Real-time and long-term behavioural monitoring systems in precision livestock farming have huge potential to improve welfare and productivity for the better health of farm animals. However, some of the biggest challenges for long-term monitoring systems relate to “concept drift”, which occurs when systems are presented with challenging new or changing conditions, and/or in scenarios where training data is not accurately reflective of live sensed data. This study presents a combined offline algorithm and online learning algorithm which deals with concept drift and is deemed by the authors as a useful mechanism for long-term in-the-field monitoring systems. The proposed algorithm classifies three relevant sheep behaviours using information from an embedded edge device that includes tri-axial accelerometer and tri-axial gyroscope sensors. The proposed approach is for the first time reported in precision livestock behavior monitoring and demonstrates improvement in classifying relevant behaviour in sheep, in real-time, under dynamically changing conditions.

Social network analysis in dairy calves has not been widely studied, with previous studies limited by the short study duration, and low number of animals and replicates. In this study, we investigated social proximity interactions of 79 Holstein–Friesian calves from 5 cohorts for up to 76 days. Networks were computed using 4-day aggregated associations obtained from ultrawideband location sensor technology, at 1 Hz sampling rate. The effect of age, familiarity, health, and weaning status on the social proximity networks of dairy calves was assessed. Networks were poorly correlated (non-stable) between the different 4-day periods, in the majority of them calves associated heterogeneously, and individuals assorted based on previous familiarity for the whole duration of the study. Age significantly increased association strength, social time and eigenvector centrality and significantly decreased closeness and coefficient of variation in association (CV). Sick calves had a significantly lower strength, social time, centrality and CV, and significantly higher closeness compared to the healthy calves. During and after weaning, calves had significantly lower closeness and CV, and significantly higher association strength, social time, and eigenvector centrality. These results indicate that age, familiarity, weaning, and sickness have a significant impact on the variation of social proximity interaction of calves.

Using an Ekomilk Bond Total Ultrasonic Milk Analyzer, the physicochemical characteristics of milk from 10 herds and 237 samples were analyzed during the rainy season (July and August; 109 samples) and dry season (January and February; 128 samples), providing the following measures: fat (%), non-fat solids (NFSs; %), density (g/mL), freezing point (°C), acidity (%), temperature (°C), lactose (%), conductivity (mS/cm), pH and water (%). In this study, there was a higher percentage of fat (2.975 vs. 5.001%) and a higher density (1033.692 vs. 1035.45 g/mL) in milk during the dry season compared to during the rainy season (p < 0.05). The rest of the milk characteristics were not significantly different relative to the study season (p > 0.05). The variables of fat and temperature of milk presented higher values in the dry season than in the rainy season, whereas the variables of density, freezing point and conductivity presented higher values in the rainy season than in the dry season (p < 0.001). Regarding the variables of NFSs, protein, acidity, lactose, pH and water percentage, the results were similar in the two study seasons (p > 0.05). Finally, density maintained significant positive correlations with the variables of protein, freezing point and lactose (p < 0.05). However, fat showed significant negative correlations with density, freezing point, acidity and conductivity (p < 0.05). In conclusion, the physicochemical characteristics of raw milk from dual-purpose cows per herd are better during the rainy season.

Qualitative Behavioural Assessment (QBA) has been used at both the herd and individual levels as an indirect measure to assess changes in the emotional state in dairy cattle but has not been used in individual cattle to explore associations between emotional state and milk production over a lactation. The study assessed 114 individual cows over a 13-month period utilising QBA and milk production data on a single farm. QBA scores were analysed using Principal Component Analysis (PCA). A decrease in Principal Component 1 (PC1) was associated with an increase in scores for positive behavioural terms such as happy, relaxed, calm and content. An increase in Principal Component 2 (PC2) was associated with an increased score for terms such as active, playful and inquisitive. Analysis with mixed models showed that milk yield was significantly negatively associated with PC1 for first-lactation cows (estimate = −0.24, p = 0.016) whilst accounting for days in milk (DIM) and temporal dependence. Variations in daily milk yields, assessed using the log-transformed variance of deviations from individual lactation curves (LnVar), were significantly positively associated with PC1 (percentage change in variability = 3.63, p = 0.004). These results indicate that increased milk yield and reduced daily yield variation were associated with more positive behavioural QBA terms. Exploratory work using clustering of PCA scores from the first 30 DIM was associated with 100-day milk yield, with cows in the more positive clusters (cluster 2 for PC1 and cluster 2 for PC2) producing the greater yield volumes (PC1 cluster 2: estimate 529.4, p = 0.02; PC2 cluster 2: estimate 1158.5, p < 0.001). These findings suggest that cows with more positive emotion scores are associated with increased milk production and provides evidence for further exploration in this research area.