For both the atmosphere and the ocean, the quality of a model prediction depends on the accurate determination of the initial state and on a sophisticated model to simulate the subsequent evolution realistically. This session will focus on the use of observations, on data assimilation techniques which are developed or implemented in meteorology and oceanography, and on the observation impact.

Active research is carried out on algorithmic aspects of data assimilation such as :
• Intercomparison and study of the complementarity between different assimilation techniques : Kalman filtering, variational assimilation, nudging techniques for frequent analysis cycles used in nowcasting, etc...
• Variational techniques with longer assimilation windows and weak constraint methods to allow for the inclusion of model error estimates.
• Ensemble based assimilation systems and flow dependent estimation of background error statistics.
• Coupling strategies for the analysis of the atmosphere, the ocean and continental surfaces.
• Model error formulations in data assimilation.

This session will also accept papers about:
• The impact of observations assessed through Observing System Experiments (OSEs), Observing System Simulation Experiments (OSSEs), or sensitivity studies based on adjoint computations...
• Targeting strategies and observation network design.
• The impact of innovative observing systems on meso-scale NWP models, especially in the context of rapidly developing convective systems.