The Southern Oscillation Index, or SOI, is a number that tracks the difference in air pressure between the western and eastern tropical Pacific Ocean. Specifically, it compares the sea-level barometric pressure at Tahiti (in the central-east Pacific) with the pressure at Darwin, Australia (in the western Pacific). That simple pressure difference turns out to be one of the most reliable gauges of whether the tropical Pacific is tilting toward El Niño, La Niña, or neutral conditions, and its influence reaches far beyond the ocean it measures.
How the SOI Is Calculated
Weather stations at Tahiti and Darwin have recorded barometric pressure for well over a century. To compute the SOI, you take the standardized anomaly of the pressure difference between these two stations. In plain terms: you find out how much the Tahiti-minus-Darwin pressure gap departs from the long-term average, then scale that departure so it can be compared across months and seasons. The result is a dimensionless number that fluctuates above and below zero.
A sustained negative SOI generally signals El Niño conditions. During El Niño, pressure drops at Tahiti (the eastern side) relative to Darwin, the trade winds weaken, and warm water pools in the central and eastern Pacific. A sustained positive SOI points to La Niña, when the trade winds strengthen, cool water upwells more vigorously in the eastern Pacific, and warm water gets pushed further west than usual. Single-month blips don’t mean much; meteorologists typically look for the SOI to stay consistently negative or positive over several months before labeling an event.
Where the Idea Came From
The concept traces back to Sir Gilbert Walker, a British mathematician who became director-general of observatories in India in the early 1900s. Walker was trying to predict the Indian monsoon after a catastrophic famine linked to monsoon failure in 1899. He gathered barometric pressure data from stations around the tropics and noticed a seesaw pattern: when pressure was high in the eastern Pacific, it tended to be low in the west, and vice versa. He called this the “Southern Oscillation.” Decades later, atmospheric scientists named the large-scale east-west circulation cell in the tropical Pacific the “Walker Circulation” in his honor. Interestingly, Walker also contributed foundational statistical methods still used today; the Yule-Walker equations in time-series analysis bear his name for work connected to the same climate-prediction research.1Project Euclid / Statistical Science. Sir Gilbert Walker and a Connection between El Niño and Statistics
Walker’s insight was that the atmosphere and ocean in the tropical Pacific behave as a coupled system. It took several more decades for scientists to formally link Walker’s Southern Oscillation to the oceanic warming pattern fishermen off Peru had long called El Niño. The combined phenomenon is now known as ENSO, short for El Niño-Southern Oscillation, and the SOI remains one of the standard ways to track its atmospheric component.
SOI Versus Other ENSO Indices
The SOI is purely atmospheric: two pressure readings on opposite sides of the Pacific. But ENSO involves the ocean too, so researchers have developed a whole family of indices. Among the most common are the Niño 3.4 sea surface temperature (SST) index, which averages ocean temperatures in a box of the central-eastern tropical Pacific, and the Multivariate ENSO Index (MEI), which blends pressure, SST, wind, and cloudiness into a single score. Researchers studying global wave climate and coastal hazards, for instance, have used the MEI to correlate ENSO phases with changes in wave power and direction along coastlines worldwide.2Journal of Geophysical Research: Oceans. El Niño‐Southern Oscillation Impacts on Global Wave Climate and Potential Coastal Hazards Studies tracking South American climate routinely consider the SOI alongside SST anomalies in four different Niño regions, the Oceanic Niño Index, the Pacific Decadal Oscillation, and others.3PubMed Central. Analysis of South American climate and teleconnection indices
So why bother with the SOI when fancier options exist? One reason is sheer length of record. Continuous pressure data from Tahiti and Darwin stretches back to the late 1800s, giving the SOI one of the longest instrumental time series of any ENSO tracker. Another is simplicity: it relies on just two well-maintained surface stations, making it straightforward to update and hard to argue with. And because it captures the atmospheric side of ENSO rather than the oceanic side, it sometimes picks up changes in the system’s behavior that SST-based indices miss, or vice versa. In practice, forecasters watch several indices simultaneously rather than relying on any single one.
Reaching Back Before Instruments
The instrumental SOI record only goes back roughly 150 years, which is a blink in climate terms. To understand how ENSO behaved in earlier centuries, scientists have turned to natural archives. A multiproxy reconstruction using tree rings, corals, and ice cores extended an ENSO index back to the year 1525, and found that the reconstructed signal correlates strongly with the modern SOI, Niño 3.4 SST, and a combined ocean-atmosphere ENSO index.4Journal of Geophysical Research: Atmospheres. A multiproxy index of the El Niño–Southern Oscillation, A.D. 1525–1982 This kind of work matters because it lets researchers test whether extreme El Niño events like those in 1997-98 or 2015-16 are unprecedented or whether similar swings occurred long before humans were measuring atmospheric pressure. The broad answer is that large ENSO swings have happened repeatedly over the past five centuries, but the frequency and intensity of events are not perfectly steady over time.
