Stratosphere-troposphere coupling: implication for prediction of anomalous weather regimes Pogoreltsev A., Ugrjumov A.. Russian State Hydrometeorological University Problems related to nowcasting and specialized hydrometeorological provision of large-scale complex sport events by an example of Olympic games 2008 Bejing, 6-11 April 2009

Previous results (motivation) Nonlinear interaction between quasi-stationary planetary wave (PW) and mean flow in the stratosphere leads to irregular oscillations of the PW amplitude and mean flow intensity, the so-called stratospheric vacillations (Holton & Mass, 1976), and to development the sudden stratospheric warming (SSW, Matsuno, 1971). The vacillations of the mean frow and its deceleration during the SSW events can be considered as the changes of the Northern Annular Mode (NAM) idex. Baldwin & Dunkerton (1999, 2001) found that NAM anomalies sometimes descend from the stratosphere into the troposphere. Hovewer, this result can not be interpret as direct evidence for downward propagation of information (Hardiman & Haynes, 2008). Couglin & Tung (2005) have shown that some of the NAM seen in the observations could be projections onto the tropospheric NAM pattern of the PW response to changes aloft. The principal mechanisms whereby stratospheric variability influences the tropospheric circulation remain unclear. One of the possible mechanisms is amplification of PW due to internal tropospheric dynamics (Song & Robinson, 2004) and/or reflection of PW in the lower stratosphere (Perwitz & Harnik, 2003).

Stratosphere-troposphere coupling due to NAM Composites of time-height development of the northern annular mode (NAM) for (А) 18 weak polar vortex events and (В) 30 strong polar vortex events (Baldwin and Dunkerton, Science, 2001, 294, 581-584).

Outline Simulation of the Northern Hemisphere winter-time general circulation of the middle atmosphere with the Middle and Upper Atmosphere Model (MUAM); Diagnostic study of the PW generation, propagation, and reflection using 3D flux of wave activity and its divergence; Analysis of the UK Met Office data with respect to the PW and SSW in the stratosphere and the horizontal wind in the troposphere during and just after the SSW events; Synoptic analysis of the sea level pressure (SLP) maps after the SSW events in the stratosphere; Conclusions

Middle and Upper Atmosphere Model (MUAM) The MUAM is a 3D nonlinear mechanistic model of the atmospheric circulation extended from the 1000 hPa surface up to the heights of the ionospheric F2-layer. It is based on the Cologne Model of the Middle Atmosphere (COMMA). The MUAM is a grid-point model with horizontal (latitude/longitude) resolution of 5°*5.625°. It has up to 60 levels in the nondimensional log-pressure height x = - ln(p/1000) with a step-size of about 0.4. The model allows to use an arbitrary number of levels (ranging from 48 to 60) with the same vertical resolution. In the present study we use 48-level version with the upper boundary at x = 19, which corresponds to the geopotential height of about 150 km. To integrate the prognostic equations, the initial Cauchy problem was split into the set of simpler problems according to the physical processes considered. To solve these simpler problems, we use the Matsuno time-integration scheme.

MUAM: the lower boundary conditions At the lower boundary of MUAM (1000 hPa pressure level) the monthly mean climatological geopotential height and temperature fields including the zonal mean state and stationary planetary waves with zonal wave numbers m=1, 2, and 3 are specified. These fields were obtained by averaging over 11 years (1992-2002) of NCEP/NCAR reanalysis data; To reproduce more correctly the zonally averaged field in the troposphere, up to the lower stratosphere heights the zonally mean temperature is relaxed to the climatological NCEP/NCAR temperature by inserting a nudging term into prognostic equation for temperature.

Sudden Stratospheric Warming (SSW) and the meridional wind in the troposphere (height=4 km) at middle latitudes of the Northern Hemisphere over Europe simulated with the MUAM

Horizontal wind distributions at the log-pressure height 1 Horizontal wind distributions at the log-pressure height 1.4 km calculated with the MUAM. Left panel shows the wind 5 days before the SSW, right panel – during the SSW.

Horizontal wind distributions at the log-pressure height 1 Horizontal wind distributions at the log-pressure height 1.4 km calculated with the MUAM. Left panel shows the wind 2 days after the SSW, right panel – 7 days after the SSW.

