LEXINGTON, Mass., Aug. 22 (UPI) -- U.S. scientists have confirmed a model that uses October Siberian snow cover to predict winter temperatures and snowfall across the Northern Hemisphere.
Atmospheric scientist Judah Cohen of AER Inc., in Lexington, Mass., and colleagues analyzed seven real-time winter forecasts and 33 winter hindcasts -- simulations of winters dating to 1972 -- to verify the model called sCast, short for seasonal forecast model.
"sCast works well in accurately predicting winter conditions over much of the eastern United States and Northern Eurasia," said Jay Fein, program director in the National Science Foundation's Division of Atmospheric Sciences, which funded the research.
"Dynamical model prediction of winter climate remains a formidable challenge, and statistical approaches such as Cohen's continue to be a valuable alternative," added Fein.
The results of the study appear in the Journal of Climate.
