Differences between the 2018 and 2019 stratospheric polar vortex split events

Two recent occurrences in February 2018 and January 2019 of a dynamic split in the Northern Hemisphere stratospheric polar vortex are compared in terms of their evolution and predictability. The 2018 split vortex was associated with primarily wavenumber-2 wave forcing that was not well predicted more than 7–10 days ahead of time, and was followed by persistent coupling to the surface with strong weather impacts. In 2019 the vortex was first displaced by slow wavenumber-1 amplification into the stratosphere, which was predictable at longer lead times and then split; the surface impacts following the event were weaker. Here we examine the role of large-scale climate influences, such as the phase of the El Niño–Southern Oscillation, the Quasi-biennial Oscillation and the Madden–Julian Oscillation, on the wave forcing, surface impacts and predictability of these two events. Linkages between the forecast error in the stratospheric polar vortex winds with the forecast error in the Quasi-biennial Oscillation and Madden–Julian Oscillation are examined.