Estudo de curvas de lactação em bovinos mestiços leiteiros na região do brejo paraibano.

Abstract

Brazil is among the largest milk producers in the world and increasingly invests in technologies to enhance production to attain the growing market demand. About 70% of the national production comes from mixed breeds that use Holstein and Gir breeds, enabling milk production in different latitudes. Animal breeding is one of the technologies that are being used to enhance production, being a tool that aids to identify the best animals and guides which mating schemes will provide better quality progenies. The objective of this work was to analyze the productive parameters of lactating cows that are from Departamento de Zootecnia, Universidade Federal da Paraíba - UFPB, to guide the general management of the herd. A total of 351 records were used, which were collected between 2007 to 2017 from 139 cows of seven genetic (Jersolando 1/2 Jersey – 1/2 Holstein), Jersey puro por cruza, 1/2 Holstein – 1/2 Gir, 1/4 Holstein – 3/4 Gir, 3/4 Holstein – 1/4 Gir, 5/8 Holstein – 3/8 Gir, and 7/8 Holstein – 1/8 Gir). Animals were kept in pasture, with concentrated supplement provided twice a day at the time of milking. The effects of genetic composition and calving season (dry and rainy) were tested and only the effect of genetic composition was significant (p <0.05) for milk production at peak of lactation, days in lactation and days in lactation post-peak. Four mathematical models for lactation curves were tested for all genetic compositions, these being the Ali Schaeffer, Incomplete Gamma, Logarithmic Quadratic and Parabolic Exponential models. The best model, according to decision criteria (Akaike, Bayesian, Coefficient of determination, and significance of curve parameters) was the Logarithmic Quadratic model. It was verified the influence of different genetic compositions concerning the variables of milk production at peak of lactation, days in lactation and days in lactation post-peak. This allows the identification of genetically superior individuals among those belonging to the same genetic compositions and allows the definition of which of these compositions are the most productive. It would also be possible to indicate which genetic compositions are the most suitable for the region under study, as was the case with animals with the greatest contribution of the Holstein breed to the genetic composition. The Quadratic Logarithmic model was the most suitable for adjusting the lactation curves for all evaluated genetic compositions, making it possible to infer, with greater precision, the production/peak of lactation peak and the magnitude of the lactation decline until the drying period.