Wavelet Transform-based Scaling Response of PM2.5 with Meteorological Factors and Other Air Pollutants

Abstract

To quantify the spatiotemporal variations of PM_2.5 and its response relationship with the driving factors (including meteorological factors and other air pollutants) in the multi-scale time–frequency domain, the monitoring data of PM_2.5 and its driving factors in Hunan Province from 2017 to 2021 were researched via four wavelet transform methods including continuous wavelet transform (CWT), discrete wavelet transform (DWT), wavelet transform coherence (WTC), and multiple wavelet coherence (MWC). Results revealed that: (1) The annual average PM_2.5 concentration exhibited a decreasing trend, with a cumulative decrease of 27.2%. The seasonal distribution pattern of PM_2.5 was winter > fall > spring > summer. (2) The mutation of PM_2.5 mainly occurred in winter and was particularly concentrated in the Chang-Zhu-Tan urban agglomeration. The periodicity of 250–280 days was the dominant cycle in the time series. In addition, 30–40- and 70–80-day cycles were observed in winter and from autumn to winter, respectively. (3) The response of PM_2.5 to its driving factors depended on the time–frequency scale and the combination of factors. For the meteorological factors, temperature (TEM) was the strongest single factor that affected PM_2.5 at all time–frequency scale. Meanwhile, the coherence increased with an increasing number of meteorological factors. The other air pollutants had higher abilities to explain PM_2.5 variations than the meteorological factors. Among them, PM₁₀ was the strongest single factor that affected the PM_2.5 at all time–frequency scale, with a significant positive coherence between the two. The tetravariate combination of PM₁₀-SO₂-O₃-CO at the large time–frequency scales showed the highest degree of explanation of PM_2.5 concentration variations among all combinations. (4) Combining meteorological and pollutant factors significantly improves PM_2.5 variation explanation, but more factors do not guarantee better results. The research results of this paper may provide a reference for more precise identification of the influencing factors of PM_2.5 and the formulation of related air pollution control policies.