Seminar: Nedjeljka Žagar: A global view of the limits of prediction skill of NWP models

Sreda, 16. december 2015, ob 13:15
Predavalnica F6, 1. nadstropje, FMF, Jadranska 19, Ljubljana

V okviru internega seminarja na katedri za meteorologijo nam bo naša profesorica, dr. Nedjeljka Žagar, predstavila svoje raziskovalno delo. Seminar je odprt za vse obiskovalce, potekal pa bo v angleškem jeziku.


The data assimilation for numerical weather prediction (NWP) has traditionally been focused on the reduction of forecast errors in the storm-track regions in the midlatitudes. On the other hand, the uncertainties in initial conditions for the NWP models and ensemble prediction systems (ENS) are largest in the tropics; the same applies to the short-range forecast errors. Recent results suggest that these properties hold even in the perfect-model framework and data assimilation with the ensemble Kalman filter and a high-quality globally homogeneous observation network.
The loss of predictability and limits of prediction skill with relevance for NWP have been estimated by using the models and variables which describe midlatitude dynamics. The ongoing work addresses the global predictability loss in relation to the global large-scale dynamics. The seminar will provide a new view of the limits of the global prediction skill associated with physical processes. New estimates of the forecast error growth and asymptotic errors are provided for different horizontal scales and dynamical regimes associated with the balanced and inertio-gravity dynamics.