[5-2]

Global Ensemble Data Assimilation for ICON

Ana Fernandez del Rio, Alexander Cress, Andreas Rhodin, Michael Denhard, Jaison Thomas Ambadan, Harald Anlauf and Roland Potthast (DWD)

The ICOsahedral Nonhydrostatic general circulation (ICON) model has been jointly developed by the Max Planck Institute for Meteorology (MPI-M) and the German Weather Service (Deutscher Wetterdienst, DWD) to replace the previous global model in operations at DWD. The Local Ensemble Kalman Filter (LETKF) in experimental use at DWD has been adapted to the ICON model and is currently being tuned for operational implementation. It is a key component of the VarEnKF hybrid system, which is currently also being implemented, as well as a natural way to provide boundary conditions for the Kilometer Scale Ensemble Data Assimilation (KENDA) for the COSMO model, the data assimilation scheme currently under development for the limited area model at DWD. The latest results will be presented, discussing the performance of the ensemble prediction system and of the deterministic 3D-VAR forecast, as well as the impact of different covariance inflation techniques (adaptive inflation, relaxation to prior perturbations, perturbed surface analysis). Special attention will be paid to assessing how good the ensemble's spread is in capturing the system's uncertainty, as well as on the impact of its use in providing boundary conditions for KENDA.