Abstract
Abstract. An Ensemble Kalman Filter (EnKF) data assimilation system was developed for a regional dust transport model. This paper applied the EnKF method to investigate modeling of severe dust storm episodes occurring in March 2002 over China based on surface observations of dust concentrations to explore the impact of the EnKF data assimilation systems on forecast improvement. A series of sensitivity experiments using our system demonstrates the ability of the advanced EnKF assimilation method using surface observed PM10 in North China to correct initial conditions, which leads to improved forecasts of dust storms. However, large errors in the forecast may arise from model errors (uncertainties in meteorological fields, dust emissions, dry deposition velocity, etc.). This result illustrates that the EnKF requires identification and correction model errors during the assimilation procedure in order to significantly improve forecasts. Results also show that the EnKF should use a large inflation parameter to obtain better model performance and forecast potential. Furthermore, the ensemble perturbations generated at the initial time should include enough ensemble spreads to represent the background error after several assimilation cycles.
Highlights
Dust storms have drawn much concern during the past two decades for the various impacts on atmospheric environment, biogeochemical cycles, radiative balance and human health
We developed a regional chemical transport model combined with ensemble Kalman filter (EnKF) data assimilation method to improve the forecast performance and to investigate the vertical structure of this super dust storm during the period of 15–25 March 2002 in East Asia
Izontal correlations, we present some examples of horizon- Two peaks of dust concentration distribution exhibited on tal correlations of surface concentrations with respect to the 20 March in Beijing with EnKF assimilation agree well with points shown with the black dots in Figs. 2 and 3
Summary
Dust storms have drawn much concern during the past two decades for the various impacts on atmospheric environment, biogeochemical cycles, radiative balance and human health. The CUACE/Dust forecast system and showed the capability of short-term forecast improvement These all indicate the important role of data assimilation to combine observations with modeling in air quality prediction. In these methods, the background error statistics, one of the most important aspects of data assimilation, are usually assumed to be spatially homogeneous, horizontally isotropic, and temporally stationary. In this study we perform ensemble Kalman filter (EnKF) data assimilation experiments during some severe dust storm episodes in China using surface observations of dust concentrations and a realistic model in order to explore the impacts on forecast skills. We made an initial effort to explore the potential problems of this issue with EnKF
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call