In dairy production, high feed efficiency (FE) is important to reduce feed costs and negative impacts of milk production on the climate and environment, yet little is known about the relationship between FE, eating behaviour and activity. This research communication describes how cows differing in FE, expressed as daily energy corrected milk production per unit of feed intake, differed in eating behaviour and activity. We used data from a study of 253 lactations obtained from 97 Holstein and 91 Jersey cows milked in an automatic milking system. Automated feed troughs recorded feed intake behaviour and cows wore a sensor that recorded activity from 5 to 200 d in milk (DIM). We used a mixed linear model to estimate random solutions for individual cows for traits of steps, lying and eating behaviour and calculated their correlation with FE during four periods (5-35, 36-75, 76-120 and 121-200 DIM). Separate analyses were performed for each breed and period. We found that individual level correlations between FE and behaviour traits were stronger in Jersey than in Holstein cows. Eating rate correlated weakly negatively to FE in Holstein cows and more strongly so in Jersey cows, such that efficient Jerseys were slower eaters. The physical activity of Jersey cows was weakly and negatively correlated to FE, but this was not the case in Holstein cows. We conclude that eating rate was consistently negatively associated with FE throughout lactation for Jersey cows, but not for Holstein cows.

•We aimed to establish common guidelines for experimental studies with cattle.•A book on “Methods in cattle physiology and behaviour research” was published.•The book is designed as an open-access living handbook and is open to everyone.•Citing the book saves space and avoids repetitions in scientific journals.•Referencing guidelines reduces apparent plagiarism among authors.

Feed intake is important to consider when studying welfare, productivity and efficiency in ruminants, particularly cattle. Over the last decade, several methods for automatic intake recording have been developed. These methods have been described in the chapter on feed and water intake of the present book of methods. Automatic feed bins do not only record feed intake, but also feeding behaviour. The duration of feed bin visits estimated by the Insentec Roughage Intake Control system (Hokofarm, Marknesse, The Netherlands) is highly correlated with that measured by direct observations and with chewing time at the feed bin, recorded by noseband pressure sensors. During a feed bin visit, the animal spends time eating, but may also stand without eating. Therefore, visit duration does not necessarily correspond to eating duration. To estimate eating duration from automatic feed bins which record intake, it is necessary to check that eating duration and feed bin visit duration are linked. We applied the checklist to validate sensor output for the recording of cattle behaviour presented in the present book of method on data from different types of automatic feed bins used to estimate eating time. From this work we specified the procedure of validation of such measurements.

AimThis study aimed to examine the effects of feeding or abdominal brushing on the release of the hormones oxytocin, ACTH and cortisol during milking in dairy cows.MethodsTwelve cows in early lactation were used (2 × 2 factorial experimental design), testing the effects of two types of sensory stimulation during milking over a 3 day period; feeding concentrate or manual abdominal brushing (1 stroke/s). Blood samples for hormone analyses were collected at time at −15, −1, 0 (onset of cluster), every min for 8 min, at 10, 12, 14, 16, 30, and 60 min. Hormone levels were assayed and AUC was calculated.ResultsMilking was associated with an immediate and significant rise of oxytocin. When milking was combined with feeding, significantly higher levels of oxytocin were observed at 2 and 4 mins (p < 0.05). No effect of brushing on oxytocin levels was observed. Milking alone was associated with a significant rise of ACTH levels. Feeding in connection with milking reduced the immediate rise of ACTH levels (p < 0.05) and AUC (p < 0.02), whereas no effects of brushing were found. Milking caused a progressive rise of cortisol levels. Concomitant feeding did not influence cortisol levels, whereas brushing significantly decreased cortisol levels at 1, 5 and 14 mins after onset of milking (p < 0.05).ConclusionFeeding increases oxytocin release in response to milking and decreases ACTH levels. Abdominal brushing did not influence these variables, but decreased cortisol levels. These data demonstrate that activation of afferent vagal nerve fibres and of cutaneous sensory nerves originating from the abdominal skin in front of the udder influence milking related hormone release differently.

