No abstract

Nimbus 7 Limb Infrared Monitor of the Stratosphere (LIMS) observations are used to study the evolution of potential vorticity in the stratosphere, January–February 1979. Daily analysis of this quantity at 850° and 1200°K provides circumstantial evidence of planetary wave “breaking” by which air parcels undergo rapid and irreversible separation from the circumpolar vortex during stratospheric warmings. Complementing this effect is the advection of subtropical, low‐vorticity air into the polar region. Temporal evolution of the size, shape, and orientation of the main circumpolar vortex is revealed very clearly by the potential vorticity field. All three factors are important, although some have been emphasized more strongly in previous literature. The size of the vortex determines the range of latitudes over which planetary Rossby waves are able to propagate vertically. Diminution of vortex area during the observed warmings is believed to precondition the flow, focusing subsequent Rossby wave activity into the polar cap, as in the major warming of late February 1979. The shape of the vortex undergoes both reversible and irreversible deformation. Examples of irreversible deformation are seen in the advective formation of extended high‐vorticity tongues over subtropical latitudes in connection with the warmings of late January, early February, and late February 1979. Two of these were recently discussed by McIntyre and Palmer. Reversible deformation is observed in the sudden cooling and concurrent wave 1, wave 2 vacillation, after the January warming. The orientation of the vortex can also be important, as in the period of rotation leading up to the major wave 2 warming of late February 1979. We suggest that the orientation of the vortex be included as part of the preconditioning process, in accord with numerical results of Butchart et al. We briefly consider the vertical structure of potential vorticity and ozone on two disturbed days in late January 1979. Meridional cross sections of potential vorticity in the middle stratosphere resemble ozone cross sections shown in the work of Leovy et al. and exhibit good vertical coherence extending into the lower mesosphere, where ozone (unlike potential vorticity) no longer serves as a tracer of the motion.

No abstract

Abstract. Analysis of rawinsonde and rocketsonde data at Ascension Island (7.6øS, 14.4øW) and Kwajalein (8.7øN, 167øE) in 1962-1991 suggests that the quasi-biennial oscillation (QBO) in the middle stratosphere is synchronized with the seasonal cycle and that descending westerly phases of the stratopause semiannual oscillation (SAO) are strongly influenced by the underlying QBO. The effect of the seasonal cycle on the QBO in the middle stratosphere is revealed in two, perhaps unrelated, observations: first, a tendency for deseasonalized QBO westerly maxima to occur in local winter (or to avoid local summer); second, a smooth, uninterrupted connection between descending SAO westerly shear zones and the formation of a new QBO westerly shear zone aloft. The timing of deseasonalized QBO westerly maxima in the middle stratosphere allows a simple composite of 2-and 3-year cycles to be constructed from the data, illustrating the effect of the QBO on descending westerly phases of the stratopause SAO. IntroductionThe quasi-biennial oscillation (QBO) of the equatorial stratosphere is so named because the average period of the oscillation is slightly longer than 2 years. Almost 20 cycles have been observed since rawinsonde data first became available in the early 1950s; the average period is now 28.4 months. Individual cycles range from ---22 to 36 months, i.e., from about 2 to 3 years.The prefix "quasi" would apply equally well to an oscillation with continuously variable period, or an oscillation consisting of discrete periods (e.g., 24 months, with a few 30-or 36-month cycles) strung together in some regular or irregular fashion. The period of the QBO is customarily measured at mandatory rawinsonde pressure levels, e.g., 50, 30, or 10 mbar [Dunkerton and Delisi, 1985;Angell, 1986;Naujokat, 1986;Dunkerton, 1990]. This approach, based on one-dimensional time series, leads to the conclusion that the QBO period is variable, tending to form a continuous distribution as time progresses. Exact synchronization with the seasonal cycle is not observed, although as noted by Dunkerton [1990], the deseasonalized mean flow acceleration, distribution of phase onsets, and period of the QBO evidently depend on the time of year. This dependence on the seasonal cycle is due, in part, to a pronounced tendency for descending QBO easterlies to "stall" near 30 mbar between July and February [Naujokat, 1986]. Modulation of the QBO by the seasonal cycle can be distinguished from exact synchronization, which would produce one or more delta functions in the distribution of onset times. Both types of behavior were simulated in a QBO model [Dunkerton, 1990] Data AnalysisData from the historical rocketsonde network were obtained from a variety of sources and processed to form monthly means 26,107

Abstract.Data