Virtual Fencing for Livestock Management: Effects on Cattle Behavior, Welfare, and Productivity

Introduction

For generations, farmers have relied on physical fences to manage their livestock. However, these traditional fences can be labor-intensive to install and maintain. Recently, a new technology called virtual fencing (VF) has emerged, allowing farmers to manage their livestock without physical barriers.

The VF system uses associative learning, where animals learn to connect a specific audio cue (sound) with a mild electrical pulse (EP) (Murphy and Lupfer, 2014). The VF boundary is created through a user-friendly interface to keep animals within a designated area. Both beef and dairy cattle learn to associate the audio cue with the EP and are generally contained within the virtual boundaries for 97-99% of the time (Wilms et al., 2024).

Virtual fencing technology includes several key components: Global Positioning System (GPS)-enabled collars worn by each animal to track their location relative to pre-set virtual boundaries; audio cues emitted by the collar when an animal approaches the virtual boundary (Goliński et al., 2022); and an aversive stimulus, such as a low-energy EP or vibration, delivered if the animal continues towards or crosses the boundary despite the audio cue (Murphy and Lupfer, 2014). The collar communicates with a base station; the data are usually stored in the cloud. Some advanced systems use machine learning to customize the stimulus level for individual animals, and smartphone apps allow farmers to adjust the virtual fence settings remotely (Goliński et al., 2022). 

Animal welfare is an important consideration for VF. While the Eps are meant to be aversive to encourage avoidance, concerns about stress arise (Wilms et al., 2024). However, research indicates that animals learn to avoid electrical pulses over time, and some studies suggest that stress levels are similar to or less than those associated with traditional fencing (Campbell et al., 2019). This technology offers flexibility in managing grazing areas and reduces labor costs associated with traditional fencing. This paper reviews the effects of using VF technology (wireless fencing) on the behavior and well-being of dairy and beef cattle.

What is Virtual Fencing?

Virtual fencing is a technology that uses a GPS-enabled collar on livestock to create invisible boundaries instead of traditional physical fences. This allows farmers to manage their livestock remotely from their mobile phones or tablets, offering greater flexibility (Goliński et al., 2022). Virtual fencing is an effective system for managing cattle, with animals learning to respond to audio cues and avoid EPs (Lee et al., 2009). Virtual fencing technology also helps manage grazing areas more efficiently and reduces labor costs associated with traditional fencing.

How Virtual Fencing Works

Virtual fencing technology replaces traditional physical barriers with virtual ones. Each animal wears a GPS-equipped collar (Figure 1). The farmer sets virtual boundaries within a geographic information system. When an animal approaches this boundary, the collar emits an audio warning signal or sound (2 s; 84 dB) (Verdon and Rawnsley, 2020; Verdon et al., 2024). If the animal continues towards the boundary, it receives a mild EP (0.5 s; 3 V, 120 mW) (Figure 1; Verdon and Rawnsley, 2020; Verdon et al., 2024). The system relies on animal learning to associate the audio warning (cue) with the EP and turn away from the boundary (Figure 1). This type of learning, known as aversion learning, helps animals avoid the unpleasant EP by responding to the audio cue alone (Verdon et al., 2024). Over time, the number of EPs typically decreases as animals become accustomed to the audio cues (Wilms et al., 2024). Virtual fencing can be applied in various agricultural settings, including rangeland management and intensive grazing systems. It optimizes pasture-based cow nutrition, reduces labor costs associated with physical fencing, and protects sensitive environmental areas (Figure 1). 

Figure 1. How virtual fencing works Adapted from Hogewerf et al. (2019)

Different Types of Systems

Today, a few different companies make VF systems. These systems usually include a collar that the animal wears and software to manage the VF lines (Figure 1). These systems send an audio cue as the animal approaches the boundary and then use a mild EP or vibration if it keeps going. Some systems use a combination of both (Goliński et al., 2022).

