Meteorological Impact on Dynamic Air Pollutant Concentrations in Different Timescales: Typical Case in Chengdu Megacity, China

Due to the rapid urbanization and climate change, one of the pressing problems confronting Chinese cities lies in air pollution. Especially during recent years, most cosmopolitan cities in China suffer from thick smog caused by PMs. It is reported that such air pollution has close relations with traffic-related pollutants, such as NO 2 . Meteorological condition contributes significantly to long-range transport sources and has remarkable influences on the distribution of the above pollutants. Thus, it is necessary to quantitatively analyze such influential patterns. In this paper, taking Chengdu as an illustrative example, the meteorological and pollutant statistics in 4947 h are collected and fitted with 8 meteorological parameters of PM 2.5 , PM 10 , and NO 2 through linear regression method. And data is analyzed and compared on the hourly, daily, weekly, and monthly basis, respectively. Results show that hourly and daily data focus on fluctuations that affected by multiple interferences other than meteorological influences. When hourly values of PM 2.5 , PM 10 , and NO 2 exceed 175 μg/m 3 , 250 μg/m 3 , and 125 μg/m 3 , respectively, discrete values that cannot be explained by meteorological influences increase notably, while weekly and monthly variations illustrate more tendencies, and communicable patterns that decisively influenced by meteorology can be extracted among 3 pollutants. By comparing the results in 4 timescales, respectively, similar principal meteorological factors are involved in fitting processes. Thirdly, visibility (V), wind speed (WS), precipitation (P), and cloud condition (CC) are the most influential factors and served as main parameters in linear regressions within 4 timescales. Air temperature (AT) and dew temperature (DT) are less influential factors by daily and weekly data but are served as main parameters when it goes to hourly and monthly data and function as variables that signify tendency variances in linear regressions, while relative humidity (RH) and barometric pressure (BP) are lest influential factors and have limited effect on fitting results. This work can provide guidance and reference for urban planning optimization and air environment protection in cities with air quality control considerations impacted by local climatic conditions.