Parâmetros de qualidade do leite e ocorrência de leite instável não ácido no Norte de Minas Gerais

Abstract

Little is known about the influence of unstable non-acidic milk (LINA) on the quality of milk in tropical regions semiarid regions. Researches in different locations and production systems can contribute to explain this change. The objective of this study was to verify physical-chemical parameters of milk quality and occurrence of LINA in northern Minas Gerais State, from the database of dairy industry. Data were collected in Northern Minas Gerais from January to December 2014, corresponding to 10,654 samples, derived units producing milk, distributed in the region. Were evaluated effects of months, routes and seasons of the year on the samples classified as normal, LINA, alkaline and acidic. Furthermore, it was found the implication of these variables on the occurrence of these variables outside of the standards, in relation to the titratable acidity, density and freezing point. Analyzes of Chi-square test of variance, correspondence and simple regression, clustering, discriminant analysis, logistic regression and correlation were performed to obtain the results discussed. The frequency of samples classified as normal was 79.46%; LINA, 12.93%; milk acid, 3.48% and alkaline, 4.14%. Higher occurrence of LINA was observed in September, October and November, without, however, occur variation between the seasons of the year. As for the distribution of the frequency of classes of milk between the routes, Ubaí showed higher percentage of LINA, 15.94%, followed by Brasília de Minas, 15.17%. In the analysis of titratable acidity between the months and seasons of the year the values ranged from 2.13% in July to 16.01% in October and were higher in the rainy season (10.57%). Percentages of samples outside of the standards were higher in the months of October to 5.38% and march with 4.88% for density, while for the freezing point in January, with 9.64% and April with 8.32%. Pearson's correlation analysis indicated negative association of average magnitude between titratable acidity and freezing point. In the study of matching classes of milk with months, the two dimensions formed explained 91.62% of accumulated inertia, respectively. September showed a greater association with the classes of milk LINA and acid, while April to normal class. As discriminant analysis all variables, classes of milk, titratable acidity, density and freezing point, proved to be important to differentiate between dry and rainy seasons, as well as, for the groups of months formed in cluster analysis. The variables group of month and routes were significant on the probability of occurrence of LINA in accordance with logistic regression analysis. In this study, although they were observed changes in the milk stability, a higher proportion of samples within the normal it was identified. The increase in the frequency of positive samples for alcohol test occurred at the end of the dry season, while LINA was associated with the transition period of the dry season to the rainy season.