Validating statistical properties of resilience indicators derived from simulated longitudinal performance measures of farmed animals.

Resilience is commonly defined as the ability of an individual to be minimally affected or to quickly recover from a challenge. Improvement of animals' resilience is a vital component of sustainable livestock production but has so far been hampered by the lack of established quantitative resilience measures. Several studies proposed that summary statistics of the deviations of an animal's observed performance from its target performance trajectory (i.e., performance in the absence of challenge) may constitute suitable quantitative resilience indicators. However, these statistical indicators require further validation. The aim of this study was to obtain a better understanding of these resilience indicators in their ability to discriminate between different response types and their dependence on different response characteristics of animals, and data recording features. To this purpose, milk-yield trajectories of individual dairy cattle differing in resilience, without and when exposed to a short-term challenge, were simulated. Individuals were categorised into three broad response types (with individual variation within each type): Fully Resilient animals, which experience no systematic perturbation in milk yield after challenge, Non-Resilient animals whose milk yield permanently deviates from the target trajectory after challenge and Partially Resilient animals that experience temporary perturbations but recover. The following statistical resilience indicators previously suggested in the literature were validated with respect to their ability to discriminate between response types and their sensitivity to various response features and data characteristics: logarithm of mean of squares (LMS), logarithm of variance (LV), skewness (S), lag-1 autocorrelation (AC1), and area under the curve (AUC) of deviations. Furthermore, different methods for estimating unknown target trajectories were evaluated. All of the considered resilience indicators could distinguish between the Fully Resilient response type and either of the other two types when target trajectories were known or estimated using a parametric method. When the comparison was between Partially Resilient and Non-Resilient, only LMS, LV, and AUC could correctly rank the response types, provided that the observation period was at least twice as long as the perturbation period. Skewness was in general the least reliable indicator, although all indicators showed correct dependency on the amplitude and duration of the perturbations. In addition, all resilience indicators except for AC1 were robust to lower frequency of measurements. In general, parametric methods (quantile or repeated regression) combined with three resilience indicators (LMS, LV and AUC) were found the most reliable techniques for ranking animals in terms of their resilience.

Inferring resource use from functional area presence in a small, single-flock of chickens in a mobile barn.

In poultry behavior research, the reliance on presence data to estimate actual resource usage has substantially increased with the advent of tracking technologies such as radio frequency identification (RFID) and image-based systems. Although such widely used technologies are fundamentally designed for presence tracking, many studies claim to use them to investigate actual resource usage. This study investigates whether the duration of chickens' presence near key resources accurately reflects their actual usage. To this end, we analyzed 210 ten-min video sequences from 5 days of recordings of 21 chickens, focusing on their proximity to and use of 6 key resources in a mobile poultry barn. Human observers manually assessed the durations of proximity-presence in defined functional areas of interest-and resource use for each individual in the video sequences. Significant correlations (Spearman's coefficient 0.83-1) were found for most resources, except the pophole (Rho = -0.30). Usage-to-presence ratios varied: perches exceeded 87%, feeder and enrichments around 66%, drinker 50%, and pophole 10%. Our findings highlight that mere proximity to resources does not always guarantee their effective use. We emphasize the need for careful interpretation of data from tracking technologies, acknowledging the distinction between mere proximity and actual resource use. Future studies should include larger sample sizes and varied conditions to ensure broader applicability.

Dataset for herding and predator detection with the use of robots.

Precision livestock farming involves the use of new technologies to improve the performance of farms with low profit margins. Since extensive livestock farming is demanding work requiring continuous supervision, it has not improved as drastically as agriculture. Furthermore, nowadays the world is more aware of the importance of respecting biodiversity and reducing the carbon footprint, for which sustainable animal production is recommended. This is the case of small livestock farms, generally located in unpopulated areas and with difficult generational replacement, due to the tasks involved. The use of robots and other devices equipped with intelligent systems can be useful to the farmer in his daily work. In this way, livestock, specifically flocks of sheep, can be monitored and the presence of potential predators such as the wolf identified. Encountering said predator can be avoided by moving the herd to other, safer pasture areas. This work presents a dataset that contains images and videos that allow detecting, classifying and analyzing flocks of sheep and one of their usual predators, wolves. The dataset includes videos of flocks in different locations, with different lighting conditions and different types of sheep. In addition, it contains images of wolves in natural spaces, which are not usually included in the most common datasets used in computer vision. This dataset can be very useful for the work being carried out in extensive precision livestock farming, to develop intelligent systems, such as a robot, that allow autonomous monitoring and control of a herd. Furthermore, it can be used to analyze animal behavior in the presence of a robot, since some of the images have been acquired with the cameras of a quadruped robot. This dataset has been split into three different Zenodo repositories due to its size. Images of sheep can be downloaded from https://zenodo.org/records/11313800 The images of classes 'Person', 'Wolf' and the depth maps for simulation are publicly available at https://zenodo.org/records/11313966 YOLO annotations are at https://zenodo.org/records/11313165.

