Search

In this study, we investigated the variation in the microbial community present in bulk tank milk samples and the potential effect of different farm management factors. Bulk tank milk samples were collected repeatedly over one year from 42 farms located in northern Sweden. Total and thermoresistant bacteria counts and 16S rRNA gene-based amplicon sequencing were used to characterize microbial community composition. The microbial community was in general heterogeneous both within and between different farms and the community composition in the bulk tank milk was commonly dominated by Pseudomonas , Acinetobacter , Streptococcus , unclassified Peptostreptococcaceae , and Staphylococcus. Principal component analysis including farm factor variables and microbial taxa data revealed that the microbial community in milk was affected by type of milking system. Milk from farms using an automatic (robot) milking system (AMS) and loose housing showed different microbial community composition compared with milk from tiestall farms. A discriminant analysis model revealed that this difference was dependent on several microbial taxa. Among farms using an automatic milking system, there were further differences in the microbial community composition depending on the brand of the milking robot used. On tiestall farms, routines for teat preparation and cleaning of the milking equipment affected the microbial community composition in milk. Total bacteria count (TBC) in milk differed between the farm types, and TBC were higher on AMS than tiestall farms (log 4.05 vs. log 3.79 TBC/mL for AMS and tiestalls, respectively). Among tiestall farms, milk from farms using a chemical agent in connection to teat preparation and a more frequent use of acid to clean the milking equipment had lower TBC in milk, than milk from farms using water for teat preparation and a less frequent use of acid to clean the milking equipment (log 3.68 vs. 4.02 TBC/mL). There were no significant differences in the number of thermoresistant bacteria between farm types. The evaluated factors explained only a small proportion of total variation in the microbiota data, however, despite this, the study highlights the effect of routines associated with teat preparation and cleaning of the milking equipment on raw milk microbiota, irrespective of type of milking system used. [ABSTRACT FROM AUTHOR]

This study was part of a larger project that aimed to understand the causes for increasing variation in cheese ripening in a cheese-producing region in northern Sweden. The influence of different on-farm factors on raw milk composition and properties was investigated and is described in this paper, whereas the monthly variation in the milk quality traits during 1 yr is described in our companion paper. The dairy farming systems on a total of 42 dairy farms were characterized through a questionnaire and farm visits. Milk from farm tanks was sampled monthly over 1 yr and analyzed for quality attributes important for cheese making. On applying principal component analyses to evaluate the variation in on-farm factors, different types of farms were distinguished. Farms with loose housing and automatic milking system (AMS) or milking parlor had a higher number of lactating cows, and predominantly Swedish Holstein (SH) breed. Farms associated with tiestalls had a lower number of lactating cows and breeds other than SH. Applying principal component analyses to study the variation in composition and properties of tank milk samples from farms revealed a tendency for the formation of 2 clusters: milk from farms with AMS or a milking parlor, and milk from farms with tiestall milking. The interaction between the milking system, housing system, and breed probably contributed to this grouping. Other factors that were used in the characterization of the farming systems only showed a minor influence on raw milk quality. Despite the interaction, milk from tiestall farms with various cow breeds had higher concentrations (g/100 g of milk) of fat (4.74) and protein (3.63), and lower lactose concentrations (4.67) than milk from farms with predominantly SH cows and AMS (4.32, 3.47, and 4.74 g/100 g of milk, respectively) or a milking parlor (4.47, 3.54, and 4.79 g/100 g of milk, respectively). Higher somatic cell count (195 × 103/mL) and lower free fatty acid concentration (0.75 mmol/100 g of fat) were observed in milk from farms with AMS than in milk from tiestall systems (150 × 103/mL and 0.83 mmol/100 g of fat, respectively). Type of farm influenced milk gel strength, with milk from farms with predominantly SH cows showing the lowest gel strength (65.0 Pa), but not a longer rennet coagulation time. Effects of dairy farming system (e.g., dominant breed, milking system, housing, and herd size) on milk quality attributes indicate a need for further studies to evaluate the in-depth effects of farm-related factors on milk quality attributes. [ABSTRACT FROM AUTHOR]

This study investigated the influence of monthly variation on the composition and properties of raw farm milk collected as part of a full-scale cheese-making trial in a region in northern Sweden. In our companion paper, the contribution of on-farm factors to the variation in milk quality attributes is described. In total, 42 dairy farms were recruited for the study, and farm milk samples were collected monthly over 1 yr and characterized for quality attributes of importance for cheese making. Principal component analysis suggested that milk samples collected during the outdoor period (June–September) were different from milk samples collected during the indoor period. Despite the interaction with the milking system, the results showed that fat and protein concentrations were lower in milk collected during May through August, and lactose concentration was higher in milk collected during April through July than for the other months. Concentrations of free fatty acids were generally low, with the highest value (0.86 mmol/100 g of fat) observed in February and the lowest (0.70 mmol/100 g of fat) observed in June. Plasmin and plasminogen-derived activities varied with sampling month without a clear seasonal pattern. The pH of farm tank milk ranged from 6.60 to 6.82, with the lowest and highest values in September and February, respectively. The highest somatic cell count was observed in August (201 × 103 cells/mL) and the lowest in April (143 × 103 cells/mL). The highest value of gel strength, was recorded in December (88 Pa) and the lowest in July (64 Pa). Rennet coagulation time and gel strength were inversely correlated, with the lowest rennet coagulation time value observed in December. Orthogonal projections to latent structures (OPLS) and discriminant analysis adaptation of OPLS identified casein micelle size and total proteolysis as the milk quality attributes with major responses to sampling month, with smaller casein micelle size and higher total proteolysis associated with the outdoor months. Using discriminant analysis adaptation of OPLS to further investigate causes behind the variation in milk traits revealed that there were factors in addition to feeding on pasture that differed between outdoor and indoor months. Because fresh grass was seldom the primary feed in the region during the outdoor period, grazing was not considered the sole reason for the observed difference between outdoor and indoor periods in raw milk quality attributes. [ABSTRACT FROM AUTHOR]