Tracking random walks.

Area of Science:

• Movement ecology
• Statistical physics
• Human mobility studies

Background:

• Empirical studies often record animal or individual trajectories in space and time.
• The impact of sampling procedures on recorded data bias remains unclear.

Purpose of the Study:

• To investigate how sampling methods affect the statistical properties of movement data.
• To analyze the bias introduced by different sampling intervals and duration distributions in movement trajectories.

Main Methods:

• Analytical calculations for ideal cases with constant sampling intervals and short-tailed duration distributions.
• Computer simulations incorporating realistic long-tailed rest duration distributions.
• Empirical validation using high-resolution GPS human mobility trajectories and communication data.

Main Results:

• Constant sampling intervals can recover at most 18% of human movements, inflating average trip length by a factor of two.
• Using sampling intervals from communication data recovers only 11% of moves, with a maximum of 16% even with optimal algorithms.
• Long-tailed rest duration distributions dramatically reduce the fraction of correctly sampled trajectories.

Conclusions:

• Current sampling procedures introduce substantial bias in movement data analysis.
• A more critical approach to data collection and interpretation is necessary for quantitative studies of individual movements.
• Findings highlight the need for developing advanced sampling strategies to accurately capture movement patterns.