Skillful Seasonal Prediction of Winter Blocking and Extreme Temperature Frequency

Atmospheric blocking is a significant producer of extreme weather. Despite its importance, seasonal prediction of blocking remains challenging. A statistical model is developed to predict the winter (Dec-Feb) seasonal blocking frequency over Eurasia using predictors derived from the September and October monthly mean Arctic sea ice, 70-hPa geopotential height, and the Atlantic sea surface temperature. The model captures over half of the interannual variance of blocking frequency during 1979-2016. Given the connection between atmospheric blocking and extreme weather, the same predictors are used to predict the winter seasonal occurrence of extreme warm and cold temperature. Skillful predictions of temperature extremes exist over Greenland and parts of the Eurasian continent. The dynamical mechanisms underlying the skillful predictions are further explored using reanalysis data.