A Predictability Study of a Polar Low Linked to a Tropopause Polar Vortex

This study investigates physical processes that influence the predictability of a polar low that is linked to a tropopause polar vortex (TPV). A database of TPVs constructed using the ERA-Interim is compared to the Sea Surface Temperature and Altimeter Synergy for Improved Forecasting of Polar lows (STARS) database of polar lows in the Nordic Seas to identify polar lows that may be linked to the TPVs in this database. Results indicate that 104 out of a total 140 polar lows, or 74.3%, may be linked to TPVs. An ERA5-based multiscale analysis is performed for a Barents Sea polar low that is linked to a TPV transported equatorward along a tropospheric-deep baroclinic zone during 10­–11 February 2011. The analysis shows that the amplification of two ridges flanking the Barents Sea likely supports the equatorward transport of the TPV, and that the interaction between the TPV and the tropospheric-deep baroclinic zone likely plays an important role in the evolution of the polar low.

The 51-member ECMWF Ensemble Prediction System is utilized to evaluate the forecast skill of the evolution of the polar low. Ensemble members initialized 30 h prior to polar low genesis are separated in two groups: the most accurate and the least accurate members in terms of combined track and intensity error of the polar low. Normalized composite differences between the most and least accurate groups suggest that differences in the position and structure of the two ridges may contribute to differences in the positions of the TPV and the tropospheric-deep baroclinic zone. These differences in the positions of the TPV and the tropospheric-deep baroclinic zone likely contribute to differences in the track and intensity of the polar low between the two groups. A statistical piecewise potential vorticity (PV) regression technique will be applied to upper-level PV differences between the two groups to elucidate how differences in the position and structure of the two ridges may contribute to differences in the position of the TPV. The extent to which the results of the evaluation of the forecast skill of the evolution of the polar low for the ensembles initialized 30 h prior to polar low genesis are similar to those for other forecast lead times will be tested.