Learning end-to-end respiratory rate prediction of dairy cows from RGB videos.

Respiratory rate (RR) is an important indicator of the health and welfare status of dairy cows. In recent years, progress has been made in monitoring the RR of dairy cows using video data and learning methods. However, existing approaches often involve multiple processing modules, such as region of interest (ROI) detection and tracking, which can introduce errors that propagate through successive steps. The objective of this study was to develop an end-to-end computer vision method to predict RR of dairy cows continuously and automatically. The method leverages the capabilities of a state-of-the-art Transformer model, VideoMAE, which divides video frames into patches as input tokens, enabling the automated selection and featurization of relevant regions, such as a cow's abdomen, for predicting RR. The original encoder of VideoMAE was retained, and a classification head was added on top of it. Further, the weights of the first 11 layers of the pre-trained model were kept, while the weights of the final layer and classifier were fine-tuned using video data collected in a tie-stall barn from 6 dairy cows. Respiratory rates measured using a respiratory belt for individual cows were serving as the ground truth (GT). The evaluation of the developed model was conducted using multiple metrics, including mean absolute error (MAE) of 2.58 breaths per minute (bpm), root mean squared error (RMSE) of 3.52 bpm, root mean squared prediction error (RMSPE; as a proportion of observed mean) of 15.03%, and Pearson correlation (r) of 0.86. Compared with a conventional method involving multiple processing modules, the end-to-end approach performed better in terms of MAE, RMSE and RMSPE. These results suggest the potential to implement the developed computer vision method for an end-to-end solution, for monitoring RR of dairy cows automatically in a tie-stall setting. Future research on integrating this method with other behavioral detection and animal identification algorithms for animal monitoring in a free-stall dairy barn can be beneficial for a broader application.

Descriptive evaluation of a camera-based dairy cattle lameness detection technology.

Lameness in dairy cattle is a clinical sign of impaired locomotion, mainly caused by painful foot lesions, compromising the US dairy industry's economic, environmental, and social sustainability goals. Combining technology and on farm data may be a more precise and less labor-intensive lameness detection tool, particularly for early detection. The objective of this observational study was to describe the association between average weekly autonomous camera-based (AUTO) locomotion scores and hoof trimming (HT) data. The AUTO data were collected from 3 farms from April 2022 to March 2023. Historical farm HT data were collected from March 2016 to March 2023 and used to determine cow lesion history and date of HT event. The HT events were categorized as a regular HT (TRIM; n = 2290) or a HT with a lesion recorded (LESION; n = 670). Events with LESION were sub-categorized based on lesion category: digital dermatitis (DD; n = 276), sole ulcer (SU; n = 79), white line disease (WLD; n = 141), and other (n = 174). The data also contained the leg of the LESION, classified as front left (FL; n = 54), front right (FR; n = 146), rear left (RL; n = 281), or rear right (RR; n = 183) leg with 6 events missing the leg. Cows' HT histories were classified as follows: cows with no previous recorded instance of any lesion were classified as TRIM0 (n = 1554). The first instance of any hoof lesion was classified as LESION1 (n = 238). This classification was retained until a subsequent TRIM occurred - recorded as TRIM1 (n = 632). The next unique instance of any lesion following a TRIM1 was classified as LESION2 (n = 86). Any LESION events occurring after LESION1 or LESION2 without a subsequent TRIM were considered a hoof lesion recurrence and classified as LESIONRE1 (n = 164) and LESIONRE2 (n = 22), respectively. TRIM events after LESION2 or LESION2RE (n = 104) or LESION events after LESIONRE1 or LESIONRE2 were classified as LESION_OTHER (n = 160). The AUTO scores from -28 to -1 days prior to the HT event were summarized into weekly scores and included if cows had at least 1 observation per week in the 4 weeks before the event. For all weeks, LESION cows had a higher median AUTO score than TRIM cows. Cows with TRIM0 had the lowest and most consistent median weekly score compared to LESION and other TRIM classifications. Before HT cows with TRIM0 and TRIM1, both had median score increases of 1 across the 4 weeks, while the LESION categories had an increase of 4 to 8. Scores increased with each subsequent LESION event compared to the previous LESION event. Cows with SU lesions had the highest median score across the 4 weeks, WLD had the largest score increase, and DD had the lowest median score and score increase. When grouping a LESION event by leg the hoof lesion was found on, the AUTO scores for four groups displayed comparable median values. Due to the difference between TRIM and LESION events, this technology shows potential for the early detection of hoof lesions.

