The Arctic Oscillation AO

In 1998, a paper appeared in Geophysical Research Letters by D. Thompson, then a Ph.D. student at the University of Washington, Seattle, and his advisor, J. Wallace (Thompson and Wallace, 1998). Follow-on papers were published shortly thereafter (Thompson and Wallace, 2000; Thompson etal., 2000; Wallace, 2000). Thompson and Wallace (1998) argue that NAO should be considered as a regional manifestation of a more basic mode of SLP variability, which has come to be known as the AO, or NAM. They define the AO as the leading mode of SLP variability from an EOF analysis for the Northern Hemisphere winter (using data for 1900-97). Like the NAO (as based on the station index), the AO is best expressed during winter. The major characteristics (Figure 11.10) are a primary center of action over the Arctic, focused in the area of the Icelandic Low, and opposing, weaker centers of action in the North Atlantic and

Figure 11.10 Normalized leading EOF of winter (December-March) SLP anomalies over the Northern Hemisphere (20-90° N). The pattern (accounting for 23.5% of the variance in Northern Hemisphere SLP) is displayed in terms of amplitude (hPa) obtained by regressing the hemispheric SLP anomalies upon the leading principal component time series. The contour interval is 0.5 hPa and the zero contour has been excluded. The data cover the period 1899-2001 (from Hurrell et al., 2003, by permission of AGU).

Figure 11.10 Normalized leading EOF of winter (December-March) SLP anomalies over the Northern Hemisphere (20-90° N). The pattern (accounting for 23.5% of the variance in Northern Hemisphere SLP) is displayed in terms of amplitude (hPa) obtained by regressing the hemispheric SLP anomalies upon the leading principal component time series. The contour interval is 0.5 hPa and the zero contour has been excluded. The data cover the period 1899-2001 (from Hurrell et al., 2003, by permission of AGU).

North Pacific. The interpretation is that if pressures are low (high) over the Arctic, they are high (low) over the North Atlantic and North Pacific. As is obvious from Figure 11.8, the Arctic and Atlantic AO centers of action correspond closely to the NAO centers of action. Looking back at Figure 11.9, the time series of the AO (NAM) and NAO are also very similar. Thompson and Wallace (1998) were not the first to perform a Northern Hemisphere analysis. As far as we are aware, the first arguments for a more fundamental mode of variability than the NAO come from the studies of Lorenz (1951), Kutzbach (1970) and Wallace and Gutzler (1981).

Thompson and Wallace (1998) interpret the winter AO as the surface signature of modulations in the strength of the circumpolar stratospheric vortex. We saw in Chapter 4 that the mean winter circulation in the stratosphere is primarily zonally symmetric (or annular), while that of the troposphere and the surface is characterized by a more wavy pattern. The primary pattern of variability has the same basic features - in the stratosphere, it is primarily in the strength of the annular vortex, while for the troposphere and surface, it is the wavy pattern. The NAO framework focuses on the Atlantic sector, where the surface and tropospheric variability is strongly expressed. However, in the AO framework, the winter stratospheric variability and that for the troposphere and surface are not considered in terms of different modes, but as elements of a single, inherently annular modal structure that are physically linked through most of the depth of the atmosphere.

The winter AO resembles a similar mode in the Southern Hemisphere, which is more symmetric from the stratosphere through to the surface. Put differently, its regional expressions are not as pronounced. The stronger asymmetry of the winter AO (seen especially in terms of the focus of the strong Arctic center of action to near the Icelandic Low) is viewed in terms of distortions upon the inherently annular structure driven by the land-sea distribution (the presence of a warm ocean extending to high northern latitudes) and orography. Were it not for these distorting influences, the AO would more closely resemble its more symmetric Southern Hemisphere counterpart. In turn, variations in the phase and strength of the winter AO, seen in terms of mass exchanges between the Arctic and mid-latitudes, are viewed in terms of controls from the stratosphere. While the details of such controls are still not well understood, one appears to involve the sudden stratospheric warmings examined in Chapter 4. Such warmings weaken or even reverse the stratospheric zonal winds. This in turn alters conditions for upward propagation of planetary waves, which ultimately may be seen in transitions from the positive to negative mode of the AO.

This emphasis on the winter stratosphere is also relevant to the interpretation of recent changes. The upward tendency of the NAO in recent decades is thought of as being forced from the surface by slowly varying North Atlantic or tropical SSTs, the latter finding support from recent modeling studies. By contrast, in the AO framework, thinking turns to linkages with stratospheric ozone depletion or the buildup of greenhouse gases that are associated with stratospheric cooling. The basic idea is that such changes can "spin up" the polar stratospheric vortex, with the effects propagating down to the surface, leading to a positive AO state (e.g., Gillett et al., 2003;

Yukimoto and Kodera, 2005; see also Chapter 4). Thompson et al. (2000) note that more than one-third of the observed winter AO trend can be linearly related to observed stratospheric ozone depletion in the Northern Hemisphere.

The concept of the AO as more fundamental than the NAO was not met with universal enthusiasm. While many in the research community were quick to hail the AO as the "new paradigm", others were skeptical. Deser (2000) concluded that the correlation between the Pacific and Azores centers of action is not significant and that the AO cannot therefore be viewed as reflecting such a teleconnection. Ambaum et al. (2001) find that even the correlation between the Pacific and Icelandic-Arctic centers of action is not significant. They argue that the AO is mainly a reflection of similar behavior in the Pacific and Atlantic basins. Other studies contributing to the debate include Fyfe (2003) and Ambaum and Hoskins (2002).

Our understanding of the physics of the NAO and AO will continue to evolve. Is the primary control from the stratosphere or the surface? While it would seem that elements of both views are inherently correct, the physical concept of the AO in terms of a stratospheric link breaks down in summer. During summer, the stratospheric circulation is easterly and the stratosphere and troposphere are isolated from each other (Chapter 4). Regardless, from a practical view, indices of the AO and NAO are very highly correlated and for many applications one can use either paradigm. The following sections take this approach. Differences in interpretation (NAO vs. AO) are addressed as needed. Further aspects of possible surface forcing on NAO/AO variability are reviewed in Section 11.5.

0 0

Post a comment