Forecasting Skill
One of the SOI’s most useful properties is that it carries predictive information. Research using autoregressive models found that SOI-based predictions show useful skill out to about 30 to 36 months.5Journal of Geophysical Research: Atmospheres. Adaptive filtering and prediction of the Southern Oscillation index That is a remarkably long forecast horizon for a climate signal. In practice, seasonal climate outlooks issued by meteorological agencies around the world incorporate ENSO phase information heavily. If the SOI has been running strongly negative for several months and ocean temperatures confirm El Niño, forecasters can say with reasonable confidence what rainfall and temperature patterns are likely in places as far from the Pacific as eastern Africa or southern Brazil.
That said, the SOI is noisy on short timescales. Day-to-day pressure fluctuations at Tahiti and Darwin reflect passing weather systems that have nothing to do with ENSO. Intraseasonal oscillations, particularly the Madden-Julian Oscillation (MJO), can also influence regional weather independently of ENSO. In the Puget Sound region of the U.S. Pacific Northwest, for example, certain MJO phases trigger extreme precipitation and temperature events at lags of just a few days, and these connections hold regardless of how the ENSO signal is handled statistically.6Journal of Geophysical Research: Atmospheres. Impact of El Niño‐Southern Oscillation and Madden‐Julian Oscillation on the US Puget Sound Regional Hydroclimate The point for anyone using the SOI is that it captures the big, slow swing of the Pacific but does not account for every weather event layered on top.
What the SOI Means for Farming
Agriculture is one of the areas where the SOI has been put to the most direct practical use. In Argentina, researchers modeled maize yields across a range of soils and climates and found that the SOI phase meaningfully sorted crop seasons into wetter and drier categories. Seasons when the SOI stayed consistently negative (El Niño conditions) brought higher rainfall and higher grain yields, while consistently positive SOI phases (La Niña conditions) brought lower rainfall and lower yields. The study also found that the best planting strategy depended on the SOI phase: during El Niño years, early sowing produced the highest yields, whereas during La Niña years sowing date didn’t make much difference. Keeping crop stubble on the soil surface helped more during La Niña years, when rainfall was below average, by conserving moisture.7Field Crops Research. Modelled yield and water use efficiency of maize in response to crop management and Southern Oscillation Index in a soil-climate transect in Argentina
This kind of result is why agricultural agencies in Australia, parts of South America, and other ENSO-sensitive regions routinely publish SOI updates for farmers. Knowing the SOI phase months before the growing season can guide decisions about which variety to plant, how densely to plant it, and whether to invest in supplemental irrigation. The SOI is not a perfect crystal ball, but it shifts the odds enough to be worth consulting.
Marine Ecosystems and the Peruvian Upwelling
The eastern tropical Pacific, particularly off the coast of Peru, is one of the most productive marine regions on the planet. Cold, nutrient-rich water wells up from depth, fueling enormous blooms of phytoplankton that support anchovy fisheries and marine food webs. ENSO disrupts this process. During El Niño, the warm water that spreads eastward suppresses upwelling, and the nutrient supply drops. Modeling of primary production in the Peruvian upwelling zone over three decades found that El Niño events drove negative anomalies in biological productivity, while La Niña phases drove positive anomalies, with the strongest effects concentrated in the top 10 meters of the water column.8Ocean Modelling. Three decades of interannual variability in modeled primary production in the Peruvian Upwelling Region (1993–2022)
For fisheries managers and the communities that depend on Peruvian anchoveta, tracking the SOI and related ENSO indices is not academic. A strong El Niño can crash anchovy stocks for a season or longer, with cascading effects on seabirds, marine mammals, and the global fishmeal market. The 1972-73 El Niño, combined with overfishing, contributed to one of the most dramatic fishery collapses in modern history.
Fire, Drought, and Terrestrial Ecology
ENSO’s fingerprints show up in wildfire records too. In the inland Pacific Northwest of the United States, researchers used fire-scarred trees dating back to 1701 to investigate the relationship between fire occurrence and climate oscillations including the SOI. They found that the percentage of trees scarred by fire and the ENSO cycle were spectrally coherent at roughly six years, matching ENSO’s approximate periodicity. However, the overall conclusion was that ENSO was only a weak driver of fire occurrence in that region over the past three centuries; the Pacific Decadal Oscillation, a longer-cycle pattern, showed a stronger connection at its own roughly 47-year cycle.9Ecological Applications. Drought and Pacific Decadal Oscillation Linked to Fire Occurrence in the Inland Pacific Northwest
This is a useful reminder that the SOI matters more in some regions than others. In Australia and much of Southeast Asia, ENSO is the dominant driver of year-to-year rainfall variation, and a positive SOI (La Niña) typically means a wetter and greener landscape with lower fire risk, while a negative SOI (El Niño) means drought and heightened fire danger. In the Pacific Northwest of North America, that ENSO signal exists but competes with other oscillations. The practical takeaway is that the SOI’s predictive power is not uniform around the globe.