From Herd Health to Public Health: Digital Tools for Combating Antibiotic Resistance in Dairy Farms.

The emergence of antimicrobial resistance (AMR) is a significant threat to global food security, human health, and the future of livestock production. Higher rates of antimicrobial use in dairy farming and the sheer lack of new antimicrobials available for use focused attention on the question of how the dairy production sector contributed to the development of AMR and paved the path toward taking action to curtail it on the targeted type of farms. This paper aims to provide an introduction to a phenomenon that has gained considerable attention in the recent past due to its ever-increasing impact, the use of antimicrobial drugs, the emergence of antimicrobial resistance (AMR) on dairy farms, and seeks to discuss the possibilities of approaches such as digital health monitoring and precision livestock farming. Using sensors, data, knowledge, automation, etc., digital health monitoring, as well as Precision Livestock Farming (PLF), is expected to enhance health control and minimize disease and antimicrobial usage. The work presents a literature review on the current status and trends of AMR in dairy farms, an understanding of the concept of digital health monitoring and PLF, and the presentation and usefulness of digital health monitoring and PLF in preventing AMR. The study also analyses the strengths and weaknesses of adopting and incorporating digital technologies and artificial intelligence for dairy farming and presents areas for further study and level of use.

Sequential Behavior of Broiler Chickens in Enriched Environments under Varying Thermal Conditions Using the Generalized Sequential Pattern Algorithm: A Proof of Concept.

Behavior analysis is a widely used non-invasive tool in the practical production routine, as the animal acts as a biosensor capable of reflecting its degree of adaptation and discomfort to some environmental challenge. Conventional statistics use occurrence data for behavioral evaluation and well-being estimation, disregarding the temporal sequence of events. The Generalized Sequential Pattern (GSP) algorithm is a data mining method that identifies recurrent sequences that exceed a user-specified support threshold, the potential of which has not yet been investigated for broiler chickens in enriched environments. Enrichment aims to increase environmental complexity with promising effects on animal welfare, stimulating priority behaviors and potentially reducing the deleterious effects of heat stress. The objective here was to validate the application of the GSP algorithm to identify temporal correlations between heat stress and the behavior of broiler chickens in enriched environments through a proof of concept. Video image collection was carried out automatically for 48 continuous hours, analyzing a continuous period of seven hours, from 12:00 PM to 6:00 PM, during two consecutive days of tests for chickens housed in enriched and non-enriched environments under comfort and stress temperatures. Chickens at the comfort temperature showed high motivation to perform the behaviors of preening (P), foraging (F), lying down (Ld), eating (E), and walking (W); the sequences <{Ld,P}>; <{Ld,F}>; <{P,F,P}>; <{Ld,P,F}>; and <{E,W,F}> were the only ones observed in both treatments. All other sequential patterns (comfort and stress) were distinct, suggesting that environmental enrichment alters the behavioral pattern of broiler chickens. Heat stress drastically reduced the sequential patterns found at the 20% threshold level in the tested environments. The behavior of lying laterally "Ll" is a strong indicator of heat stress in broilers and was only frequent in the non-enriched environment, which may suggest that environmental enrichment provides the animal with better opportunities to adapt to stress-inducing challenges, such as heat.

The potential of virtual fencing technology to facilitate sustainable livestock grazing management.

