Wavelet Multiscale Analysis Reveals PM_2.5 and PM₁₀ Temporal Dynamics and Meteorological Coupling in Guiyang, a...

Background: Understanding the long-term, multiscale temporal dynamics of particulate matter (PM_2.5 and PM₁₀) and its complex interactions with meteorological factors is critical for developing effective air pollution control strategies, particularly in cities with unique topographical characteristics. This study investigates the decadal evolution of PM pollution in Guiyang, a typical karst highland city, to unravel its underlying mechanisms and inform targeted mitigation policies. Methods: We analyzed continuous daily PM_2.5 and PM₁₀ concentration data from 2015 to 2024 in Guiyang, along with concurrent meteorological data (temperature, relative humidity, wind speed, precipitation, and air pressure). Multiscale wavelet analysis, employing Daubechies 6 and Morlet wavelets, was used to decompose the time series, identify dominant periodicities, and quantify the scale-dependent correlations between PM concentrations and meteorological drivers. Results: Over the last decade, Guiyang has achieved significant improvement in air quality, with the annual average concentrations of PM_2.5 and PM₁₀ decreasing by 51.05% and 49.60%, respectively, compared to 2015. The annual average PM_2.5 concentration exceeded the grade II of the National Ambient Air Quality Standard (35 μg/m³) in 2015 and 2016. A stable "U"-shaped annual cycle (winter > spring > autumn > summer) was identified, which wavelet analysis further resolved into three distinct dynamic stages and attempt to analyze its dominant factors: a winter-spring fluctuation period (January-May) dominated by atmospheric inversions and anthropogenic emissions, a summer stabilization period (May-October) driven by clean air masses and enhanced wet deposition, and an autumn-winter accumulation period (October-January) triggered by regional biomass burning and stagnant weather. Extreme pollution events, such as the 2024 Chinese New Year, caused PM_2.5 and PM₁₀ to surge by over 130% on a single day. The PM-meteorology coupling exhibited strong scale dependence: at short scales (<16 d), relative humidity and precipitation were the primary drivers; at medium scales (16-64 d), wind speed and temperature (promoting secondary aerosol formation) became dominant; while at large scales (>64 d), correlations weakened. Conclusion: While long-term policies have been effective, Guiyang's air quality is challenged by persistent winter pollution and extreme events, governed by a multistage annual pattern and constrained by its unique karst topography. We recommend a three-pronged strategy: (1) prioritizing winter-spring controls focusing on clean heating, (2) implementing a dynamic "festival-meteorology" early warning system for extreme events, and (3) developing terrain-adapted strategies to optimize emission source layouts and enhance urban ventilation. These findings also provide crucial insights for improving air quality models in mountainous regions by incorporating multiscale dynamics, particle-specific responses, and terrain-specific parametrizations.