One of the major advances in weather prediction of the past three decades has been the provision of reliable uncertainty estimations. This has been achieved by shifting from issuing a single to an ensemble of forecasts. This paradigm shift allows us not only to forecast of the most likely future state, but also to have reliable estimates of its accuracy, for example expressed in terms of a range of possible future scenario, or in probabilities. Similarly, today we have ensembles of analyses and reanalyses, that allows us to estimate the full probability distribution function of current and past states. In this talk, I will review how we developed the earlier ensembles, and the key characteristics of today’s operational ensembles. I will also discuss how work is progressing towards the development of more reliable, coupled Earth-system ensembles of data assimilations and forecasts, which should help us further advancing forecast skill.
Per partecipare al Colloquio cliccare sul seguente Link identifier #identifier__171314-1Link
Link identifier #identifier__53206-2Locandina
This post is also available in: Link identifier #identifier__3758-5Eng