El Niño–Southern Oscillation

El Niño–Southern Oscillation (ENSO) is a recurring fluctuation of the coupled ocean and atmosphere over the tropical Pacific. It cycles among a warm phase (El Niño), a cold phase (La Niña) and a neutral state. Sea surface temperature anomalies in the central and eastern equatorial Pacific, often only a few degrees Celsius, shift rainfall, drought and storm patterns across continents far from the Pacific. ENSO is the largest source of year-to-year climate variability on Earth, and it is a major source of skill in seasonal climate prediction. The name combines the ocean warming known as El Niño with the "Southern Oscillation," an atmospheric seesaw in surface pressure between the eastern Indian Ocean and the Pacific that was described in the 1920s.

The phenomenon rests on a feedback loop between the atmospheric Walker circulation and the ocean's surface temperature and thermocline structure. A weakening of the easterly trade winds allows warm water to spread eastward, which further weakens the winds. Heat stored in the equatorial Pacific then discharges and reverses the cycle, so the system oscillates irregularly, usually every two to seven years. The cycle links to monsoon systems, tropical cyclone activity, marine ecosystems and temperature and precipitation patterns on every inhabited continent, which makes ENSO central to both physical climate science and applied forecasting.

As of NOAA's Climate Prediction Center diagnostic discussion of 8 October 2026, an El Niño Advisory was in effect and the event was still strengthening, with forecasters assigning high odds to a strong-to-very strong peak in late 2026 or early 2027. This entry covers the definition and indices of ENSO, the atmospheric and oceanic mechanisms, the phases and their global impacts, the paleoclimate and climate-change context, the observing and forecasting systems, the history of research, and the questions that remain open.

Definition, Phases and Indices

El Niño–Southern Oscillation (ENSO) is a coupled phenomenon. The sea surface of the tropical Pacific and the overlying atmosphere continually adjust to each other, so the cycle cannot be described by the ocean or the atmosphere alone. Warm water shifts the location of rising air, and changes in the winds then reshape the ocean's surface temperatures and subsurface structure.

ENSO has three phases: El Niño (warm), La Niña (cool) and neutral. Each is defined by sustained sea surface temperature (SST) anomalies in the central and eastern equatorial Pacific, measured against a long-term average. The threshold most often used is about ±0.5 °C. In NOAA's operational practice, the anomaly must reach that threshold for at least five consecutive overlapping three-month periods before an event is classified as El Niño or La Niña.

Several fixed boxes serve as monitoring regions. Niño-1+2 covers the coastal eastern Pacific off Peru and Ecuador, Niño-3 the eastern equatorial Pacific, Niño-4 the central-to-western equatorial Pacific, and Niño-3.4 the central equatorial Pacific between about 170°W and 120°W. Niño-3.4 is the most widely used of these. The Oceanic Niño Index (ONI) is the three-month running mean of Niño-3.4 anomalies, computed relative to a 30-year climatology whose base period is updated periodically. Under the base period used in NOAA's real-time record, which a 2026 WMO assessment gives as 1996–2025, the index is refreshed as the reference climate shifts.

Warming of the whole tropical ocean complicates this approach, because a fixed baseline can make a given Niño-3.4 anomaly look stronger than it is relative to the surrounding tropics. The Relative Oceanic Niño Index (RONI) addresses this by subtracting the tropical-mean SST anomaly, usually taken over about 20°S–20°N, from the Niño-3.4 anomaly. The result is then rescaled to the variance of the original Niño-3.4 series and smoothed with a three-month mean. According to NOAA-related sources, RONI became NOAA's primary classification index in February 2026, while ONI remains in use as a companion measure. The trend correction matters in practice: a WMO analysis estimates that tropical SST has been rising by roughly 0.11 °C per decade, enough to shift the effective El Niño threshold by about 0.3 °C over three decades.

The Southern Oscillation Index (SOI) takes the atmospheric side of the cycle. It is based on the difference in sea-level pressure between Tahiti in the eastern Pacific and Darwin in northern Australia, normally standardized so that its sign and size can be compared over time. Sustained negative SOI values generally accompany El Niño, and positive values accompany La Niña. Because the SOI measures pressure rather than temperature, it is often used together with SST-based indices to check whether the ocean and the atmosphere are behaving in step.

