Vertical distribution of PM2.5 and interactions with the atmospheric boundary layer during the development stage of a heavy haze pollution event

Sci Total Environ. 2020 Feb 20:704:135329. doi: 10.1016/j.scitotenv.2019.135329. Epub 2019 Nov 22.

Abstract

Vertical profiles of PM2.5 (i.e., particulate matter with an aerodynamic diameter of 2.5 µm or less) and meteorological variables (e.g., potential temperature, specific humidity) are crucial to understand formation mechanism including accumulation and dispersion process of PM2.5, as well as interactions between aerosols and the atmospheric boundary layer (ABL). In this study, vertical distributions of PM2.5 are characterized through comprehensive analyses of vertical profiles measured by unmanned aerial vehicle (UAV), Micro Pulse LiDAR, and other surface observational data of a heavy aerosol pollution episode occurring on December 22-25, 2017 in Nanjing, China. Results show that PM2.5 profiles are characterized by a clear three-layer structure with near constant within the mixed layer, a transition layer with a large local gradient in the entrainment zone, and a layer with low concentration and small gradient in the free atmosphere, which shows a large similarity to that of specific humidity. The accumulation of aerosols is found near top of the ABL with the largest increase rate. Vertical distributions of PM2.5 and their evolution are largely constrained by the ABL thermodynamics during daytime, but show much less dependence on the ABL evolution at nighttime. PM2.5 provides an important feedback on the nocturnal boundary layer (NBL) leading to significant modification of vertical distributions of potential temperature and water vapor. Moreover, this study suggests that the current boundary layer parameterization scheme needs refinement with aerosol radiative effect included to further improve the ABL height (ABLH) and air quality predictions.

Keywords: Aerosol pollution; Atmospheric boundary layer height (ABLH); PM(2.5); Unmanned aerial vehicle (UAV); Vertical profiles.