MADDEN-JULIAN OSCILLATION

Main Reference：Chidong Zhang 2005 (b)

This article is an introduction to Madden-Julian-Oscillation(also called intraseasonal oscillation) according to Zhang's 2005 review paper. What is Madden-Julian Oscillation? Actually we can't define MJO exactly only from its feature. We still need to use some reanalysis data to define the signal of MJO, and this is also a proper method for us doing research objectively. But in order to give a simple introduction of MJO, I will discussion from meteorology phenomena to climate dynamic. Zhang's 2005 review will be my guide book this time.

      MJO has some significant features
(1)30-90 period (move around tropical earth )
(2)~5m/s propagation speed
(3) large scale circulation combine convection
(4) maximum signal will appear around maritime continent
(5) first occur when the strong westerly burst
(6) air-sea interaction will the key point of its mechanism
(7) Kelvin coupled Rossby wave sometimes induce tropical cyclone
(8) influence ENSO period (induce ocean Kelvin Wave)

 The article will goes through four topics
1.basic observed features  2.mechanism 3.numerical simulation 4.air-sea interaction

-------------------------------------------------------------------------------------------
CH2 Basic observed features


There are two main phases of MJO-active phase and inactive phase. The first one companies with strong convection and precipitation and the second phase is followed by suppressed weather phenomena(subsidence). Both of these two phases occur on equatorial earth, and this also means the transient state of these two phase is also the transient state of Walker circulation.   When MJO occurs, there will be strong westerly wind burst. According to Emmaunel's Quasi-Equilibrium, most of tropical atmosphere remains first baroclinic mode. It implies at least two things (1)strong convection everywhere (2)inverse phase between upper troposphere and lower troposphere. We can check picture 1 and picture 2.

                                   fig 1                                                           fig2

fig 1：this picture shows the wind direction, convection center and sea level pressure.
fig 2：(a)zonal wind (2.5*2.5 degree, ms^-1) at 850 hPa(roughly 1.5km above sea level) from NCEP[Kalnay et al., 1996] (b)precipitation(1*1 degree mmd^-1) from GPCP combined data set [Huffman et al., 1997]from June 2000 to May 2001, it's 10N~10S meridional average. The straight lines mark each MJO event. We can estimate its propagation speed from the slopes. Some faster signal are eastward moving gravity waves, such as Kelvin wave.
Even without filtering, we still can see strong signal of MJO from reanalysis data. (But it's a difficult work for Madden in 1971 when he only had some sounding data)。

MJO cloud covers extends 12000~20000km. Occasionally, two different phases(active phase at Indian ocean and suppressed  phase at Pacific ocean) will occur at the same time.[e.g., Wheeler and Hendon, 2004] Comparing with circulation, convection signal is more like "peak" or "spectrum". Fig1 and Fig 2 can see the difference. Circulation always range from wave number1 to wave number 2, but convection can exceed wave number 3.

For propagation speed, MJO is much more like 4th baroclinic mode (5ms^-1) and its much slower than convectively coupled Kelvin waves(17 ms^-1)[e.g., Wheeler and Kiladis, 1999] The phase speed of MJO varies with different stage. When convection signal vanish in eastern Pacific, propagation is more like free wave(uncoupled wave) and the wave speed can reach 30~35 m/s(easterly).

Concerning the large scale circulation. We can use simple wave dynamics to separate MJO into two parts(but the true condition is more complicated). East of the convection center,  low level easterly companies with Kelvin patter circulation. West of the convection center, low level easterly(about three times stronger than easterly)  companies Rossby waves circulation.(twin cyclones or anticyclones). Both of these two structure are important for MJO and they were combined together by convection.See fig3...

Some observation revealed convection doesn't always locate at the same place referring to circulation. In Indian ocean, convection center is close the transient field between easterly and westerly. When MJO move to  the Pacific, convection prevails westerly.

to be continued......