Observations and modeling of OH and HO2 radicals in Chengdu, China in summer 2019

This study reports on the first continuous measurements of ambient OH and HO2 radicals at a suburban site in Chengdu, Southwest China, which were collected during 2019 as part of a comprehensive field campaign 'CompreHensive field experiment to explOre the photochemical Ozone formation mechaniSm in summEr - 2019 (CHOOSE-2019)'. The mean concentrations (11:00-15:00) of the observed OH and HO2 radicals were 9.5 x 10(6) and 9.0 x 10(8) cm(3), respectively. To investigate the state-of-the-art chemical mechanism of radical, closure experiments were conducted with a box model, in which the RACM2 mechanism updated with the latest isoprene chemistry (RACM2-LIM1) was used. In the base run, OH radicals were underestimated by the model for the low-NO regime, which was likely due to the missing OH recycling. However, good agreement between the observed and modeled OH concentrations was achieved when an additional species X (equivalent to 0.25 ppb of NO mixing ratio) from one new OH regeneration cycle (RO2 + X -> HO2, HO2 + X -> OH) was added into the model. Additionally, in the base run, the model could reproduce the observed HO2 concentrations. Discrepancies in the observed and modeled HO2 concentrations were found in the sensitivity runs with HO2 heterogeneous uptake, indicating that the impact of the uptake may be less significant in Chengdu because of the relatively low aerosol concentrations. The ROx (= OH - HO2 + RO2) primary source was dominated by photolysis reactions, in which HONO, O-3, and HCHO photolysis accounted for 34%,19%, and 23% during the daytime, respectively. The efficiency of radical cycling was quantified by the radical chain length, which was determined by the NO to NO2 ratio successfully. The parameterization of the radical chain length may be very useful for the further determinations of radical recycling. (C) 2021 Elsevier B.V. All rights reserved.

Automated summary

This study presents the first continuous measurements of ambient OH and HO2 radicals in a suburban site in Chengdu, Southwest China. The analysis shows a discrepancy between the observed and modeled OH concentrations, which could be improved by adding an additional species X into the model. Additionally, the model was able to reproduce the observed HO2 concentrations but discrepancies were found in sensitivity runs with HO2 heterogeneous uptake.