The name "El Niño" comes from Spanish-speaking fishermen on the coasts of Peru and Ecuador. They used the term for a warm coastal current that appeared around Christmas, and it was associated with empty nets and a disrupted fishery. Accounts differ on when the usage began: some place it in the 1600s, while Britannica dates its use by northern Peruvian fishermen to the 19th century. Over time, the term came to refer to the irregular, stronger warm episodes rather than to the annual current. Scientists later showed that these episodes were tied to changes across the entire tropical Pacific and beyond, which led to the use of the broader term ENSO. The name La Niña, meaning "the girl," was coined for the opposite, cool phase.

The Walker Circulation and the Southern Oscillation

The atmospheric half of ENSO was identified decades before its link to the ocean was understood. In 1904 the British climatologist Gilbert Walker began analysing weather records to find out whether the Indian monsoon could be forecast, and he found that surface pressure across the tropical Indo-Pacific fluctuated in a coherent way. He called this fluctuation the Southern Oscillation. In a series of papers in the 1920s and 1930s, he showed that pressure at Tahiti and Darwin tended to move in opposite directions, so that when pressure rose in the east it usually fell in the west. The pattern is often described as a "seesaw," and its statistical signal later became the basis of the Southern Oscillation Index (SOI), which is discussed in the definition and indices section. Walker's work was based on station data and correlation analysis rather than any physical model, and he did not initially connect the seesaw to El Niño.

The overturning circulation that accompanies this pressure pattern is now called the Walker circulation. It is a zonal cell that runs along the equator. Air rises over the warm western Pacific, where the pressure is low and convective clouds build. The air then flows eastward at upper levels, sinks over the cooler eastern Pacific, where the surface pressure is high and skies are comparatively clear, and returns westward near the surface as the trade winds. The circulation is driven mainly by the east–west contrast in sea surface temperature, which makes it a thermally direct cell. Because it is tied to sea surface temperature, it responds strongly to changes in the ocean.

The western Pacific warm pool is the anchor of this system. Persistent easterly trade winds push surface water westward, so warm water accumulates around Indonesia and the Maritime Continent, and the thermocline there is deeper than in the east. Warm, moist air over this pool produces the deepest and most persistent convection in the tropics, which is why rainfall is concentrated over the Maritime Continent. During El Niño, the warm water spreads eastward and the rising branch of the Walker cell follows it. Rainfall and cloud move with the convection toward the central and sometimes eastern Pacific, while Indonesia, Southeast Asia and northern Australia lose rainfall and become drier. The eastward displacement of deep convection is therefore the atmospheric expression of the event, not merely a side effect of it.

The Walker cell also varies from year to year, and this variability forms the atmospheric component of the ENSO cycle. Its strength is closely tied to the surface pressure gradient across the Pacific: a larger pressure difference between the eastern and western ends of the basin is associated with stronger trade winds, while a reduced difference is associated with weaker winds. Negative SOI values, which indicate a weaker gradient, therefore accompany weakened trades and the warm-water shift that defines El Niño, whereas positive values accompany strengthened trades, enhanced upwelling and La Niña conditions. Because the trade winds are the main link between the atmosphere and the ocean, the strength of the Walker cell is one of the clearest indicators of which phase of the cycle the system is in. This is also why the Walker circulation is sometimes described as ENSO's atmospheric counterpart.

Walker's seesaw was thus a statistical description of a pattern whose physical cause remained unknown for decades. Walker recognised that the pressure pattern influenced rainfall and temperature across many parts of the world, and this observation laid the foundation for later work on teleconnections. The physical link between the seesaw and the warm events in the eastern Pacific was established only in the late 1960s, when Jacob Bjerknes showed that the two phenomena are the atmospheric and oceanic expressions of a single coupled system. The mechanisms of that coupling are described in the coupled feedback section.

Coupled Ocean–Atmosphere Feedback and the ENSO Cycle

The cycle rests on a positive feedback between the equatorial Pacific sea surface and the trade winds, which is known as the Bjerknes feedback. Along the equator, the eastern Pacific normally has cold surface water (the "cold tongue") while the western Pacific holds a warm pool. This temperature contrast maintains the Walker cell and the easterly trade winds, and those winds in turn drive ocean currents and upwelling that keep the east cool. If an initial warm anomaly appears in the east, the trade winds weaken. Weaker winds reduce upwelling and allow the thermocline, the layer separating warm surface water from cold deep water, to deepen. The surface then warms further, which weakens the winds again. The loop can therefore amplify a small anomaly over several months. Jacob Bjerknes described this coupling in the 1960s and 1969, and it remains the core growth mechanism in most conceptual models.

