Not exact matches
In online science discussions, the fate of this years summer sea
ice has been the focus of a significant betting pool, a test of expert
prediction skills, and a week - by - week (almost) running commentary.
Yuan et al. (LDEO Columbia University), 5.08 (+ / - 0.51), Statistical The
prediction is made by statistical models, which are capable to predict Arctic sea
ice concentrations at grid points 3 - month in advance with reasonable
skills.
«Contributed to the
skill of retrospective decadal sea
ice predictions»?.
External radiative forcing contributes to the
skill of retrospective decadal sea
ice predictions
However, initialized
prediction ensembles using CESM can skillfully predict low - frequency modulations in the decadal trends of Arctic sea
ice, and the significant
skill scores for Atlantic sector sea
ice extent, in particular, suggest that CESM DP future forecasts merit serious consideration.
Stroeve et al., «Predicting September Sea
Ice — Ensemble
Skill of the Search Sea
Ice Outlook 2008 — 2013» (GRL 2014) observes that in some years
predictions are pretty good and in others they stink.
Cross-validation
skill, measured by the correlation between two - month lead
predictions and observations of September
ice extent, is 0.82, while the RMS error of
predictions is 0.72 million square km.
Cross-validation
skill, measured by correlation between four - month lead
prediction and observation of September
ice extent, is 0.77.