Variabilidade espacial da composição e qualidade do leite cru refrigerado no estado de Alagoas e no agreste pernambucano

Abstract

The aim of this study was to analyze the spatial variability of the composition and quality of refrigerated raw milk in the state of Alagoas and in the middle region of Pernambuco Agreste, in the years 2014 and 2015. Thus, were obtained from the three companies in the dairy sector submitted to the federal inspection service (SIF) that collect milk in the state of Alagoas and in the middle region of Pernambuco Agreste, 3863 monthly official reports on somatic cell count (SCC) and total bacterial (CBT), fat, protein, lactose, total solids and nonfat dry extract of 432 different tanks direct expansion milk in the years 2014 and 2015. the data were analyzed for spatial dependence and geographically weighted regression, using the ArcGIS 10.3 software. It was observed predominantly spatial distribution of CBT log 5.07 to 5.19 and diffuse areas with values from 5.20 to 5.54. It was observed predominant value of somatic cell score (SCS) from 2.81 to 2.99; sites with from 1.87 to 2.80; places with 3.00 to 3.16, and isolated points with peaks from 3.17 to 3.39. It was identified low influence of altitude, rainfall and altitude interaction x rainfall on the ECS and the CBT log. The analysis of spatial dependence was observed that there is high spatial dependence between variables fat, lactose and total solids. Still, protein, ECS and CBT log have moderate spatial dependence and the dry extract low spatial dependence. The altitude, rainfall and altitude interaction x rainfall have low influence on the fat, lactose, protein, total solids, nonfat dry extract, ECS and CBT log. Finally, geostatistics is an important tool for the analysis of milk quality, helping the classical statistics for better understanding of the phenomenon when the variable is regionalized.