Human population growth and intensified livestock production drive environmental degradation, biodiversity loss, and animal welfare concerns.
In Norway, free-ranging livestock traditions combined with virtual fencing technology (GPS collars) presented Erik Versluijs with an opportunity to study cattle behaviour in the boreal forest.
"I studied how environmental, climatic, and temporal drivers influenced cattle behaviour using virtual fence collars equipped with GPS and accelerometers", he tells inn.no.
In addition, he explored how virtual fence techniques, behavioural classification, and the effects of drivers on behavioural states can contribute to the study of animal welfare.
"I assessed the performance of virtual fence collars in remote landscapes, revealing that GPS performance in rugged terrain is negatively affected. Using high-resolution accelerometer data, a behavioural classification model was developed to identify 10 common behaviours, which were used to explore the relationships between cattle behaviours and various drivers."
Results showed daylight was favoured for foraging and social behaviours, while resting dominated at night. Higher temperature increased foraging under dense canopy cover.
Versluijs explains that his work demonstrates the potential of accelerometer data to provide detailed insights into cattle behaviour.
"Integrating GPS and accelerometer technologies offers practical frameworks to improve animal welfare, monitor herds, and optimise farming practices in shared landscapes", he says.
Erik Versluijs has completed his research in the PhD programme Applied Ecology and Biotechnology. His thesis is titled "Free-Ranging Cattle in the Boreal Forest: Predicting behaviours and exploring behavioural drivers with virtual fencing technology".
The main findings of the doctoral thesis:
- Virtual fence collars in remote areas do have slightly lower accuracy (on average 11 meters). However, smart fence design can compensate for this by avoiding rugged areas as fence areas.
- A minimal size of five hectares is recommended to avoid fence related accuracy issues in the boreal forest.
- Up to ten most common behaviours can be predicted using high resolution acceleration data (10hz).
- Free-ranging cattle increases the probability of foraging under canopy cover with higher temperature, likely to avoid heat stress.
- Free-ranging cattle has the highest probability of resting during the night and are most likely to forage around sunset and sunrise.
- The methods used in this dissertation gives the possibility to monitor and study behaviours, and animal welfare at a finer scale.