• Introduction
• The composite results
• The waveguide modulations
• The impact on the troposphere
• Summary
• References

FIGURES

• Figure 1
• Figure 2
• Figure 3
• Figure 4

Introduction

It has been a controversy whether, and how, stratospheric changes have a direct effect on tropospheric weather and climate. Recently, Baldwin and Dunkerton (1999) have found downward propagation of AO (Arctic Oscillation) anomaly in low-pass-filtered data with a propagating time of three weeks from 10 hPa to surface. They also noticed that not every AO anomaly in the upper stratospheric propagated down, only those with large amplitude and persistence could have a clear signature through the troposphere. The purpose of this study is to find out under what conditions an upper stratospheric anomaly can propagate down to the troposphere, and when this happens, how the tropospheric circulations are affected. We use polar temperature anomaly (70N-90N) as an indicative feature of downward propagation, which is calculated at different pressure levels from 1000 hPa to 10 hPa based on 22 years (1978-1999) of NCEP/NCAR Reanalysis daily data (Kalnay et al., 1996). We mainly focus on winter-spring seasons because in summer stratosphere temperature anomaly is relatively small. The anomaly is normalized by its standard deviation at different level to minimize the density effect. Then we select those warming episodes with large amplitudes in the upper stratosphere, and divide them in two categories: a propagating feature is defined as the case in which temperature anomaly is greater than 2-standard deviation at 10 hPa and followed by a temperature anomaly greater than 1.5-standard deviation at 200 hPa, and a non-propagating feature as that temperature anomaly is also greater than 2-standard deviation at 10 hPa but followed by a temperature anomaly smaller than 1-standard deviation at 200 hPa.

The composite results

Instead of studying individual episodes, we use composite method to study common characteristics of each category. Although the lasting time varies in different episodes, we select each period consisting of 40 days and arrange those periods according to their maximal amplitudes at 10 hPa without lose of generality. The composite results include temperature anomaly, zonal mean wind, Eliassen-Palm flux and its divergence. The E-P diagnostics are used to examine wave-mean flow interactions (Andrews and McIntyre, 1976; Edmon et al., 1980), which are very important links between stratosphere and troposphere. All of the quantities are calculated from the NCEP/NCAR Reanalysis daily data. The composite figures of propagating case are given in Figure 1. The downward propagating feature seems to consist of two stages. The initial stage is faster, taking just a few days down from 10 hPa to 50 hPa level, and the following stage is slower, with about three weeks of propagating time from 50 hPa to 200 hPa. In the first stage, the upper stratosphere westerly wind reversed to easterly rapidly. The "critical line" (zero wind line) descended to below 50 hPa. Because waves cannot propagate in easterly wind (Charney and Drazin, 1961), the altitude of wave transport also descended with the "critical line." The largest zonal wind deceleration took place in the same time and same place of the maximum wave forcing, as shown by the convergence of E-P flux. Very large poleward heat flux ( which is proportional to the vertical component of E-P flux) occurred in this stage. After a short interruption, a second pulse of wave flux took place, which again forced the polar wind to decelerate and the "critical line" to descend. Although the second pulse of wave is not so strong as the first pulse, it plays a deciding role for warm temperature propagating downward, as will be discussed later. Figure 2 shows the composites of the non-propagating case. Compared to the propagating case, there is no significant warming below 100 hPa although the warm anomaly at 10 hPa is about as large in the two categories. There is no clear downward propagating feature in the upper stratosphere. The zonal wind was weakened but did not change direction. This is consistent with the weaker wave forcing in the upper stratosphere, with the convergence of E-P flux only half as large as that in the propagating case. It was also limited at higher altitudes and lasted for a shorter period. The amplitude of vertical component of E-P flux was initially as large as that in the propagating case, but faded away earlier near the tropopause. There was no second pulse of wave flux following up, therefore, the upper warm anomaly did not extend to lower altitudes.

Figure 1. Composites for the propagating warm anomalies. All quantities shown are 70N-90N average and smoothed by a 5-day running average. (a) Polar temperature anomaly. The contour interval is 2 K; (b) Zonal wind. The contour interval is 5 m/s and the heavy purple line indicates zero wind ; (c) E-P flux divergence. The contour interval is 2 m/s per day, zero line is omitted, and negative values are shown in blue; and (d) Vertical component of E-P flux. The contour interval is 2 kg/m/s/s, and contours with value greater than 16 are omitted.

Figure 2. Same as Figure 1 except for the non-propagating warm anomalies.

