A LETKF algorithm with divergence constraint for convective-scale radar data assimilation in the scenario of OSSEs

Abstract

The constrained Ensemble Kalman filter (EnKF) has not been applied for the storm-scale radar data assimilation (DA) problem before. In this work, the Local Ensemble Transform Kalman Filter (LETKF) algorithm with constraint of divergence equation is implemented within the Advanced Regional Prediction System (ARPS) ensemble DA framework. This constrained LETKF system is tested by assimilating simulated radar observations of a supercell storm through observing system simulation experiments (OSSEs). The most remarkable improvement of performance could be seen on the analysis of perturbation pressure field. The results of experiments with different ensemble sizes show that, when using less ensemble members, the divergence constraint can improve the performance of LETKF, especially on the wind component variables, through the data assimilation cycles by reducing the increments of forecast error after the analysis with constraint. It indicates that the divergence constraint can enhance the property of balance among the wind and mass fields in the analysis, which could be weakened by sampling noise, especially due to a relatively small ensemble. Furthermore, with the experiments with sensitivity to observation type, it is found that when only radar radial wind observations are analyzed, the divergence constraint make the major and positive impact on the analysis performances of wind components. It shows the capability of divergence constraint to improve the analysis of wind fields when only assimilating radial wind data, which is incompetent to the retrieval of three dimensional wind fields.