The Madden-Julian Oscillation (MJO) is a broad-scale wave-like convective phenomena centred on the equator and also the main source of tropical predictability on the monthly time scale. It is important as it identifies where tropical deep convection interacts with the general atmospheric circulation, and it is observed mainly in a sector spanning the Indian and Pacific oceans. The feature is typically initiated over the Indian Ocean, and usually advances steadily eastwards, as areas of relatively organised convection, bringing a significant enhancement of rainfall, followed in their wake by much less active convection. The MJO may be monitored using a combination of lower and upper tropospheric zonal flow patterns and out-going long wave radiation (OLR) or simulations thereof.
Fig8.2.2.1: Schematic representation of the global teleconnection patterns associated with the Madden-Julian Oscillation (MJO).
The Madden-Julian Oscillation affects the North Atlantic Oscillation. The mid-latitude jet streams in turn can affect the flow over the North Atlantic which has a strong impact on the tracking of depressions (e.g. towards Europe). The cross-Atlantic flow can be categorised as positive or negative phases or regimes of the North Atlantic Oscillation (NAO).
Fig8.2.2.2: Schematic diagram of the effects of positive NAO and negative NAO phases in the European and Mediterranean areas. Positive NAO phase tends to steer depressions towards Northern Europe bringing wetter conditions while the Mediterranean region tends to be drier. Negative NAO phase tends to steer depressions towards Southern Europe bringing wetter conditions while Northern Europe tends to be drier.
Fig8.2.2.3: To view Hovmoeller diagrams:
An MJO Hovmoeller diagram displays the time evolution of the ensemble mean anomaly. The longitude within the equatorial band (15S-15N) is shown on the x-axis. Time is shown on the y-axis. The analyed data at the data time of each extended range forecast run is shown above the horizontal line. The forecast mean ENS values are shown below the horizontal line. Several MJO forecast products are available:
Fig8.2.2.4: Time-longitude section (Hovmoeller diagram) of ensemble mean anomalies averaged over a tropical band (15N-15S, shown in map section at the bottom). Analysed values above the horizontal black line, forecast values below it.
Left figure: Anomalies of velocity potential at 200hPa. Note the observed and predicted progression eastwards (diagonally towards bottom right of the diagram) of the maxima (red, yellow) and minima (blue, green) of the 200hPa velocity potential.
Central figure: Anomalies of zonal wind at 850hPa. Note the forecast of a dipole of positive anomaly (red: stronger westerly flow than ER-M-climate) and negative anomaly (blue, green: weaker westerly flow than ER-M-climate) suggesting an area of anomalous convergence in between. The diagonal pattern suggests it has progressed and will continue to progress slowly eastwards.
Right figure: Anomalies of outgoing long wave radiation. Note the area of negative anomaly (blue, green: weaker outgoing radiation) indicative of enhanced cloud cover with cold radiating cloud-tops and inhibition of transmission of warmer terrestial radiation.
The Hovmoeller diagrams show the recently observed and predicted movement and intensities of three parameters:
Commonly, negative velocity potential at 200hPa and/or indication of 850hPa convergence (and hence upward motion in the atmosphere) coincide with a reduction in outgoing long wave radiation (due to existence of deeper cloud with colder tops). Broadly, this can be seen on the plots above (e.g. at 150E at T+0 (i.e. 00UTC 25 Jan 2018) marked by the horizontal line. The greater is the magnitude of each of these components, and the more coincident they are, the greater will be, in general, the magnitude of the MJO.
The Wheeler-Hendon MJO index plot displays the time evolution, location and magnitude of the MJO predicted by ENS and described by a multivariate MJO index (Wheeler and Hendon 2004 Mon. Wea. Rev. vol. 132, 1917-1932). The magnitude of the MJO is essentially proportional to the distance of the point from the centre of the diagram.
Fig8.2.2.5: Wheeler-Hendon MJO diagram. Time evolution of the MJO as predicted by the ENS using zonal winds at 850hPa, 200hPa velocity potential and simulated OLR, all averaged between 15S and 15N. Individual ensemble member values at day 1, 5, 10, 15 and 20 are represented respectively by red, pink, orange blue and green circles. The ensemble mean values (black triangles) are joined by a solid black line, and the analysis values of the preceding 30 days are joined by a grey line (with grey dots added every 5 days). Points representing sequential values trace anti-clockwise trajectories around the origin,indicating systematic eastward propagation of the MJO. Large amplitudes (outside of the circle) signify strong cycles of the MJO, while weak activity appears as rather random motion near the origin. The phases (broadly geographical regions) are numbered 1 to 8. See Wheeler and Hendon 2004 Mon. Wea. Rev. vol. 132, 1917-1932 for details.
In the example in Fig8.2.2.5 there is initially a strong indication of a large amplitude MJO (red, Day1; pink, Day5) progressing eastwards but this is forecast to probably weaken and become less identifiable (Day15, blue) over the western Pacific, but the ENS spread becomes large by then implying large uncertainty with some members maintaining a strong signal (plots far from the centre of the diagram), others weak (plots towards the centre and inside the circle). Compare with the Hovmoeller diagrams (Fig8.2.2.4) where:
Individual values of these parameters for the different ENS members give varying intensities of MJO on the diagram and the spread becomes large. Note also that by construction, the amplitude of the forecast ENS mean MJO on this diagram (as shown by black triangles) will tend to reduce, on average, at longer lead-times.
(Note: In older material there may be references to issues that have subsequently been addressed)