Virtual fencing (VF) technology is gaining interest due to its potential to facilitate sustainable grazing management. It allows farmers to contain grazing livestock without physical fences, thereby reducing the time and labour associated with the implementation of conventional fences. From a conservation perspective, some sensitive areas within uplands should not be grazed during certain periods of the year, and VF provides an invisible and moveable fence line that can exclude livestock from these areas. However, there are also concerns associated with its use, including animal welfare impacts, cost-effectiveness, and public perception. The extent to which VF can contribute to make livestock systems more sustainable remains to be investigated. To address this gap, this study investigates the potential of VF to promote sustainable grazing management using the Efficiency, Substitution, and Redesign framework, which has been used for the first time in this context. The framework is particularly relevant in taking an active and normative approach to identify key aspects to focus on to help achieve sustainability. We consulted stakeholders including farmers, wildlife inspectors, veterinarians, policy officers, researchers, NGOs, farm advisors or certification managers, through focus groups (N = 4) and in-depth, semi-structured interviews (N = 5). Stakeholders have highlighted the potential of VF to provide new opportunities to increase the efficiency and sustainability of livestock grazing systems, enabling their redesign, and contributing to improved environmental and animal welfare outcomes, as well as higher financial and social performance. However, there are important aspects that remain to be addressed to achieve such redesign, including issues of reliability due to poor network signal, animals' ability to learn, biosecurity and safety issues related to the absence of physical fences, farm suitability and farmers' ability to use the systems effectively. This study highlights the need to ensure that the development and uptake of VF are mutually beneficial to farmers, animals, and the wider farming industry. This includes a highlight on the importance of participative approaches to involve key stakeholders to address concerns and maximise the potential of the technology.

Combining genetic markers, on-farm information and infrared data for the in-line prediction of blood biomarkers of metabolic disorders in Holstein cattle.

BACKGROUND: Various blood metabolites are known to be useful indicators of health status in dairy cattle, but their routine assessment is time-consuming, expensive, and stressful for the cows at the herd level. Thus, we evaluated the effectiveness of combining in-line near infrared (NIR) milk spectra with on-farm (days in milk [DIM] and parity) and genetic markers for predicting blood metabolites in Holstein cattle. Data were obtained from 388 Holstein cows from a farm with an AfiLab system. NIR spectra, on-farm information, and single nucleotide polymorphisms (SNP) markers were blended to develop calibration equations for blood metabolites using the elastic net (ENet) approach, considering 3 models: (1) Model 1 (M1) including only NIR information, (2) Model 2 (M2) with both NIR and on-farm information, and (3) Model 3 (M3) combining NIR, on-farm and genomic information. Dimension reduction was considered for M3 by preselecting SNP markers from genome-wide association study (GWAS) results. RESULTS: Results indicate that M2 improved the predictive ability by an average of 19% for energy-related metabolites (glucose, cholesterol, NEFA, BHB, urea, and creatinine), 20% for liver function/hepatic damage, 7% for inflammation/innate immunity, 24% for oxidative stress metabolites, and 23% for minerals compared to M1. Meanwhile, M3 further enhanced the predictive ability by 34% for energy-related metabolites, 32% for liver function/hepatic damage, 22% for inflammation/innate immunity, 42.1% for oxidative stress metabolites, and 41% for minerals, compared to M1. We found improved predictive ability of M3 using selected SNP markers from GWAS results using a threshold of > 2.0 by 5% for energy-related metabolites, 9% for liver function/hepatic damage, 8% for inflammation/innate immunity, 22% for oxidative stress metabolites, and 9% for minerals. Slight reductions were observed for phosphorus (2%), ferric-reducing antioxidant power (1%), and glucose (3%). Furthermore, it was found that prediction accuracies are influenced by using more restrictive thresholds (-log10(P-value) > 2.5 and 3.0), with a lower increase in the predictive ability. CONCLUSION: Our results highlighted the potential of combining several sources of information, such as genetic markers, on-farm information, and in-line NIR infrared data improves the predictive ability of blood metabolites in dairy cattle, representing an effective strategy for large-scale in-line health monitoring in commercial herds.

Feed intake in housed dairy cows: validation of a three-dimensional camera-based feed intake measurement system.

