Aerosol-radiation interaction in the operational atmospheric chemistry model GRAPES_Meso5.1/CUACE and its impacts on mesoscale NWP in Beijing-Tianjin-Hebei, China

Abstract

The aerosol-radiation interaction mechanism (ARI) is important for air quality modeling, and its absence in atmospheric chemistry models may cause considerable uncertainties. In this study, the real-time calculated aerosol optical parameters, i.e., mass extinction coefficient (Kext), single scattering albedo (ω), and asymmetry factor (g) from the simulated aerosol concentrations are introduced into the Global/Regional Assimilation and Prediction System 5.1 version coupled with the Chinese Unified Atmospheric Chemistry Environment model (GRAPES_Meso5.1/CUACE) to establish the ARI mechanism for the two-way feedback between aerosols and weather processes. Focusing on Beijing-Tianjin-Hebei (BTH), China, in January 2017, ARI impacts on meteorology and haze-fog are simulated by using this atmospheric chemistry model. The results show that during the severe pollution, ARI can significantly alter the meteorology, especially for the heavily polluted central and southern BTH: The surface solar radiation (SSR), the temperature at 2 m (T2), the vertical diffusion coefficient (VDC), and the planetary boundary layer height (PBLH) are reduced by up to 50 W m−2, 2.5°C, 4 m2 s−1, and 100 m, respectively. In addition to surface meteorology, ARI affects the vertical structure of the planetary boundary layer (PBL), by cooling and humidifying the lower air and heating and drying the higher air. These modifications stabilize the PBL, resulting in an increase of 8% in relative humidity at 2 m (RH2), a key factor influencing secondary aerosol generation and aerosol hygroscopic growth. Stabilized PBL and increased RH2 ultimately result in a 100 μg m−3 increase in PM2.5 concentrations and a 1.5 km reduction in atmospheric visibility (VIS). By considering the ARI, the model generally improves numerical weather prediction (NWP) and haze-fog prediction. For the entire BTH region, the mean biases for T2 and RH2 are reduced by up to 2°C and 4% respectively. This more accurate NWP will lead to an overall improvement in haze-fog prediction, particularly in a heavily polluted city like Shijiazhuang (SJZ), the biases for PM2.5 and VIS are reduced by 28% and 125% respectively. The results show the positive feedback between meteorology and haze-fog, and demonstrate the importance of ARI in NWP and haze-fog prediction.