CMCC Webinar
21 February 2023, 12.00 pm CET
To join the webinar, register here

Speaker
Iuliia Polkova, University of Hamburg

Moderator
Panos Athanasiadis, CMCC – CSP Division

Abstract
Decadal climate predictions provide information about relevant climate indicators up to ten years into the future. Previously, it has been shown that climate predictions are sensitive to the starting climate conditions and the underlying initialization/data-assimilation procedures. Recently, the progress on finding the “best” initialization procedure for decadal climate predictions has stalled, suggesting that minor modifications to “conventional” initialization procedures only lead to minor improvements, if at all, in climate predictions. The presentation thus covers different initialization procedures for climate predictions in general terms, the ideas behind different initialization and data assimilation methods for decadal climate predictions, and speculations about why they fulfilled or not fulfilled the expectations of their developers. The presentation also mentions whether climate projections can replace climate predictions in the future and discusses what we can learn from simpler climate models (i.e., Lorenz model and intermediate complexity models) to advance the progress on initialization and data assimilation for decadal climate predictions. The presentation can be of interest to prediction system developers, and more broadly to anyone who would like to know how ocean and atmosphere observations enter climate prediction systems and the implications of this procedure for the quality of climate predictions.

WATCH THE VIDEO