Measuring feed intake accurately is crucial to determine feed efficiency and for genetic selection. A system using three-dimensional (3D) cameras and deep learning algorithms can measure the volume of feed intake in dairy cows, but for now, the system has not been validated for feed intake expressed as weight of feed. The aim of this study was to validate the weight of feed intake predicted from the 3D cameras with the actual measured weight. It was hypothesised that diet-specific coefficients are necessary for predicting changes in weight, that the relationship between weight and volume is curvilinear throughout the day, and that manually pushing the feed affects this relationship. Twenty-four lactating Danish Holstein cows were used in a cross-over design with four dietary treatments, 2 × 2 factorial arranged with either grass-clover silage or maize silage as silage factor, and barley or dried beet pulp as concentrate factor. Cows were adapted to the diets for 11 d, and for 3 d to tie-stall housing before camera measurements. Six cameras were used for recording, each mounted over an individual feeding platform equipped with a weight scale. When building the predictive models, four cameras were used for training, and the remaining two for testing the prediction of the models. The most accurate predictions were found for the average feed intake over a period when using the starting density of the feed pile, which resulted in the lowest errors, 6% when expressed as RMSE and 5% expressed as mean absolute error. A model including curvilinear effects of feed volume and the impact of manual feed pushing was used on a dataset including daily time points. When cross-validating, the inclusion of a curvilinear effect and a feed push effect did not improve the accuracy of the model for neither the feed pile nor the feed removed by the cow between consecutive time points. In conclusion, measuring daily feed intake from this 3D camera system in the present experimental setup could be accomplished with an acceptable error (below 8%), but the system should be improved for individual meal intake measurements if these measures were to be implemented.

Exploring individual responses to welfare issues in growing-finishing pig feeding behaviour.

The feeding behaviour of individual growing-finishing pigs can be continuously monitored using sensors such as electronic feeding stations (EFSs), and this could be further used to monitor pig welfare. To make accurate conclusions about individual pig welfare, however, it is important to know whether deviations in feeding behaviour in response to welfare issues are shown only on average or by each individual pig. Therefore, this study aimed (1) to quantify the individual variation in feeding behaviour changes in response to a range of welfare issues, and (2) to explain this individual variation by quantifying the responses to welfare issues for specific subgroups of pigs. We monitored four rounds of 110 growing-finishing pigs each (3-4 months per round). We collected feeding behaviour data using IVOG® EFSs and identified health issues and heat stress using climate sensors and twice-weekly health observations. For each pig, a generalised additive model was fitted, which modelled feeding behaviour through time and estimated the effect of each welfare issue that the pig had suffered from. The range of these effect estimates was compared between pigs to study the individual variation in responses. Subsequently, pigs were repeatedly grouped using physical and feeding characteristics, and, with meta-subset analysis, it was determined for each group whether a deviation in response to the welfare issue (i.e. their combined effect estimates) was present. We found that the range in effect estimates was very large, approaching normal distributions for most combinations of welfare issues and feeding variables. This indicates that most pigs did not show feeding behaviour deviations during the welfare issue, while those that did could show both increases and reductions. One exception was heat stress, for which almost all pigs showed reductions in their feed intake, feeding duration and feeding frequency. When looking at subgroups of pigs, it was seen that especially for lameness and tail damage pigs with certain physical characteristics or feeding strategies did consistently deviate on some feeding components during welfare issues (e.g. only relatively heavier pigs reduced their feeding frequency during lameness). In conclusion, while detection of individual pigs suffering from heat stress using feeding variables should be feasible, detection of (mild) health issues would be difficult due to pigs responding differently, if at all, to a given health issue. For some pigs with specific physical or behavioural characteristics, nevertheless, detection of some health issues, such as lameness or tail damage, may be possible.

Digital and Precision Technologies in Dairy Cattle Farming: A Bibliometric Analysis.

The advancement of technology has significantly transformed the livestock landscape, particularly in the management of dairy cattle, through the incorporation of digital and precision approaches. This study presents a bibliometric analysis focused on these technologies involving dairy farming to explore and map the extent of research in the scientific literature. Through this review, it was possible to investigate academic production related to digital and precision livestock farming and identify emerging patterns, main research themes, and author collaborations. To carry out this investigation in the literature, the entire timeline was considered, finding works from 2008 to November 2023 in the scientific databases Scopus and Web of Science. Next, the Bibliometrix (version 4.1.3) package in R (version 4.3.1) and its Biblioshiny software extension (version 4.1.3) were used as a graphical interface, in addition to the VOSviewer (version 1.6.19) software, focusing on filtering and creating graphs and thematic maps to analyze the temporal evolution of 198 works identified and classified for this research. The results indicate that the main journals of interest for publications with identified affiliations are "Computers and Electronics in Agriculture" and "Journal of Dairy Science". It has been observed that the authors focus on emerging technologies such as machine learning, deep learning, and computer vision for behavioral monitoring, dairy cattle identification, and management of thermal stress in these animals. These technologies are crucial for making decisions that enhance health and efficiency in milk production, contributing to more sustainable practices. This work highlights the evolution of precision livestock farming and introduces the concept of digital livestock farming, demonstrating how the adoption of advanced digital tools can transform dairy herd management. Digital livestock farming not only boosts productivity but also redefines cattle management through technological innovations, emphasizing the significant impact of these trends on the sustainability and efficiency of dairy production.