Growth is only half of the story, and the ocean's response to wind changes sets the timing. Weakened easterlies, and especially westerly wind bursts, excite downwelling oceanic Kelvin waves that travel eastward along the equator. These waves deepen the thermocline in the east and push warm water toward the coast of South America. Off the equator, oceanic Rossby waves carry a similar adjustment westward. When they reach the western boundary of the basin, they reflect as Kelvin waves that return eastward along the equator. Because these waves move far more slowly than the atmosphere responds, the adjustment arrives with a lag of several months. Upwelling is the key link here. In normal conditions, cold water rises from below the thermocline along the Peruvian and Ecuadorian coast and spreads westward along the equator. During El Niño, the thermocline sits deeper, so less cold water reaches the surface. The surface then receives less cooling from upwelling, and it gains heat from the atmosphere, although that heat gain acts as a damping force that is too weak to offset the ocean's warming.

The growth phase needs a delayed negative feedback to end an event. Several conceptual models have been proposed to supply it. The delayed oscillator relies on Rossby waves that reflect at the western boundary and return as Kelvin waves, which reverse the thermocline anomaly after a delay. The recharge–discharge oscillator focuses on the total heat content of the equatorial upper ocean. During El Niño, heat is discharged from the equatorial band, which lowers the heat content and eventually allows cold conditions to take hold. Between events, the heat content gradually rebuilds, which sets up the next transition. Other models emphasize Sverdrup transport, western-Pacific wind-forced waves, or advection at the eastern boundary. Most current frameworks incorporate the Bjerknes feedback together with one or more of these delayed processes. The delay is what turns a warm event into a cold one, and a cold event into a warm one, which gives ENSO its oscillatory character.

The cycle is not a clock. Its period varies from event to event because the system is driven by irregular inputs. Fast atmospheric disturbances, including stochastic westerly wind bursts and the Madden–Julian Oscillation, can trigger or intensify warm episodes, but they are not themselves a product of ENSO. Nonlinear processes also matter. The response of the thermocline to a wind anomaly depends on the mean state, so warm and cold phases do not mirror each other exactly. Models that treat the feedbacks as linear tend to misrepresent this asymmetry. Modeling studies have also reported that some climate models underestimate the strength of key elements of the Bjerknes feedback. Together, these sources of randomness and nonlinearity explain why events differ in timing, intensity and duration even when the underlying feedbacks are the same.

Phases, Event Types and Variability

An El Niño event usually follows a recognisable seasonal sequence. Warm anomalies begin to build in the equatorial Pacific in boreal spring or summer, strengthen through northern summer and autumn, and peak in late autumn or winter. By the following spring they generally decay toward neutral conditions. La Niña develops in a broadly mirrored way, with cold anomalies that peak in winter and fade in spring. The cycle is a tendency rather than a schedule: some events peak early, others linger, and a fraction persist into a second winter.

The two phases produce opposite signatures across the tropical Pacific. During El Niño, the warmest surface water shifts eastward, and deep convection and heavy rainfall move with it toward the central and eastern basin. Weaker trade winds reduce the upwelling of cold, nutrient-rich water along the coasts of Peru and Ecuador, so surface waters there warm further. La Niña reverses these tendencies: trade winds strengthen, cold water wells up more strongly in the east, and convection concentrates over the western Pacific and the Maritime Continent. The phases are not symmetrical in magnitude, and the strongest warm events are considerably larger than the strongest cold ones. Strong El Niño events are often followed by La Niña conditions. The 1997–98 event, for example, was followed by a La Niña that persisted into 2001, while the 2015–16 event was followed by a short, weaker La Niña in 2016–17.