Dengue, Health, and Disease Spread
The connection between ENSO and human health might seem indirect, but it runs through temperature and rainfall. Warmer, wetter conditions expand the habitat and breeding season for mosquitoes that carry diseases like dengue fever. A continental-scale analysis across the Americas found that among the five years with the highest dengue case counts in their study period (1995-2005), four coincided with El Niño activity.10Geospatial Health. Geographical distribution of the association between El Niño South Oscillation and dengue fever in the Americas: a continental analysis using geographical information system-based techniques The researchers used SOI data from the Australian Bureau of Meteorology to classify years by ENSO phase.
The association is not deterministic. Dengue outbreaks depend on urban infrastructure, mosquito control programs, population immunity, and many other factors. But ENSO phase, as tracked by the SOI or its companion indices, gives public health agencies a heads-up: an incoming El Niño year in a dengue-prone region warrants ramping up vector-control efforts and hospital preparedness. Similar associations have been studied for cholera, malaria, and other climate-sensitive diseases, though the strength of the link varies by region and pathogen.
Energy Grids and Hydropower Vulnerability
Latin American countries rely heavily on hydropower, which makes their electricity supply vulnerable to ENSO-driven rainfall swings. During El Niño, parts of northern South America and Central America experience drought, reducing reservoir levels and hydroelectric output. A study evaluating complementarity between solar, wind, and hydropower across the region found that adding roughly 136 gigawatts of wind and solar capacity with high complementarity could compensate for hydropower fluctuations during drought ENSO phases. The benefits were substantial: up to a five-fold increase in the minimum threshold of renewable power available each month and up to a 65 percent reduction in generation variability, at a cost comparable to new natural gas plants.11Renewable Energy. Evaluating the complementarity of solar, wind and hydropower to mitigate the impact of El Niño Southern Oscillation in Latin America
This is an increasingly important planning consideration as countries push to decarbonize. If your grid depends on rain filling reservoirs, and you know the SOI can give you months of warning about whether it will be a dry year or a wet one, you can design a generation mix that is resilient to both phases. Wind and solar resources often peak at different times or in different ENSO phases than hydropower, providing a natural hedge.
What Climate Change May Do to ENSO
One of the open questions in climate science is whether global warming will change the behavior of ENSO, and by extension the usefulness of the SOI. The evidence here is genuinely uncertain. An analysis of future projections from the latest generation of climate models found that about half of the simulations shifted toward more El Niño-like conditions in the tropical Pacific, associated with a weakening of the east-west temperature gradient across the equatorial ocean. None of the models projected a significant shift toward La Niña-like conditions. But there was no consensus on whether ENSO variability, meaning the size of the swings back and forth, would increase or decrease.12Journal of Geophysical Research: Atmospheres. Future Projections of the El Niño—Southern Oscillation and Tropical Pacific Mean State in CMIP6
A separate study using large model ensembles found that ENSO amplitude projections are nonlinear over time and differ substantially between models. A majority of models showed an increase in ENSO amplitude, but some showed a decrease, and several showed a nonmonotonic pattern where amplitude first rises, then plateaus, then falls.13Earth System Dynamics. The future of the El Niño–Southern Oscillation: using large ensembles to illuminate time-varying responses and inter-model differences The honest summary is that we should expect ENSO to keep oscillating, the background state of the tropical Pacific may shift in an El Niño-like direction, but the question of whether future El Niños and La Niñas will be bigger or smaller than today’s remains unresolved.
For the SOI specifically, a shift in the mean state would change what “normal” pressure looks like at Tahiti and Darwin. Since the SOI is defined as an anomaly from the long-term average, the index itself can be recalibrated. But if the underlying ENSO dynamics change in character, the relationships between the SOI and regional weather patterns that farmers, fisheries managers, and public health agencies rely on could shift too. That is the deeper concern: not that the index would break, but that the playbook for interpreting it might need updating.
How to Read the SOI in Practice
If you want to check the SOI yourself, the Australian Bureau of Meteorology publishes monthly values along with a 30-day running average. Sustained values below about -7 suggest El Niño-like conditions; sustained values above about +7 suggest La Niña. Values between those thresholds are considered neutral. The Bureau also categorizes recent months into SOI “phases” based on how the index has behaved over the preceding couple of months, giving a quick shorthand for whether conditions are trending warm, cool, or staying flat.
When reading the SOI, patience matters. A single month at -10 does not make an El Niño. The atmosphere is inherently variable, and one strong weather system passing through Tahiti can temporarily push the index around. Most agencies require several consecutive months of sustained values in El Niño or La Niña territory, coupled with consistent oceanic signals, before declaring an event. If you are using the SOI for any kind of planning, whether for planting decisions, energy portfolio management, or disease surveillance, the multi-month trend is what counts, not the latest single reading.