Long-range sea surface temperature forecasts

Updated fortnightly

Sea surface temperature graphs

Australian climate is influenced by sea surface temperature patterns in the Pacific and Indian oceans. Specific regions are monitored, as they can indicate the presence, or potential development, of ENSO (El Niño/La Niña) and Indian Ocean Dipole (IOD) events.

Sea surface temperature graphs

NINO34 predictions for the next 5 months.

Long-range forecast graph for selected SST region

NINO34 probabilities

Product code: IDCK000073

Sea surface temperature maps

Sea surface temperature maps are not available for forecasts before Spring 2018

Global sea surface temperature forecasts for the months and season ahead. Anomalies indicate the difference from normal.

Sea surface temperature maps (select map for larger view)

SST forecasts for the next 3 months

About the graphs

The plume graphs show long-range forecast scenarios for sea surface temperatures (SSTs) averaged over particular regions of the Pacific and Indian oceans. The SSTs in these regions are related to different phases of the El Niño-Southern Oscillation (ENSO) and the Indian Ocean Dipole (IOD); climate drivers that can influence Australian rainfall and temperature patterns.

The graphs show 99 possible scenarios (grey lines), that are produced by the Bureau's climate long-range forecast model, which represent the range of outcomes that may occur over the forecast period. For example, they may show the SSTs in the NINO3.4 region to be warming, cooling, or remaining mostly steady. At times the long-range forecast might suggest a shift towards (or away from) values typically associated with El Niño or La Niña events. Each of the 99 scenarios is based on current conditions in the global oceans/atmosphere and how the model anticipates their likely development over the long-range forecast period, with each given slightly different treatment to provide a range of likely possibilities. This technique allows us to see the range of what is possible, with a small spread in the range of scenarios meaning more confidence in the likely path, and a larger spread meaning less confidence.

The green line is the average of all these 99 scenarios, often known as the ensemble mean. The solid black line shows the observations (based on the Bureau's SST observation analysis for each region) for the previous months.

The graphs are updated fortnightly. As a result, the value given for the 'current month' can vary depending on at what point in the month the forecast is being issued. Forecasts made on the 1st to the 11th of the month show a forecast value for the current month. For forecasts made after the 11th of the month, a month-to-date observation (shown by an open circle and dashed line), based on weekly observational data, will be used for the current month as a preliminary value until the final monthly data is available.

About the maps

While the climate model runs a set of 99 possible scenarios, it can be useful to look at the ensemble mean (the average of these forecasts) to see the most likely scenario. The global SST maps show the most likely SST anomaly for the months and seasons ahead. This can be useful to see how ENSO and IOD look spatially.

These graphs and maps are also included in the Bureau's Climate Driver Update and Model Summary webpages.

Base period

For maps and graphs since 22 October 2021, the SST anomalies show the difference from the 1981–2018 average. Earlier charts use the 1990–2012 base period.

About the long-range forecast model

The SST long-range forecasts are generated by the Bureau's climate model, ACCESS–S (Australian Community Climate Earth-System Simulator–Seasonal). ACCESS–S is the Bureau of Meteorology's dynamical (physics-based) weather and climate model used for monthly, seasonal and longer-lead climate forecasts. Prior to August 2018, climate forecasts (including these graphs) were produced by the Bureau's earlier model, POAMA.


Creative Commons By Attribution logo Unless otherwise noted, all maps, graphs and diagrams in this page are licensed under the Creative Commons Attribution 4.0 International Licence