Events are also classified by where the warming peaks. In the eastern Pacific (EP) type, the largest warm anomalies lie close to the South American coast, and the anomalous convection shifts toward Peru and Ecuador. In the central Pacific (CP) type, which is often called the dateline or Modoki type, warming is confined to the central equatorial Pacific, and the anomalous convection develops west of the EP location. The difference matters for impacts. EP events are linked to the catastrophic floods that affect northern Peru and Ecuador, and these are largely absent in CP events. In the Asian monsoon region, EP events are associated with reduced precipitation, whereas moderate CP events in 2002 and 2004 coincided with serious dry spells in India, and the very strong 1997–98 event had an unusually limited effect on the Indian summer monsoon. The type also shapes the North American winter response: according to one study based on reanalysis and model experiments, the surface temperature pattern over the United States rotates by roughly 90 degrees between EP and CP events. Many events are mixed, so the two types are best viewed as end members of a continuum.

Strength is usually assigned from peak values of a three-month Niño-3.4 anomaly, using the ONI or, in NOAA's newer practice, the RONI. Unofficial thresholds commonly used in climate communication are about 1.5 °C for "strong" and 2.0 °C for "very strong" or "historically strong". NOAA's own categories are weak, moderate, strong and very strong, and its monthly outlooks give the probability of each category for successive overlapping three-month periods, so that public statements about strength generally refer to these forecast odds rather than to a single observed value. The term "super El Niño" is an informal media expression for a very strong event, with a threshold of about +2.0 °C. It has no official definition, and it is sometimes applied to forecast ensemble averages before any observed peak exists, which can overstate what the observations later show.

ENSO's spectral power is concentrated in an irregular band of roughly two to seven years, so the cycle is better described as a broad range of recurrence times than as a fixed clock. This irregularity is visible in the time series of Pacific temperatures, where some intervals contain a single brief event and others contain sustained warm or cold episodes. Multi-year, or "double-dip", episodes are an important part of this variability. The 2010–12 La Niña spanned two consecutive winters, and the 2020–23 La Niña was a "triple-dip" event lasting three winters. Such prolonged episodes can have cumulative effects on rainfall, drought and ecosystems that exceed those of a single season with a similar peak.

Global Teleconnections and Weather Impacts

A teleconnection is a link between weather or climate at distant locations, carried through the atmosphere. ENSO is the largest source of interannual teleconnections because it moves the zone of deep tropical convection. During El Niño, the convection and the Hadley circulation expand eastward from the western Pacific, sometimes reaching the west coast of South America. The shift changes where air rises and sinks around the globe, which moves the subtropical jet stream and the storm tracks that carry weather into the middle latitudes. The effects therefore appear far from the Pacific.

The tropical teleconnections are the most direct. Rainfall tends to decline over Indonesia and northern Australia, where droughts and fire risk increase, while the coasts of Peru and Ecuador receive heavier rain as warm water and enhanced convection move eastward. In northeastern Brazil, El Niño generally reduces rainfall and brings drought, whereas southern Brazil tends toward wetter conditions. Brazil's southeast and central-west show no uniform signal and are usually assessed case by case. The pattern is therefore a shift in the odds across a region, not a continent-wide rule.

The Asian and Australian monsoons are weakened on average during El Niño. Indian summer monsoon rainfall shows a statistically significant inverse relationship with ENSO, although the correspondence is not one-to-one. Analyses of the past century found that most of India's severe droughts coincided with El Niño, but a substantial number of droughts occurred without it. The mechanisms run through changes in the Walker circulation and through the "atmospheric bridge" proposed by Lau in 1996, which links tropical Pacific sea surface temperature anomalies to the Indian and Atlantic Oceans. Interactions with the Indian Ocean Dipole, discussed in a separate section, can reinforce or offset these effects.

Atlantic tropical cyclone activity is also affected, chiefly through vertical wind shear, the change in wind speed and direction with height. El Niño strengthens upper-level westerlies over the tropical Atlantic. The added shear tears developing storms apart, and El Niño also increases sinking motion and atmospheric stability, so fewer hurricanes form during the August–October peak. La Niña weakens the upper-level westerlies, reduces shear, and generally favors more and stronger Atlantic hurricanes. The eastern and central Pacific show the opposite tendency, with cyclone activity generally increasing during El Niño. The responses differ from event to event, depending on local conditions that can amplify or dampen the ENSO signal.

Extratropical teleconnections over North America are strongest in the cold half of the year, roughly October through March. El Niño tends to produce wetter winters in the southern United States and drier, warmer conditions in parts of the north, with the strongest signals appearing from the West to the South. Western signals generally become apparent in January. In the Intermountain West, ENSO effects on winter precipitation are shaped by topography: the high-elevation western slopes show a much larger range of anomalies than lower terrain. The southwestern precipitation link was strongest in the late twentieth century and has weakened in recent decades, and some years with El Niño have not produced the expected wet anomaly. These fluctuations are one reason the impacts are described in probabilistic terms.