The objective of the study was to describe the feeding behaviour of primiparous and multiparous Jersey cows compared to Holstein cows housed in separate groups in the same barn. Such information could help farmers to optimise management with respect to welfare and production. Yet, it remains limited for Jersey cows over the entire period of lactation. Feeding data of 116 Danish Jersey (mean parity 2.14 ± 1.32) and 218 Danish Holstein cows (mean parity 1.90 ± 1.16) were assessed using automatic feeders from day 15 to 252 of lactation. Total eating duration, duration of eating per visit, intervals between meals, number of visits per day and the eating rate were analysed using linear mixed effects models. The cows were kept in a loose-housing system, with cubicles and automatic milking robots, and the group composition was dynamic. Compared to Holstein cows, Jersey cows visited the feeder significantly more often with shorter between meal intervals. However, the visit duration and total daily eating time and eating rates were significantly shorter for Jersey cows. There was no difference between breeds in the daily eating time and eating rate of older cows. Younger Jersey cows had significantly lower eating rates than older Jersey cows. No other difference in parity was found within Jersey cows. Weeks in milk significantly affected the eating time per day, number of visits per day and eating rate. The trajectories of outcome variables during lactation did not differ between the two breeds. In conclusion, we found substantial differences in the feeding behaviour of Jersey and Holstein cows, however, these differences could also be related to a group effect.

To answer specific questions regarding cow behaviour or affective states, it is sometimes necessary to study the cow’s responses to specific stimuli in a test situation. This can be done either in the animal’s home environment or in a test arena. As specific research questions determine the test design, it is not possible to provide a general guideline that fits all types of behavioural test. However, there are several considerations common for most behaviour tests in cattle. These guidelines will assist when planning a test conducted either in the home environment or in a test arena but is by no means an exhaustive description of behavioural tests in cattle.

Standing and lying behaviours are well-defined. Lying is often described as when the flank or sternum of the animal is in contact with the ground, and end of lying when all four legs are perpendicular to the body. The transition from lying to standing and vice versa requires only a few seconds, and therefore differences in the description of lying do not greatly affect the calculation of time spent lying or standing. By contrast, if the transition between lying and standing is the subject under study, then the description of the behaviour is crucial, especially when one wants to compare results from different studies. For instance, if one investigates the comfort of cubicles, the total duration of lying may reflect the comfort of the cubicle when the animal is lying, while a low frequency of lying bouts may reflect difficulties in lying down/getting up due to poor design of separations. Likewise, standing and walking can be defined in different ways. Eating behaviour can be defined as the whole process by which the animal ingests feeds able to satisfy organic needs and rejects non-alimentary or toxic compounds. In practice, measuring eating behaviour often consists of assessing the number and duration of eating bouts over a specified time interval. New equipment for automatic recording of lying/standing/eating behaviour has been developed for both research and commercial use.

A process of validation assesses the appropriateness and usefulness of a tool for its intended purpose within a specific context. Ideally, the validation of a tool should describe the range of purposes and contexts in which it is appropriate. This generally cannot be done completely. Therefore, when such a wide validation cannot be done, the validation process needs to refer clearly to the purpose of the use of the tool and to which animals it is going to be applied. The technological development has led to an increase in the number of sensors and devices that are available for measuring cattle behaviour. The technical solutions behind the sensors can differ substantially, sensors use accelerometer technology, sound recording or image analysis to mention some, and even within each solution, there is a large amount of factors that can influence the quality of the sensor’s information output or battery capacity. A careful evaluation of the purpose and validation of the outputs from sensors for measuring cattle behaviour is required before use. [...] Many behavioural tests are conducted in different research units. Regardless of the nature and design of the experiment, the basic aspects of a standardised behavioural test must be followed to generate data, which is unambiguous and accurate.

