Application of WRF-Chem to Forecast PM10 Concentrations Over Poland

The meteorological and chemical transport model WRF-Chem has been implemented to forecast PM10 concentrations over Poland. WRF-Chem version 3.5 was configured with three one way nested domains and the simulations were driven by the GFS meteorological data and the TNO MACC II emissions. The Regional Acid Deposition Model (RADM2) gas phase chemistry and the Modal Aerosol Dynamics Model for Europe with Secondary Organic Aerosol Model (MADE/SORGAM) aerosol module were applied in WRF-Chem. Forecasts were run 48 hours ahead for each day during a winter period from the 1st January to 28th February 2014. We have found that the model for most stations captures the variability of observed PM10 concentrations, whereas the highest observed peaks are often underestimated. Several of the peaks coincide with either stable atmospheric condition during nighttime in the lower part of the planetary boundary layer or on days with very low surface temperatures. This suggests that a key to a good simulation of the peak concentrations is to focus on the simulation of PBL processes and the distribution of emissions with high resolution in WRF-Chem