Two models and two emission databases – evaluation of the PM10 and PM2.5 concentrations modelled with WRF-Chem and EMEP4PL

Some European countries in Eastern or Central Europe, such as Poland, have serious problems with air quality. High concentrations of particulate matter (PM) in winter are often related to high coal and wood combustion for residential heating. Meteorological conditions, i.e. low air temperature and anticyclones, provide favourable conditions for the accumulation of air pollution, rendering it harmful to people.  PM concentrations during the warmer period are much lower, however there are episodes with elevated concentrations related to e.g. long-range transport of pollutants from biomass burning areas. Policy makers in Poland put a lot of effort to improve air quality as well as inform and aware people on harmful effects of air pollution. One of the relevant tools which provides information on the past, current and future state of the air pollution are chemical transport models.

In this study we aim for validation of PM10 and PM2.5 concentrations from two different chemical transport models – WRF-Chem and EMEP4PL and two different emission databases – a) a regional EMEP database, and b) a local database provided by the Chief Inspectorate of Environmental Pollution. Modelled PM10 and PM2.5 concentrations were compared with observations from Polish stations for the year 2018. The results show a clear seasonal variation of the models performance with the lowest correlation coefficients in summer. Higher seasonal variability is observed for WRF-Chem than EMEP, which is probably related to differences in calculations of boundary layer height. Application of local database improves the results for both models. For several months, the performance of WRF-Chem and EMEP is clearly different, which shows that an ensemble approach with an application of these two models could improve the modelling results. The differences in the model performance significantly influence the results of the population exposure assessment.