movetrack

movetrack reconstructs animal flight paths from automated radio-telemetry data using a hidden Markov model framework that combines coarse geometric position estimates with movement dynamics to distinguish behavioral states.

Key Features:

• Hidden Markov Model Framework: Reconstructs animal trajectories by accounting for measurement error, temporal gaps, and movement dynamics.
• Geometric Position Estimation: Generates coarse position estimates based on antenna bearing and signal strength from radio-telemetry receivers.
• Behavioral State Classification: Distinguishes behavioral states including migratory flight, local movement, and stopovers by analyzing directional persistence and speed.
• Uncertainty Quantification: Produces flight path estimates with quantifiable uncertainty.
• Motus Wildlife Tracking System Compatibility: Processes data collected by the Motus Wildlife Tracking System and similar automated radio-telemetry networks.

Scientific Applications:

• Aerial Animal Movement Ecology: Enables study of small animal migration for species such as songbirds, bats, and insects that cannot carry GPS transmitters.
• Environmental Change Response: Supports research on how aerial animals adjust and adapt migration in response to environmental and human-induced changes.
• Anthropogenic Impact Assessment: Facilitates formulation of conservation measures related to conflicts with artificial light at night, pesticides, and human-made structures.

Methodology:

movetrack uses a hidden Markov model that combines coarse geometric position estimates derived from antenna bearing and signal strength with movement dynamics. The model analyzes directional persistence and speed to classify behavioral states. Validation was performed using controlled low-altitude aircraft flights simultaneously recorded with GPS and radio-telemetry to simulate migratory flights of aerial species.