the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Offsetting Impacts of Aromatic Autoxidation on Organic and Inorganic Aerosols Modelled by GEOS-Chem-TOMAS during Winter Haze in Nanjing, China
Abstract. Chemical transport models (CTMs) often underestimate secondary organic aerosols, especially during severe haze events. Recent research has shown that autoxidation is a key pathway for rapid organic aerosol (OA) growth, yet few studies to date have attempted to implement this mechanism into large-scale models. This study implemented an aromatic autoxidation mechanism (AAM) produced by the automated mechanism generator, autoAPRAM-fw, into the global CTM, GEOS-Chem, coupled with the TwO-Moment Aerosol Sectional (TOMAS) module, representing one of the first implementations of such a mechanism into a large-scale model. We compared simulations with and without the AAM to measurements in Nanjing, China during winter haze in early 2023. Results demonstrated that the AAM increased modelled OA mass by 3.94 μg/m3 (12 %), reducing the gap between modelled and measured OA. Furthermore, modelled particulate nitrate and ammonium, which were overestimated by GEOS-Chem in China, decreased after introducing the AAM, reducing the bias. These two impacts had an offsetting effect over Nanjing, limiting the AAM's net influence on fine particulate matter (PM2.5) mass. The autoxidation chemistry, especially the offsetting effect, was particularly sensitive to NOx concentration. The AAM produced higher particle number concentrations in the smallest size ranges and improved 3–10 nm particle growth rates when compared to the measurements. We additionally ran a sensitivity test with 50 % reduced aromatic VOC emissions, which reduced PM2.5 but did not significantly affect ultrafine particle number concentration. This demonstrates that under high-haze conditions, lower condensation sink offsets aromatic vapour reductions, favouring ultrafine particle growth.
- Preprint
(39831 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 31 Dec 2026)