SPATIO-TEMPORAL EVOLUTION OF THE AIR POLLUTION SPECTRUM IN CHINA AT THE PREFECTURE LEVEL (2015-2020)

Air pollution is a critical global issue affecting human health. This paper introduces the Air Pollution Spectrum (APS), a multi-pollutant evaluation model (CO, NO2, PM10, PM2.5, O-3, and SO2) that surpasses the Air Quality Index (AQI) by reflecting both spatial quality and pollutant structure. This paper uses the Bayesian spacetime hierarch piecewise regression model (BSTHPRM) to study the spatio-temporal evolution characteristics of the APS in China at the prefecture level (2015-2020). At the same time, the spatio-temporal characteristics of the APS component structure are also explored. We found higher levels of air pollution in Central, North, Northwest, and East China and lower air pollution levels in Southwest, South, and Northeast China. The change in APS values in Chinese cities during the study period is divided into two stages: in the first stage, the local change trend is rapid in Northwest, Southwest, and North China and slow in Central and Northeast China; in the second stage, the local change trend is fast in Northeast China and slow in Northwest and Southwest China. Spatial distribution and variation trends suggest that regional differences in APS values are narrowing. The APS component structure also has significant spatio-temporal distribution characteristics. O3 has gradually become one of the main pollutants involved in air pollution, and its importance is greater in North China, West China, and South China and smaller in Central China and East China. PM10 accounts for the largest, most stable proportion of air pollution in China and, in general, plays an important role in air pollution in most regions except some western regions. The importance of PM2.5 in air pollution in China has declined, with its importance greater in the central and eastern regions and smaller in the northern, western, and southern regions. The component structure of air pollutants at the prefecture level in China is changing significantly, with pollutants becoming more diverse. Parallel research on multiple pollutants has become an inevitable trend in air pollution research.