Using Bayesian Networks to Investigate the Influence of Subseasonal Arctic Variability on Midlatitude North Atlantic Circulation

Abstract Recent enhanced warming and sea ice depletion in the Arctic have been put forward as potential drivers of severe weather midlatitudes. Evidence a link between midlatitude atmospheric circulation is growing, but role processes relative to other remains unknown. Arctic–midlatitude connections North Atlantic region are particularly complex important due frequent occurrence winters recent decades. Here, dynamic Bayesian networks with hidden variables introduced field assess their suitability for teleconnection analyses. Climate constructed analyze variability at 5-day monthly time scales during winter months years 1981–2018. The inclusion number Arctic, midlatitude, tropical allows an investigation into influence compared internal remote drivers. A robust covariability regions amplified two definitions found occur entirely within submonthly scales. Hidden incorporated represent distinct modes stratospheric polar vortex variability, capturing periodic shift average conditions slower anomalous flow. Barents–Kara Seas on Oscillation be strongest 5- 10-day averages, while strongly influences jet