Graf Roland
Analysing flight behaviour of roe deer (Capreolus capreolus) using GPS and accelerometer data
Project Number: CH-6142
| Project Type: |
Master |
| Project Duration: |
01/01/2016 - 12/16/2016 project completed |
| Funding Source: |
other , |
| Leading Institution: |
ZHAW Wädenswil |
| Project Leader: |
Dr. Roland Graf
Institut Umwelt & Natürliche Ressourcen (IUNR)
ZHAW
Grüntal
8820 Wädenswil
Phone: ; +41 (0) 58 934 55 78
FAX: +41 (0) 58 934 59 11
e-Mail: roland.graf(at)zhaw.ch
http://www.unr.ch/ |
Research Areas:
Disciplines:
Keywords:
roe deer, Capreolus capreolus, flight behaviour, GPS
Abstract:
Human-caused perturbations on wildlife are increasing in multiple ways and trigger natural anti- predatory behaviours among prey. Analysing these behavioural responses can help understand how individual animals perceive and respond to human disturbance. This knowledge can help evaluate the effects of human activities on specific wildlife and provide the basis for management strategies.
Within the scope of the Roe Deer Project Sihlwald, fifteen animals in habitats of different anthropogenic usage intensities were equipped with gps-gsm-collars and observed over a period of 216-697 (562±194) days, totalling 7’869 days for all animals. In addition to the position data in a three hour interval, biaxial accelerometer (acc) data was sampled every four minutes over the whole observation period for fourteen of the fifteen animals. Furthermore, so called ‘reaction experiments’ were conducted in the vicinity of known roe deer locations: Participants exercising different activities (hunting, mountain biking, tracking and orienteering) in immediate proximity of the roe deer were likewise equipped with gps-sensors. During these experiments gps-sampling interval was reduced to 5 minutes in order to enable movement analysis on a finer spatiotemporal scale.
Within this study, fine grained position data of both humans and animals were used to develop and implement a method to identify human-animal encounters. The spatiotemporal patterns of the flight trajectories subsequent to these encounters were analysed: Over all experiment types, 32 % of the encounters lead to immediate flight reactions. Flight trajectories were shortest during bike experiments (median 149 m) and highest during sport events (median 634 m).
An additional goal was to identify distinctive patterns within the acc-data to enable the detection of flight reactions outside the reaction experiments, where information both on human activity and detailed animal movement is lacking. Once the flight reactions can be inferred from the long-term acc-dataset, spatial and temporal patterns of flight reactions are analysed.
This goal was tightly constrained by the resolution and the accuracy of the available data. None the less, a distinct pattern in the acc data was apparent during flight reactions in the form of a threshold value. Using this value, flight reactions were inferred on the long-term acc-dataset. The hereby hypothesized flight reactions occur predominantly during dusk and dawn in summer as well as in winter. Notably, the findings indicate high levels of disturbances during late hours in remote areas.
Publications:
Ratnaweera, N. (2016): Analysing flight behaviour of roe deer (Capreolus capreolus) using GPS and accelerometer data. Masterthesis, Department Life Sciences and Facility Management, Zurich University of Applied Sciences (ZHAW), Wädenswil.
Last update: 6/8/20
Source of data: ProClim- Research InfoSystem (1993-2024)
Update the data of project: CH-6142
Go Back