Indian Ocean Dipole

Tracking sea-surface temperature gradients across the tropical Indian Ocean.

About

Background on the Indian Ocean Dipole and what the forecast predicts.

The Indian Ocean Dipole (IOD) describes the year-to-year swing in the east–west sea-surface temperature (SST) gradient across the tropical Indian Ocean. In a positive phase the western basin is anomalously warm while the waters off Sumatra and Java are anomalously cool; in a negative phase the pattern reverses. Because the gradient drives where deep convection sets up over the basin, the IOD shifts rainfall across a wide swath of the Indian Ocean rim – East Africa, the Indian subcontinent, maritime Southeast Asia and Australia.

The Dipole Mode Index

The IOD's state is summarised by the Dipole Mode Index (DMI): the difference between the SST anomaly averaged over the western equatorial Indian Ocean (50°E–70°E, 10°S–10°N) and the SST anomaly averaged over the south-eastern equatorial Indian Ocean (90°E–110°E, 10°S–0°). A positive DMI means the west is warm relative to the east, a negative DMI the opposite. The forecast on the Forecast page is a prediction of this index.

Seasonality

The IOD is strongly phase-locked to the annual cycle. Events typically begin to develop in boreal summer, peak in September–November, and decay rapidly through December as the monsoon circulation reverses. That life cycle limits how far ahead the index can usefully be predicted: forecasts initialised in the first half of the year carry much less skill than those initialised as an event is emerging, and skill drops sharply for target seasons past the boreal-winter decay.

Relationship to ENSO

The IOD and ENSO are related but distinct. Positive IOD events often, though not always, coincide with El Niño and negative events with La Niña, and part of the predictability of the DMI comes by way of the Pacific. Sizeable IOD events do occur independently of ENSO, however, so the two are forecast and interpreted separately.

Interpreting the forecast

The forecast figure shows the individual model predictions of the DMI together with the probabilities of the index falling in each category, so both the spread across models and the consensus can be read from it. A wide spread indicates genuine uncertainty about the coming season rather than a weak event. See the Methodology page for which prediction systems contribute and how they are combined.

Disclaimer: The CCSR/IRI IOD forecast is a research product. Please consult your country's national meteorological service for the official forecast for your country.