Search Results

You are looking at 1 - 10 of 28 items for

  • Author or Editor: Anne K. Smith x
  • All content x
Clear All Modify Search
Anne K. Smith

Abstract

The major stratospheric sudden warming of February 1979 was characterized by the interaction of the zonal flow with a wavenumber 2 perturbation that was eastward traveling in the early stages of the warming, and slowed to stationary around the time of the wind reversal. Steady state calculations by Smith and Avery 1987 of wave 2 structure for the observed wind fields during this prewarming period indicate a sharp resonance of eastward traveling wave 2. The frequency range for which wave 2 is resonant shifts towards lower (stationary) values with increasing time. This suggests that resonant self-tuning may have played a role in the 1979 sudden warming. In self-tuning, a resonant, or near resonant wave, grows and interacts with the basic state; the change in the basic state modifies the conditions for resonance and the wave frequency then shifts in such a way that it continues to stay near the resonant frequency and to grow.

A quasi-linear time-dependent model is used to investigate several questions about the role of resonance in the 1979 warming. The model is global and has boundaries at the earth's surface and at 100 km to minimize the impact of artificial boundaries on resonance. The basic state is initialized with observations based on the LIMS data for February 1979. A vorticity perturbation with a constant phase speed is applied in the upper troposphere.

Model sensitivity studies are performed to separate those conditions that lead to a sudden warming from those that do not. These tests indicate that the warming simulation has a weak dependence on the frequency of the wave forcing and on the day used for initializing the basic state. Eastward phase progression of the wave 2 forcing is necessary to simulate the warming (for a realistic amplitude of wave forcing), although the model results are not sensitive to the exact frequency. The forcing frequencies for which warmings develop in these tests are close to the resonant frequencies found in steady state calculations using the observed winds. The range of wave frequencies that lead to a sudden warming shifts toward lower frequencies with progressive initialization days.

The zonal wind fields computed by the model during the simulation differ in some ways from the observed wind changes, particularly in the extreme suddenness of the model warming, but overall the winds have a similar structure to those observed. Some of the model-generated zonal wind fields permit very strong resonance for a steady-state traveling wave, thus indicating that an artificial boundary at the tropopause is not necessary for resonance in this simple model. Zonal wind fields generated by the model begin to exhibit resonances one or two days after the beginning of the model integration. The resonance frequency then shifts with increasing time toward the frequency of the prescribed forcing for several days, and then shifts back towards low frequency. The instantaneous frequency of the evolving wave in the model integration follows the same pattern, but remains consistently to one side of the steady-state resonant frequency until the warming has begun. This behavior is a clear signal that self-tuning is occurring in the model warming simulation. The mean flow and wave frequency evolve rapidly during this self-tuning process, and therefore the question remains of how far the analogy to resonance, which is essentially a steady-state phenomenon, can be pushed.

Full access
Anne K. Smith

Abstract

A three-dimensional primitive equation model of the stratosphere is used to investigate the phenomenon of preconditioning for sudden warmings. A flow is said to be preconditioned when the stratosphere is in a state such that a realistic planetary wave pulse from below will lead to a major sudden stratospheric warming. Two observed sudden warmings (February 1979 and February 1989) are simulated by the model. The initial fields are based on zonal-mean observations for individual days during these two winters. The observed wave evolution over a 20-day period is imposed at the lower boundary (near 250 mb). All of the days chosen for initialization and for the beginning of the lower-boundary wave evolution correspond to a time about nine days prior to either one of the major sudden warmings or a minor sudden warming of the same winter. For comparison, model runs are initialized with hybrid initial fields constructed by splicing together upper-level (above 10 mb) data from one day and lower-level data from another time during the same winter. The results of the model runs indicate that a major warming occurs for certain initial fields regardless of whether the lower-boundary wave evolution corresponds to that observed or was taken from another period with a minor warming. In most of the model integrations, the occurrence of a warming in the model depends only on the lower-level winds and is independent of the upper-level winds.