The daily rumination pattern in cattle is influenced by different factors such as feeding frequency, physical and chemical characteristics of the diet, feeding time, fasting, photoperiod and grazing management. Studies have observed an apparent decrease in rumination activity in ruminal acidosis or mastitis challenging dairy cattle. Social and physical environment can also affect cattle’s rumination behaviour. Manual observation of eating and rumination in individual animal is time-consuming and labour intensive. Therefore, the need for developing innovative and non-intrusive techniques to assess rumination behaviour is important. The IGER behaviour recorder was the first commercially available tool, initially introduced by Penning, and further developed via integrating microcomputer-based systems for the digital recording of jaw movements. The IGER system has some limitations, such as feasibility for longer experiments or difficulties when interpreting acquired data. The Hi-Tag system is a neck-collar based rumination monitoring tool, which records data based on acoustic biotelemetry. The applicability of this system has been proven, however the accuracy of the collected data relies on the correct positioning of the collar on the animal’s body. The recently developed RumiWatch system combines data from a built-in pressure sensor and a triaxle accelerometer to track different behavioural characteristics in cattle. This system has been validated in dairy cows under different housing management systems.

Feed intake and time spent eating at the feed bunk are important predictors of dairy cows' productivity and animal welfare, and deviations from normal eating behavior may indicate subclinical or clinical disease. In the current study, we developed a random forests algorithm to predict dairy cows' daily eating time (of a total mixed ration from a common feed bunk) using data from a 3-dimensional accelerometer and a radiofrequency identification (RFID) prototype device (logger) mounted on a neck collar. Models were trained on continuous focal animal observations from a total of 24 video recordings of 18 dairy cows at the Danish Cattle Research Centre (Foulum, Tjele, Denmark). Each session lasted from 21 to 48 h. The models included both the present time signal and observations several seconds back in time (lag window). These time-lagged signals were included with the purpose of capturing changes over time. Because of the high costs of installing an RFID antenna in the feed bunk, we also investigated a model based solely on 3-dimensional accelerometer data. Furthermore, to address the trade-off between prediction accuracy and reduced model complexity and its implications for battery longevity, we investigated the importance of including observations back in time using lag window sizes between 8 and 128 s. Performance was evaluated by internal leave-one-cow-out cross-validation. The results indicated that we obtained accurate predictions of daily eating time. For the most complex model (a lag window size of 128 s), the median of the balanced accuracy was 0.95 (interquartile interval: 0.93 to 0.96), and the median daily eating time deviation was 7 min 37 s (interquartile interval: −6 to 15 min). The median of the average daily eating time during sessions was 3 h 41 min with an interquartile interval of 2 h 56 min to 4 h 16 min. Exclusion of RFID data resulted in a considerable decrease in prediction accuracy, mainly due to a decreased sensitivity of locating the cow at the feed bunk (median balanced accuracy of 0.87 at a lag window size of 128 s). In contrast, prediction accuracy only slightly decreased with decreasing lag window size (median balanced accuracy of 0.94 at a lag window size of 8 s). We suggest a lag window size of 64 s for further development of the prototype logger. The methodology presented in this paper may be relevant for future automatic recordings of eating behavior in commercial dairy herds.

Dairy farming is a complex production system involving biological, technological and human inputs. Therefore, `general knowledge of cause and effect' often seems inadequate to identify and implement optimal management procedures. To solve herd-specific problems, this paper explores the potential of planned experiments for internal use at the farm level to take advantage of local causal relationships. The shift towards larger dairy herds with access to automatic data recordings of a large number of relevant inputs and performance indicators supports the development of management tools that are able to estimate the effect of changes made in daily management on individual farms. The concept of EVolutionary OPeration (EVOP) implies making small systematic changes in production factors or procedures while running the production and continuously evaluating the results. The aim of this study was to develop and evaluate the feasibility of implementing EVOP in commercial dairy herds as an integral part of herd management. The concept of EVOP-Dairy is based on five principles: (1) farmer-driven identification of areas for improvement; (2) herd-specific goals for the interventions to be evaluated in EVOP trials; (3) a short EVOP trial period; (4) simple, but statistically sound, EVOP designs including data access and (5) regular estimation of intervention effects and frequent reporting to the farmer. The project involved three activities: first, visiting a number of dairy farms with the aim of identifying areas for management improvements and to define potential EVOP interventions and relevant designs of EVOP trials; second, conducting a series of EVOP trials to develop data registration, statistical models, analysis and reporting; third, interviewing the farmers to obtain their opinion of the conceptual idea and the process. These activities were documented for the twelve project farms, and five different EVOP trials are described in detail to illustrate the concept. In conclusion, the farmers found the concept a useful management improvement tool, although the EVOP trials created additional work. The EVOP-Dairy statistical models need to include dynamic multilevel data and control for confounding factors when estimating intervention effects, as design with randomization was not feasible in the majority of the identified EVOP trials. Therefore, future development for the EVOP-Dairy should focus on (i) easy to implement and execute interventions, (ii) guidelines to interpret intervention effect when practical conditions hinder fully randomized and well replicated interventions, and (iii) automation of the data analysis and reporting part of the concept.

