Spatial-Temporal Distribution Variation of Ground-Level Ozone in China's Pearl River Delta Metropolitan Region

Area of Science:

• Environmental Science
• Public Health
• Atmospheric Chemistry

Background:

• Ground-level ozone is a severe air pollutant in China's Pearl River Delta Metropolitan Region (PRD).
• Long-term exposure to ozone pollution threatens residents' physical and mental health.
• Understanding spatial-temporal ozone exposure patterns is crucial for public health management.

Purpose of the Study:

• To accurately reveal the spatial-temporal distribution characteristics of ozone pollution exposure in the PRD in 2015.
• To compare the performance of spatial-temporal kriging (STK) models and ordinary kriging (OK) for ozone concentration simulation.
• To identify the optimal model for predicting ozone exposure patterns.

Main Methods:

• Utilized daily maximum 8-h ozone concentration data from 55 air quality monitoring stations in the PRD for 2015.
• Employed six spatial-temporal kriging (STK) models, including the Bilonick model (BM), and ordinary kriging (OK) for spatial-temporal interpolation.
• Selected the best-performing model to analyze ozone exposure characteristics.

Main Results:

• The Bilonick model (BM) demonstrated the highest simulation precision among the six STK models.
• STK models significantly outperformed the OK model in simulating ozone concentrations.
• Annual average ozone concentrations in 2015 exhibited high spatial variation (north and east) and low variation (south and west).
• Ozone concentrations peaked in summer and autumn, and were lowest in winter and spring.
• Ozone concentration's center of gravity shifted seasonally, migrating north/west, then south, and finally east.
• Significant positive spatial autocorrelation was observed, characterized by high-high and low-low clustering, with seasonal temporal migration and conversion.
• Spatial autocorrelation was most significant during winter.

Conclusions:

• The Bilonick model (BM) within the STK framework is superior for simulating and predicting ozone pollution in the PRD.
• Ozone pollution in the PRD displays distinct spatial and temporal variations, with higher concentrations in summer/autumn and specific geographical areas.
• Understanding these complex spatial-temporal patterns and clustering is vital for targeted public health interventions and air quality management.