A supervised model to identify wolf behavior from tri-axial acceleration
Abstract Background In wildlife studies, animal behavior serves as a key indicator of the impact of environmental changes and anthropogenic disturbances. However, wild animals are elusive and traditional GPS studies only provide limited insight into their daily activities. To address this issue, beh...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-01-01
|
| Series: | Animal Biotelemetry |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s40317-025-00400-w |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571838017830912 |
|---|---|
| author | Charlotte Lorand Léa Bouet Olivier Devineau Marianna Chimienti Alina L. Evans Peggy Callahan Mark Beckel Timothy G. Laske Ane Eriksen |
| author_facet | Charlotte Lorand Léa Bouet Olivier Devineau Marianna Chimienti Alina L. Evans Peggy Callahan Mark Beckel Timothy G. Laske Ane Eriksen |
| author_sort | Charlotte Lorand |
| collection | DOAJ |
| description | Abstract Background In wildlife studies, animal behavior serves as a key indicator of the impact of environmental changes and anthropogenic disturbances. However, wild animals are elusive and traditional GPS studies only provide limited insight into their daily activities. To address this issue, behavior classification models have increasingly been used to detect specific behaviors in wildlife equipped with tri-axial accelerometers. Such models typically need to be trained on data from the target species. The present study focuses on developing a behavioral classification model tailored to the grey wolf (Canis lupus) and encompassing a variety of ecologically relevant behaviors. Methods We collected data from nine captive wolves equipped with collar-mounted tri-axial accelerometers recording continuous acceleration at 32 Hz (“fine-scale”) and averaged acceleration over 5-min intervals (“activity”). Using simultaneous video observations, we trained Random Forest models to classify wolf acceleration data into specific behaviors. We investigated the potential limits to the generalizability of these models to unlabeled data through individual-based cross-validation. Results We present: (1) a model classifying fine-scale acceleration data (32 Hz) into 12 distinct behaviors (lying, trotting, stationary, galloping, walking, chewing, sniffing, climbing, howling, shaking, digging and jumping) with a class recall of 0.77–0.99 (0.01–0.91 in cross-validation), (2) a model classifying activity data (5-min averages) into 3 behavior categories (static, locomotion and miscellaneous) with a class recall of 0.43–0.91 (0.39–0.92 in cross-validation). Although classification performance decreased following cross-validation, recall scores for lying, trotting, stationary, galloping, walking and chewing individual behaviors (as well as static and locomotion categories) remained above 0.6. Classification performance was consistently poorer for rare behaviors, which constituted less than 1.1% of the training dataset. Conclusions We demonstrate the use of collar-mounted accelerometer to distinguish between 12 behaviors and 3 behavior categories in captive wolves, at fine-scale (32 Hz) and averaged 5-min resolutions, respectively. We also discuss the generalizability of the two models to free-ranging settings. These models can be employed to support future behavioral studies examining questions such as conflict mitigation, wolf responses to human disturbances, or specific activity budgets. |
| format | Article |
| id | doaj-art-0bb6135dcfaf498e8e1c19362806d680 |
| institution | Kabale University |
| issn | 2050-3385 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | BMC |
| record_format | Article |
| series | Animal Biotelemetry |
| spelling | doaj-art-0bb6135dcfaf498e8e1c19362806d6802025-02-02T12:13:36ZengBMCAnimal Biotelemetry2050-33852025-01-0113111310.1186/s40317-025-00400-wA supervised model to identify wolf behavior from tri-axial accelerationCharlotte Lorand0Léa Bouet1Olivier Devineau2Marianna Chimienti3Alina L. Evans4Peggy Callahan5Mark Beckel6Timothy G. Laske7Ane Eriksen8Department of Forestry and Wildlife Management, University of Inland NorwayDepartment of Forestry and Wildlife Management, University of Inland NorwayDepartment of Forestry and Wildlife Management, University of Inland NorwaySchool of Ocean Sciences, Bangor UniversityDepartment of Forestry and Wildlife Management, University of Inland NorwayWildlife Science CenterWildlife Science CenterAtrial Fibrillation Solutions, Medtronic plcDepartment of Forestry and Wildlife Management, University of Inland NorwayAbstract Background In wildlife studies, animal behavior serves as a key indicator of the impact of environmental changes and anthropogenic disturbances. However, wild animals are elusive and traditional GPS studies only