Full access
Anne K. Smith

Abstract

Madden and Labitzke reported an exceptionally large 18-day traveling wave 1 in the troposphere and lower stratosphere during January 1979. Observations from the LIMS instrument on Nimbus 7 indicate that during this period the regular westward phase regression extended upward as far as the lower. mesosphere. Alternate destructive and constructive interference between the transient wave and the time mean wave 1 resulted in large variations in wave amplitude with time. Analysis of the zonal wind deceleration during the minor sudden warming of that month indicates that this wave 1 transience was the primary cause of the warming.

Prior to the large wave 1 amplitude growth immediately preceding the sudden warming (18-22 January), there were two other pulses of exceptionally large wave 1 amplitude (>2000 m peak amplitude) separated by about ten days. Thew earlier pulses were not associated with the traveling wave, which first appeared around 10 January. The change in the zonal wind associated with these first two pulses was much smaller. The amplitude growth during these earlier pulses did not persist as long, and was smaller in the lower stratosphere. Either of these factors may be the reason that these pulses did not cause strong wind deceleration.

Full access
Anne K. Smith

Abstract

Data from the LIMS instrument for January 1979 are used to provide further evidence for the often observed vacillation between the amplitudes of waves 1 and 2 in the stratosphere. The vacillation is shown to result primarily from nonlinear wave-wave interactions within the stratosphere. Two ways of interacting nonlinearity are discussed. In the first, the basic state is defined to include large amplitude waves as well as the mean zonal wind. A forced wave propagates with respect to this asymmetric basic state, which can lead to changes in the conventional zonal wavemumber measured at one latitude. The other view of nonlinearity, interaction of wave with the zonal flow and with other wavenumber are considered separately. Wave-wave interactions among waves 1, 2 and 3 are calculated.

The derivation and computation of wave-wave interaction terms in the potential enstrophy balance are given. The observations indicate that enstrophy transfer among waves can be substantial even when the amplitude of one of the contributing waves is small. The computed enstrophy balance also demonstrates that wave-wave interactions can have a large effect on the interaction of waves with the men flow.

Full access
Anne K. Smith

Abstract

Satellite observations indicate that quasi-stationary planetary waves often exist to at least 100 km in the winter mesosphere. Waves are also seen in the summer upper mesosphere. A three-dimensional numerical model was used to simulate these waves and to diagnose the physical processes involved. The waves simulated in the model closely resemble observed waves. Several model runs that isolate specific processes are used to determine the relative importance of two forcing mechanisms. In the model, planetary waves that propagate from below are significantly damped at the altitude where gravity wave drag becomes large (about 75 km in the winter midlatitudes) or below if a reversal in the mean wind is encountered. Momentum forcing associated with breaking gravity waves that have been filtered by planetary-scale wind variations below acts to generate planetary waves in the middle and upper mesosphere. The amplitude from in situ forcing by gravity wave breaking exceeds the amplitude from the upward-propagating Rossby wave above 80 km.

Full access
Anne K. Smith
Open access
Anne K. Smith

Abstract

Quasi-stationary planetary-scale longitudinal variations are found in the upper mesospheric winds measured during winter by the HRDI satellite instrument. These are negatively correlated with eddy winds in the stratosphere. Two different mechanisms are proposed to explain the mesospheric perturbation winds and their anticorrelation with stratospheric winds: 1) Planetary waves propagate through the stratosphere and into the mesosphere, with a phase shift of one-half cycle and 2) mesospheric perturbations are forced in situ by gravity waves whose spectrum has longitudinal asymmetries due to filtering by planetary waves in the stratosphere. The first mechanism is more consistent with the observations during Southern Hemisphere late winter (August) and also may explain the observations during Northern Hemisphere early winter (December). The second mechanism gives a more consistent explanation for the Northern Hemisphere late winter observations, as previously shown by the author. The hemispheric differences are reproduced in two linear model calculations in which the only difference is the background zonal mean wind.

