A Study of Characteristics, Origins, and Fates of Ultrafine Particles Over Taichung City, Taiwan

Abstract

Ultrafine particles (UFPs, <0.1 μm) pose serious health risks and remain challenging to accurately monitor and model due to complex atmospheric processes. This study investigates UFP dynamics over Taichung City, Taiwan, utilizing April 2021  observational data from a Scanning Mobility Particle Sizer (SMPS) and simulations from the GEOS-Chem-TOMAS model. Observations included particle number size distributions (analyzed from 11.8 to 593.5 nm), sulfuric acid concentrations, and meteorological variables. The model employed 15 size bins and evaluated five nucleation schemes (Basecase, Binary, Binary ion, Ternary, Ternary ion) to assess new particle formation (NPF) sensitivity. Results indicate that Basecase, Ternary, and Ternary ion schemes simulated sporadic nucleation bursts; Binary schemes showed minimal activity. Despite visual  coincidence with sulfuric acid peaks, simulated nucleation rates exhibited a weak linear correlation (Pearson R ≈ −0.04), underscoring that NPF is a complex, non-linear process influenced by multiple atmospheric parameters beyond sulfuric acid, including ammonia, amines, and organic vapors, The model consistently and substantially underestimated observed particle number concentrations across all modes (NMBs ranging from -8.5% to -90.9%; correlation coefficients (absolute values) from 0.06 to 0.21) except for accumulation mode that the model display relatively good agreement of magnitude. The current model struggled to capture the overall magnitude and variability of particle number size distribution, underscoring the need for improving our understanding of the emission inventory as well as the boundary layer meteorology.  With regard to new particle formation, crucially no clear classical "banana-shaped" NPF events were observed Importantly, no clear classical "banana-shaped" NPF events were observed in either the observational data or the model simulations throughout the study period. These findings underscore the critical need for improved nucleation and growth parameterizations, higher-resolution meteorological data, and refined local emission inventories to enhance the accuracy of urban aerosol modeling in subtropical environments.