Assimilation of Surface Ozone Measurements to WRF-Chem—Impact on the Model Capability to Predict Peak Concentrations

Ozone in the troposphere has negative effects on vegetation and human health, which causes that forecasts of ozone concentrations are developed and used by policy makers and society to assess air quality. In this study we used the WRF-Chem model and GSI assimilation tool to analyse the impact of surface 3D-Var data assimilation on modelled O3 concentrations over Central Europe, with a focus on Poland. We run two simulations for summer 2015: (1) BASE—no data assimilation, and (2) SURF—the WRF-Chem output from the previous day was modified by the GSI assimilation system. The results show that data assimilation has not a significant positive impact on mean error measures such as factor of two, mean bias, and mean gross error. However, the results also show that the SURF simulation better reflects peaks of high O3 concentrations, which is especially important in terms of air pollution forecasting and prevention of the harmful effects of pollutants.