Inferring symmetric and asymmetric interactions between animals and groups from positional data.

Area of Science:

• Ecology
• Animal Behavior
• Spatial Analysis

Background:

• Interactions between domestic and wild species are a growing global concern.
• Global Position Systems (GPS) generate extensive animal movement data, necessitating advanced analytical methods.
• Understanding species interactions is crucial for wildlife management and conservation.

Purpose of the Study:

• To assess various methods for inferring animal interactions and their symmetry.
• To evaluate the efficiency of different data transformations and analytical techniques.
• To provide tools for behavioral ecologists to analyze spatial data from various tracking technologies.

Main Methods:

• Utilized simulated and real-world GPS data from sheep and deer.
• Applied time series transformations (e.g., convex hull area, nearest neighbor distance).
• Employed analytical techniques including cross-correlation, Granger causality, and network analysis.

Main Results:

• Methods were more effective at inferring group-level interactions than individual-level interactions.
• Different data transformations yielded varied insights into interaction dynamics.
• Both sheep and deer exhibited consistent within-species grouping structures.
• No clear evidence of inter-species interaction via niche partitioning was found.

Conclusions:

• Complex interactions exist between sheep and deer, influenced by data transformation methods.
• The study offers guidelines for selecting appropriate transformations based on research goals.
• Developed tools can aid in inferring animal interactions and symmetry from positional data.