Scientific literature on virtual fencing and virtual herding of cattle
There is now a body of scientific literature on the efficacy of virtual fencing and herding technology to manage cattle. Across these studies – spanning multiple commercially available virtual fencing technologies, production systems, and research groups – researchers consistently reported that animals adapted effectively to the technology, and no significant adverse effects in the behavioural or physiological welfare indicators were measured within those studies’ reported durations.
Most of these published studies span at least several weeks, with a few extending to several months. This aligns with the widely accepted understanding in the virtual fencing literature that the early learning and adaptation period is the phase of highest theoretical welfare risk. If adverse welfare effects were likely to occur, they would be expected during this initial period of training and adaptation, rather than after animals have learned the cues.
The consistent findings across these studies demonstrate that animals successfully adjust during this training period. This is consistent with Halter’s extensive real-world evidence base from commercial farms, where animals typically learn within days and reinforcement rates remain low once trained.
- Verdon et al, (2024); The effectiveness of a virtual fencing technology to allocate pasture and herd cows to the milking shed. J. Dairy Sci. 107(8), 6161-6167
- Simonsen P, et al, (2024); Effects of Social Facilitation and Introduction Methods for Cattle on Virtual Fence Adaptation. Animals, 14(10),1456
- Hamidi et al, (2024); Training cattle for virtual fencing: Different approaches to determine learning success. Applied Animal Behaviour Science 273 (2024) 106220
- Fuchs P, et al (2024): Stress indicators in dairy cows adapting to virtual fencing. Journal of Animal Science, 102, 1–17
- Dozler K, et al, (2025): Influence of virtual fence on heart rate response in beef cattle. Rangelands, 47(1):72–83
- Confessore A, et al, (2024): A matter of age? How age affects the adaptation of lactating dairy cows to virtual fencing. Journal of Animal Science, 102, skae137
- Sonne C, et al, (2022); Cortisol in Manure from Cattle Enclosed with Nofence Virtual Fencing. Animals, 12,3017
- Hamidi D, et al, (2022): Heifers don’t care: no evidence of negative impact on animal welfare of growing heifers when using virtual fences compared to physical [electric] fences for grazing. Animal; 16(9).
- Confessore A, et al, (2022); Application of Virtual Fencing for the management of Limousin cows at pasture. Livestock Science 263 105037
- Bruijn B, et al, (2025); Effects of virtual fencing on behavior, cortisol concentrations, feed intake, and milk yield of lactating dairy cows in different grazing systems Journal of Animal Science 103, ska363
- Fuchs P, et al, (2025); Effectiveness of virtual fencing in a mountain environment and its impact on heifer behaviour and welfare Animal 19, 101600
Benefits of virtual fencing
Virtual fencing allows farmers to set and adjust boundaries in real-time very easily. This advancement in animal and pasture management enables farmers to shift animals virtually via the app as required. Cows can be moved to ensure nutritional intake is optimised and can be moved to safety during a weather event e.g. heavy rainfall or flooding. Virtual fences free farmers from the labour intensive management and maintenance of physical fences allowing them to commit more time on cow nutrition, safety and welfare. Halter increases herd productivity while reducing the need for electric fences, dogs and motorbikes for mustering cows, potentially leading to less stress and fewer injuries on farm.
GPS accuracy
Every Halter collar has a GPS tracker to continuously track the location of each cow. The guidance cues are informed by the location and behaviour of each cow. “GPS drift” is a minor inaccuracy between the location detected on a device and the device’s true location. It is common to all GPS-enabled technologies. For Halter, this inaccuracy occurs infrequently and is caused by external factors like the number and position of the satellites, as well as occasional interference from buildings, trees or atmospheric conditions.
Currently, the level of accuracy with Halter’s collars is approximately 1.5 metres. This can result in the infrequent occurrence of either overallocation or underallocation of feed using virtual fencing by this margin. To mitigate this, the Halter system has various filters to discard incorrect GPS position data. Halter also recommends to farmers to not ‘feed out’ along a virtual boundary, and to not graze mobs in long, thin breaks.
