detected shapes can be considered as derivations of the two classical groups of typical or atypical curves, this practice could compromise the variability among lactation curves inside a herd, therefore further studies are needed.

Guardado en:
Detalles Bibliográficos
Autores principales: Palacios Espinosa, Alejandro, González-Peña Fundora, Dianelys, Guerra Iglesias, Danilo, Espinoza Villavicencio, José Luis, Ortega Pérez, Ricardo, Guillén Trujillo, Ariel, Ávila Serrano, Narciso
Formato: Digital revista
Idioma:spa
Publicado: Instituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias 2016
Acceso en línea:https://cienciaspecuarias.inifap.gob.mx/index.php/Pecuarias/article/view/4147
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!