Multilevel sensor for monitoring external and internal environment of eggs.

Although it is well known that incubation environment has a great influence on embryogenesis and post-hatching performance of birds, not much is known about how external thermal, sound and light stimuli are isolated by eggshells and perceived by embryos. In this context, this study aimed to develop, calibrate and evaluate a multilevel sensor for integrated monitoring of the external (incubator) and internal environment of eggs. The variables of interest for the external environment were air temperature and relative humidity. For the internal environment, shell temperature, internal temperature, luminosity and sound pressure level were considered. The sensor was developed with an ATmega328 microcontroller, in open-source prototyping, using electronic components which are compatible with the egg's physical structure. Calibrations were carried out in a controlled environment, comparing the multilevel sensor with commercial equipment, obtaining coefficients of determination of R 2 > 0.90 for all variables studied. The multilevel sensor was also validated, simulating a commercial incubation situation and comparing eggs with 2 shell colors (white and brown) and internal volume (intact and empty). Validation results showed that white-shelled eggs insulate less external light (P < 0.001) and full eggs presented higher internal temperatures, greater light and lower sound pressure levels compared to empty eggs (P < 0.001). The multilevel sensor developed here is an innovative proposal for monitoring, simultaneously and in real time, different variables of interest in the commercial incubation environment.

Pinpointing pigs: performance and challenges of an ultra-wideband real-time location system for tracking growing-finishing pigs under practical conditions.

Real-Time Location Systems (RTLSs) are promising precision livestock farming tools and have been employed in behavioural studies across various farm animal species. However, their application in research with fattening pigs is so far unexplored. The implementation of these systems has great potential to gain insight into pigs' spatial behaviour such as the use of functional areas and pigs' proximity to each other as indicators for social relationships. The aim of this study was therefore to validate the accuracy, precision, and data quality of the commercial Noldus Information Technology BV TrackLab system. We conducted different measurement sets: first, we performed static measurements in 12 pens at four different locations in each pen at three heights each using a single ultra-wideband tag (UWB). We recorded unfiltered x- and y-coordinates at 1 Hz. We repeated these measurements with six tags aligned in a 2 × 3 grid with varied spacing to test interference between the tags. We also tested dynamic performance by moving the tags along the centre line of the pens. Finally, we measured the data quality with 55 growing pigs in six pens, including the identification of location 'jumps' from the inside to the outside of the pen. Each pen housed ten animals fitted with a UWB tag attached to their farm ear tag. We collected data for 10 days and analysed seven 24-h periods of raw and filtered data. The mean accuracy of the RTLS measurements was 0.53 m (precision: 0.14 m) for single and 0.46 m (precision: 0.07 m) for grouped tags. Accuracy improved with increasing measurement height for single tags but less clearly for grouped tags (P [height single] = 0.01; P [height grouped] = 0.22). When tags were fitted to animals, 63.3% of the filtered data was lost and 21.8% of the filtered location estimates were outside the pens. Altogether, the TrackLab system was capable of fairly accurate and precise assignment of the functional areas where individual animals were located, but was insufficient for the analysis of social relationships. Furthermore, the frequent occurrence of gaps in signal transmission and the overall high data loss rates presented significant limitations. Additionally, the challenging hardware requirements for attaching sensors to the animals underline the need for further technological advances in RTLS for the application with growing-finishing pigs.

A matter of age? How age affects the adaptation of lactating dairy cows to virtual fencing.

