Ensemble Adjustment Kalman Filter Data Assimilation for a Global Atmospheric Model

摘要

This work describes the implementation and evaluation of an Ensemble Adjustment Kalman Filter (EAKF) with a global atmospheric zoom model (version 5) of the Laboratoire de Meteorologie Dynamique (LMDZ5, Z stands for zoom). An interface has been developed to use Data Assimilation Research Testbed (DART), a community EAKF system, with LMDZ5 model. The NCEP PREBUFR real observation data have been assimilated to evaluate the performance of newly developed LMDZ5-DART system. It has been demonstrated with the help of a numerical experiment that LMDZ5-DART system successfully assimilates real observations. A one month LMDZ5-DART analysis has been created using assimilation of NCEP PREBUFR observation data, and the assimilated fields are compared with NCEP CDAS reanalysis. Results show that LMDZ5-DART produces remarkably similar reanalysis to NCEP products. This is therefore a very encouraging result towards a long-term goal of creating a high quality analysis over the Indian subcontinent from the assimilation of local satellite products.