Stress indicators in dairy cows adapting to virtual fencing.

Virtual fencing (VF) enables livestock grazing without physical fences by conditioning animals to a virtual boundary delimited with an audio tone (AT) and an electric pulse (EP). The present study followed the adaptation process of lactating dairy cows to a VF system with changing virtual boundaries and investigated its impact on animal welfare. Twenty cows were divided into stratified groups (2× VF; 2× electric fencing, EF) of five individuals. Each group grazed half-days in a separate EF paddock of comparable size during 3 d of acclimation (P0), followed by 21, 14, 14, and 7 d of experimental treatment (P1 to P4). At the start of the trial, all cows were equipped with an IceQube pedometer (Peacock Technology Ltd, Stirling, UK) and a VF collar (Nofence AS, Batnfjordsøra, Norway). During P0, cows were accustomed to their first paddock with a deactivated virtual boundary and wearing the sensors. In P1 to P4, an active virtual boundary for the VF groups, and a second EF for the EF groups was set up parallel to an outer EF within their paddock. Throughout the trial, the sensors continuously tracked cow positions and activity behavior at 15-min intervals. From P1 onwards, the VF collars additionally recorded each AT and EP per cow with a georeferenced time stamp. During P0 to P4, daily feed intake, body weight, and milk yield were recorded in the barn. A total of 26 milk samples were collected per cow to determine milk cortisol levels. Behavioral observations were conducted for 2 h on day 23 to record agonistic behaviors, vocalizations, and excretions. The total number of stimuli per cow ranged from 37 to 225 ATs (mean ±â€…SD: 1.9 ±â€…3.3 per day) and 3 to 11 EPs (mean ±â€…SD: 0.1 ±â€…0.7 per day) throughout the trial. The maximum number of EPs per day was 8 for an individual cow and occurred once on D1. Mean EP/AT decreased by 55% during the first three half-days of grazing and with each paddock change from 0.2 EP/AT in week 1 to 0.03, 0.02, and 0 EP/AT in weeks 4, 6, and 8, respectively. Linear and generalized mixed effects models revealed that milk yield and cortisol, feed intake, body weight, and activity and lying behavior did not significantly differ between VF and EF groups. A higher number of agonistic behaviors were observed in the VF groups when the VF system was activated. However, due to the short observation periods only few contacts were observed in total. Overall, all cows adapted to the VF system without evidence of lasting adverse effects on animal welfare.

Virtual fences are commercially available but face restrictions in some countries due to animal welfare concerns. For virtual fencing (VF), animals are equipped with collars that emit audio tones (ATs) followed by electric pulses (EPs) when they cross a virtual boundary tracked by global navigation. Existing studies have so far not covered the aspect of longer-term learning, impacting possibly VF suitability. The present study followed therefore the learning process of dairy cows with changing virtual boundaries and examined behavior and stress indicators in dairy cows during an 8-wk adaptation to VF across four experimental periods. Four control and treatment groups of five cows each were investigated. EPs occurred most frequently on days 1 to 3 and remained low for the remaining experiment. In the latter two experimental periods, almost no EPs were recorded while ATs were still triggered, indicating that it took the animals two introductions to a new fence line to respond to the ATs only. Animal welfare was assessed by monitoring cow activity and lying behavior, milk yield, milk cortisol, feed intake, body weight, and frequencies of agonistic interactions, vocalizations, and excretions. All cows adapted to the VF system without compromising animal welfare during the study period.

Development of a data-driven method for assessing health and welfare in the most common livestock species in Switzerland: The Smart Animal Health project.

Improving animal health and welfare in livestock systems depends on reliable proxies for assessment and monitoring. The aim of this project was to develop a novel method that relies on animal-based indicators and data-driven metrics for assessing health and welfare at farm level for the most common livestock species in Switzerland. Method development followed a uniform multi-stage process for each species. Scientific literature was systematically reviewed to identify potential health and welfare indicators for cattle, sheep, goats, pigs and poultry. Suitable indicators were applied in the field and compared with outcomes of the Welfare Quality® scores of a given farm. To identify farms at risk for violations of animal welfare regulations, several agricultural and animal health databases were interconnected and various supervised machine-learning techniques were applied to model the status of farms. Literature reviews identified a variety of indicators, some of which are well established, while others lack reliability or practicability, or still need further validation. Data quality and availability strongly varied among animal species, with most data available for dairy cows and pigs. Data-based indicators were almost exclusively limited to the categories "Animal health" and "Husbandry and feeding". The assessment of "Appropriate behavior" and "Freedom from pain, suffering, harm and anxiety" depended largely on indicators that had to be assessed and monitored on-farm. The different machine-learning techniques used to identify farms for risk-based animal welfare inspections reached similar classification performances with sensitivities above 80%. Features with the highest predictive weights were: Participation in federal ecological and animal welfare programs, farm demographics and farmers' notification discipline for animal movements. A common method with individual sets of indicators for each species was developed. The results show that, depending on data availability for the individual animal categories, models based on proxy data can achieve high correlations with animal health and welfare assessed on-farm. Nevertheless, for sufficient validity, a combination of data-based indicators and on-farm assessments is currently required. For a broad implementation of the methods, alternatives to extensive manual on-farm assessments are needed, whereby smart farming technologies have great potential to support the assessment if the specific monitoring goals are defined.