Virtual Fencing (VF) can be a helpful technology in managing herds in pasture-based systems. In VF systems, animals wear a VF collar using global positioning, and physical boundaries are replaced by virtual ones. The Nofence (Nofence AS, Batnfjordsøra, Norway) collars used in this study emit an acoustic warning when an animal approaches the virtual boundaries, followed by an aversive electrical pulse if the animal does not return to the defined area. The stimuli sequence is repeated up to three times if the animal continues to walk forward. Although it has been demonstrated that animals successfully learn to adapt to the system, it is unknown if this adaptation changes with animal age and thus has consequences for VF training and animal welfare. This study compared the ability of younger and older dairy cows to adapt to a VF system and whether age affected activity behavior, milk yield, and animal long-term stress under VF management. The study was conducted on four comparable strip-grazing paddocks. Twenty lactating Holstein-Friesian cows, divided into four groups of five animals each, were equipped with VF collars and pedometers. Groups differed in age: two groups of older cows (>4 lactations) and two groups of younger ones (first lactation). After a 7-d training, paddock sizes were increased by successively moving the virtual fence during four consecutive grazing periods. Throughout the study, the pedometers recorded daily step count, time spent standing, and time spent lying. For the determination of long-term stress, hair samples were collected on the first and last day of the trial and the hair cortisol content was assessed. Data were analyzed by generalized mixed-effect models. Overall, age had no significant impact on animal responses to VF, but there were interaction effects of time: the number of acoustic warnings in the last period was higher in younger cows (P < 0.001), and the duration of acoustic warnings at training was shorter for older cows (P < 0.01). Moreover, younger cows walked more per day during the training (P < 0.01). Finally, no effects on milk yield or hair cortisol content were detected. In conclusion, all cows, regardless of age, adapted rapidly to the VF system without compromising their welfare according to the indicators measured.

For dairy farmers, pasture management is a difficult task, including feeding the herd on demand, improving pasture use efficiency, and dealing with high labor costs. Virtual Fencing (VF) is an innovative technology that can help farmers to solve these issues. In a VF system animals wear a tracking collar. Physical boundaries are replaced by virtual ones using a smartphone application. The collars emit an acoustic warning when the animal reaches the virtual boundaries, further accompanied by an aversive electrical pulse if the animal does not return to the predefined area. Previous studies have shown that cattle learned to adapt to the system easily, but it is still unclear if older animals can adapt just as quickly. Thus, this is the first study investigating the effect of dairy cow age on learning VF in a strip-grazing trial. The results showed that older and younger cows adapted to the system equally fast, with no differences in activity behavior or changes in daily milk yield. Moreover, hair cortisol levels did not indicate lasting stress in the cows associated with the VF management during the trial. These results demonstrate the potential of VF in the management of lactating grazing cows of all ages.

Comparison of 3 mathematical models to estimate lactation performance in dairy cows.

Milk yield dynamics and production performance reflect how dairy cows cope with their environment. To optimize farm management, time series of individual cow milk yield have been studied in the context of precision livestock farming, and many mathematical models have been proposed to translate raw data into useful information for the stakeholders of the dairy chain. To gain better insights on the topic, this study aimed at comparing 3 recent methods that allow one to estimate individual cow potential lactation performance, using daily data recorded by the automatic milking systems of 14 dairy farms (7 Holstein, 7 Italian Simmental) from Belgium, the Netherlands, and Italy. An iterative Wood model (IW), a perturbed lactation model (PLM), and a quantile regression (QR) were compared in terms of estimated total unperturbed (i.e., expected) milk production and estimated total milk loss (relative to unperturbed yield). The IW and PLM can also be used to identify perturbations of the lactation curve and were thus compared in this regard. The outcome of this study may help a given end-user in choosing the most appropriate method according to their specific requirements. If there is a specific interest in the post-peak lactation phase, IW can be the best option. If one wants to accurately describe the perturbations of the lactation curve, PLM can be the most suitable method. If there is need for a fast and easy approach on a very large dataset, QR can be the choice. Finally, as an example of application, PLM was used to analyze the effect of cow parity, calving season, and breed on their estimated lactation performance.

Early detection of bovine respiratory disease in pre-weaned dairy calves using sensor based feeding, movement, and social behavioural data.

Previous research shows that feeding and activity behaviours in combination with machine learning algorithms has the potential to predict the onset of bovine respiratory disease (BRD). This study used 229 novel and previously researched feeding, movement, and social behavioural features with machine learning classification algorithms to predict BRD events in pre-weaned calves. Data for 172 group housed calves were collected using automatic milk feeding machines and ultrawideband location sensors. Health assessments were carried out twice weekly using a modified Wisconsin scoring system and calves were classified as sick if they had a Wisconsin score of five or above and/or a rectal temperature of 39.5 °C or higher. A gradient boosting machine classification algorithm produced moderate to high performance: accuracy (0.773), precision (0.776), sensitivity (0.625), specificity (0.872), and F1-score (0.689). The most important 30 features were 40% feeding, 50% movement, and 10% social behavioural features. Movement behaviours, specifically the distance walked per day, were most important for model prediction, whereas feeding and social features aided in the model's prediction minimally. These results highlighting the predictive potential in this area but the need for further improvement before behavioural changes can be used to reliably predict the onset of BRD in pre-weaned calves.