Teleconnections should therefore be understood as shifts in probability. Official outlooks, such as those of the U.S. Climate Prediction Center, express how much more or less likely temperature and precipitation are to fall above or below average. Local outcomes depend on the strength of the event, the state of other climate modes such as the North Atlantic Oscillation, and short-lived weather systems, including atmospheric rivers, that can either reinforce or override the canonical pattern.

Ecological, Agricultural and Health Effects

Marine ecosystems and fisheries

Marine life responds quickly to El Niño. Prolonged warmth drives symbiotic algae out of coral tissue, leaving colonies pale, and corals will expel their algae partners causing them to turn white in a process called coral bleaching. Sometimes corals can recover from this event, but not always. Reef surveys at Isla del Caño, Costa Rica, recorded about half of the corals dying during the 1982–83 event, as the coral reef record shows. In northern Venezuela, a marked decrease in coral extension rates coincides with a baseline shift to more negative values in oxygen and carbon isotopic composition after 1997–98, while a neighboring coral recovered to pre-bleaching extension rates simultaneously.

The Peruvian anchoveta fishery shows the economic side of these events. Catches fell sharply during the 1972 event, and the drop was steep: the fishery declined 55% in 1972 and 51% in 1973, leading to severe economic hardship. Overfishing made the damage worse, since the previously abundant population of anchovetas began to heavily deplete as a result of overfishing from previous seasons and as a result of that year's strong El Niño current. The anchoveta feeds on plankton that depend on nutrient-rich upwelling, which El Niño weakens. In recent strong events, quota cuts and season closures have followed, and this has led to record-high fishmeal prices which will make aquaculture food more expensive and lead to higher fish farming costs globally.

Terrestrial agriculture

Crop yields respond to ENSO-driven rainfall and temperature shifts in several major growing regions. El Niño usually brings drier summers to southern Africa, where maize harvests are highly exposed to drought, and weakens the monsoon rains that sustain South Asian crops. Eastern Australia tends to receive less rain during El Niño, which reduces wheat and barley output. In South America, the same event often brings wetter conditions to the Argentine Pampas, while the reverse tends to occur during La Niña. Effects vary by season, soil and irrigation access, so the same event can raise yields in one region and cut them in another.

Fire, smoke and peatlands

Indonesia's peatlands are among the most sensitive terrestrial systems to ENSO. Devastating forest fires struck Indonesia in 1997 and 1998, burning nearly 10 million hectares of forest. Estimates based on burn-depth measurements put the carbon released from peat combustion in Central Kalimantan in 1997 at 0.19–0.23 gigatonnes (Gt) of carbon. The vulnerability is largely a product of land management: Indonesian peatlands rarely experienced fire until recent decades, and drainage and deforestation of extensive areas of peatland in Indonesia make the naturally fire-resilient peatland susceptible to fire. Smoke from these fires has moved across borders, and the haze has reached neighbouring Malaysia, Singapore and the Philippines.

Vector-borne and nutritional health effects

Changes in temperature and rainfall alter the habitats of disease-carrying insects, and this shapes malaria risk. In Venezuela, El Niño is associated with low rainfall, and the biggest increases in malaria incidence occur a year after recognized El Niño events. Documented epidemic jumps have also occurred in Peru, Rwanda, Pakistan and Sri Lanka, among other places. For dengue, the evidence is more mixed. The extrinsic incubation period (EIP) is shortened at higher temperatures, thus potentially increasing the proportion of mosquitoes that become infectious at a given time. Most regional studies link El Niño with more reported dengue, but a recent study on Solomon Islands estimated a risk reduction in dengue-like-illness.

Food insecurity is another pathway. In Central America, the WHO's 2023 public-health analysis expected lower harvests to feed into rising malnutrition due to reduced agricultural yields, which compounds the risks of malnutrition and public health strain. Floods and droughts linked to ENSO thus affect both the availability and the affordability of food.

