Figures and text based on Zhang (2003) ; review of MJO in Journal of Geophysical Research. And George Kiladis (personal communication)

1
MJO Lecture
Figures and text based on Zhang (2003) review
of MJO in Journal of Geophysical Research. And
George Kiladis (personal communication)
2
1. Observations

• Longitude-height schematic of MJO based on
  Madden and Julian (1972)
• Organised planetary scale system, influencing
  all of the tropics.
• Moves eastwards at about 5m/s
• Convective signal strongest in Indian Ocean and
  West/Central Pacific.
• Dynamic signal seen throughout the tropics.

3
1. Observations

• MJO characterised by convectively active and
  inactive phases
• Phases connected by deep overturning zonal
  circulations
• Zonal winds reverse between lower and
  upper-levels

4
1. Observations
Zonal wind (2.5N-2.5S) Precipitation (1N-1S)
MJO seen in unfiltered fields
Straight white lines MJOs Black dashed lines
convectively coupled Kelvin waves White arrows
indicate westward propagating Rossby or mixed
Rossby-gravity waves
5
1. Observations

• Positive period eastward Negative period
  westward
• Note clear peaks of the MJO at 30-100 days in
  ppn and zonal wind at 850hPa
• Wide range reflects highly episodic nature, and
  seasonal to interannual variability

6
1. Observations
MJO composite based on regression of equatorial
band-pass (30 - 90 days) filtered 850 hPa zonal
wind (contours, interval 0.2 m s-1) and
precipitation (colors, mm day -1) upon 850 hPa
zonal wind of the MJO at 160E and the equator.
The MJO zonal wind was extracted from the
band-pass filtered time series using its four
leading modes of SVD (singular vector
decomposition) (Zhang and Dong 2004). The
straight cyan lines indicate the eastward phase
speed of 5 m s-1.

• MJO phase speed of 5m/s distinguishes it from
  the fast convectively coupled Kelvin waves which
  propagate at greater speeds (15-17m/s).
• MJO moves faster when it does not have a
  convective signal (30-35m/s)

7
1. Observations

• Large-scale wind structure is often described in
  terms of equatorial waves coupled to deep
  convection.
• Equatorial Kelvin wave to east, Equatorial
  Rossby wave to west both considered essential to
  MJO.

8
1. Observations
Diabatic heating

• Immediately ahead of convective center are
  low-level convergence, ascending motions and
  low-level moistening drying and low-levels to
  west.
• Encourages eastward propagation

9
1. Observations

• Eastward moving convective center of active
  phase of MJO, made up of many higher frequency
  small scale convective systems moving in all
  directions
• Includes coupled Kelvin waves, and westward
  moving 2-day and 5-day disturbances

• Longitude-time diagrams of deep cloud clusters
  (cloud top infrared temperature lt 208 K) over 0
  - 10S for (a) 1 - 31 December 1992 during which
  an MJO event propagated through the eastern
  Indian and western Pacific Ocean (Yanai et al.
  2000) (b) Details for 20 - 31 December as marked
  by the lower right box in (a) (c) Details for 22
  - 28 December as marked by the box in (b). Sizes
  of ovals are proportional to the actual sizes of
  cloud clusters. (From Chen et al 1996)

10
1. Observations

• MJO signals in convection confined to Indian and
  Western Pacific Oceans
• Associated with warm SSTs known as the warm
  pool
• Note MJO signal in East Pacific north of cold
  tongue, in boreal summer again emphasising the
  significance of warm SSTs.
• MJO undergoes string seasonal cycle peaking in
  boreal winter/spring when strongest signals are
  immediately south of equator

Variance of the MJO (contours) in (a) 850 hPa
zonal wind and (b) precipitation during December
March, (c) 850 hPa zonal wind and (d)
precipitation during June September, overlaid
with mean SST (C). Contour intervals are 1 m2
s-2 for the wind starting from 2 m2 s-2 and 2 mm2
day-2 for precipitation starting from 2 mm2
day-2. See Zhang and Dong (2004) for details of
defining the MJO in this figure.
11
2. Mechanisms

• Since Kelvin wave is only eastward propagating
  equatoial wave and it resembles the MJO east of
  heating, the Kelvin wave has been taken as the
  backbone of the MJO from day one
• BUT, coupled Kelvin waves propagate eastwrds too
  fast
• Therefore key questions that must be addressed
  by any MJO theory are
• What are the mechanisms that distinguish the
  MJO from convectively coupled Kelvin waves?
• What processes must take place to supply
  energy against dissipation to the MJO?
• Few theories answer these questions.
• There are two major schools of though on the
  energy source of the MJO
• (I) Eastward propagation and coupling between
  convection and wind are secondary by-products of
  the atmospheric response to convection
• (II) The MJO creates its own energy source
  through atmospheric instability

12
2. Mechanisms (Atmospheric Response to
Independent Forcing)

1. Intraseasonal variations in the Asian Monsoon
   have been proposed to be a forcing for MJO.
   Observations have suggested the existence of
   intraseasonal standing oscillations in
   convection, but these are NOT statistically
   significant. Idealised modelling studies also
   refute this hypothesis.
2. Tropical Stochastic Forcing a localised
   stochastic heat source can give rise to
   oscillations at intraseasonal timescales. The
   maximum growth however is at smaller scales
   (zonal wave numbers gt 4)
3. Lateral Forcing Intraseasonal perturbations
   coherent with the MJO exist in the extratropics
   and may force MJOs. Eastward moving extratropical
   disturbances can excite a variety of equatorial
   waves.

