Analysis of high level ozone concentrations using nonparametric methods.

Area of Science:

• Environmental Science
• Atmospheric Chemistry
• Statistics

Background:

• Ozone (O3) is a secondary air pollutant with significant health and environmental impacts.
• Controlling air pollution and setting air quality standards are critical governmental responsibilities.
• Classical extreme value theory (EVT) using parametric Generalized Extreme Value (GEV) distributions is common in air pollution studies.

Purpose of the Study:

• To propose and evaluate nonparametric methods for estimating extreme ozone values and trends.
• To compare the performance of nonparametric estimators against classical parametric approaches.
• To provide improved tools for air quality management and risk assessment.

Main Methods:

• Application of nonparametric statistical methods for extreme value analysis.
• Estimation of exceedance probabilities, quantiles, return levels, and mean return periods.
• Analysis of trends in very high ozone values using nonparametric estimators.

Main Results:

• Nonparametric methods provide satisfactory estimations for extreme ozone values.
• Nonparametric estimators generally outperform classical parametric GEV models.
• The methods were successfully applied to real-world ozone data from UK monitoring stations.

Conclusions:

• Nonparametric approaches offer a robust alternative to parametric methods for extreme ozone analysis.
• These methods can enhance the accuracy of air quality assessments and environmental protection strategies.
• The study demonstrates the practical utility of nonparametric statistics in environmental monitoring.