Abstract The aims of this study were to evaluate the yield and composition of dual-purpose wheat pasture BRS Tarumã managed with various urea nitrogen (N) doses and validate an exponential model and compare nutrient production costs. The completely randomized design had four replications per treatment (0, 150, 250, 350, and 450 kg N ha-1). For the 350 and 450 kg ha-1 treatments, the cycle was 212 d whereas that of the control was 167 d. The control accumulated 1,771 kg ha-1 dry matter. In contrast, the 450 kg ha-1 treatment accumulated 7,011 kg ha-1 DM. Topdressing nitrogen (150, 250, 350, and 450 kg ha-1) increased the traditional average daily accumulation rate by 586% relative to the control. However, the degree-days method determined a daily accumulation rate 652% higher than the control. The levels of dry matter and other nutrients in BRS Tarumã wheat pasture were influenced by the doses of nitrogen in the topdressing under the same environmental conditions (temperature and rainfall). An exponential model explained the dynamics of nutrient accumulation and was based on the thermal sum of each nitrogen dose impacting the cost per kilogram of pasture produced.

ABSTRACT. This study was realized to evaluate the monthly production, composition and quality of milk (total and defatted dry extract, lactose, fat and protein) and their relation to somatic cell count (SCC) and total bacterial count (TBC) using multivariate statistical analyses. The data are from a dairy farm for the period of two years (from January 2015 to December 2016). The SCC and TBC variables were transformed to somatic cell score (SCS) and log10 (LogTBC). Factor analysis, discriminant analysis and cluster analysis were used. Through factor analysis, it was found two factors that together explained 69.5% of the total data variation. The first factor represented the inverse relationship between lactose versus fat and protein content, while the second factor represented the inverse relationship among monthly milk yield versus SCS and LogTBC. The discriminant analysis identified that lactose and protein contents and SCS were the variables that had the greatest participation in the separation of the groups formed by the cluster analysis. The groups differed mainly by the monthly production of milk, composition and SCS. Finally, there are important multivariate relations between the variables milk production, composition and quality.

Abstract Our objective was to quantify the relationship between seasons of the year, milk production, and milk composition of a dairy farm based on data for 48 consecutive months, using multivariate statistical analyses. The dataset contained information on productive indexes and milk composition from the bulk tank milk, which was measured from milk samples, collected monthly and used to determine the total dry extract and defatted dry extract, lactose, fat, protein, somatic cell count, and total bacterial count. Seasons of the year and milk production/hectare were also considered. Factor, cluster, and discriminant analysis were used to study the relationships between the above-mentioned variables. A positive relationship was noted between season and total dry extract, defatted dry extract, milk fat, and protein, with higher values being observed in winter and spring. Similarly, a positive relationship was noted between season and milk production/hectare, lactose content, with an increase in milk production and lactose content in winter and spring, which was negatively related to the somatic cell count and total bacterial count. Milk production and composition varied mainly with seasons. Multivariate analyses facilitated a better understanding of the relationship between milk production and composition on this dairy farm.