Abstract:

Aims: For elusive and difficult to observe wild cats, camera trap and satellite tracking collar are widely used sampling methods to determine daily activity patterns. While satellite tracking collars can continuously monitor a limited number of captured individuals, camera traps have the potential to monitor all population members in a given area. The detection probability of camera traps for wild cat activity patterns is likely influenced by the selected spatial placement of the cameras. The activity patterns are constructed based on the binary classification of target animals’ active status (i.e., active or inactive). Although the different methods of monitoring wild cats may lead to different activity patterns, few empirical studies have examined the difference between the patterns. The Chinese mountain cat (Felis bieti), a small felid endemic to China, is distributed in the eastern and northern edge of the Qinghai-Tibet Plateau at low density. The Chinese mountain cat was uplisted to Class-I National Key Protected Wildlife of China in 2021, is listed as Vulnerable (VU) by the IUCN Red List, and is considered Critically Endangered (CR) in China’s Red List. However, the existing surveys and data are insufficient to carry out conservation of Chinese mountain cat. Therefore, urgent field investigations about the ecological activities of the Chinese mountain cat are needed.
Methods: From June 2020 to December 2021, we investigated activity patterns of Chinese mountain cats in Qilian Mountains of Qinghai Province using camera traps and satellite-tracking collars. We compared daily activity patterns derived from camera traps placed at den sites (2 sites, 173 independent detections) and animal trails (23 sites, 423 independent detections), to satellite-tracking collars fitted on Chinese mountain cats (10 individuals, 62,942 fixed locations).
Results: Results show Chinese mountain cats were active all day long, with peak activity from 17:00 to 19:00. A daily activity curve produced by camera traps showed a high level of overlap with results produced by the activity sensor implanted in the collars; the coefficient of overlap is 0.89. However, the camera traps showed significantly higher levels of activity from 16:00 to 21:00. Detection of activity from 16:00 to 21:00 was primarily concentrated at a small range of den sites. Concurrently, activity records of the collars were relatively lower. Based on the ethogram of Chinese mountain cat, activity patterns recorded from the camera traps placed at den sites predominantly represents social behaviors related to reproduction and nursing while activity records from the trail camera traps generally represent movement and foraging. The satellite-tracking collar was limited to distinguishing just the movement and still of an individually tracked animal.
Conclusions: Our results indicate considerable differences between the activity patterns of wild cats obtained via camera-trapping and a satellite collar tracking. The binary classification of “active” or “inactive” used to construct activity patterns reflect different behaviors in data derived from different methods. Caution should be taken for researchers when comparing the activity patterns derived from different studies. Proper and accurate interpretation of data should be made on the basis of in-depth and comprehensive understanding of the target animal’s behaviors.

Key words: animal behavior, daily activity pattern, small cats, Chinese mountain cat, camera trap, satellite tracking collar