Can we detect patterns in behavioral time series of cows using cluster analysis?

Time series analysis can facilitate the detection of complex behavioral patterns and potentially provide new opportunities to assess animal welfare. The aim was to investigate whether dairy cows exhibit daily, individual patterns in activity and in area use in the barn. We predicted that behavioral patterns will be more consistent (1) within than between cows, (2) when area categorization is more specific and, thus, allows the detection of individual preferences for areas, and (3) during the night. We conducted the study at an experimental farm with 20 lactating Brown Swiss and Swiss Fleckvieh cows. The animals were housed in cubicles, and they received feed and were milked twice daily. Activity was recorded with IceTag pedometers (IceRobotics Ltd.), and area use with the SMARTBOW sensor system (Zoetis). Data were collected for 55 consecutive days and analyzed at 1-min intervals. To investigate the behavioral time series, we performed a hierarchical clustering analysis. A clustering process calculated distances between days, which were compared within and between cows based on t-tests and analyses of variance. Dendrograms of activity and area use showed that days of individual cows could not be grouped more closely together than those of different cows. A slightly better grouping was achieved with a more specific area categorization, but not during a specific time period. However, the average distances between days were always smaller within (mean ± SD; activity: 95.62 ± 76.88, lying areas: 0.14 ± 0.03, functional areas: 0.12 ± 0.01) than between cows (activity: 109.62 ± 75.33, lying areas: 0.16 ± 0.02, functional areas 0.13 ± 0.01). Considering that the time series of individual cows were slightly but always more similar compared with those between cows, and that more consistent patterns were found when the area categorization was more specific, it can be concluded that the cows exhibited weak individual preferences in area use and also weak daily individual patterns in activity and area use. Because the visual exploratory and empirical approaches used in this study do not account for variability, they do not seem to be suitable for the detection of patterns in animals that display greater plasticity in their temporal structure of activity. Thus, although determining the temporal structure of activity and area use bears the potential to assess the behavior and, in turn, for example, the physiological state and health status of cows, it does not seem to be achievable with a cluster analysis. Therefore, time series methods that account for temporal fluctuations in behavior should be further explored.

Drying-off practices on Swiss dairy farms: Status quo and adoption potential of integrating incomplete milking.

Drying-off practices to reduce milk production before dry-off are gaining attention because high milk yields at dry-off are becoming more common and increase the risk to cow health and welfare during the dry period. Incomplete milking for the last days before dry-off is one approach for reducing milk production. We conducted an online survey to determine the currently used drying-off practices on Swiss dairy farms and to identify the adoption potential of integrating incomplete milking before dry-off. In March 2021, the online survey was sent to a representative sample of 1,974 Swiss dairy farmers. A total of 518 completed questionnaires were analyzed. The mean number of dairy cows per farm was 39 (range: 11-140 cows). Thirty-five percent of cows produced considerable quantities of milk (>15 kg/d) at dry-off, and milk yield at dry-off increased with increasing annual milk yield. Abrupt dry-off was applied on 45% of the farms. The participants reported observing behavioral changes of cows such as increased vocalizations and decreased lying time associated with dry-off. Selective dry cow therapy was applied on 74% of the farms, and 44% of the participants indicated the use of antibiotics at dry-off as being "rather often," "often," or "always." Correlation analysis revealed that with increasing annual milk yields, the frequency of observed behavioral changes and antibiotic use at dry-off increased as well. Therefore, drying-off approaches that reduce milk production while supporting cow welfare are needed. We found that farmers showed an interest in testing the presented drying-off approach of incomplete milking. In addition, the farmers indicated that they would be more willing to test incomplete milking before dry-off if it became available for automated use in milking parlors or robots. Uncertainties regarding udder health appeared to be the main barrier for the adoption potential of this approach.

The adoption of precision agriculture enabling technologies in Swiss outdoor vegetable production: a Delphi study.

