The Functional Spatio-Temporal Statistical Model with Application to O₃ Pollution in Beijing, China

Area of Science:

• Environmental Science
• Statistics
• Atmospheric Chemistry

Background:

• Ground-level ozone (O3) is a significant air pollutant, exacerbated by industrialization and energy consumption.
• Air quality data exhibit strong spatial and temporal correlations, necessitating advanced analytical methods.
• Understanding ozone pollution formation and diurnal patterns is crucial for effective mitigation strategies.

Purpose of the Study:

• To propose and validate a functional spatio-temporal statistical model for analyzing air quality data, specifically ground-level ozone (O3).
• To investigate the influence of covariates, including other pollutants and meteorological variables, on ozone pollution formation.
• To explore the diurnal cycle of O3 pollution using functional data analysis.

Main Methods:

• Development of a functional spatio-temporal statistical model incorporating spatial and temporal correlations.
• Inclusion of covariate effects to elucidate ozone pollution formation mechanisms.
• Application of functional data analysis to capture the diurnal patterns of O3.

Main Results:

• The proposed spatio-temporal model demonstrates significant potential compared to existing models for analyzing air quality data.
• Covariate effects, such as other pollutants and meteorological factors, on O3 pollution were identified.
• The diurnal cycle of O3 pollution was successfully analyzed and discussed.

Conclusions:

• The functional spatio-temporal model provides a robust framework for analyzing complex air quality data, including ground-level ozone.
• The model effectively explains the impact of various factors on ozone pollution and its daily variations.
• This approach offers valuable insights for air quality management and pollution control strategies.