Devices that record behavior automatically have made it possible to accurately measure the lying and eating behavior of large numbers of dairy cows. During lactation, weight, feed intake, and production of cows change; however, longitudinal studies of how the behavior of dairy cows is correlated with production traits during lactation are limited. This study describes changes in duration of lying and eating behavior throughout lactation and how these variables are related to changes in milk yield, live weight, and feed intake in lactating primi- and multiparous Holstein and Jersey cows. Data were from 255 cow lactations (43 primi- and 80 multiparous Jersey cows, and 56 primi- and 76 multiparous Holstein cows) from 5 to 200 d in milk. Leg-mounted tags were used to record lying time and steps; ad libitum feed intake (of a partial mixed ration) variables were recorded from feed bins on weight cells; and milk yield and live weight were recorded during automatic milking, all on a daily basis. The lactation trajectory was split into 4 segments. Data were analyzed using mixed effects linear models. Holstein cows spent more time lying and eating than Jersey cows, whereas Jersey cows had a greater number of steps (25-37%). First-lactation cows spent less time eating and had more steps than older cows. Average daily lying time was approximately 1 h longer during February than the shortest lying time, which was observed in August. Both Holstein and Jersey multiparous cows had longer lying times than cows in first parity after parturition; however, the lying time of multiparous cows decreased, whereas that of primiparous cows increased in the beginning of lactation. Later in lactation, older cows tended to increase duration of lying more than younger cows did. The daily change in behavior (lying, eating, and steps) and milk yield, live weight, and dry matter intake, characterized as slopes in the lactation period for each cow, were not strongly correlated. However, we found a moderate correlation between changes in milk yield and dry matter intake, and between changes in eating time and rate of eating. An increase in eating rate in multiparous Holstein cows was correlated with increasing lying time. In conclusion, the use of automated behavior recording enabled thorough investigations of relationships between a range of behavior traits and frequently recorded production traits, and revealed that patterns of change during lactation are strongly affected by breed and parity.

Enting concluded lameness, from an economic perspective, as the third most costly health disease, following mastitis and reproductive failure issues, in cattle units. Archer estimated the incidence rate of lameness in the United Kingdom cattle herds roughly 50 cases/100 cows in a year; nevertheless, due to poor correlation between incidence rates and records of treatments in farms, the actual number seems to be higher. Surprisingly, the significance of lameness associated with cattle welfare, health and profitability of the unit has been greatly underestimated. Recent works have shown a clear link between BCS and hook condition of cows with the development of lameness in these animals. Lameness is a multifactorial and progressive issue where different detriments contribute to its development via complex interactions. Detection of lame cattle can be facilitated through description of the animals' gait characteristics in a numerical scaling system known as locomotion scoring. The total number of visual (manual) locomotion scoring systems can reach up to 25, where differences lie mostly in the used scales, characterization of cows' gait, and posture. Automated locomotion scoring tools would be a big advantage for regular monitoring of lameness in the herd. Three methods that are commonly engaged with automated systems are: kinetic, kinematic and indirect. The kinetic and kinematic approaches measure the forces, involved in locomotion, and time and distance of variables, associated to limb movement, respectively. The indirect method simply exploits behavioral or production data as indicators for impaired locomotion. The automated tools/instruments that will be developed, based upon either of the aforementioned approaches, need to be validated with a ‘reference’ method. This usually is done by comparing with manual scoring; however, it is noteworthy that manual scoring systems have their own set of limitations.

