Sensitivity of the blocking-North Atlantic Oscillation relationship to index

Abstract

North Atlantic blocking and the North Atlantic Oscillation (NAO) are two phenomena that have been extensively studied due to their significant spatio-temporal overlap. This work presents an index comparison study applied to this relationship, bringing light to how the strength of it varies considerably depending on blocking index choice and why this could be leading to discrepancies found in previous works. A PV–θ blocking index is used alongside a direction of breaking metric to classify blocking as either cyclonic or anticyclonic based on the Rossby wave breaking occurring at onset. These results are compared against those found using an absolute geopotential height (AGP) index. The analysis is performed using both area-averaged blocking count during winter and at each grid-point across the North Atlantic for all seasons. The study demonstrates that the choice of method significantly affects the results when correlating wintertime blocking and NAO. Blocks found using the AGP index show a much stronger correlation with the NAO compared to those found with the PV–θ index. Other analyses, such as frequency, duration, and composites, suggest that the AGP algorithm detects stronger, more mature, and persistent blocks, which promote a higher correlation with the NAO compared to the shorter-lived PV–θ blocks. Based on this analysis, it can be concluded that different blocking events are being measured between the two indices, contributing to the stark differences in the correlation analysis, each with their own advantages and disadvantages in relating blocking to the NAO.

Appendix

Figures 

Fig. 9

DJF composites of anomalous Z500mb (black contours), precipitation (brown-green shading), and SSTs (blue-red shading) during all blocked days within a given subregion, by index and for the PV–θ blocks, by classification. Classified cyclonic PV–θ blocks by the Greenland, European, and Ural regions are in panels a–c, d–f for PV–θ anticyclonic blocks by subregion, and g, h for AGP blocks. Daily NAO composites are also provided for context and comparison in (k). N represented the number of days included in the composite

9,

Fig. 10

The same as Fig. 9 showing daily blocking event and NAO composites by subregion and index during DJF, but for anomalous sea level pressure (contours) and two-meter temperature (blue-red shading). Contours are every 50 mb

10,

Fig. 11

the same as Fig. 9 but for MAM, showing daily blocking event and NAO composites by subregion and index of anomalous Z500mb (black contours), precipitation (brown-green shading), and SSTs (blue-red shading)

11,

Fig. 12

The same as Fig. 10 but for MAM, showing daily blocking event and NAO composites by subregion and index for anomalous sea level pressure (contours) and two-meter temperature (blue-red shading). Contours are every 50 mb

12,

Fig. 13

The same as Figs. 9 and 11 but for JJA showing daily blocking event and NAO composites by subregion and index of anomalous Z500mb (black contours), precipitation (brown-green shading), and SSTs (blue-red shading)

13,

Fig. 14

The same as Figs. 10 and 12 but for JJA, showing daily blocking event and NAO composites by subregion and index for anomalous sea level pressure (contours) and two-meter temperature (blue-red shading). Contours are every 30 mb

14,

Fig. 15

The same as Figs. 9, 11, and 13 but for SON showing daily blocking event and NAO composites by subregion and index of anomalous Z500mb (black contours), precipitation (brown-green shading), and SSTs (blue-red shading)

15,

Fig. 16

The same as Figs. 10, 12, and 14 but for SON, showing daily blocking event and NAO composites by subregion and index for anomalous sea level pressure (contours) and two-meter temperature (blue-red shading). Contours are every 50 mb

16 and

Fig. 17

Mean seasonal blocking frequency from 1980 to 2019 representing the average percent of blocked days per season when applying the absolute geopotential height (AGP) index. Contours are every 1%

17.