Some challenges in mesoscale data assimilation for numerical weather prediction. Speaker: Nedjeljka Žagar University of Ljubljana, Slovenia ATTENTION!, The Starting time has changed in 15.30 hours)

Mesoscale models for numerical weather prediction (NWP) in Europe have reached a horizontal resolution close to 1 km.

Data assimilation for mesoscale NWP is expected to provide initial conditions for these km-scale forecasts that outperform the global model forecasts (e.g., ECMWF) on the short range (few hours to 1 day). A lack of frequent and accurate high-resolution observations, especially wind observations, makes the challenge of mesoscale data assimilation and forecasting only greater. Furthermore, forecasts from limited areas models are exposed to an additional source of uncertainty with respect to the global models, which is associated with the update of the prognostic variables at the lateral boundaries. Challenges of mesoscale data assimilation were the focus of a recently held workshop at the University of Ljubljana, http://meteo.fmf.uni-lj.si/en/workshop. In this seminar, I will present two topics discussed at the workshop: the growth of forecast errors in two mesoscale NWP models due to lateral boundary perturbations and the role of the background-error constraints and balance issues in data assimilation.