The future sustainability of cattle production will require improved resource use efficiency, reduced GHG emissions, and improved animal health and welfare. Facing this challenge needs far more complex animal traits than previously and they need to be assessed under a range of conditions. For example, the concepts of feed efficiency, robustness and sensitivity to health disorders are more difficult to include in selection indices than simple productivity traits. It is now time to look for means to investigate complex animal traits using smart technologies and rapid analytical methods in a standardised way applied in many contexts. At the same time, the European Strategy Forum on Research Infrastructures (ESFRI) roadmap clearly identified the need for improved coordination, harmonisation and access to European research infrastructures (RIs) on farm animals. The SmartCow project (www.smartcow.eu) answering the call H2020-INFRAIA2016-2017 was selected by the European Commission for 4 years funding starting from 1st February 2018. Three types of activities are developed to increase the phenotyping capabilities of the cattle European sector. Networking activities will create, thanks to an inventory and an interactive map, a unique portal to key European cattle RIs. The project will ensure that existing guidelines are adopted [e.g., the ICAR Guidelines for Bovine Functional Traits (section 7) are cited in distinct guidelines generated by the Networking activities]. When no international standard exists (e.g. feed efficiency, digestive, behavioural traits…), the project works towards the use of unified measurement methods through common standards and guidelines. The development of the cattle ontology of traits (ATOL and EOL; www.atol-ontology.com) through SmartCow will also be an important step to unify research methodologies and link definition of traits with standardized methods. A cloud-based database platform developed by Agrimetrics using web semantic will ensure integration, sharing and interoperability of data generated by the project leading to an open European database on cattle traits and phenotypes. Joint research activities will generate innovations for the research community on cattle towards the use of less-invasive methods and high-throughput phenotyping. Refining in vivo methods to evaluate feed efficiency and emissions will generate innovations in experimental design and planning for more accuracy. The development of new biomarkers (proxies) that can be easily measured in milk, faeces, urine, or blood through rapid analytical methods (NIRS) will bring new phenotyping capacities. The development of tools to generate new and improved information from animal sensors and other routinely collected data (e.g. prediction of individual cow status in terms of health and welfare) will also enable a more efficient phenotyping and genetic selection of cattle. Finally, the project organizes transnational access to major RIs: INRA in France, Scotland’s Rural College and University of Reading in the UK, Wageningen University and WUR/DLO in the Netherlands, FBN-Leibniz in Germany, Teagasc in Ireland, Aarhus University in Denmark and IRTA in Spain. It provides access to around 2500 dairy and 1000 beef cattle and facilitate up to 30 research projects to be financed by the SmartCow project after selection through specific calls. Eleven projects have already been selected after the first call.

The present study investigated the effect of straw yard housing during the dry period and 2 d of additional maternity pen housing postcalving on lying and feeding behavior and calving difficulty in Holstein dairy cows. In this study, 122 multiparous cows were moved to either a straw yard or into freestall housing 4 wk before their expected calving date. Cows that had been housed in straw yards stayed in the maternity pen for an additional 2 d after their calving day, but cows that had been housed in freestalls were moved to the general lactation group the morning after calving. Lying time, lying bouts, feeding time, number of feeder visits, feed intake, feeding rate, and assisted calvings were recorded. Observations were divided into 2 periods: precalving (the 4-wk dry period before calving) and postcalving (the day of calving and the 2 d after). During the precalving period, cows housed in straw yards showed a higher number of lying bouts but no difference in lying time compared with cows housed in freestalls. Cows that were housed longer in the straw-bedded maternity pen postcalving spent more time lying during the 2 d postcalving and had a higher number of lying bouts on the day of calving than cows moved to the freestall area on the day postcalving. Additionally, cows that were housed longer in the maternity pen had a slower feeding rate and longer total feeding time during the 2 d after calving than cows with a shorter stay in the maternity pen. We found no difference in the number of assisted calvings. This study suggests that straw yard housing during the dry period may facilitate the transition between standing and lying. Furthermore, the extended stay in the maternity pen postcalving increased lying time, the number of lying bouts, and feeding time, but decreased feeding rate compared with cows that were moved to the general lactation group on the day postcalving. These results suggest potential recovery benefits with an extended stay in a maternity pen postcalving. However, further studies are needed to separate the effects of housing in the dry period and the effects of an extended housing in individual maternity pens.