Several VF technologies have been utilized in various studies, each with its own design and functionality. These technologies generally employ a combination of audio cues and aversive stimuli, such as a low-energy electrical pulse, to manage animal movement. 

eShepherd 

This technology, developed in Australia by Gallagher, is frequently featured in published research. It has been used in multiple studies evaluating the impact of age, breed, moving virtual fences, and containment of cattle (Campbell et al., 2017; Campbell et al., 2018a,b; Lomax et al., 2019; Langworthy et al., 2021; Verdon et al., 2021a).

NoFence 

This system, from Norway, is another prominent technology in VF research (Aaser et al., 2022; Hamidi et al., 2022; Staahltoft et al., 2023).

Halter 

This New Zealand-based technology is specifically designed for intensive pastoral dairy production. It is unique in that it includes virtual herding capabilities in addition to VF (Verdon et al., 2024).

Boviguard 

This system from Agrifence in the UK, employs an induction cable laid on the ground (Wilms et al., 2024).

Pre-Commercial Prototype 

A pre-commercial prototype was used in a number of studies that evaluated a range of aspects, including animal behavior and welfare (Campbell et al., 2019; Lomax et al., 2019; Langworthy et al., 2021; Verdon et al., 2021b)

Custom VF Systems 

Some studies have also utilized custom-built VF systems. These systems were often developed for specific research purposes and are not commercially available (Lee et al., 2007; McSweeney et al., 2020; Colusso et al., 2021).

Most of the early VF research focused on extensive grazing systems, particularly rangelands. However, growing interest is in adapting this technology for more intensive systems, such as pastoral dairy production. The Halter system, for example, is specifically marketed for intensive dairy farming (Verdon et al., 2024). The development of VF technology is ongoing, and the systems continue to evolve. The effectiveness of different VF systems can vary and depends on several factors, including the specific technology, the animals' learning, and the grazing context. Further research is needed to evaluate the long-term impact of different VF technologies on animal behavior and welfare, particularly in intensive dairy systems.

Effects of Virtual Fencing on Cattle Behavior, Welfare, and Productivity

The effectiveness of VF relies on animals learning to associate an audio signal (cue) with a potential electrical pulse (EP), prompting them to move away from the virtual boundary to avoid the pulse (Lee et al., 2009; Umstatter, 2011). Understanding these factors can help farmers optimize the use of VF technology.

How Cattle Learn with Virtual Fences

Cattle are capable of associative learning, meaning they can connect cues and locations and learn from other herd members (conspecifics) (Keshavarzi et al., 2020). For VF to work effectively, cattle need to learn that an audio cue is followed by an electrical pulse (EP) (Lee et al., 2009). This understanding helps them avoid the virtual boundary by responding to the audio cue.

Rapid Learning

Studies show that cattle can learn to respond to VF within a few days. Most animals quickly associate the acoustic signal with the electric pulse (EP), learning to respond to the audio cue and avoid the electrical stimulus (Lee et al., 2009). For instance, dairy cows reduced the percentage of EPs received after the audio cue from 65% to 32% over a three-day training period (Langworthy et al., 2021). Similarly, beef heifers showed a reduction in the number of EPs received during the learning period (Campbell et al., 2017). An eight-week study indicated that most EPs occurred in the first three days, after which the animals were contained by audio tones alone (Verdon et al., 2021b). This suggests that animals learn quickly and can adapt to VF after initial exposure. However, individual responses to VF vary, and social learning and other factors can affect the learning process (Campbell et al., 2017). 

Social Learning 

Cattle can learn from each other, which helps the whole herd quickly adapt to VF (Keshavarzi et al., 2020). If one cow receives an electrical pulse (EP) and moves away, others tend to follow, learning from that experience (Keshavarzi et al., 2020). Social learning can facilitate the herd's adaptation to VF. However, cows trained in groups might rely more on their herd mates' reactions than their own experiences with the stimuli, so it's important for all individuals to learn the association between the audio cue and the EP (Colusso et al., 2020). 

