Periodic Reporting for period 2 - ClearFarm (Co-designed Welfare Monitoring Platform for Pig and Dairy Cattle)
Berichtszeitraum: 2021-04-01 bis 2022-09-30
Animal welfare has become a fundamental aspect of modern livestock production as citizens in Europe and elsewhere demand that animals are reared, transported and slaughtered as humanely as possible. Animal welfare is first and foremost an ethical issue derived from the fact that animals are sentient beings. Also, improving farm animal welfare has positive effects on human health. Important as it is, however, animal welfare should not be seen as an stand-alone issue, but in relationship with other societal concerns, including environmental issues.
If farm animal welfare is to make a significant progress, we need to have scientifically validated systems to assess the welfare of animals. These assessment systems shall provide reliable information to consumers so that they can make informed choices when buying animal foods and, ideally, information on animal welfare and on the environmental impact of farming should be combined.
In this context, the ClearFarn project aims to develop and validate a system to assess the welfare of pigs and dairy cattle using information that can be automatically gathered on the production farms and with a particular focus on welfare issues that have an impact on the environment.
We aim to validate, under controlled conditions, the use of existing sensors to identify the welfare status of pigs (welfare indicators). DOL supplies the instrumentation and the data infrastructure.
If farm animal welfare is to make a significant progress, we need to have scientifically validated systems to assess the welfare of animals. These assessment systems shall provide reliable information to consumers so that they can make informed choices when buying animal foods and, ideally, information on animal welfare and on the environmental impact of farming should be combined.
In this context, the ClearFarn project aims to develop and validate a system to assess the welfare of pigs and dairy cattle using information that can be automatically gathered on the production farms and with a particular focus on welfare issues that have an impact on the environment.
We aim to validate, under controlled conditions, the use of existing sensors to identify the welfare status of pigs (welfare indicators). DOL supplies the instrumentation and the data infrastructure.
So far, we have gathered detailed information on consumers' and other stakeholders views on the potential and perceived risks of using automatically recorded information to assess the welfare of farm animals. Our results indicate that although stakeholders have a generally positive view of such an approach, they also perceive some risks, including the fear that the integration of automatically recorded data will introduce more industrialisation into livestock farming production; the concern that such data are vulnerable to misuse and cyber-crime, and that it is not communicated adequately to allow informed purchase decisions. These research findings provide directions for members of the animal-based food value chain to make informed decisions to improve their sustainability, social responsibility and credibility by endorsing the acceptance automatically recorded data to assess farm animal welfare amongst European consumers.
We have also reviewed the existing validated and commercially available technologies that could be used for sensor-based welfare assessment in dairy cattle and pigs. Our findings show that existing technologies are potential tools for on-farm animal welfare assessment in pig and dairy cattle production. However, validation studies are lacking for an important percentage of market available tools. In particular research and development need to focus on identifying technologies to assess the animals' affective states or emotions, both positive and negative.
We performed twelve trials on pig (6) and dairy (6) farms during one and a half years to validate several automatic recording systems that hopefully will provide valid information on the behaviour and affective state of farm animals taking into account consumers' expectations. In pigs trials, work has included automatic feeding stations that record feeding behaviour of pigs, automatic recording of play behaviour as a positive welfare indicator and automatic recording of pig vocalizations. Data from sensors has been compared with behavioural recordings and environmental data, as well as with performance and biochemical parameters. In dairy cattle automatic recordings of cows’ activity have been compared with the results of in situ welfare assessment as well as with performance and biochemical parameters. We collected data on the environmental impact of pig and dairy farming using life cycle analysis and bioeconomic models that are being developed within the project.
The consortium started to work on the development of the ClearFarm platform based on data provided by sensors installed in pig and dairy farms.
Finally, the focus has been on the investigation to understand the business opportunities for the exploitation of ClearFarm technologies through mapping the market and identifying the options offered by the most promising business models.
DOL has in coordination with AU, Wageningen and Luke focused on establishing an environment to validates pigs play behavior and curly tails as positive welfare indicators. DOL supplies the instrumentation and the data infrastructure to observe and systemize these welfare indicators.
DOL supplies video cameras including IR video to observe and store behavior and movement patterns. Each pig is identified using RFID tag technology. This equipment also allows measuring daily weight (growth) at the individual level. DOL also installs NH3-sensors and CO2-sensors and measures the temperature and humidity to determine the air quality and thermal conditions at pig height.
