Every breath you take, I‘ll be watching you: Automated measurement of breath rate from mobile phone videos as a severity assessment parameter in wild great tits
Grantee: Caroline Deimel
Institutions: Max Planck Institute, Germany
Grant amount: $19,200
Grant type: Small grants
Focal species: Great tits (Parus major)
Conservation status: Least concern
Disciplines: Animal behavior, animal welfare science, physiology
Research location: Germany
Project summary
Breath rate (BR) is increasingly used as a non-invasive proxy of stress that is fast, cheap, and field-friendly. However, BR has not been evaluated in a bird welfare context, and it is unclear how it relates to established physiological proxies of stress, like glucocorticoid measurements in blood. Also, the currently used protocols to measure BR lack standardization, scalability, and validation. This project aims to provide a validated, non-invasive reference tool for standardizing BR measurements following capture and physical sampling protocols in wild birds by implementing computer vision capabilities and other improvements in software the researchers have recently developed to estimate BR metrics from mobile phone videos. This will allow the research community to gather objective, reproducible, and comparable data, and provide institutions tasked to oversee animal welfare with objective and feasible monitoring requirements. The project will also evaluate BR as a welfare indicator in great tits (Parus major), a songbird extensively used in wild animal research across Europe, by analyzing an existing, five-year dataset to test whether BR corresponds to simultaneous glucocorticoid measurements from free-living great tits. These data will provide reliable baselines and ranges for glucocorticoids and BR, and benchmarks for video recording lengths.
Why we funded this project
We are generally interested in developing non-invasive ways of measuring indicators of wild animal welfare. One of the key advantages to non-invasive measurement, besides the obvious of not causing unnecessary fear or pain to animals, is that stress induced by the measurement process can obscure the animal’s baseline stress levels if the method is too invasive or not carried out properly. We also tentatively believe that instilling a norm of minimizing animal harm within welfare biology research will increase the likelihood that researchers act as scientist-advocates for implementation of wild animal welfare interventions. This project’s focus on making breath rate easy and inexpensive to measure in a consistent way also fits well with our desire for a greater volume and accessibility of welfare data collection.