Individual Variation 

Despite their ability to learn, individual animals respond differently to VF (Campbell et al., 2017). Some cows may need more audio cues and electrical pulses (EPs) than others, indicating varying learning rates within a herd (Campbell et al., 2017; McSweeney et al., 2020). Cows with a longer habituation period to electric fences are quicker to associate a visual cue with aversive electrical stimuli than those learning from audio cues alone (McDonald et al., 1981; Uetake & Kudo, 1994). Higher stocking densities can increase animal interaction with the VF (Marini et al., 2018). Lactating dairy cows with higher nutritional needs might be more motivated to cross a virtual fence than non-lactating heifers (Lomax et al., 2019). Additionally, daily changes in the paddock and virtual front-fence locations can initially increase the number of audio cues that cows receive (Langworthy et al., 2021).

Stress and Welfare 

A significant concern surrounding VF is its effect on animal welfare. While acute stress is expected during the initial learning phase, this should be minimal once animals learn to avoid the electrical stimulus (EP) (Lee et al., 2018). Once animals associate the audio cue with the EP and can avoid it, predictability and controllability are restored to their environment (Lee et al., 2018). Research has produced positive results, indicating that VF does not negatively affect animal behavior or stress levels compared to traditional electric fencing for beef and dairy cattle (Wilms et al., 2024). Measures such as pasture utilization and herbage consumption are similar between electric and VF in sheep (Marini et al., 2022). Similarly, in dairy cows, milk production, cortisol concentrations, rumination and grazing time, activity behavior, and total energy intakes are comparable between electric and VF periods (Langworthy et al., 2021). In line with previous studies, Hamidi et al. (2022) found no significant differences in stress indicators like cortisol levels, heart rate, or lying and eating behaviors between growing beef heifers managed with virtual and traditional electric fencing. Eftang and Bøe (2019) found that when cattle received an EP, their heart rate increased immediately and then returned to normal within minutes, which can also occur independently of virtual fence situations. Behavioral measures, such as lying time and steps per hour, can be helpful to assess any changes.

Grazing

Research indicates that animals graze similarly in areas enclosed by virtual and electric fences, suggesting that virtual fences do not discourage grazing up to the fence line (Marini et al., 2022). Virtual fencing can support natural grazing behavior, leading to more equitable pasture intake, especially in large herds, without increasing stocking rates (Langworthy et al., 2021). However, dairy cows in intensive grazing systems may experience some short-term disruptions in behavior when using VF. One study showed a decrease in grazing time and activity, along with an increase in rumination time, during the initial days (4-6) of using a virtual fence compared to an electric fence, indicating a potential disruption in their behavioral time budgets (Langworthy et al., 2021). This was accompanied by higher milk cortisol concentrations on day 5 with a virtual fence, although this finding was not statistically significant after a conservative statistical adjustment (Langworthy et al., 2021).

These behavioral changes in dairy cows may also be related to skin abrasions observed on their jaws, which were not seen in beef cattle studies (Langworthy et al., 2021). The skin abrasions could contribute to the reduction in grazing time. It's important to note that the study with dairy cattle involved daily changes in the location of the virtual fence, which may reduce environmental predictability and increase stress (Langworthy et al., 2021). In contrast, a study with Angus steers in 6-ha paddocks with static virtual fences showed a decrease in fecal cortisol metabolites over a 4-week period (Wilms et al., 2024).

 Nutrient Intake

Studies have indicated that virtual fences can effectively control grazing dairy cow movement even when pasture availability is limited (Colusso et al., 2021). This is because virtual fences do not discourage cows from grazing up to the fence line. As a result, cows can achieve similar total energy intakes, whether managed with virtual or electric fences (Colusso et al., 2021).

Milk Yield

Research indicates that VF does not negatively affect milk production or composition (Dias et al., 2019; Wilms et al., 2024). Studies have shown that milk production is similar for cows managed with virtual fences compared to those managed with traditional electric fences. Studies showed that milk yield did not differ between virtual and electric fence strip-grazing management systems (Langworthy et al., 2021; Verdon et al., 2021b; Colusso et al., 2021; Wilms et al., 2024). These studies demonstrate the application of VF technology in strip-grazing systems for dairy cattle. Strip-grazing involves moving the fence to provide a new strip of pasture, often daily, for efficient pasture use and is studied in combination with VF.

Milk Composition

Studies suggest that using VF does not impact milk composition, including fat and protein content (Wilms et al., 2024). This indicates that this technology does not compromise the nutritional quality of milk.

