A comparison of tracking methods for extreme cyclones in the Arctic basin

Dramatic climate changes have occurred in recent decades over the Arctic region, and very noticeably in near-surface warming and reductions in sea ice extent. In a climatological sense, Arctic cyclone behaviour is linked to the distributions of lower troposphere temperature and sea ice, and hence the monitoring of storms can be seen as an important component of the analysis of Arctic climate. The analysis of cyclone behaviour, however, is not without ambiguity, and different cyclone identification algorithms can lead to divergent conclusions. Here we analyse a subset of Arctic cyclones with 10 state-of-the-art cyclone identification schemes applied to the ERA-Interim reanalysis. The subset is comprised of the five most intense (defined in terms of central pressure) Arctic cyclones for each of the 12 calendar months over the 30-yr period from 1 January 1979 to 31 March 2009. There is a considerable difference between the central pressures diagnosed by the algorithms of typically 5–10 hPa. By contrast, there is substantial agreement as to the location of the centre of these extreme storms. The cyclone tracking algorithms also display some differences in the evolution and life cycle of these storms, while overall finding them to be quite long-lived. For all but six of the 60 storms an intense tropopause polar vortex is identified within 555 km of the surface system. The results presented here highlight some significant differences between the outputs of the algorithms, and hence point to the value using multiple identification schemes in the study of cyclone behaviour. Overall, however, the algorithms reached a very robust consensus on most aspects of the behaviour of these very extreme cyclones in the Arctic basin.