Access to mechanical brushes enables grooming behaviour in dairy cows and has shown benefits for cow welfare, including improved cleanliness, comfort, stress reduction. Brush-use may also promote a positive emotional state. Reduced brush use has been associated with health issues, suggesting its potential for automated health monitoring. This study aimed at evaluating whether data generated by pose estimation could be used to assess brush use patterns in loose-housed dairy cows. It presents an approach for automatically identifying the body segment being brushed as an application of pose estimation. Data collection was carried out at the Swedish Livestock Research Centre in a loose housing system equipped with an automatic milking system and two mechanical rotating brushes. Recordings spanned 25:30 h and used three cameras, at different positions, monitoring a single mechanical brush placed in a passageway between cubicle rows. One human observer with access to recordings from all three synchronized cameras annotated the data-set on a second-by-second basis. The observer recorded: (1) the number of cows using the brush; (2) the anatomical segment being brushed; and (3) whether brushing resumed after a pause. The same video recordings were processed with object detection and pose estimation, which predicted the location of bounding boxes for cows and for the brush as well as corresponding keypoints. Using the brush and cow keypoint locations, we attempted to detect brushing by anatomical region. In a first stage, machine-learning models were trained to predict brushing state (independent of location) using keypoint distance to the brush, achieving an accuracy of 86.3 %. To mitigate the risk of error propagation, we relied on human annotations to segment the video to confirmed brushing bouts for analysis in the second stage. To identify the anatomical location of brushing, two methods were evaluated: (1) simply assigning the brushing location to the closest keypoint, achieving 73 % average accuracy across classes, and (2) projecting brush and anatomical keypoints onto a spline modelling the cow's backline, resulting in 87 % accuracy. Misclassifications were predominantly limited to adjacent body segments. Given that intra-observer reliability was 90 %, the spline-based method was deemed sufficiently reliable for research applications to accurately monitor the specific body segments being brushed.

Social interactions in a dairy herd are essential to maintain the herd's structure. Disturbances in social relationships can be stressful and may impact both animal welfare and production. Pathological and physiological changes, biological variations, but also the social environment induced by surrounding cows can affect variation in the daily milk production. This study aims to investigate the social interplay between cows during milking by examining the milking order in a milking parlour and determining if the individuals a cow stands next to will affect its daily milk yield. Milking order data from 234 individuals was collected from a two-sided herringbone parlour twice a day for 35 days. The indirect effect of the neighbour cows in the milking parlour was studied by fitting a linear mixed model to the daily milk yield residuals. The estimated indirect effects on milk yield ranged from -1.07 kg to 0.85 kg. We described a weak negative correlation of -0.26 (SE: 0.09) between direct and indirect effect estimates. The average of the indirect effects of neighbouring cows differed between different lactation stages and regrouped cows changed to a more negative estimated indirect effect in their new group. Our results show individual variation in the average indirect effect on the milk yield of the neighbour, with some individuals having a positive effect on their group mates, while others have a more negative effect. Further investigation of these effects would be helpful in selecting the best individuals in a herd and optimising group composition and milking routines.

A common management practice of commercial dairy farms is grouping cows according to their nutritional needs. Having different lactation groups also implies the regrouping of animals during their lactation period. Real-time locating systems based on ultra-wideband devices provide information about the cow's position inside barns. This information can be used to infer time budgets and animal locations on a continuous basis. The present study aimed to evaluate the behavioural changes that could be induced in dairy cows when regrouping between lactation groups by using location data obtained from real-time locating systems technology. This study is based on observational data from a commercial dairy farm, where 12 regrouping events, each involving between 3 and 15 cows, were recorded without any experimental intervention. The analyses included six months of data from two lactation groups with around 100 cows in each group. The effects of regrouping on time budgets and home range were analysed by comparing changes in regrouped cows before and after the process, and by comparing regrouped cows with their matched controls. The results showed that regrouping had a slight impact on the feeding and resting times of dairy cows. In contrast, parity was associated with opposing effects on feeding and resting times. Furthermore, regrouped cows tended to cover a larger area of the barn. The extent to which regrouped cows increased their use of barn space was also affected by the number of cows regrouped simultaneously. Our analysis of location data from a commercial dairy herd provides new insights into the impact of regrouping events on cow behaviour and welfare in commercial settings.

Cattle develop preferential relationships with other individuals in the herd. These social interactions between individuals have a significant impact on both animal welfare and production. Given the relevance of social behaviour in dairy cattle, scientific studies have focused on understanding social interactions among cattle. These may also be influenced by individual area preferences, particularly when animals are housed in confined spaces. Therefore, investigating the relationship between individual area preferences and social interactions is essential for understanding social behaviour in dairy cattle. Real-time location systems provide the opportunity to monitor individual area preferences and social contacts at the same time. This study aims to assess the impact of dairy cows’ area preferences on their daily social contacts and to determine the potential implications of overlooking individual area preferences in social behaviour studies. The individual position of the lactating cows was automatically collected once per second for two months on a Swedish commercial farm housing dairy cows inside a free-stall barn. The location data of 243 lactating cows was used to construct the social networks and to estimate the similarity of the area utilisation distributions between these individuals. The effect of utilisation distribution similarity in social networks was investigated by applying separable temporal exponential random graph mixed models. The role of different cow characteristics in the similarity of the utilisation distributions was assessed through a linear mixed model. Our analyses stressed the importance of similarity of area preference, parity, kindergarten effect, and filial relatedness in shaping daily social contacts in dairy cattle. The kindergarten effect refers to the effect on cow behaviour of being grouped together in the early stages of their lives. Similarity of area preference was influenced by the kindergarten effect and relatedness by pedigree, which favoured interactions between these individuals. The described approach allowed to disassociate the area preference from the social contacts between cows, providing more accurate results of the importance of the cow’s characteristics on their social behaviour. 