Areas that Need Further Research

More studies are needed to confirm that VF does not compromise animal welfare, regardless of cattle type and grazing system (Wilms et al., 2024). More investigation is also needed to assess the long-term effects on the welfare of dairy cattle, especially in intensive systems (Wilms et al., 2024).

Individual variation is an important factor, as some animals may be more sensitive to the VF system than others (Aaser et al., 2022). Some animals respond to social cues from herd mates rather than directly receiving stimuli themselves (Keshavarzi et al., 2020). 

Long-term studies are necessary to fully assess the effects of VF over extended periods on animal welfare. Additionally, most studies focus on simple grazing regimens, so more research is needed on how animals will adapt to more complex VF systems.

To reduce stress, it's important to introduce animals to VF through training programs. Animals should have the opportunity to associate the signals in a controlled environment before being exposed to more complex situations. This may include a stepwise implementation of the VF system.

Considerations for Farmers

Before adopting VF, farmers should consider the following points:

Initial costs: The initial investment in VF involves purchasing collars for your animals and a base station for communication. This cost can be substantial, with collar prices ranging from around $33 to $195 per unit (Umstatter, 2011). The cost of a base station or GPS tower can be around $10,000, and some systems may also require a software purchase (Goliński et al., 2022). 

While these upfront costs are significant, it’s important to compare them to traditional fencing methods. For instance, traditional fencing can cost around $9,500 per mile when considering labor. Virtual fencing systems eliminate the need for traditional fencing materials and the associated maintenance costs, which may offer savings over time (Goliński et al., 2022). For example, a beef cattle farm with 420 heads on 1,380 hectares (3410 ac) would pay approximately $104,020 to install 14.9 km (48884.5 ft) of internal fencing. In comparison, setting up a VF system would cost about $42,800, saving around $61,220 (Reichelt and Nettle, 2023).

For a dairy farm, a VF system for 290 cows could cost around $31,100, compared to the labor costs of moving electric fences daily, which would be around $10,654 per year (Langworthy et al., 2021). Therefore, although the initial costs can be high for VF, the long-term savings can be substantial for most farmers. However, the cost of VF can vary significantly based on the system, with a UK estimate of £200,000 for 100 animals (Waterhouse, 2023). This difference highlights that the cost-effectiveness of VF will depend on the scale of your operation, the number of animals, and the specific technology you choose (Goliński et al., 2022). More so, it is important to consider the potential for improved pasture usage and animal performance that could result from more flexible grazing management (Goliński et al., 2022).

Training: To ensure animals understand the audio and electric pulse signals, proper training is essential. A structured training schedule can help animals learn how the system works. For example, Verdon and Rawnsley (2020) recommend training dairy heifers on VF technology close to calving age rather than earlier in their development. Further, it is recommended to re-train livestock after an extended period without using VF technology. 

Technology: The system must have a reliable GPS signal and battery life to function effectively. However, a recent study suggested that the widespread adoption of VF may be limited by network connectivity issues, especially persistent network connection problems in rural areas (Harland et al., 2024). Additionally, environmental conditions, such as winter with short daylight hours and low temperatures, could reduce the overall effectiveness of VF collars (Harland et al., 2024). This highlights the importance of proper collar fitting and having contingency plans to manage lost and disconnected collars.

Maintenance: Ensure the VF collars are properly fitted, charged, and maintained. A study showed that fitting VF collars to bulls proved difficult due to their neck circumference, which often exceeded the size of their heads, allowing the collars to slip off more easily (Harland et al., 2024). However, collar failures were addressed through manual resets and collar retrieval and redeployment.

Animal welfare: Continuously monitor animals for any signs of distress or discomfort.

Long-term impact: Research on the long-term effect of VF is still needed, particularly when more complicated grazing systems are used.

