Modeling study of aerosol-meteorology feedback during winter haze events over the north China plain

Abstract

An online-coupled regional atmospheric chemistry/aerosol-climate model RIEMS-Chem was utilized to investigate the aerosol radiative feedback effect during winter hazes over the North China Plain. Three severe winter haze events (8–January 15, 2013, 20–February 26, 2014, and 15–December 22, 2016) were studied. Model comparison against observations showed a reduction of absolute model bias by 16–21% for PM2.5 when aerosol radiative effect (ARE) was included. Due to ARE, the period- and regional-mean changes in 2 m air temperature (T2), planetary boundary layer height (PBLH), and PM2.5 concentration were −0.4∼-1.8 °C (−14%∼-60%), −30.9∼-183.6 m (−12%∼-31%), and 37.7–61.1 μg/m3 (26%–47%), respectively. The magnitudes of changes in downward solar radiation (SWD), T2, and PBLH were proportional to AOD (aerosol optical depth) and PM2.5 column level but not to surface PM2.5 level. The response of PM2.5 to aerosol radiative feedback was dependent on both PM2.5 level and ARE. The average Integrated Process Rate (IPR) for PM2.5 in Beijing was estimated to be 11.9 μg/m3/hr during the growth stage of the three severe hazes. Aerosol radiative feedback caused a net increase of IPR for PM2.5 by 2.1–9.5 μg/m3/hr during the growth stages, mostly contributed by chemical processes. Linear relationship can be derived for the PM2.5-ΔPM2.5 data pairs over the study region. The high correlation coefficient (0.95–0.96) suggested PM2.5 could be an indicator of the feedback-induced ΔPM2.5. A threshold PM2.5 concentration of 75 μg/m3 was found above which more than 10% PM2.5 change would occur due to aerosol radiative feedback, but this value differed among haze events and cities.