Work

Improving ozone episode predictions in the Great Bay Area: An evaluation of the contribution of gas-phase chemical mechanisms

Year: 2025

Type: article

Abstract: Improving model forecasting capability of high surface ozone (O 3 ) concentrations is critical for air quality management. Accurate predictions of O 3 and other criteria air pollutants depend heavily on the chemical mechanisms used in models. In this study, the Weather Research Forecast /Community Multiscale Air Quality (WRF/CMAQ) modeling system is utilized to quantify the impact of two chemical mechanisms, the Carbon Bond Mechanism 6 revision 3 (CB6r3) and the State Air Pollution Research Center Version 07 with toluene (T), iodine (I) and chlorine (C) chemistry (SAPRC07) on O 3 predictions in the Guangdong-Hong Kong-Macau Greater Bay Area of China (GBA). Three-month simulations were conducted with the two gas-phase mechanisms over the four nested domains in the GBA for July, August, and September 2021. The simulations are evaluated extensively with observations from surface meteorology and air quality monitoring networks, including three volatile organic compounds (VOC) observational sites in Hong Kong. The evaluations show that the SAPRC07 mechanism has a stronger capability than the CB6r3 mechanism in predicting O 3 peak concentrations during exceedance events with mean bias (MB) of -2.15 ppbv, correlation coefficient (CR) of 0.86, and hit rate of 0.58, which are higher than CB6r3 with MB of -11.54 ppbv, CR of 0.81, and hit rate of 0.35. The improvement of O 3 predictions is largely attributed to the more detailed treatment of VOC species by SAPRC07 mechanisms, as evidenced by our evaluation showing its superior performance in reproducing VOC concentration changes compared to CB6r3 during O 3 episodes. The study highlights the potential of implementing detailed VOC chemical mechanisms, such as SAPRC07, in real-time forecast and sensitivity studies to support the development of effective emission reduction strategies, given the rapid advancement of computer technologies. • SAPRC07 outperforms CB6 in predicting peak O3 during exceedances. • Detailed VOC treatment in SAPRC07 enhances O3 prediction accuracy. • Implementation of detailed VOC mechanisms improves air quality modeling.

Source: Atmospheric Environment X

DOI: 10.1016/j.aeaoa.2025.100392