Modelling VOC source impacts on high ozone episode days observed at a mountain summit in Hong Kong under the influence of mountain-valley breezes

A top-down approach to evaluate high ozone (O3) formation, attributed to different emission sources, is developed for anti-cyclonic conditions in a region of Hong Kong influenced by meso-scale circulations. A near-explicit photochemical model coupled with the Master Chemical Mechanism (MCMv3.2) is used to investigate the chemical characteristics in the region. Important features have been enhanced in this model including the photolysis rates, simulated by the National Center for Atmospheric Research (NCAR) Tropospheric Ultraviolet and Visible (TUV) Radiation Model, as well as hourly variation of Volatile Organic Compound (VOC) concentration input from on-site sampling. In general, the combined model gives a reasonably good representation of high O3levels in the region. The model successfully captured a multi-day O3event in the autumn of 2010. Source apportionment via Positive Matrix Factorization (PMF) was carried out on the sampled VOC data, to determine the major sources in the region. Based on the outcomes of the PMF source apportionment, a sensitivity analysis using the developed photochemical model was conducted and used to estimate O3reduction under different source removal regimes. Results indicate that vehicular emissions are the dominant VOC source contributing to O3formation. This study has demonstrated a potentially efficient secondary pollutants control methodology, using a combined field measurements and modelling approach.