Madden-Julian Oscillation (MJO)

MJO phase diagram


*Note: There are missing satellite observations from 16/3/1978 to 31/12/1978.

The MJO phase diagram illustrates the progression of the MJO through different phases, which generally coincide with locations along the equator around the globe. RMM1 and RMM2 are mathematical methods that combine cloud amount and winds at upper and lower levels of the atmosphere to provide a measure of the strength and location of the MJO. When the index is within the centre circle the MJO is considered weak, meaning it is difficult to discern using the RMM methods. Outside of this circle the index is stronger and will usually move in an anti-clockwise direction as the MJO moves from west to east. For convenience, we define 8 different MJO phases in this diagram.

Average weekly rainfall probabilities

These maps show average weekly rainfall probabilities and expected 850 hPa (approximately 1.5 km above sea level) wind anomalies for each of the 8 MJO phases. Green and blue shading indicates higher than normal rainfall would be expected, while red and orange shading indicates lower than normal rainfall would be expected. The direction and length of the arrows indicate the direction and strength of the wind anomaly. The darker the arrow, the more reliable the information is. The relationship of the MJO with Australian rainfall and winds changes with the season (which can be selected at the top).

Average outgoing longwave radiation (OLR)

Outgoing longwave radiation (OLR) is often used as a way to identify tall, thick, convective rain clouds. These maps show the difference from expected cloudiness based on the position of the MJO. The violet and blue shading indicates higher than normal, active or enhanced tropical weather, while orange shading indicates lower than normal cloud or suppressed conditions. The direction and length of the arrows indicate the direction and strength of the wind anomaly. The darker the arrow, the more reliable the information is. The relationship of the MJO with tropical weather patterns changes with the season (which can be selected above the maps).

Global maps of outgoing longwave radiation (OLR)


Global maps of outgoing longwave radiation (OLR) highlight regions experiencing more or less cloudiness. The top panel is the total OLR in Watts per square metre (W/m²) and the bottom panel is the anomaly (current minus the 1979-1998 climate average), in W/m². In the bottom panel, negative values (blue shading) represent above normal cloudiness while positive values (brown shading) represent below normal cloudiness.

Regional maps of outgoing longwave radiation (OLR)

Click on the boxes to view a timeseries of cloudiness for that region.
Map of regional cloudiness Dateline Vanuatu Coral Sea Fiji Nauru & Tuvalu Solomon Islands New Guinea Northern Australia Micronesia Malaysia & Indonesia Guam & Marianas Philippines Indochina Southern India & Sri Lanka

Below: OLR totals over the dateline

Click to see full-size graph of OLR totals over the dateline.

The graphs linked to this map show the OLRs for the different regions within the Darwin RSMC area. The horizontal dashed line represents what is normal for that time of year (based on the 1979 to 1998 period). The coloured curve is the 3-day moving average OLR in W/m². Below normal OLR indicates cloudier than normal conditions in this particular area, and is shown in blue shading. Above normal OLR indicates less cloudy conditions and is shown in yellow shading.

Daily averaged OLR anomalies

OLR Archive:   

Westerly wind anomalies

Winds Archive:

Time-longitude plots of daily averaged OLR anomalies (left) and 850 hPa (approximately 1.5 km above sea level) westerly wind anomalies (right) are useful for indicating the movement of the MJO.

How to read the Time-Longitude plots

The vertical axis represents time with the most distant past on the top and becoming more recent as you move down the chart. The Horizontal axis represents longitude.

Eastward movement of a strong MJO event would be seen as a diagonal line of violet (downward from left to right) in the OLR diagram, and a corresponding diagonal line of purple in the wind diagram. These diagonal lines would most likely fall between 60°E and 150°E and they would be repeated nearly every 1 to 2 months.

Average to slightly above average number of tropical cyclones likely for Australia in 2021–22

The Bureau released its seasonal Tropical Cyclone Outlook on 11 October. It predicts a 65% chance of an above-average number of tropical cyclones (TC) in the Australian Region during the coming season which runs between November 2021 and April 2022. The average number of TCs per season since 1970 is 11. However, it should be noted that TC numbers across the Australian region have declined in recent decades, with an average number of 9 per season since 2000. With this trend towards lower numbers, the current outlook suggests an average to slightly-above-average number of tropical cyclones are likely this season. Likewise for tropical lows (systems which either don't intensify sufficiently to be classified as TCs, or that remain over land), conditions expected during the coming season favour average to above-average numbers.

The main influence on this season's outlook are the water temperature patterns around northern Australia and the Pacific Ocean. The current, weak negative Indian Ocean Dipole (IOD) and a developing La Niña are the broadscale climate drivers associated with these ocean temperature patterns.

The South Pacific Tropical Cyclone Outlook was also released, with predictions of close-to-average TC numbers in both the eastern and western sub-regions in 2021–22.


The El Niño–Southern Oscillation (ENSO) remains neutral. However, the Bureau recently raised its ENSO Outlook status to La Niña ALERT, due to the cooling in the tropical Pacific Ocean and an increase in the number of climate models showing sustained La Niña conditions over the coming months. Historically, when La Niña ALERT criteria have been met, La Niña has subsequently developed around 70% of the time. A 70% chance of an event is approximately triple the normal likelihood.

Seasonal climate drivers such as ENSO can also influence the timing of the start of the tropical cyclone season in Australia. For example, both La Niña and a negative Indian Ocean Dipole (IOD) are associated with an early start to TC activity in the Australian region. Since 1970, when either of these climate drivers were active, the first TC was typically observed about 3 weeks earlier than average; mid-November instead of early December.

Read more about the current climate drivers in the Climate Driver Update

The influence of climate drivers on the northern wet season

In addition to its influence on tropical cyclone activity, the ENSO state can significantly affect the timing of the start of the Australian monsoon. Based on the monsoon onset date at Darwin, La Niña usually leads to an early start to the monsoon. Since 1957, the average monsoon onset date in a normal year is the last week of December, and about 1-2 weeks earlier during La Niña years.

Rainfall totals across northern Australia are also affected by ENSO state. While the influence of the IOD typically reduces markedly early in the wet season (by November or December), La Niña can influence rainfall patterns across the northern wet season. As a result, wet-season rainfall totals are typically above average across northern Australia when La Niña is active.

The Tropical Climate Note has changed

On 18 May, the Tropical Climate Note was renamed the Tropical Climate Update. It is published fortnightly, on alternate weeks to the Climate Driver Update.

The Climate Driver Update provides a summary of the major climate drivers affecting Australia, including tropical climate drivers.

Product code: IDCKGEW000

ACKNOWLEDGEMENT: Interpolated OLR data provided by the NOAA/OAR/ESRL PSD, Boulder, Colorado, USA.

Product Code: IDCKGEWWOO