An excerpt from

A new weather-forecasting model based on autumn snowfall in Siberia could help meteorologists predict winter temperatures and snowfall in the United States and Europe.

The model results, reported this week in the Journal of Climate, could help make climate prediction more accurate and reliable for fields such as agriculture, water management and general weather risks. At least $3 trillion of the U.S. economy is sensitive to weather conditions, estimates the National Science Foundation (NSF).

Scientists led by Judah Cohen of Atmospheric and Environmental Research, Inc. (AER Inc.) verified the sCast (short for “seasonal forecast”) model with seven real-time winter forecasts and 33 simulations of winters going back to 1972.