Modulation of synoptic circulation to dry season PM2.5 pollution over the Pearl River Delta region: An investigation based on self-organizing maps

Abstract

Abstract Fine particle (PM2.5) pollution is of concern for the Pearl River Delta (PRD) region, particularly during the dry season (October to March), when the PM2.5 concentrations can exceed Chinese national standards. In part due to sparse observations, the dispersion/transport mechanisms that lead to PM2.5 pollution over the PRD region are not yet fully understood, and the potential synoptic controls have not been investigated. In this study, systematic analyses were conducted using the multisite surface meteorological observations, two-site radiosonde measurements, and regional PM2.5 observations, based on the application of self-organizing map neural network to large-scale mean sea-level pressure (MSLP) data. The results suggested that the relative position of the PRD to high-pressure systems exerted significant effects on the dispersion conditions and the PM2.5 distribution over the PRD region. When cold high pressure invaded, the PRD region was dominated by synoptic-scale northerly flow, which caused southward pollution transport and hence positive PM2.5 anomalies in the coastal area. When cold high pressure weakened and moved eastward, the PRD region was dominated by meso/local-scale flows, resulting in strong atmospheric recirculation and elevated PM2.5 concentrations in the inland area. However, the regional average of PM2.5 concentrations was not sensitive to the changes in synoptic circulation patterns, which implied that the synoptic circulation played roles most in redistributing the air pollutants within the PRD region and hence a regional PM2.5 pollution emission reduction is needed to improve regional air quality.