This study compared two types of (offline vs. online) bias correction models applied to correct the mismatch between the predicted daily averaged PM10 concentration by the deterministic air quality forecasting system WRF-EURAD and the measured concentration at the air quality monitoring station in Porto, Portugal. The WRF-EURAD is a Eulerian system consisting of a Weather Research Forecasting (WRF) model and a European Air Pollution Dispersion (EURAD) model. Both bias correction models were linear statistical models developed with the same set of input variables. The major difference between the online or the offline models is the adaptiveness. While the coefficients of the offline bias correction model are fixed after training with the ordinary least squares, those in the online bias correction model are updated adaptively with the Kalman filter. Comparison of these bias correction models was made at an urban traffic station Senhora da Hora in Porto, Portugal within 2013. The fractional bias of the daily PM10 forecast by the WRF-EURAD was found significantly improved from -38% to 7% and 3% after the correction by the offline or the online bias correction model, respectively. In addition, further comparison of the overall performance indicators (root-mean-squared error and correlation coefficient) and the indicators focusing on the days of the PM10 episodes (episode detection rate, false alarm rate, and critical success index) revealed that the online correction model could lead to more improvement of the WRF-EURAD compared to the offline model. By comparing the temporal variation of each estimated model coefficient and the 95% confidence interval between both correction models, the model coefficient corresponding to the raw forecast of the WRFEURAD in the online model has varied significantly in the second half of the year. It was concluded that the air quality system in Porto may be time-varying and further investigation is necessary to find out the leading cause.