3D flux of the wave activity Plumb R.A. On the Three-Dimensional Propagation of Stationary Waves, J. Atmos. Sci., 1985, 42, 217-229.

Zonally averaged flux of wave activity (Fy, Fz. 50) Zonally averaged flux of wave activity (Fy, Fz*50)*EXP(z/H) in the troposphere and lower stratosphere 2 days before (left panel) and during the SSW (right panel)

Zonally averaged flux of wave activity (Fy, Fz. 50) Zonally averaged flux of wave activity (Fy, Fz*50)*EXP(z/H) in the troposphere and lower stratosphere 2 and 7 days after the SSW event (left and right panels, respectively)

Vertical flux of wave activity (Fz. 50 Vertical flux of wave activity (Fz*50*EXP(z/H)) at 7 km 2 days before and during SSW (left and right panels)

Vertical flux of wave activity (Fz. 50 Vertical flux of wave activity (Fz*50*EXP(z/H)) at 7 km 2 and 7 days after the SSW (left and right panels)

Divergence of the wave activity flux (m/s/day) at 1 Divergence of the wave activity flux (m/s/day) at 1.4 km 2 days before and during SSW (left and right panels)

Divergence of the wave activity flux (m/s/day) at 1 Divergence of the wave activity flux (m/s/day) at 1.4 km 2 and 7 days after the SSW (left and right panels)

Wave activity conservation low

Monthly mean divergence of the wave activity flux and change in time of the wave activity density (m/s/day) at 1.4 km (left and right panels, respectively) 2 days after the SSW event

Amplitude of the zonal harmonic with wave number m=1 in the geopotential height (left panel), zonal mean wind (right panel blue line) at 60N, and the temperature above the Northern pole (right panel red line), 10 hPa, January-February 2001

Horizontal wind distributions at the 681 hPa from UKMO data Horizontal wind distributions at the 681 hPa from UKMO data. Left panel shows the wind 3 days before the SSW, right panel shows the wind during the SSW event in February 2001.

Horizontal wind distributions at the 681 hPa from UKMO data Horizontal wind distributions at the 681 hPa from UKMO data. Wind 3 and 12 days after the SSW event in February 2001 (left and right panels, respectively).

Open circles show the positions and invasions of the anticyclones during 5 days time interval (9-13 February 2001) during the SSW

Amplitude of the zonal harmonic with wave number m=1 in the geopotential height (left panel), zonal mean wind (right panel blue line) at 60N, and temperature above the Northern pole (right panel red line), 10 hPa, January-February 2006

Horizontal wind distributions at the 681 hPa from UKMO data Horizontal wind distributions at the 681 hPa from UKMO data. Left panel shows the wind 3 days before the SSW, right panel shows the wind during SSW event at the end of January 2006.

Horizontal wind distributions at the 681 hPa from UKMO data Horizontal wind distributions at the 681 hPa from UKMO data. Wind 7 and 12 days after the SSW event at the end of January 2006 (left and right panels, respectively)

Amplitude of the zonal harmonic with wave number m=1 in the geopotential height (left panel), zonal mean wind (right panel blue line) at 60N, and temperature above the Northern pole (right panel red line), 10 hPa, January-February 1979

Open circles show the positions and invasions of the anticyclones during 6 days time interval (7-12 February 1979) after the SSW

Conclusions Results of numerical simulation with the MUAM show an increase in the southward meridional wind in the troposphere at high and middle latitudes after the SSW events in the longitudinal sectors over Europe and Siberia; Latitude-height vector of the wave activity flux shows a substantial downward reflection of PW at the high latitude and 3D divergence of this flux demonstrates an increase of wave activity density in the troposphere during and especially just after the SSW; Analysis of UK Met Office and SLP data supports the results of simulation and shows that after the SSW events in the stratosphere there exists an increase of the southward meridional flow in the troposphere and invasions of anticyclones from polar region into the middle latitudes over Europe and Siberia; The time-scale and time-delay of extratropical stratosphere-troposphere interaction events suggest that stratospheric conditions could be used for middle-range weather forecasts.

Thank you for your attention!