The aim of this study was to investigate the effects of a concentrate strategy based on the individual cow partial mixed ration (PMR) intake compared with a flat rate. Half of the cows were fed the individual concentrate strategy, whereas the other half were fed the control strategy. The individually fed cows were offered a concentrate proportion equal to 30% of the ad libitum intake of the PMR in the automatic milking system, and the control cows were offered 3 kg of concentrate/d in the automatic milking system. The cows (83 Holstein and 64 Jersey), kept in 2 separate groups, were blocked between the treatments according to expected calving date, breed, and parity and were randomly divided between treatments. Lactation was divided into 3 periods (early, mid, and late lactation), and the MIXED and GLIMMIX procedures in SAS (SAS Institute Inc., Cary, NC) were used to analyze the production and behavioral responses. The response trajectories during lactation were analyzed with a MIXED procedure fitted by Wilmink parameters. The individually fed cows had a higher concentrate intake and a lower PMR intake than control cows. Moreover, the total dry matter intake and energy-corrected milk yield did not differ between concentrate strategies. The actual average concentrate intake reached 19% of the PMR intake during mid lactation and not 30% of the PMR intake, as intended with the strategy. This was due to leftovers and lower allocation than intended. The cow behavior did not differ between concentrate strategies. However, variation in PMR intake and eating rate between cows fed individually was lower during mid lactation and lower for lying bouts during early lactation compared with control cows. This may indicate that an individual cow concentrate strategy results in a more stable PMR intake and time budget, which potentially could improve welfare, but this needs further investigation. The overall conclusion is that the cows were robust to adjustments in the concentrate allowance.

The aim of this study was to investigate the short-term responses of dairy cows during periods of change in the concentrate allowance in an automatic milking system. The experiment had a design with a 2 × 2 factorial arrangement including 2 types of concentrates and 2 amounts of concentrates (type O: mix of pelleted concentrate and steamrolled, acidified barley; type S: pelleted) in amounts of 3 and 6 kg/d. The experiment length was 11 wk. The concentrate type changed between wk 6 and 7 and included both increase and decrease in concentrate allowance for each concentrate type. The concentrate allowance was changed by 0.5 kg/d over 6 d. The 96 cows (48 Danish Jersey, 48 Danish Holstein) included in the experiment were blocked according to breed, parity, and days in milk, and randomly divided into 8 groups of treatment order. The cows visited the automatic milking unit more often when concentrate type O was offered, but not when an increased concentrate allowance was provided. The changes in concentrate intake and partial mixed ration (PMR) eating time showed a symmetrical pattern between the periods of increasing allowance and decreasing allowance. However, PMR intake and milk yield varied in the magnitude of the responses, indicating that these responses may not be driven by the same underlying mechanisms during increase and decrease in concentrate allowance. The daily lying time increased and the PMR eating rate decreased during periods of both increase and decrease in concentrate allowance. We found no significant change in milk yield during increase in concentrate allowance, despite a higher milk yield during periods with constant concentrate allowance at the high concentrate amount; however, the milk yield decreased during periods of decrease in concentrate allowance. Visit frequency, lying time, and steps changed during periods of changes in concentrate allowance without showing any differences at the constant concentrate allowance. In conclusion, these results indicate that it may be difficult to adjust the individual concentrate allowance based on the short-term responses of the cow.