Mixed POT-BM Approach for Modeling Unhealthy Air Pollution Events

Area of Science:

• Environmental Science
• Atmospheric Science
• Statistics

Background:

• Unhealthy air pollution events pose significant environmental and health risks.
• Accurate modeling of extreme air pollution events is crucial for effective risk management.
• Existing methods may struggle with data dependency and precision in extreme value analysis.

Purpose of the Study:

• To introduce and evaluate a novel data selection technique, the mixed peak-over-threshold-block-maxima (POT-BM) approach.
• To improve the statistical modeling of extreme air pollution events.
• To address data dependency issues in extreme value analysis.

Main Methods:

• The study employed the peak-over-threshold (POT) technique to identify extreme data points exceeding a defined threshold (u).
• A declustering technique was applied to mitigate dependency issues among POT blocks.
• The block maxima (BM) concept was integrated to select maximum data points within each POT block.

Main Results:

• The mixed POT-BM approach successfully identified extreme data points that satisfy independence properties.
• The fitted models demonstrated satisfactory precision in representing extreme air pollution events.
• The method achieved a balanced trade-off between bias and variance in statistical modeling.

Conclusions:

• The mixed POT-BM approach is a robust technique for modeling extreme air pollution events.
• This method enhances the reliability and precision of extreme value statistical analysis.
• The case study in Klang, Malaysia, validates the approach for real-world air quality monitoring.