provide limited insight into their daily activities. To address this issue, behavior classification models have increasingly been used to detect specific behaviors in wildlife equipped with tri-axial accelerometers. Such models typically need to be trained on data from the target species. The present study focuses on developing a behavioral classification model tailored to the grey wolf (Canis lupus) and encompassing a variety of ecologically relevant behaviors. Methods We collected data from nine captive wolves equipped with collar-mounted tri-axial accelerometers recording continuous acceleration at 32 Hz (“fine-scale”) and averaged acceleration over 5-min intervals (“activity”). Using simultaneous video observations, we trained Random Forest models to classify wolf acceleration data into specific behaviors. We investigated the potential limits to the generalizability of these models to unlabeled data through individual-based cross-validation. Results We present: (1) a model classifying fine-scale acceleration data (32 Hz) into 12 distinct behaviors (lying, trotting, stationary, galloping, walking, chewing, sniffing, climbing, howling, shaking, digging and jumping) with a class recall of 0.77–0.99 (0.01–0.91 in cross-validation), (2) a model classifying activity data (5-min averages) into 3 behavior categories (static, locomotion and miscellaneous) with a class recall of 0.43–0.91 (0.39–0.92 in cross-validation). Although classification performance decreased following cross-validation, recall scores for lying, trotting, stationary, galloping, walking and chewing individual behaviors (as well as static and locomotion categories) remained above 0.6. Classification performance was consistently poorer for rare behaviors, which constituted less than 1.1% of the training dataset. Conclusions We demonstrate the use of collar-mounted accelerometer to distinguish between 12 behaviors and 3 behavior categories in captive wolves, at fine-scale (32 Hz) and averaged 5-min resolutions, respectively. We also discuss the generalizability of the two models to free-ranging settings. These models can be employed to support future behavioral studies examining questions such as conflict mitigation, wolf responses to human disturbances, or specific activity budgets.https://doi.org/10.1186/s40317-025-00400-wCanis lupusWolfRandom ForestClassificationMulti-classAcceleration |
| spellingShingle | Charlotte Lorand Léa Bouet Olivier Devineau Marianna Chimienti Alina L. Evans Peggy Callahan Mark Beckel Timothy G. Laske Ane Eriksen A supervised model to identify wolf behavior from tri-axial acceleration Animal Biotelemetry Canis lupus Wolf Random Forest Classification Multi-class Acceleration |
| title | A supervised model to identify wolf behavior from tri-axial acceleration |
| title_full | A supervised model to identify wolf behavior from tri-axial acceleration |
| title_fullStr | A supervised model to identify wolf behavior from tri-axial acceleration |
| title_full_unstemmed | A supervised model to identify wolf behavior from tri-axial acceleration |
| title_short | A supervised model to identify wolf behavior from tri-axial acceleration |
| title_sort | supervised model to identify wolf behavior from tri axial acceleration |
| topic | Canis lupus Wolf Random Forest Classification Multi-class Acceleration |
| url | https://doi.org/10.1186/s40317-025-00400-w |
| work_keys_str_mv | AT charlottelorand asupervisedmodeltoidentifywolfbehaviorfromtriaxialacceleration AT leabouet asupervisedmodeltoidentifywolfbehaviorfromtriaxialacceleration AT olivierdevineau asupervisedmodeltoidentifywolfbehaviorfromtriaxialacceleration AT mariannachimienti asupervisedmodeltoidentifywolfbehaviorfromtriaxialacceleration AT alinalevans asupervisedmodeltoidentifywolfbehaviorfromtriaxialacceleration AT peggycallahan asupervisedmodeltoidentifywolfbehaviorfromtriaxialacceleration AT markbeckel asupervisedmodeltoidentifywolfbehaviorfromtriaxialacceleration AT timothyglaske asupervisedmodeltoidentifywolfbehaviorfromtriaxialacceleration AT aneeriksen asupervisedmodeltoidentifywolfbehaviorfromtriaxialacceleration AT charlottelorand supervisedmodeltoidentifywolfbehaviorfromtriaxialacceleration AT leabouet supervisedmodeltoidentifywolfbehaviorfromtriaxialacceleration AT olivierdevineau supervisedmodeltoidentifywolfbehaviorfromtriaxialacceleration AT mariannachimienti supervisedmodeltoidentifywolfbehaviorfromtriaxialacceleration AT alinalevans supervisedmodeltoidentifywolfbehaviorfromtriaxialacceleration AT peggycallahan supervisedmodeltoidentifywolfbehaviorfromtriaxialacceleration AT markbeckel supervisedmodeltoidentifywolfbehaviorfromtriaxialacceleration AT timothyglaske supervisedmodeltoidentifywolfbehaviorfromtriaxialacceleration AT aneeriksen supervisedmodeltoidentifywolfbehaviorfromtriaxialacceleration |