Description
High labour requirements for fencing and animal monitoring appear to be general obstacles for the wider use of pastureland for grazing livestock. Virtual fencing (VF) enables a less laborious pasture management. Fence lines can be easily drawn and moved using GPS data. The advantages of VF for reducing the labour inherent to controlled grazing management are obvious. Potential additional animal monitoring opportunities arising from such a VF system that uses real-time GPS data have not yet been studied. Lying is seen as an indicator for assessing comfort or restlessness of cattle. Therefore, we focus on lying behavior in this evaluation. Based on data from conventional GPS collars, lying and standing often cannot be distinguished. The VF collars (® Nofence, AS, Batnfjordsøra Norway) used in this study detect low movement via an integrated accelerometer and then send the same GPS position during this time of low movement, in order to save battery life. We tested whether this battery life save function could be suitable for the detection of lying behavior. To address this question in a two-step-approach, we first compared observational data with IceTag pedometer (IceRobotics Ltd. Edinburgh, Scotland) data. In the second step, the pedometer data were compared to VF collar data via a confusion matrix. With 93% precision, 89% accuracy and 83% recall in this second step, the use of the VF collars can be recommended for a valid measure of lying behavior monitoring on pasture, which would be an added benefit to VF technology.
DOI
https://doi.org/10.13023/nxep-sy30
Citation
Hamidi, D.; Komainda, M.; Grinnell, N.; Reisch, F.; Horn, J.; Hamidi, M.; Traulsen, I.; and Isselstein, J., "More Than a Fencing System? Testing the Validity of Virtual Fencing Collars for Animal Monitoring on Pasture" (2024). IGC Proceedings (1993-2023). 40.
https://uknowledge.uky.edu/igc/XXV_IGC_2023/Sustainability/40
Included in
Agricultural Science Commons, Agronomy and Crop Sciences Commons, Plant Biology Commons, Plant Pathology Commons, Soil Science Commons, Weed Science Commons
More Than a Fencing System? Testing the Validity of Virtual Fencing Collars for Animal Monitoring on Pasture
High labour requirements for fencing and animal monitoring appear to be general obstacles for the wider use of pastureland for grazing livestock. Virtual fencing (VF) enables a less laborious pasture management. Fence lines can be easily drawn and moved using GPS data. The advantages of VF for reducing the labour inherent to controlled grazing management are obvious. Potential additional animal monitoring opportunities arising from such a VF system that uses real-time GPS data have not yet been studied. Lying is seen as an indicator for assessing comfort or restlessness of cattle. Therefore, we focus on lying behavior in this evaluation. Based on data from conventional GPS collars, lying and standing often cannot be distinguished. The VF collars (® Nofence, AS, Batnfjordsøra Norway) used in this study detect low movement via an integrated accelerometer and then send the same GPS position during this time of low movement, in order to save battery life. We tested whether this battery life save function could be suitable for the detection of lying behavior. To address this question in a two-step-approach, we first compared observational data with IceTag pedometer (IceRobotics Ltd. Edinburgh, Scotland) data. In the second step, the pedometer data were compared to VF collar data via a confusion matrix. With 93% precision, 89% accuracy and 83% recall in this second step, the use of the VF collars can be recommended for a valid measure of lying behavior monitoring on pasture, which would be an added benefit to VF technology.
