Study on the evolution patterns and predictive modeling of ambient air quality in oasis cities of arid regions: a case study of Urumqi

TL;DR

This study examines air quality patterns and future trends in Urumqi, an oasis city in an arid region, using data and modeling to inform pollution management strategies.

Contribution

The study introduces a comprehensive analysis of air quality in arid oasis cities, combining observational data with predictive modeling.

Findings

Urumqi's AQI in 2022 ranged from 24 to 363, with notable percentages of days experiencing moderate to severe pollution.

Dust events contributed 6.5% to PM concentrations and 1.1% to the air quality composite index.

Meteorological factors like wind speed and temperature showed significant correlations with pollutant levels, and ozone concentrations are projected to rise by 2025.

Abstract

This study investigates the evolution patterns and future trends of ambient air quality in oasis cities within arid regions, with Urumqi as a representative case.

Utilizing observational data from eight urban monitoring stations, we comprehensively analyze air quality variations and project future scenarios through the Air Quality Index (AQI), Spearman’s rank correlation coefficient, and Grey correlation modeling. Our aim is to elucidate the contributions of atmospheric pollutants to ambient air quality in arid oasis cities.

The results show that: (1) In 2022, Urumqi’s AQI ranged from 24 to 363, with exceed rates of 2.5% for severely polluted weather, 6.15% for heavily polluted conditions, 5.8% for moderate pollution, and 10% for mild pollution. (2) Dust events elevated inhalable particulate matter (PM) concentrations by 5 μg·m−3, contributing 6.5% to pollution levels, while the ambient air quality composite index reached 4.45. Dust’s specific contribution to this index was 0.05(1.1%).(3) Meteorological factors—precipitation, wind speed, temperature, and vapor pressure—exhibited significant negative correlations (p < 0.05) with PM₂.₅, PM₁₀, SO₂, NO₂, and CO concentrations, but a positive correlation with O₃. Wind speed showed a strong negative association with NO₂ (p < 0.01), while temperature and vapor pressure were positively linked to O₃ (p < 0.01). (4) The GM (1,1) model demonstrated high predictive accuracy, with errors between 1.0 and 4.2%. Projections indicate a rising trend in ozone concentrations, with the 90th percentile O₃-8 h potentially exceeding 160 μg·m−3 by 2025.

These results provide critical insights into the spatiotemporal dynamics of air pollution in Urumqi and its natural drivers, offering a scientific basis for regional air quality management and pollution mitigation strategies.

Figures