Other partners supply equipment to observe the number of visits and duration at feed troughs and water trough to estimate the feed and water intake and methods to obtain physiological stress indicators from each pig by collecting and analyzing salivary.
In WP 2 behavior, movement patterns and stress indicators are sampled manually while the remaining welfare parameters are sampled automatically. DOL supplies and open data structure allowing all data collected from both DOL supplied equipment and equipment from other partners to be correlated on time and basis and per pig capita. Certain production related events during the lifetime of the pigs which are expected to influence their well-being can be given as timestamped input data. These events could be when the piglets are weaned or when the pigs are moved or regrouped.
The instrumentation setup and the data infrastructure allow validation of welfare indicators. The validation process maps observed differences as correlations between positive welfare indicators - play behavior and curly tails - and negative indicators - poor growth, low feed and/or water consumption, high stress, poor air quality and thermal environment.
We have also reviewed the existing validated and commercially available technologies that could be used for sensor-based welfare assessment in dairy cattle and pigs. Our findings show that existing technologies are potential tools for on-farm animal welfare assessment in pig and dairy cattle production. However, validation studies are lacking for an important percentage of market available tools. In particular research and development need to focus on identifying technologies to assess the animals' affective states or emotions, both positive and negative.
We performed twelve trials on pig (6) and dairy (6) farms during one and a half years to validate several automatic recording systems that hopefully will provide valid information on the behaviour and affective state of farm animals taking into account consumers' expectations. In pigs trials, work has included automatic feeding stations that record feeding behaviour of pigs, automatic recording of play behaviour as a positive welfare indicator and automatic recording of pig vocalizations. Data from sensors has been compared with behavioural recordings and environmental data, as well as with performance and biochemical parameters. In dairy cattle automatic recordings of cows’ activity have been compared with the results of in situ welfare assessment as well as with performance and biochemical parameters. We collected data on the environmental impact of pig and dairy farming using life cycle analysis and bioeconomic models that are being developed within the project.
The consortium started to work on the development of the ClearFarm platform based on data provided by sensors installed in pig and dairy farms.
Finally, the focus has been on the investigation to understand the business opportunities for the exploitation of ClearFarm technologies through mapping the market and identifying the options offered by the most promising business models.
DOL has in coordination with AU, Wageningen and Luke focused on establishing an environment to validates pigs play behavior and curly tails as positive welfare indicators. DOL supplies the instrumentation and the data infrastructure to observe and systemize these welfare indicators.
DOL supplies video cameras including IR video to observe and store behavior and movement patterns. Each pig is identified using RFID tag technology. This equipment also allows measuring daily weight (growth) at the individual level. DOL also installs NH3-sensors and CO2-sensors and measures the temperature and humidity to determine the air quality and thermal conditions at pig height.
Other partners supply equipment to observe the number of visits and duration at feed troughs and water trough to estimate the feed and water intake and methods to obtain physiological stress indicators from each pig by collecting and analyzing salivary.
In WP 2 behavior, movement patterns and stress indicators are sampled manually while the remaining welfare parameters are sampled automatically. DOL supplies and open data structure allowing all data collected from both DOL supplied equipment and equipment from other partners to be correlated on time and basis and per pig capita. Certain production related events during the lifetime of the pigs which are expected to influence their well-being can be given as timestamped input data. These events could be when the piglets are weaned or when the pigs are moved or regrouped.
The instrumentation setup and the data infrastructure allow validation of welfare indicators. The validation process maps observed differences as correlations between positive welfare indicators - play behavior and curly tails - and negative indicators - poor growth, low feed and/or water consumption, high stress, poor air quality and thermal environment.
The results of our on-farm trials suggest that animal welfare can be monitored using PLF data from sensors in pig and dairy cattle farms. The information from these technologies, if apropriately integrated, bring an unprecedented level of understanding of the status of farm animals at an individual (cows) and group (pigs) level. The platform that will arise from ClearFarm is expected to assess in a feasible and valid way the welfare of pigs and dairy cattle using automatically recorded information. This, in turn, will be a valuable tool for European consumers and will contribute to the sustainability of the European livestock industry.
An important future result would be automatic characterization of play behavior, tail pose and patterns of movement in terms of current degree of well-being.
Correlation between video collected data and more basic sensor inputs should enable determination of welfare index from pig production sites with simpler instrumentation.
An important future result would be automatic characterization of play behavior, tail pose and patterns of movement in terms of current degree of well-being.
Correlation between video collected data and more basic sensor inputs should enable determination of welfare index from pig production sites with simpler instrumentation.