How do grazing beef and dairy cattle respond to virtual fences? A review.

Virtual fencing (VF) is a modern fencing technology that requires the animal to wear a device (e.g., a collar) that emits acoustic signals to replace the visual cue of traditional physical fences (PF) and, if necessary, mild electric signals. The use of devices that provide electric signals leads to concerns regarding the welfare of virtually fenced animals. The objective of this review is to give an overview of the current state of VF research into the welfare and learning behavior of cattle. Therefore, a systematic literature search was conducted using two online databases and reference lists of relevant articles. Studies included were peer-reviewed and written in English, used beef or dairy cattle, and tested neck-mounted VF devices. Further inclusion criteria were a combination of audio and electrical signals and a setup as a pasture trial, which implied that animals grazed in groups on grassland for 4 h minimum while at least one fence side was virtually fenced. The eligible studies (n = 13) were assigned to one or two of the following categories: animal welfare (n studies = 8) or learning behavior (n studies = 9). As data availability for conducting a meta-analysis was not sufficient, a comparison of the means of welfare indicators (daily weight gain, daily lying time, steps per hour, daily number of lying bouts, and fecal cortisol metabolites [FCM]) for virtually and physically fenced animals was done instead. In an additional qualitative approach, the results from the welfare-related studies were assembled and discussed. For the learning behavior, the number of acoustic and electric signals and their ratio were used in a linear regression model with duration in days as a numeric predictor to assess the learning trends over time. There were no significant differences between VF and PF for most welfare indicators (except FCM with lower values for VF; P = 0.0165). The duration in days did not have a significant effect on the number of acoustic and electric signals. However, a significant effect of trial duration on the ratio of electric-to-acoustic signals (P = 0.0014) could be detected, resulting in a decreasing trend of the ratio over time, which suggests successful learning. Overall, we conclude that the VF research done so far is promising but is not yet sufficient to ensure that the technology could not have impacts on the welfare of certain cattle types. More research is necessary to investigate especially possible long-term effects of VF.

Virtual fencing is a GPS-enabled fencing technology with the potential for improved livestock and pasture management, as well as socioeconomic and environmental benefits. However, the missing visual cue of a physical fence and the use of electric signals to ensure animals stay within the invisible boundary raise ethical and animal welfare concerns regarding the animal's ability to understand and learn the technology and the stress and anxiety associated with these processes. In this review, data from studies investigating the welfare and learning behaviors of virtually fenced animals were collected and analyzed to give an overview of this research field. It shows that the welfare of cattle in extensive systems is not adversely affected by the virtual fencing system, and the animals learn to avoid the electric signals. However, more research is necessary, especially over longer periods of time and with cows in intensive grazing systems, to ensure the welfare of virtually fenced cattle.

Strategies of maize silage supplementation of grazing dairy cows: Effects on milk production, pasture intake, grazing behaviour and methane emissions.

Supplementation strategy and grazing management can strongly influence dairy cow feeding behaviour, herbage intake, milk production and methane emissions. Two studies were conducted to investigate (1) the level of supplementation with partial mixed rations (PMR) and (2) the timing of maize silage feeding (morning vs. evening) for cows that have access to pasture either only during the day or day and night. A dataset was built that includes all individual cow measurements from both studies. It consists of 18 Microsoft® Excel files that correspond to several scales of information. The main file, "GrASTech_04_CowMeasurements", contains individual weekly measurements of milk production and composition, body weight, supplement and herbage dry matter intake measured using the n-alkane method and grazing behaviour measured using Lifecorder Plus, for a total of 168 cow × week datapoints. Five Excel files provide supplementary information at larger scales: periods, experimental treatments, feeds offered and their chemical composition, pasture characteristics and grazing management, and cow characteristics. The remaining 12 Excel files provide information at the daily scale on weather (1 file), methane concentrations and emissions (1 file), the grazing schedule (1 file) and grazing behaviour (9 files). The files related to grazing behaviour include the daily pattern of grazing time every 2 min as determined by Lifecorder Plus, as well as the daily pattern of grazing time, rumination, overactivity, other activity, rest and standing every 5 min as determined by Feed'Live. This dataset can be used to better understand and investigate relations among and the influence of animal characteristics, grazing management, the supplementation strategy and weather conditions on daily herbage intake, grazing behaviour, milk production and methane emissions at a weekly scale. The detailed information on feeding and grazing behaviour can also be used to study between-cow and between-day variability in daily cow activities.