Full access
Anne K. Smith

Abstract

Mesospheric horizontal wind data from the High Resolution Doppler Image (HRDI) satellite instrument are used to investigate the longitudinal variation of nontidal winds in the upper mesosphere and their relationship with winds in the stratosphere. The period covered by the study is late winter (February–mid-March) for 1992–1994. The results indicate that there is a negative correlation between longitudinal variations of the zonal winds in the stratosphere and those in the mesosphere, and between longitudinal variations of the meridional winds in the stratosphere and mesosphere. The data are examined to determine which of several possible causes is responsible. The possibilities are 1) that planetary-scale disturbances are produced in situ by gravity wave drag in the mesosphere that is not zonally symmetric because of filtering by the planetary waves in the stratosphere, 2) that asymmetries are caused by longitudinal variations in gravity wave drag due to azonal source distributions that are coincidentally tied to the standing wave patterns in stratospheric winds; or 3) that a planetary Rossby wave propagates from the atmosphere to the mesosphere and undergoes a phase shift. The observations are overall most consistent with the first hypothesis, that is, that filtering of gravity waves by zonal asymmetries in the stratospheric winds leads to in situ generation of planetary-scale structures in the winter upper mesosphere.

Full access
Anne K. Smith and Susan K. Avery

Abstract

A simple numerical model of the stratosphere has been used to examine the possibility that a resonant growth of wave 2 was responsible for the 1979 major sudden warning. The model solves for linear steady state solutions to the quasi-geographic wave equation in the presence of realistic damping. The basic state is taken from observations (NMC and LIMS), and the frequency of the wave forcing is varied over a wide range. The model results show that in the days during the initial observed amplification of wave 2 (14–15 February), a clear resonant mode existed. The maximum response is for a wave moving eastward with a period of 12–16 days. Another peak at very low frequency (period greater than 100 days) occurs on 22 February. Other days during the period 12–24 February show weaker, but nevertheless significant peaks for particular frequencies. The frequency of the maximum is lower for later days and is nearly stationary at the height of the warming around 21 February. This frequency shift found in the model corresponds closely to the observed wave behavior.

Although the details of the results vary with changes in the model resolution or lower boundary position, the resonant wave does not disappear. However, when the wave forcing is applied at the earth&'s surface rather than in the tropopause region, no resonance occurs. To test the effect of the lower boundary, the troposphere-stratosphere model was run with an internal vorticity forcing similar to the structure of the observed wave 2 in the troposphere. In this case the frequency dependence of the amplitude within the stratosphere was similar to that of the model with a tropopause boundary, although the magnitude was considerably smaller. This suggests that for resonance to have occurred, a planetary scale disturbance that did not propagate from the surface must have been maintained in the upper troposphere. The two well-developed blocking ridges present in the troposphere during this period may have contributed enough to planetary wave 2 to provide the necessary boundary conditions.

Full access
Lawrence V. Lyjak and Anne K. Smith

Abstract

Three-dimensional winds derived from LIMS satellite observations for the 1978/79 winter are used to compute the mean Lagrangian motion in the winter stratosphere. Material tubes of air parcels are initialized every 4 days and followed for periods of 10 days each. The initial positions of the tubes are chosen so that they lie along contours of constant geopotential and potential temperature. Maps of air parcel distributions give a qualitative picture of the degree of deformation of the material tubes with time. In addition, quantitative measures of the Lagrangian mean velocity and dispersion are computed.

During quiet periods, when the zonal wind is strong and the vortex is nearly axisymmetric, the air parcel tubes tend to remain coherent for the full 10 days of the integration. When the wave amplitudes are large, many of the tubes break and the parcels disperse. During the observed minor sudden warming, those tubes closest to the vortex center remained coherent with little distortion. In contrast, during the major sudden warming every material tube in the stratosphere was broken, and there was extensive mixing between air parcels from low and high latitudes.

Lagrangian mean vertical motion tended to be smaller than the motion in the transformed Eulerian coordinate system, which is sometimes used to represent the mean Lagrangian flow. The horizontal velocities determined from the Lagrangian parcel trajectories do not in general correspond well with the transformed Eulerian velocities. Largest differences in horizontal winds occur for situations during which the tubes underwent extensive deformation and the dispersion of air parcels was large. This suggests that the transformed Eulerian circulation is not capable of representing the horizontal Lagrangian motion when a large part of the latter is due to dispersive motion rather than to net displacements of coherent material tube.

Full access