Quasi-Biennial Fluctuations in Sea Level Pressures over the Northern Hemisphere

Abstract

An investigation is made into the presence of quasi-biennial oscillations (QBOs) in sea-level pressure fields over the Northern Hemisphere. Using 55 years (1925–79) of seasonally averaged sea-level pressure anomalies, a series of analyses has been performed in order to systematically isolate and describe the QBOs. A standard empirical orthogonal function (EOF) analysis of the seasonal anomalies reveals significant nonrandomness at QBO periods. The data were then band-pass filtered in order to focus on the QBO, and a second EOF analysis was performed. Finally, a new complex EOF analysis technique was applied to the data. Complex EOFs have the advantage of permitting both standing and propagating modes. The QBO variance in six standard EOFs is essentially confined to four EOFs in the filtered data set and compressed into three complex EOFs. The latter account for 56% of the filtered variance. The dominant mode of the complex EOF analysis was common to all analyses and has been found in many other studies. It is essentially a standing wave pattern corresponding to a high-latitude zonal index with departures in pressure of opposite sign in low and high latitudes. This mode includes elements of the North Atlantic Oscillation, the Pacific–North American teleconnection pattern and the Southern Oscillation, but also differs somewhat from all three. It tends to be phase locked to the annual cycle. The second complex EOF mode C2 exhibits clear propagating characteristics as it evolves in time. The third mode C3 also has some propagating characteristics and, at times, both modes strongly resemble wave trains of quasi-stationary Rossby waves. Neither C2 nor C3 is phase locked to the annual cycle. It appears likely that all three complex EOFs are normal mode responses of the atmosphere to different kinds of forcing. Statistical evidence for phase locking of all three complex EOFs to the QBO in zonal winds in the equatorial stratosphere is not convincing, and the origin of the preferred QBO periodicity remains to be determined.