13
2. Mechanisms (Atmospheric Instability)

• Instability theories tend to suffer the same
  problem in that the most unstable solutions tend
  to be at smallest scales
• Special tricks are required to remedy this
  including ve only heating and time-lags between
  the energy input and convective heating
• Moisture Convergence
• These mechanisms are based on CISK (Conditional
  Instability of the Second Kind) where convective
  heating is related to low-level moisture
  convergence. For ve only heating unstable modes
  move at 16-19m/s comparable to observed coupled
  Kelvin waves (not the MJO!). Growth rates are
  greatest on smallest scales. CISK often
  criticised as unphysical. Inclusion of Rossby
  wave slows moist Kelvin wave to more realistic
  values.
• (B) Surface Evaporation
• Wind-induced surface heat exchange (WISHE) has
  been proposed as a growth mechanism. Requires
  mean surface easterlies then surface fluxes and
  convection peaks east of convective center (in
  warm phase of Kelvin wave, hence growth). BUT
  observations indicate that surface fluxes peak in
  or west of convective center. And mean low-levels
  winds rarely easterly in Indian Ocean and West
  Pacific!!!

14
2. Mechanisms (Atmospheric Instability)
Temperature Structure of a Dry Kelvin Wave
Direction of Motion
W
C
W
C
15
2. Mechanisms (Atmospheric Instability)
Other Factors to consider Radiation Water
Vapor Sea Surface Temperature Scale
Interaction Heating Profile
16
3. More Observations from Kiladis (2006)
See Animation
17
The Madden-Julian Oscillation (MJO)

• Discovered by Rol Madden and Paul Julian at NCAR
  in 1971
• Characterized by an envelope of convection
  10,000 km wide moving eastward at around 5 m/s
• Most active over regions of high sea surface
  temperature
• (gt 27? C)
• Can have a profound impact on the extratropical
  circulation
• Is poorly represented in general circulation
  models, if at all
• Composed of a variety of higher frequency,
  smaller scale disturbances

18
OLR power spectrum, 19792001 (Symmetric)
from Wheeler and Kiladis, 1999
19
OLR power spectrum, 19792001 (Symmetric)
Westward Inertio-Gravity
Kelvin
Equatorial Rossby
Madden-Julian Oscillation
from Wheeler and Kiladis, 1999
20
OBSERVATIONS OF KELVIN WAVES AND THE
MJO Timelongitude diagram of CLAUS Tb
(2.5S7.5N), JanuaryApril 1987
Kelvin waves (15 m s-1)
21
3. More Observations from Kiladis (2006)
OBSERVATIONS OF WAVES WITHIN THE
MJO Timelongitude diagram of CLAUS Tb
(5Sequator), February 1987
22
OLR power spectrum, 19792001 (Symmetric)
from Wheeler and Kiladis, 1999
23
Regression Models

• Simple Linear Model
• y ax b
• where x predictor (filtered OLR)
• y predictand (OLR, circulation)

