Shift towards recognizing the importance of social network and happy animals

Technology for identifying deviation in behaviour patterns to detect animals that need attention have been used and validated for many years. These technologies are increasingly also being used as tools for welfare assessments but still with focus on lack of welfare and not positive welfare. There is thus a gap between what can be seen as good welfare and what is currently measured by technologies.

During my lecture I will guide you through my voyage from using simple technological solutions for collecting animal based measures such as a simple calf milk feeder or an of the shelf accelerometer, towards using state of the art technology namely video or UWB. These technologies are facilitating a shift of focus towards assessing the emotional state of an animal with the emphasis on developing management system and routines that will increase the general happiness for animals. Furthermore, I will give you insight into the new world of grazing without physical fences.

Groups of animals are structured by dominance/subordination relationships and preferential relationships. Preferential relationships are observed, where animals stay close to each other, synchronise their activities (e.g. eating, walking, resting), and exchange cohesive interactions such as sniffing and allo-grooming. However, very little is known about social networks and how important it is for an animal’s welfare. Some of the highlights from examining social networks in dairy cattle is that we now know that some cows are more introvert than others, while others have many connections and have a more extrovert personality. What we are examining now is if these connections are stable and remain stable between lactations. Using a combination of video surveillance, machine learning and basic ethology we will soon also be able to determine if these social interactions can have both a positive and negative effect on animals and eventually be able to perform a qualitative behaviour assessment using video image analysis.

Virtual fencing is a relatively new method for keeping grazing animals confined to a specific area using GPS for positioning and sound signals followed by an electrical pulse as means for training and keeping the animals within the virtual border. This technology is not yet allowed in Sweden, but together with a college at RISE we have during this year grazing season conducted several studies with the aim of examining the potentially use of virtual fencing for cattle and sheep. Our focus has been on collecting data on how quickly the animals learn to use the system and how this can be used for managing animals and grazing areas and for protecting ecological sensitive habitats. During this season we also managed to examine if there is a difference in the learning process if naïve animals are trained together with animals that already know how to use the system.

Based on my experience working together with engineers, programmers and industrial partners, developing and validating technologies for assessing animal based measures combining technical expertise and ethology, my plan for the future is to continue being a bridge between these different professions and thereby enabling more technical development within animal production and animal welfare for the good of farmers and animals.