Stochastic and Self-Organisation Patterns in a 17-Year PM₁₀ Time Series in Athens, Greece.

Area of Science:

• Environmental Science
• Statistical Physics
• Complexity Science

Background:

• Particulate Matter (PM10) pollution poses significant environmental and health risks.
• Understanding the complex dynamics of PM10 time series is crucial for effective air quality management.
• Previous studies have explored PM10 trends, but a comprehensive analysis integrating entropy and fractal methods is lacking.

Purpose of the Study:

• To investigate stochastic and self-organization trends in a 17-year PM10 time series from Athens, Greece.
• To identify periods where PM10 behavior deviates from stochasticity and exhibits critical self-organized tendencies.
• To combine entropy analysis with fractal and Self-Organized Criticality (SOC) techniques for a novel assessment of air quality dynamics.

Main Methods:

• Utilized statistical methods (lumping, sliding windows) to analyze PM10 time series.
• Applied Boltzmann and Tsallis entropy measures with sliding and symbolic dynamics.
• Employed a published method to detect non-stochastic behavior with critical self-organization.
• Integrated 15 different fractal and SOC analysis techniques with entropy results.

Main Results:

• Identified decreasing trends in PM10 levels across all five stations between specific window intervals.
• Detected several periods exhibiting non-stochastic behavior and critical self-organized tendencies.
• Found nine common two-month windows across at least three stations indicating significant patterns.
• Highlighted two distinct areas exhibiting non-stochastic, fractal, long-memory, and self-organization patterns.
• Observed significantly lower block-entropy in these critical areas compared to other non-stochastic regions.

Conclusions:

• This study pioneers the use of entropy analysis for PM10 time series, combined with fractal methods.
• The findings reveal complex dynamics in PM10 pollution, including periods of critical self-organization.
• The integrated approach provides a more robust framework for understanding and potentially predicting air quality behavior.