Conclusion

Virtual fencing is a promising technology that could revolutionize grazing management and help farmers balance both productivity and sustainability. Virtual fencing appears to have minimal negative impacts on animal welfare, as long as the animals learn the association between the audio and electrical stimuli. Many studies indicate that both beef and dairy cattle can adapt to VF without significant welfare issues, dairy cows in intensive grazing systems, particularly with frequent changes in virtual fence location, may exhibit short-term stress and behavioral changes.  The existing research suggests a need for more focused studies on lactating dairy cows with more complex grazing systems to thoroughly evaluate the long-term impact of VF on their welfare. As the technology advances and more research is conducted, VF has the potential to become a standard tool for livestock management, offering more sustainable practices for farmers.

References

• Aaser, M. F., Staahltoft, S. K., Korsgaard, A. H., Trige-Esbensen, A., Alstrup, A. K. O., Sonne, C., Pertoldi, C., Bruhn, D., Frikke, J., & Linder, A. C. (2022). Is virtual fencing an effective way of enclosing cattle? Personality, herd behaviour and welfare. Animals, 12(7), 842. https://doi.org/10.3390/ani12070842
• Campbell, D. L., Lea, J. M., Farrer, W. J., Haynes, S. J., & Lee, C. (2017). Tech-savvy beef cattle? How heifers respond to moving virtual fence lines. Animals, 7(9), 72.
• Campbell, D. L., Lea, J. M., Haynes, S. J., Farrer, W. J., Leigh-Lancaster, C. J., & Lee, C. (2018a). Virtual fencing of cattle using an automated collar in a feed attractant trial. Applied Animal Behaviour Science, 200, 71-77.
• Campbell, D. L., Haynes, S. J., Lea, J. M., Farrer, W. J., & Lee, C. (2018b). Temporary exclusion of cattle from a riparian zone using virtual fencing technology. Animals, 9(1), 5.
• Campbell, D. L., Lea, J. M., Keshavarzi, H., & Lee, C. (2019). Virtual fencing is comparable to electric tape fencing for cattle behavior and welfare. Frontiers in Veterinary Science, 6, 445.
• Colusso, P. I., Clark, C. E., & Lomax, S. (2020). Should dairy cattle be trained to a virtual fence system as individuals or in groups?. Animals, 10(10), 1767.
• Colusso, P. I., Clark, C. E., Ingram, L. J., Thomson, P. C., & Lomax, S. (2021). Dairy cattle response to a virtual fence when pasture on offer is restricted to the post-grazing residual. Frontiers in Animal Science, 2, 791228.
• Dias, K., Garcia, S., Islam, M., & Clark, C. (2019). Milk yield, milk composition, and the nutritive value of feed accessed varies with milking order for pasture-based dairy cattle. Animals, 9(2), 60.
• Eftang, S., & Bøe, K. E. (2019). Virtual fences on cattle–studies on heifers (2018). Report: 33pp.
• Goliński, P., Sobolewska, P., Stefańska, B., & Golińska, B. (2022). Virtual fencing technology for cattle management in the pasture feeding system—a review. Agriculture, 13(1), 91.
• Hamidi, D., Grinnell, N. A., Komainda, M., Riesch, F., Horn, J., Ammer, S., ... & Isselstein, J. (2022). Heifers don't care: no evidence of negative impact on animal welfare of growing heifers when using virtual fences compared to physical fences for grazing. Animal, 16(9), 100614.
• Harland, A. J., Novais, F. J., Durunna, O. N., Fitzsimmons, C. J., Church, J. S., & Bork, E. W. (2024). Evaluation of the technical performance of the Nofence virtual fencing system in Alberta, Canada. Smart Agricultural Technology, 10, 100713. https://doi.org/10.1016/j.atech.2024.100713
• Keshavarzi, H., Lee, C., Lea, J. M., & Campbell, D. L. (2020). Virtual fence responses are socially facilitated in beef cattle. Frontiers in Veterinary Science, 7, 543158.
• Langworthy, A. D., Verdon, M., Freeman, M. J., Corkrey, R., Hills, J. L., & Rawnsley, R. P. (2021). Virtual fencing technology to intensively graze lactating dairy cattle. I: Technology efficacy and pasture utilization. Journal of Dairy Science, 104(6), 7071-7083.