Social interactions between cows play a fundamental role in the daily activities of dairy cattle. Real-time location systems provide on a continuous and automated basis information about the position of individual cows inside barns, offering a valuable opportunity to monitor dyadic social contacts. Understanding dyadic social interactions could be applied to enhance the stability of the social structure promoting animal welfare and to model disease transmission in dairy cattle. This study aimed to identify the effect of different cow characteristics on the likelihood of the formation and persistence of social contacts in dairy cattle. The individual position of the lactating cows was automatically collected once per second for 2 wk, using an ultra-wideband system on a Swedish commercial farm consisting of almost 200 dairy cows inside a freestall barn. Social networks were constructed using the position data of 149 cows with available information on all characteristics during the study period. Social contacts were considered as a binary variable indicating whether a cow pair was within 2.5 m of each other for at least 10 min per day. The role of cow characteristics in social networks was studied by applying separable temporal exponential random graph models. Our results revealed that cows of the same parity interacted more consistently, as well as those born within 7 d of each other or closely related by pedigree. The repeatability of the topological parameters indicated a consistent short-term stability of the individual animal roles within the social network structure. Additional research is required to elucidate the underlying mechanisms governing the long-term evolution of social contacts among dairy cattle and to investigate the relationship between these networks and the transmission of diseases in the dairy cattle population. .

In modern freestall barns where large groups of cows are housed together, the behavior displayed by herd mates can influence the welfare and production of other individuals. Therefore, understanding social interactions in groups of dairy cows is important to enhance herd management and optimize the outcomes of both animal health and welfare in the future. Many factors can affect the number of social contacts in a group. This study aimed to identify which characteristics of a cow are associated with the number of contacts it has with other group members in 2 different functional areas (feeding and resting area) to increase our understanding of the social behavior of dairy cows. Inside 2 herds housed in freestall barns with around 200 lactating cows each, cow positions were recorded with an ultra-wideband real-time location system collecting all cows' positions every second over 2 wk. Using the positioning data of the cows, we quantified the number of contacts between them, assuming that cows spending time in proximity to one another (within a distance of 2.5 m for at least 10 min per day) were interacting socially. We documented in which barn areas these interactions occurred and used linear mixed models to investigate if lactation stage, parity, breed, pregnancy status, estrus, udder health, and claw health affect the number of contacts. We found variation in the number of contacts a cow had between individuals in both functional areas. Cows in later lactation had more contacts in the feeding area than cows in early lactation. Furthermore, in one herd, higher parity cows had fewer contacts in the feeding area than first parity cows, and in the other herd, cows in third parity or higher had more contacts in the resting area. This study indicates that cow characteristics such as parity and days in milk are associated with the number of contacts a cow has daily to its herd mates and provides useful information for further research on social interactions of dairy cows. 

Heat stress in dairy cows can cause an increase in body temperature and respiration rate, and a decreased feed intake leading to reduced production. Dairy cows are better at handling heat when they have access to shade. Therefore, this study aimed to determine the effects of providing shade to high-yielding dairy cattle during the summer in the Swedish climate. Twenty high-yielding Swedish Red dairy cows, held on pasture, were divided into two groups, one with access to shade (S) and one without (NS). Milk production was recorded daily and shade temperature and relative humidity were recorded at 10 min intervals at pasture. A major effect of heat stress was found in cows in early lactation in the NS group. In this group, a high mean temperature two days before and a high THI two days before affected the milk production negatively (p < 0.001), which was the same for the maximum temperature and maximum THI measured on the same day (p < 0.001). Increases in the mean temperature and THI two days before also affected milk production negatively (p < 0.05) for cows in early lactation in the S group, though to a lesser extent. This study suggests that dairy cows in early lactation benefit from access to shade during summer.

Virtual fences have been developed over the last 20 years with focusing enabling an easy and flexible pasture management for goats, sheep, and cattle. The focus on research have been on the animal’s ability to learn to associate an audio cue with a future electrical stimulus and their ability to learn a correct avoiding behaviour in response to the audio cue. No research has focused on what the presence of already trained sheep have on the learning curve of naïve sheep. Thus, the aim of this study was to examine if sheep have a faster learning curve when trained together with sheep that already know how to react to the sound cues in comparison to a group of naïve sheep.

Fifteen, one year old sheep, where divided into two groups of five (group 1) and ten (group 2). Additional five sheep of the same age, that had been trained on virtual fence two months before, was included in group 1. One day 1 the sheep was moved to two rectangular pastures with 40 meters from the back of the fence line to the virtual fence line. On day 3 the virtual fence line was move additional 30 meter to simulate strip pasture. Number of audio and electrical stimuli was collected by the system.

A success rate was calculated as ((#audio cues – #electrical stimuli)/ #audio cues) as a measure of how well the animal have learnt to react correctly to the system. Data was analyses using a mixed model in SAS, with #audio cues, #electrical stimuli and success rate as dependent variables and group, day, and their interaction as fixed effects. Animal was defined as repeated measure.

Day affected average number of sound cues, with an increasing number of sound cues from day 1 to day 5 (2.95 vs 8.25 ± 1.2, P<0.05). Furthermore, Group 1 had a higher success rate than Group 2 (0.94 ± 0.05 vs 0.75 ± 0.04, P<0.05). No other significant effects were found.

The results conclude that sheep learn to use a virtual fence and they get better over time. Additionally, it can be concluded that social transmission occur when sheep are being trained to use a virtual fence together with already trained sheep.