Regional PM2.5 pollution confined by atmospheric internal boundaries in the North China Plain: Analysis based on surface observations

Abstract

There are plenty of mesoscale meteorological discontinuities in the atmosphere, acting as atmospheric internal boundaries (AIBs). In conjunction with the atmospheric boundary layer in the vertical direction, they form confined three-dimensional structures that significantly affect air pollution. However, the role of AIBs in regional pollution has not been systematically elucidated. Based on surface observations, this study investigates PM2.5 pollution distributions under the forcing of various AIBs in the North China Plain. A total of 98 regional pollution episodes are identified during the autumn and winter of 2014–2020, and are further classified according to the impact of AIBs. In the pollution formation-maintenance stage, there are three categories. The frontal category (with a frequency of 41%), including the frontal trough type and frontal inverted trough type, displays the most polluted air masses along the mountains. The frontal AIB defines the lateral border of the pollution zone and forms a frontal inversion above, creating a closed and stable structure wherein the highest concentration of PM2.5 accumulates. The wind shear category (29%) is decided by the dynamic convergence AIB, which causes lighter PM2.5 pollution with diverse spatial patterns corresponding to west-southwest shear, southeast-east shear, and south-north shear. The topographic obstruction category (14%) presents as a narrow arc-shaped pollution belt at the foot of the windward mountains, resulting from the cold air damming AIB with dynamical obstruction and thermal stratification. Pollution diffuses in three ways: northwest, west, and northeast, respectively. The first one is the strongest and most frequent (42%), with both strong horizontal wind and vertical mixing. The second category is relatively rare (17%), characterized by foehn-induced active vertical ventilation. The last one is frequent (41%), but relatively weak, mainly relying on horizontal diffusion. Some evolution details of the AIB affecting PM2.5 pollution are also illustrated by a typical case.