The Quasi Biennial Oscillation

Examining the link between equatorial winds and the flow regime of the

wintertime polar stratosphere

Charlotte Pascoe

Layout of Talk

Introduction QBO history How does the QBO work? Why is the QBO important? Polar vortex and planetary waves The Unified Model • Experiments • ResultsSummary

QBO History

1883 Krakatau debris circles the globe from east to west in two weeks:

Krakatau Easterlies

1908 Berson launches balloons from Lake Victoria in Africa and finds lower stratospheric winds blowing from west to east:

Berson’s Westerlies

QBO History1960 Reed (US) and Elbon (UK) “The circulation of the stratosphere”

Balloon measurements reveal alternate bands of easterly and westerly winds originating above 30km and moving downwards through the stratosphere at ~1km per month.

Bands appear at 13 month intervals 26 months required for a complete

cycle

QBO History1960s Lots of meteorologists get

sun tans whilst releasing balloons to measure this strange new phenomenon. All find slightly different cycle periods.

1964 Angell and Korshover give the cycle the name:

Quasi Biennial Oscillation

The Quasi Biennial Oscillation

top panel: equatorial zonal winds from rocketsondemiddle panel: de-seasonalisedbottom panel: broad-band filtered (18-36 month)

height

time

20 km

60 km

1965 1987

40 m/s

-40 m/s

30 m/s

-30 m/s

QBO phase denotes wind direction in the lower stratosphere

How does the QBO work?

Wavy blue and red lines indicate the penetration of easterly and westerly waves

Holton and Lindzen (1972) proposed a model of the QBO based on vertically propagating waves. The mechanism was further explained by Plumb (1977).

Equatorially trapped Kelvin waves provide westerly momentum and Rossby-gravity waves provide easterly momentum to produce the QBO oscillation.

Why is the QBO important?Hurricane Forecasts West: Increased activity in the Atlantic and NW Pacific East: Increased activity in the SW Indian basin

Stratospheric Winter Warmers Holton and Tan (1980) West: Cold undisturbed polar vortex More stratospheric Ozone loss East: Warm disturbed polar vortex

More tropospheric `cold snaps’

Example of a Stratospheric Sudden Warming

PV on the 1250K isentropic surface (~42 km)

Planetary wave of wave number one

Vertical propagation of planetary waves Planetary waves (aka Rossby waves) drift to the

west relative to the background flow at typical speeds of a few metres per second.

The vertical propagation of planetary waves is only possible under the condition that the zonal wind is within the range:

0<u<B/(k2 + l2)

Under conditions of easterly background flow no vertical propagation of planetary waves can occur. (Westerly flow is never strong enough for the upper limit to be reached)

Charney and Drazin (1961) found no stratospheric planetary waves in summer when the background flow is easterly.

QBO as wave guideThe QBO phase determines the position of the zero-line

in the subtropics which acts as a critical line for planetary waves propagating into the stratosphere.

Planetary wave activity is confined to high northern latitudes

Increased heat and momentum transport into the polar vortex region

WEAK POLAR VORTEX

QBO EAST Critical line is in northern subtropics

QBO WEST Critical line is in southern subtropics

Planetary waves are free to move into the Southern Hemisphere

Less wave activity close to the pole

STRONG POLAR VORTEX

However…The Holton-Tan relationship is not exact, there are many exceptions to this rule of thumb. Gray, Drysdale, Dunkerton and Lawrence (2001) have suggested that equatorial winds in

the stratopause region are also important and may help understand polar vortex variability.

Holton-Tan

Negative correlation between polar temperature and equatorial winds

Significant correlation in stratopause region where QBO and SAO interfere

J-F Polar temperature North of 62.5oN at 24km correlated with equatorial winds

Model Description

• UKMO Unified Model (version 4.5)• Hydrostatic primitive-equation model • Run in atmosphere only mode• 64 vertical levels: 1000-0.01 hPa (0-80

km)• X-direction (E-W): 96 columns (3.75o)• Y-direction (N-S): 73 rows (2.5o)• Rayleigh friction imposed above 50 km• Ocean climatology repeated each year

Experiments

3 QBO profiles (period 27 months)1 SAO profile (period 6 months)

QBO Thick: Large overlap with SAO QBO Thin: No overlap with SAO QBO Normal: Moderate overlap with SAO

AlgorithmU=U–timestep/rlxtime(U-(UQBO+USAO))

Experiments

Latitude dependence

0

0.2

0.4

0.6

0.8

1

1.2

-60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60

latitude

mag

nit

ud

e

QBO + SAO amplitude functions

0

10

20

30

40

50

60

70

0 5 10 15 20 25 30 35 40

Magnitude (m/s)

Heig

ht

(km

)QBO and SAO forcing

amplitudes wrt height and latitude

Experiments Relaxation time scale wrt height and latitude

Relaxation Time Scale

0

10

20

30

40

50

60

70

0 2 4 6 8

time (days)

hei

gh

t (k

m)

Latitude Dependence

1

10

100

1000

10000

-60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60

latitude

y

Latitude Dependence

0

50

100

150

200

250

300

-60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60

latitude

y

Results

QBO

SAO

Results 40hPa Equatorial wind & 10hPa Polar temperature

WEST

EAST

Results January and February zonal wind composites

Results J-F Polar temperature North of 60oN at 10hPa (~30km@70oN) correlated with equatorial winds

Negative correlation in lower stratosphere

Positive correlation in upper stratosphere

Summary• Need to run the simulation for longer • We are finding the expected

negative correlation between lower stratospheric equatorial winds and polar temperatures

• There is also a positive correlation in the upper stratosphere

• No asymmetry about the mid summer months (June and July) but this should be fixed by including an annual cycle over the equator