Eddie Bokkers (WAGENINGEN UNIVERSITY & RESEARCH)
Dr. Eddie Bokkers is an associate professor at the Animal Production Systems group of Wageningen University & Research (WUR). Specialised in animal welfare and sustainable development of animal production systems, he leads work package 2 of the ClearFarm project, that aims to test and validate data from sensor technology as welfare indicators in the pig production chain.
With ClearFarm pilot tests in farms now ongoing, we asked him about the development of trials during these first months.
Could you please explain how technology can help to make livestock farming more efficient and improve animal welfare?
Sensor technology has the potential to assess and monitor different indicators associated with both negative and positive aspects of animal welfare, longitudinal and often at the individual level. Early detection of (potential) welfare issues can prevent these to become severe issues when accurate measures are taken. Preventing severe welfare issues can also prevent production losses which are relevant from an environmental and economic perspective.
So technology can help to improve the management and monitoring of animals, however, a risk is to rely too much on technology. A situation should be avoided where ‘computer says yes or no’ is the only way to manage a livestock production system and to monitor animal welfare.
Which information on animal welfare are you collecting on farms?
On the farms involved in this project we are collecting animal-based data, e.g. behaviour, health, physiology, as the main source of information to assess animal welfare. In addition resource and management data are collected.
Which technologies are you using for data collection?
We collect data via automated feeding stations for sows and fattening pigs, video recordings (using specific software to analyse behaviour), climate sensors, sound sensors. However we are not only using technology to collect data. We check the health of the animals manually on a regular base, and conduct behavioural observations from video recordings to validate automatically recorded data.
How can farmers take advantage of it?
Although in pig production the application of sensor technology is relatively new and often not focused on the individual animals, recent developments create increasing opportunities for farmers to make use of this. Our research will hopefully contribute to this development since it can help farmers in their farm management. It should give them the opportunity to earlier take action at the individual or group level to prevent welfare problems.
How many farms are participating in the pilot test? Are you considering different systems, production stages and geographical divergences?
Three countries are involved in the trials. In Denmark the experimental facilities of Aarhus University are used. They study mainly the weaning phase and focus especially on the positive aspects of animal welfare via play behaviour. In Spain three conventional farms are involved with weaners and growing-fattening pigs and in the Netherlands we study growing-fattening pigs (1 conventional farm) and sows (two farms, one organic and one conventional).
How farmers are reacting to large volumes of data that you are generating?
Most farmers are interested in more information to better manage their farm. Generally farmers don’t receive just a lot of data, the data is already processed in a way that it results in useful, practical information for the farmers.
Can you explain how do you work with the technological partners of the project to develop the ClearFarm platform with the methodology and algorithms developed to assess the welfare of animals?
We are currently discussing with our technological partners what kind of data flows we have, how to process them and what kind of formats should we use. The developments of algorithms has still to be started.
Which are the next steps?
At present data is collected with different sensor systems and manually and a start has been made with processing and analysing data. This should result in the first output of our work package. Thereafter we are going to prepare for the next experiments.