Digital technologies are a promising means to tackle the increasing global challenges (e.g., climate change, water pollution, soil degradation) and revolutionising agricultural production. The current research used a two-stage Delphi study with 34 experts from various domains, including production, advisory and research, to identify the key drivers and barriers, the most promising technologies and possible measures to support technology adoption in Swiss outdoor vegetable production. Combining these experts' views, the method provides realistic scenarios for future development. In Round 1, open-ended questions were used to collect the experts' opinions. These were then transformed into closed-ended questions for Round 2, where controlled feedback was provided to the experts. Twenty-six experts participated in both rounds, resulting in an overall response rate that was comparably high (76%). It was found that economic factors were important drivers and barriers in technology adoption and, consequently, the experts recommended financial measures to support this adoption. The practical relevance of new technologies provided through communication and education holds further potential in terms of their promotion. These findings are valuable beyond the research field. Educators and policy makers can build on the results and optimally align their efforts to target technology adoption and contribute to more sustainable agriculture.

Do we automatically detect health- or general welfare-related issues? A framework.

The early detection of health disorders is a central goal in livestock production. Thus, a great demand for technologies enabling the automated detection of such issues exists. However, despite decades of research, precision livestock farming (PLF) technologies with sufficient accuracy and ready for implementation on commercial farms are rare. A central factor impeding technological development is likely the use of non-specific indicators for various issues. On commercial farms, where animals are exposed to changing environmental conditions, where they undergo different internal states and, most importantly, where they can be challenged by more than one issue at a time, such an approach leads inevitably to errors. To improve the accuracy of PLF technologies, the presented framework proposes a categorization of the aim of detection of issues related to general welfare, disease and distress and defined disease. Each decision level provides a different degree of information and therefore requires indicators varying in specificity. Based on these considerations, it becomes apparent that while most technologies aim to detect a defined health issue, they facilitate only the identification of issues related to general welfare. To achieve detection of specific issues, new indicators such as rhythmicity patterns of behaviour or physiological processes should be examined.

Type of teat cup liner and cluster ventilation affect vacuum conditions in the liner and milking performance in dairy cows.

The optimal milking cluster should milk as gently as possible to minimize the mechanical effect on the teat tissue at an optimal milking performance and milk quality. The objective of this study was to investigate the influence of liner shape (round vs. triangular) and type of cluster ventilation (claw vs. mouthpiece chamber; MPC) on milking performance and vacuum at the teat end and in the MPC. Our hypothesis was that liner shape and cluster ventilation affect milking performance and MPC vacuum. Six Holstein Friesian cows were milked twice daily over 12 d with a bucket milker, using 4 different cluster types that combined liner shape and type of cluster ventilation at 3 different system vacuum settings (35, 42, and 50 kPa) in an incomplete randomized block design. Milk flow and vacuum in the MPC, at the teat end (measured in the short milk tube), and in the short pulse tube were continuously recorded during milking. Milk flow was higher, and hence main milking time was shorter, with the round than with the triangular liners. The MPC vacuum was lower in round than triangular liners, which was caused by higher air leakage between teat and liner barrel in the triangular liners. The MPC vacuum, as well as its cyclic fluctuations, increased at the end of milking (immediately before cluster detachment) in all cluster types, with the highest amplitude of fluctuation in the triangular liners with MPC ventilation. The MPC ventilation reduced the MPC vacuum in both liner types at the end of milking, and also in triangular liners during peak milk flow. Despite the observed differences of MPC vacuum, the ventilation type did not affect milking performance. However, milking with triangular MPC-ventilated liners caused an increased proportion of foamed milk, which could potentially have a negative effect on milk quality.

Setup, Test and Validation of a UHF RFID System for Monitoring Feeding Behaviour of Dairy Cows.

Feeding behaviour can be used as an important indicator to support animal management. However, using feeding behaviour as a tool for dairy cow management an automatic sensor system is needed. Hence, the objective of this study was to setup, test and validate a ultra-high frequency (UHF) radio-frequency identification (RFID) system for measuring time dairy cows spent at the feed fence using two types of passive UHF ear tags. In a first experiment, the reading area of the system was evaluated in two antenna positions. Subsequently, the UHF RFID system was validated with video observations and compared to the measurements of chewing time of a noseband pressure sensor and of the time spent at the feed fence registered by a sensor system with real-time localisation. Differences in the reading area were detected between the two antenna positions and types of ear tag. The antenna position leading to less false positive registrations was chosen for the experiment with cows. The validation with video data showed a high average sensitivity (93.7 ± 5.6%, mean ± standard deviation), specificity (97.8 ± 1.1%), precision (93.8 ± 2.3%) and accuracy (96.9 ± 0.9%) of the UHF RFID system for measuring the time spent at the feed fence. The comparison with the noseband pressure sensor and the real-time localisation resulted in high correlations with a correlation coefficient of r = 0.95 and r = 0.93, respectively. However, substantial absolute differences between the three systems pointed out differences between direct and indirect measures of feeding behaviour in general and between the different sensors in particular. Thus, detailed considerations are necessary before interpreting automatically measured feeding data generally.