The waveguide modulations

Comparing Figures 1 and 2, it is evident whether an upper stratospheric warm anomaly propagates down depends on how planetary waves interact with stratospheric zonal mean flow. In particular, a continuous wave transport is essential in the period following the upper stratospheric warming. As well-known, large-scale planetary waves originate in the lower troposphere, and the largest wave amplitudes are observed in the subpolar latitudes centered near 60N. Prior to stratospheric sudden warming events, the zonal mean flow is in a preferred state for large waves propagating upward and poleward. However, the condition of zonal mean flow is changed during the warming, which consequently affects the way of wave propagation, because waveguides are modulated by zonal mean flow. In Figures 3 we compare the quasi-geostrophic refractive index for wavenumber 1 stationary wave, in the post-warming period (Day 21-40) for the propagating and non-propagating cases. Note that waves tend to propagate toward large positive values of the index and avoid negative values of the index. The refractive index became very large in the 60N-70N upper troposphere in the propagating case, so that mid-latitude waves were strongly refracted poleward. The increase in the value of refractive index was mainly due to the large decrease in zonal mean zonal wind associated with the reversal of polar westerly flow and the descending of "critical line." In the non-propagating case, however, the refractive index was smaller in the subpolar troposphere, and there was little poleward wave transport.

Figure 3. (a) Average quasi-geostrophic refractive index for wavenumber 1 stationary wave (contours) and E-P flux (vectors) of Day 21 - 40 for the propagating case. (b) Same as (a) except for the non-propagating case. Contour levels are -100, -50, -25, 0, 25, 50, 100, ... and contours greater than 400 are omitted.

The impact on the troposphere

In Figure 4 we compare the average wind speed at 200 hPa in the first and second half of the 40-day period, for the propagating and non-propagating cases respectively. To highlight the jet stream we only plot the contours greater than 30 m/s. In the propagating case, the axis of Atlantic jet stream shifted to south by about 5 degrees of latitude, and the alignment of the axis became more zonal, in the post warming period. The Pacific jet stream did not change much except for somewhat south movement near the exit of the jet (140W-170W). On the other hand, in the non-propagating case the position of Atlantic jet stream remained almost unchanged, and the Pacific jet stream also moved somewhat to south near the exit. It seems that the Atlantic jet stream was more sensitive to the downward propagating stratospheric warm anomaly than the Pacific jet stream. The shift of the upper tropospheric jet stream over Atlantic Ocean affected tropospheric weather systems such as Atlantic storm tracks.

Figure 4. (a) Averaged wind speed at 200 hPa in the propagating case. Blue contours are Day 1- 20 average and red contours are Day 21 - 40 average. Contour interval is 10 m/s. Contours of wind speed less than 30 m/s are omitted. (b) Same as (a) except for the non-propagating case.

Summary

From the Northern Hemisphere polar temperature anomaly data we find that the stratosphere behaves sometimes like a "conductor" which allows a warm anomaly to propagate from the upper stratosphere to the troposphere, and sometimes like a "resistor" which prohibits downward propagation. The "conductivity" of polar atmosphere is determined by the strength of wave activity and the structure of zonal mean flow. If the initial wave forcing is so strong and persistent that could reverse the polar westerly wind, then the warm anomaly would descend with the "critical line" because waves cannot propagate into easterly wind. Further downward propagation of the warm anomaly requires continuous energy supply by tropospheric waves, which can be induced by the positive feedback due to changes in zonal mean flow in the mid-high latitudes. In the propagating case, the zonal mean flow in the post-warming period changed to a state favorable for poleward transport of tropospheric waves, which induced a second pulse of wave energy flux to reinforce the warm anomaly in the lower stratosphere and extend its downward propagation. We may call this condition of polar atmosphere as "conductive," compared with the "resistive" atmosphere which lacks descending "critical line" and continuous supply of wave energy. The impact of downward propagating warm anomaly could be felt in lower tropospheric weather systems through the link of upper tropospheric jet streams. Changes in the strength and position of subtropical jet are often indicative of changes in weather patterns. In addition, the south shift of the North America and Atlantic jet stream by the downward propagating stratospheric polar warm anomaly is consistent with the negative phase of AO or NAO (Thompson and Wallace, 1999).

References

Andrews, D. G., and M. E. McIntyre, 1976: Planetary waves in horizontal and vertical shear: The generalized Eliassen-Palm relation and the mean zonal acceleration, J. Atmos. Sci., 33, 2031-2048.

Baldwin, M. P., and T. J. Dunkerton, 1999: Propagation of the Arctic Oscillation from the stratosphere to the troposphere, J. Geophys. Res., 104, 30937-30946.

Charney, J. G., and P. G. Drazin, 1961: Propagation of planetary-scale disturbances from the lower into the upper atmosphere, J. Geophys. Res., 66, 83-109.

Edmon, H. J., Jr., B. J. Hoskins, M. E. McIntyre, 1980: Eliassen-Palm cross sections for the troposphere, J. Atmos. Sci., 37, 2600-2616.

Kalnay, M. E., and coauthor, 1996: The NCEP/NCAR Reanalysis project, Bull. Amer. Meteorol. Soc., 77, 437-471.

Thompson, D. W. J. and J. M. Wallace, 1999: Annual modes in the extratropical circulation, Part I: Month-to-month variability, J. Climate, 13, 1000-1016.