Effects of virtual fencing on behavior, cortisol concentrations, feed intake, and milk yield of lactating dairy cows in different grazing systems.

Virtual fencing system (VF) is suggested to optimize grazing management on dairy farms. The objective of this study was to evaluate the application of VF for lactating dairy cows in a daily or weekly rotational grazing system and its effect on behavior, cortisol concentrations, feed intake, and milk yield. A completely randomized block design with 64 lactating dairy cows was used with four treatments within a 2 × 2 factorial design. Treatments were: VF with weekly rotation to a new plot (VFW), VF with daily rotation to a new plot (VFD), physical electric fencing (EF) with weekly rotation to a new plot (EFW) and EF with daily rotation to a new plot (EFD). The cows were naïve to VF prior to the study. At night cows were housed indoors separately per treatment and a partial mixed ration was fed. During the day all cows grazed at the pasture. After a 4-d adaptation period on the pasture with EF for all cows, the VF cows were trained over 4 d with the VF (® Nofence, AS, Batnfjordsøra Norway). After the training period, all cows grazed in separate plots per treatment within the specific fencing and grazing system for four weeks (measurement period). Number of auditory and electrical cues, success, success ratio and confidence ratio (which weighs the success ratio against the proportion of auditory cues) were calculated from the VF data. Behavior was continuously recorded by activity sensors (SensOor, Harmelen, the Netherlands) for all cows. Behavioral observations were performed throughout the study. Milk samples were taken on day 2, 6, 9, 13, 16, 20, 27, and 34, and milk cortisol concentrations were analyzed using ELISA. At the start and end of the study hair samples were taken and hair cortisol concentrations were measured. Daily milk production, feed intake indoors and at the pasture were recorded throughout the study. The study showed no differences between cows in VF or EF in behavior, cortisol concentrations in hair and milk, feed intake, and milk yield parameters. Regardless of the grazing system, an increase in the number of successes, success ratio, and confidence ratio was observed for all cows with VF during the measurement period. Cows in VFD had a higher success and confidence ratio than VFW cows, indicating that more frequent shifts of the VF enhanced learning ability and confidence in the VF. In conclusion, VF did not affect behavior, cortisol concentrations, feed intake, and milk yield of lactating dairy cows, irrespective of the grazing system.