Application of the Ensemble Transform Kalman Filter technique at the Hungarian Meteorological Service: Preliminary results

摘要

The provision of the best background error estimates for data assimilation is crucial to derive the best initial conditions for a numerical weather prediction model. Using the Ensemble Transform Kalman Filter (ETKF) method a forecast ensemble can be generated at each analysis time, which can be used to estimate the background error statistics. The error growth implied by the forecast ensemble, which is measured with the forecast differences, characterizes the reliability of the actual background. In the present paper, the theoretical background of the Ensemble Transform Kalman Filter method is presented first. The ETKF transformation technique was applied in our tests for generating analysis ensemble containing the initial conditions for the model, then experiments were run with different setups to create forecast ensembles. At this stage of the research, estimation of the background errors was left aside, and as primary validation of this technique, the ensemble generation scheme and the transformation itself were studied. According to the results it can be said that the ensembles generated with the ETKF resealing scheme have too small spread, which should be further improved with the proper application of the inflation procedure.