24
OLR and 850 hPa Flow Regressed against
MJO-filtered OLR (scaled -40 W m2) at eq, 155?E,
1979-1993
Day 0
Streamfunction (contours 4 X 105 m2 s-1) Wind
(vectors, largest around 2 m s-1) OLR (shading
starts at /- 6 W s-2), negative blue
25
OLR and 850 hPa Flow Regressed against
MJO-filtered OLR (scaled -40 W m2) at eq, 155?E,
1979-1993
Day-16
Streamfunction (contours 4 X 105 m2 s-1) Wind
(vectors, largest around 2 m s-1) OLR (shading
starts at /- 6 W s-2), negative blue
26
OLR and 850 hPa Flow Regressed against
MJO-filtered OLR (scaled -40 W m2) at eq, 155?E,
1979-1993
Day-12
Streamfunction (contours 4 X 105 m2 s-1) Wind
(vectors, largest around 2 m s-1) OLR (shading
starts at /- 6 W s-2), negative blue
27
OLR and 850 hPa Flow Regressed against
MJO-filtered OLR (scaled -40 W m2) at eq, 155?E,
1979-1993
Day-8
Streamfunction (contours 4 X 105 m2 s-1) Wind
(vectors, largest around 2 m s-1) OLR (shading
starts at /- 6 W s-2), negative blue
28
OLR and 850 hPa Flow Regressed against
MJO-filtered OLR (scaled -40 W m2) at eq, 155?E,
1979-1993
Day-4
Streamfunction (contours 4 X 105 m2 s-1) Wind
(vectors, largest around 2 m s-1) OLR (shading
starts at /- 6 W s-2), negative blue
29
OLR and 850 hPa Flow Regressed against
MJO-filtered OLR (scaled -40 W m2) at eq, 155?E,
1979-1993
Day 0
Streamfunction (contours 4 X 105 m2 s-1) Wind
(vectors, largest around 2 m s-1) OLR (shading
starts at /- 6 W s-2), negative blue
30
OLR and 850 hPa Flow Regressed against
MJO-filtered OLR (scaled -40 W m2) at eq, 155?E,
1979-1993
Day4
Streamfunction (contours 4 X 105 m2 s-1) Wind
(vectors, largest around 2 m s-1) OLR (shading
starts at /- 6 W s-2), negative blue
31
OLR and 850 hPa Flow Regressed against
MJO-filtered OLR (scaled -40 W m2) at eq, 155?E,
1979-1993
Day8
Streamfunction (contours 4 X 105 m2 s-1) Wind
(vectors, largest around 2 m s-1) OLR (shading
starts at /- 6 W s-2), negative blue
32
OLR and 850 hPa Flow Regressed against
MJO-filtered OLR (scaled -40 W m2) at eq, 155?E,
1979-1993
Day12
Streamfunction (contours 4 X 105 m2 s-1) Wind
(vectors, largest around 2 m s-1) OLR (shading
starts at /- 6 W s-2), negative blue
33
Specific Humidity at Truk (7.5?N, 152.5?E)
Regressed against MJO-filtered OLR (scaled -40 W
m2) for 1979-1999
OLR
Pressure (hPa)
OLR (top, Wm-2) Specific Humidity (contours, 1 X
10-1 g kg-1), red positive
from Kiladis et al. 2005
34
Q1 Regressed against MJO-filtered OLR over the
IFA during COARE
from Kiladis et al. 2005
35
Morphology of a Tropical Mesoscale Convective
Complex in the eastern Atlantic during GATE (from
Zipser et al. 1981)
Storm Motion
36
Observed Kelvin wave morphology (from Straub and
Kiladis 2003)
Wave Motion
37
Two day (WIG) wave cloud morphology (from
Takayabu et al. 1996)
38
Equatorial Wave Cloud Morphology

• Consistent with a progression of shallow to deep
  convection, followed by stratiform precipitation
  for the Kelvin, Westward Inertio-gravity (2-day)
  Waves, and Easterly Waves
• This was also observed during COARE for the MJO
  (e.g. Lin and Johnson 1996 Johnson et al. 1999
  Lin et al. 2004)
• This evolution is similar to that occurring on
  the Mesoscale Convective Complex scale

39
Convection in General Circulation Models

• Question How well do GCMs do in characterizing
  intraseasonal tropical convective variability?
• Jialin Lin et al. (2006) applied identical
  space-time spectral techniques to observed and
  modeled tropical precipitation
• Models used are the 14 coupled ocean-atmosphere
  GCMs used for intercomparison in the 4th
  Assessment Report of the Intergovernmental Panel
  on Climate Change (IPCC)

40
Rainfall Power Spectra, IPCC AR4 Intercomparison
15S-15N, (Symmetric)
Observations
from Lin et al., 2006
41
Rainfall Power Spectra, IPCC AR4 Intercomparison
15S-15N, (Symmetric)
from Lin et al., 2006
42
Rainfall Spectra/Backgr, IPCC AR4
Intercomparison 15S-15N, (Symmetric)
Observations
from Lin et al., 2006
43
Rainfall Spectra/Backgr, IPCC AR4
Intercomparison 15S-15N, (Symmetric)
from Lin et al., 2006
44
Rainfall Spectra at 5S-5N, 85E from IPCC AR4
Intercomparison
45
4. Numerical Modeling (more comments from Zhang,
2005)

• Modeled eastward propagation speeds often closer
  to observed coupled convectively coupled Kelvin
  waves than MJO
• When eastward propagating signals are reproduced,
  they are too weak and structures unrealistic.

46
4. Numerical Modeling (more comments from Zhang,
2005)
U at 850hPa PPN
Obs

• All models (selected) produce some MJO signals
• Realistic spectra does not guarantee realistic
  structure (see next slide)

47
4. Numerical Modeling (more comments from Zhang,
2005)

• Common problem ve PPn anomalies tend to be in
  regions of low-level easterlies contrary to
  observations (in westerlies)
• Few models can reproduce observed MJO structures

48
5.Concluding Remarks

• Much progress has been made in past decade
• Still major challenges
• need to better observe and understand vertical
  structure
• need to understand why some idealised models
  simulate MJOs better than more realistic GCMs
• Key research topics
• scale interactions
• air-sea interaction
• prediction
• interaction with ENSO
• modulation of tropical cyclones
• interaction with monsoons
• influences on high latitude weather

49
(No Transcript)
50
(No Transcript)
51
(No Transcript)
52
(No Transcript)
53
(No Transcript)
54
(No Transcript)
55
(No Transcript)
56
(No Transcript)
57
(No Transcript)
58
(No Transcript)
59
(No Transcript)
60
(No Transcript)
61
(No Transcript)