Abstract

Aerosols play an important role in the extreme events of air pollution and climate change. Despite the decade’s efforts, air pollution models are still unable to simulate the heavy pollution events very well due to the complicated aerosol chemical composition, size distribution and spatial and temporal variations. Recently, the aerosol mass spectrometer (AMS) was deployed in Dublin to determine the chemical composition and concentration of submicron aerosols at high temporal resolution. The organic aerosol (OA) was found to dominate the aerosol composition (>50 %) in the heavy pollution events. The particular matter concentration (PM1) was measured to exceed 200 μg m-3 on 19 November 2016 and 22 January 2017.

In this study, we use the WRF-Chem model to simulate the heavy organic aerosol episodes in Dublin. However, the model could not rebuild the heavy pollution events with current emission datasets (EDGAR, EMEP, TNO, NEI or RETRO) we use. To have better simulation results, we have examined the sensitivities of simulation results to the emissions, meteorological parameters or spatial and temporal resolutions. We find that the enhancement of local emissions with the reference of measurements including diurnal variation and temperature dependence is the most effective way to rebuild the heavy pollution events. The heavy organic aerosol episodes in Dublin are found to be contributed mostly from local emission sources (peat burning). The lack of accurate speciation and temporal profiles in current emission datasets should be greatly concerned, particularly the significant underestimation of organic carbon (OC) emissions. The uncertainties in current emission datasets will hamper us to forecast the heavy air pollution events accurately and improve the early warning system from forecasts, hence it is imperative to improve the emission inventories using the new method of integrating top-down and bottom-up approaches with available measurements.