Economic exposure and risk management

ENSO affects many economic sectors. Hydropower output drops when river flows fall, as happened at Lake Kariba during the 2015–16 drought. Water supply and navigation can also be disrupted, and the 2015–16 event forced the Panama Canal to restrict transits. Insurers and governments use ENSO forecasts to price drought risk and design index-based cover, while fisheries managers, water utilities and agricultural planners use them to adjust quotas, reservoir operations and planting schedules.

Paleoclimate and Long-Term Variability

Instrumental records cover only a few decades of reliable ENSO observations, so most knowledge of long-term behaviour comes from paleoclimate proxies. The amplitude of ENSO variability affects the occurrence and predictability of climate extremes around the world, but the instrumental record is too short to characterize its natural variability. Proxy archives fall into two groups. The first lies in the core region itself, mainly fossil coral and mollusk records and some marine sediments. The second sits in remote areas that feel ENSO through teleconnections, such as lake sediments, tree rings, speleothems and ice cores. Each archive has its own sensitivities, so they can be difficult to intercompare, and records from several archives are usually combined.

Coral and shell records are the most direct. Oxygen isotopes in coral skeletons track sea surface temperature, but most published fossil coral records do not extend to time intervals prior to the Holocene. Giant clam (Tridacna) shells from the western Pacific warm pool have been dated to more than 60,000 years, and some rarely preserved corals from Papua New Guinea suggest that the ENSO variability has existed for the past 130,000 years of glacial and interglacial times. Lake sediments give a different kind of evidence. At Laguna Pallcacocha in Ecuador, hundreds of mineral-rich flood laminae result from eastern Pacific El Niño events, when convective rainstorms drive erosion and terrigenous sediment transport in the watershed. The Galápagos El Junco record instead uses grain-size data to infer past lake levels and ENSO frequency.

For the Holocene, the proxies agree on a broad pattern but differ in detail. The majority of these records suggest that ENSO frequency was substantially reduced during the early to mid-Holocene, and increased in frequency sometime in the last few thousand years. A study of Tridacna shells from the South China Sea found that ENSO variability in the early Holocene was comparable to that of today, and that a minimum in the frequency and intensity of ENSO activity occurred in the mid Holocene. A Line Islands coral network indicates a significant decrease in ENSO variance of ~20% from 3,000 to 5,000 years ago, coinciding with changes in spring/fall precessional insolation. The Pallcacocha record also shows a minimum that is not present in the high-intensity flood stratigraphies from the other two lakes, which shows that single archives can give conflicting pictures. Many climate models reproduce the mid-Holocene weakening qualitatively, but most of the simulations fail to capture the magnitude of the ENSO reduction inferred in the proxy, which could amount to 30–60%.

Evidence from earlier periods is sparser and more contested. Coral records from the Last Glacial Maximum are rare, and constraints during the LGM (19–24 ka) in particular are virtually non-existent. Foraminifera-based temperature reconstructions indicate that a deep equatorial thermocline altered the dynamics in the eastern equatorial cold tongue, resulting in reduced ENSO variability during the LGM compared to the Late Holocene. Other surface and subsurface proxies, however, have been read as showing stronger variability at the LGM, so proxy records so far show contradictions on the strength of ENSO during the glacial and deglacial periods. A Peruvian sediment study estimated that the amplitude of ENSO events was 50 to 190 % larger during the deglaciation compared to the Late Holocene, which the authors linked to meltwater discharge. For warmer interglacials, a record from northeastern Australia found that during globally warmer periods there are significantly larger amplitudes in high-frequency ENSO spectral range (3-8 years), which are absent from the record during the glacial stages, suggesting that ENSO variance depends on mean climate.

Paleo records also show multidecadal modulation of ENSO, which the short instrumental record cannot resolve well. A tree-ring-based reconstruction over 1,100 years found that ENSO amplitude exhibits a quasi-regular cycle of 50-90 years. A proxy index spanning 1525–1982 showed considerable amplitude and frequency modulation in the 310 year band on multidecadal time scales. Such variations are hard to separate from sampling noise, because the instrumental record is too short to estimate the uncertainties for fossil records, which sometimes exceed 100 years in length. The interdecadal Pacific oscillation may also change how ENSO affects precipitation, and variables such as precipitation and streamflow also appear to vary at interdecadal frequencies in the instrumental record and in longer-term paleo reconstructions.

