Decoding the urban-suburban ozone disparities in Shandong, China: A machine learning-driven attribution analysis...

Area of Science:

• Environmental Science
• Atmospheric Chemistry
• Public Health

Background:

• Urban-suburban disparities in ozone (O3) pollution are a growing concern for environmental equity and public health.
• Understanding the drivers of these disparities, especially the differences between coastal and inland cities, is crucial but remains limited.

Purpose of the Study:

• To investigate the spatiotemporal variability of O3 pollution.
• To identify and quantify the drivers of urban-suburban O3 disparities in Shandong Province.
• To explore contrasting mechanisms in coastal versus inland cities.

Main Methods:

• Utilized observational data from 267 air quality monitoring sites (2019-2023).
• Applied the de-weathered method, ensemble machine learning models, and SHAP analysis.
• Differentiated analysis for coastal and inland urban-suburban areas.

Main Results:

• Ozone concentrations (MDA8 and mean) were consistently higher in urban than suburban areas, particularly in summer and at night.
• Urban-suburban O3 disparities (Δ[MDA8] and Δ[O3]) increased by 0.4 and 1.0 μg/m³ respectively between 2019-2023.
• Emission source variations, especially nitrogen dioxide (NO2), explained 84% of the increase in Δ[O3]. PM2.5 disparities amplified O3 differences in coastal regions. Meteorological factors showed distinct impacts on coastal (temperature, wind speed) and inland (mean sea level pressure) cities.

Conclusions:

• Emission controls, particularly for NO2 in inland areas and PM2.5 in coastal cities, are recommended to reduce O3 disparities.
• Targeted emission reduction strategies are necessary to address the complex drivers of urban-suburban O3 pollution.
• Mitigating O3 disparities will contribute to improved air quality across Shandong Province.