• Lee, C., Prayaga, K., Reed, M., & Henshall, J. (2007). Methods of training cattle to avoid a location using electrical cues. Applied Animal Behaviour Science, 108(3-4), 229-238.
• Lee, C., Henshall, J. M., Wark, T. J., Crossman, C. C., Reed, M. T., Brewer, H. G., ... & Fisher, A. D. (2009). Associative learning by cattle to enable effective and ethical virtual fences. Applied Animal Behaviour Science, 119(1-2), 15-22.
• Lee, C., Colditz, I. G., & Campbell, D. L. (2018). A framework to assess the impact of new animal management technologies on welfare: A case study of virtual fencing. Frontiers in Veterinary Science, 5, 187.
• Lomax, S., Colusso, P., & Clark, C. E. (2019). Does virtual fencing work for grazing dairy cattle?. Animals, 9(7), 429.
• Marini, D., Llewellyn, R., Belson, S., & Lee, C. (2018). Controlling within-field sheep movement using virtual fencing. Animals, 8(3), 31.
• Marini, D., Cowley, F., Belson, S., Lee, C., & Wilson, C. (2022). Comparison of virtually fencing and electrically fencing sheep for pasture management. Animal Production Science. 62(10–11), 1000–1005. https://doi.org/10.1071/AN21459
• McDonald, C. L., Beilharz, R. G., & McCutchan, J. C. (1981). Training cattle to control by electric fences. Applied Animal Ethology, 7(2), 113-121.
• McSweeney, D., O'Brien, B., Coughlan, N. E., Férard, A., Ivanov, S., Halton, P., & Umstatter, C. (2020). Virtual fencing without visual cues: Design, difficulties of implementation, and associated dairy cow behaviour. Computers and Electronics in Agriculture, 176, 105613.
• Murphy, E. S., & Lupfer, G. J. (2014). Basic principles of operant conditioning. The Wiley Blackwell handbook of operant and classical conditioning, 165-194.
• Reichelt, N., & Nettle, R. (2023). Practice insights for the responsible adoption of smart farming technologies using a participatory technology assessment approach: The case of virtual herding technology in Australia. Agricultural Systems, 206, 103592. https://doi.org/10.1016/j.agsy.2022.103592
• Staahltoft, S. K., Aaser, M. F., Jensen, J. N. S., Zadran, I., Sørensen, E. B., Nielsen, A. E., ... & Pertoldi, C. (2023). The effectiveness of virtual fencing of bull calves in a holistic grazing system. Animals, 13(5), 917.
• Umstatter, C. (2011). The evolution of virtual fences: a review. Computers and Electronics in Agriculture, 75(1), 10-22. https://doi.org/10.1016/j.compag.2010.10.005
• Uetake, K., & Kudo, Y. (1994). Visual dominance over hearing in feed acquisition procedure of cattle. Applied Animal Behaviour Science, 42(1), 1-9.
• Verdon, M., Horton, B., & Rawnsley, R. (2021a). A case study on the use of virtual fencing to intensively graze angus heifers using moving front and back-fences. Frontiers in Animal Science, 2, 663963.
• Verdon, M., Langworthy, A., & Rawnsley, R. (2021b). Virtual fencing technology to intensively graze lactating dairy cattle. II: Effects on cow welfare and behavior. Journal of dairy science, 104(6), 7084-7094.
• Verdon, M., & Rawnsley, R. (2020). The effects of dairy heifer age at training on rate of learning and retention of learning in a virtual fencing feed attractant trial. Frontiers in Animal Science, 1, 618070.
• Verdon, M., Hunt, I., & Rawnsley, R. (2024). The effectiveness of a virtual fencing technology to allocate pasture and herd cows to the milking shed. Journal of Dairy Science. https://doi.org/10.3168/jds.2023-24537
• Waterhouse, T. (2023). Virtual fencing systems: balancing production and welfare outcomes. Livestock, 28(5), 227-234. https://doi.org/10.12968/live.2023.28.5.227
• Wilms, L., Komainda, M., Hamidi, D., Riesch, F., Horn, J., & Isselstein, J. (2024). How do grazing beef and dairy cattle respond to virtual fences? A review. Journal of Animal Science, skae108. https://doi.org/10.1093/jas/skae108