Machine Learning Based Prediction of Insufficient Herbage Allowance with Automated Feeding Behaviour and Activity Data.

Sensor technologies that measure grazing and ruminating behaviour as well as physical activities of individual cows are intended to be included in precision pasture management. One of the advantages of sensor data is they can be analysed to support farmers in many decision-making processes. This article thus considers the performance of a set of RumiWatchSystem recorded variables in the prediction of insufficient herbage allowance for spring calving dairy cows. Several commonly used models in machine learning (ML) were applied to the binary classification problem, i.e., sufficient or insufficient herbage allowance, and the predictive performance was compared based on the classification evaluation metrics. Most of the ML models and generalised linear model (GLM) performed similarly in leave-out-one-animal (LOOA) approach to validation studies. However, cross validation (CV) studies, where a portion of features in the test and training data resulted from the same cows, revealed that support vector machine (SVM), random forest (RF) and extreme gradient boosting (XGBoost) performed relatively better than other candidate models. In general, these ML models attained 88% AUC (area under receiver operating characteristic curve) and around 80% sensitivity, specificity, accuracy, precision and F-score. This study further identified that number of rumination chews per day and grazing bites per minute were the most important predictors and examined the marginal effects of the variables on model prediction towards a decision support system.

Effect of milking stall dimensions on upper limb and shoulder muscle activity in milkers.

Increasing societal awareness for animal welfare can promote changes in legislation. Some of these changes may also affect the person that interacts with the animal in a shared workspace, such as in milking stalls. Swiss milking stalls were designed many years ago, when cows were smaller than they are today. A recent animal-based study indicated that welfare decreased in cows exposed to restricted space allowance in milking stalls, which had resulted from increasing body size without adjustment of milking stall dimensions. However, changing the milking stall dimensions without considering the milker may be detrimental. For many years, health issues, particularly of the upper limb and shoulders, have affected milking personnel. The current study investigated the effect of large and standard milking stall dimensions on muscle activity in milkers (as a measure of workload) during milking. This assessment is fundamental to ensure that legislation improving animal welfare does not jeopardize human health. The study took place in an experimental milking parlor that allowed for size adjustment of the individual milking stall. Nine milkers performed 2 shifts of milking in a herringbone and 2 shifts in a side-by-side milking parlor. The milking stall dimensions were large on one side and standard on the other side of the parlor; the 2 sides were switched between milking shifts. We used surface electromyography to monitor bilateral muscle activity of forearm (flexor carpi ulnaris), arm (biceps brachii), and shoulder (deltoideus anterior; upper trapezius) muscles. Statistical analysis was performed separately for the herringbone and the side-by-side parlor for each muscle using mean and maximum muscle activity as the target variables in a linear mixed-effects model. The analysis showed that the different milking stall dimensions did not consistently affect activity of the measured muscles. Our results suggest that milking stall dimensions are not a primary risk factor for poor ergonomics in parlor workers.

Lower working heights decrease contraction intensity of shoulder muscles in a herringbone 30° milking parlor.

Musculoskeletal disorders have been a main concern in milkers for many years. To improve posture, a formula was developed in a previous study to calculate ergonomically optimal working heights for various milking parlor types. However, the working height recommendations based on the formula for the herringbone 30° parlor were broad. To clarify the recommendations for the optimal working height, we investigated the effect of working height on upper limb and shoulder muscle contraction intensities. We evaluated 60 milking cluster attachment procedures in a herringbone 30° milking parlor in 7 men and 9 women. Specifically, we examined the effect of working height on muscle contraction intensity of 4 arm and shoulder muscles bilaterally (flexor carpi ulnaris, biceps brachii, deltoideus anterior, and upper trapezius) by using surface electromyography. The working heights (low, medium, and high), which reflect the ratio of the subject's height to the height of the udder base, were used in the milking health formula to determine and fit individual depth of pits. Data were evaluated for each muscle and arm side in the functions holding and attaching. Statistical analysis was performed using linear mixed effects models, where muscle contraction intensity served as a target variable, whereas working height coefficient, sex, subject height, and repetition were treated as fixed effects, and repetition group nested in working height nested in subject was considered a random effect. Contraction intensities decreased with decreasing working height for the deltoideus anterior and upper trapezius, but not for the flexor carpi ulnaris or the biceps brachii muscles in both holding and attaching arm functions. We found that milking at a lower working height reduced muscle contraction intensities of the shoulder muscles. Women showed higher contraction intensities than men, whereas subject height had no effect. The study demonstrated that a lower working height decreased muscular load during milking. These lower working heights should be used within the recommendations made by the milking health formula for the herringbone 30°. Working heights could be adjusted effectively for milkers of varying body height. Future studies should therefore use the milking health formula as a tool to objectively compare and improve the accuracy of the working height coefficients.