Evaluation of Blocking Indices Using Their Relationship with Temperature Extremes

Atmospheric blockings are quasi-stationary anticyclones that persist for several days to weeks and block or divert the jet stream. They cause persistent weather patterns and can lead to weather extremes such as heat waves and cold spells. A number of blocking indices have been developed in the past 60 years to identify or measure blocking; however, these indices often use different definitions for blocking events and produce conflicting blocking statistics and trends. The statistics and trends can also be sensitive to the choices of some of the parameters of these indices such as thresholds, latitudes. Furthermore, the connection between the events identified by these blocking indices and the weather extremes, an important motivating factor for the renewed interest in blocking, is not clear and remains to be examined. In this work, we evaluate several blocking indices by calculating the correlation between interannual variations in the hemispheric area experiencing near-surface extreme warm (cold) temperature averaged over a summer (winter) and the hemispheric area experiencing blocking averaged over the same season in the ERA-interim reanalysis data. This evaluation could help us choose a skillful blocking index (in the context of weather extremes), and help us quantify and understand future changes of temperature extremes due to changes in blocking activity.

We have used several types of 2D blocking indices from the literature that use Z500 reversal (Scherrer et al 2006 and Masato et al 2013), PV (Pfahl and Wernli 2012), and Z500 reversal and anomaly (Dunn-Sigouin and Son 2013 and Hassanzadeh et al 2014) in this study. Depending on the index, blocking-extremes correlations as high as 0.68 (0.53) for hot (cold) extremes and as low as -0.04 (0.00) are obtained. The framework is also used to investigate the sensitivity of the correlations to some of the parameters used in these indices and to improve the performance of these indices in identifying extreme-causing blocking patterns.