Paleo records also serve as a baseline for judging recent changes. ENSO appears to have operated on Earth for at least 3 million years, and the existing observations support the view that variations in ENSO amplitude and frequency arise primarily from processes internal to the climate system. Reconstructions show that several reconstructions show broadly similar patterns, including periods of reduced ENSO variability in the mid 14th century, mid 18th and 19th centuries. Because the Holocene and last interglacial have been used to test model responses, comparing modern activity against these intervals helps establish whether recent changes exceed natural variability.

Climate Change and Projected ENSO Behavior

Warming of the tropical Pacific is already altering the mean state in which ENSO operates. The western tropical Pacific has warmed dramatically, while unusually persistent upwelling of cool subsurface water has produced a slight drop in average sea surface temperature over much of the eastern tropical Pacific. The result is a strengthening west-to-east temperature contrast that increasingly resembles La Niña. Because the Walker circulation is driven by the zonal sea surface temperature difference across the basin, the direction of future change matters. If the western Pacific keeps warming faster, the background circulation could become more La Niña-like; if the east warms faster, the whole circulation could shift toward an El Niño-like state, and ocean events would then develop against a different background from today's (climate change trend pattern, see the NOAA analysis of the trend pattern).

Theory points in both directions. The eastern Pacific could warm more due to cloud feedbacks, evaporation feedbacks, or a weakening of the Walker circulation. Against this, the ocean could also oppose warming in the east because increased stratification limits the upward transport of cold water. Most climate models project a long-term pattern featuring faster warming in the eastern equatorial Pacific than the surrounding oceans, and across models the east–west warming difference explains about half of the inter-model spread in how the eastern and central Pacific ENSO signals diverge. The two ENSO flavors respond differently. Rainfall variability is projected to increase more in the eastern Pacific region than in the central Pacific region, and CMIP6 models show that central Pacific ENSO has a robust decrease of persistence and predictability under global warming, notably in passing through the boreal spring, while the predictability of Eastern Pacific ENSO will not change.

Projections of ENSO amplitude and frequency remain highly uncertain. Model diversity has narrowed over successive generations, since ENSO amplitude diversity has been reduced by a factor of two from CMIP3 to CMIP5, yet the inter-model disagreement on the ENSO amplitude change persists in CMIP6 models. In one large-ensemble analysis, the model projections exhibit a large spread, ranging from increasing standard deviation of up to 0.6 °C to diminishing standard deviation of up to − 0.4 °C by the end of the twenty-first century, and the ensemble-mean ENSO amplitude change is close to zero. Internal variability dominates the uncertainty in the first three decades, while model uncertainty dominates thereafter. Event counts show a similar pattern: an ensemble of 13 models gives a total of El Niño and La Niña events that remains broadly stable across future scenarios compared to the historical baseline, but some models project pronounced increases in very strong El Niño and La Niña events under future warming, whereas others show more moderate changes. Systematic model biases feed directly into these disagreements. All models overestimate the westward extension of ENSO SST anomalies and either overestimate or underestimate the ENSO amplitude, and the representation of stratospheric ozone can by itself have a first‐order impact on ENSO amplitude projections.

Because projected changes in ENSO amplitude are uncertain, projected changes in its impacts are too. Several studies agree on one point: the fraction of extreme precipitation days occurring during El Niño years increases in the future for all models, while precipitation extremes are predicted to be nearly the same or less frequent during La Niña years in all models. Attributing observed changes to human influence is harder. The tropical Pacific has been trending toward a La Niña-like pattern for long enough that it's gotten tougher to ascribe to natural variability, yet the Pacific Decadal Oscillation, which could account for part of it, is largely influenced by ENSO itself, so the two signals are hard to separate. The short instrumental record makes this separation harder still.

ENSO also interacts with other ocean basins. The Indian Ocean Dipole is identified as a mode of variability internal to the Indian Ocean, but is also partially driven by ENSO, with positive and negative IOD events associated with El Niño and La Niña, respectively. The IOD can also feed back: the Indian Ocean Dipole is an Indian Ocean internal mode that can actively affect ENSO. Under future forcing, the anticorrelation between ENSO rainfall and the rainfall in the expanded eastern IOD node strengthens from −0.39 (HIST) to −0.59 (21C), which implies that ENSO would drive a larger share of Indian Ocean rainfall variability. A related link runs to the Atlantic, where positive IOD weakens the easterly trade winds over the tropical Atlantic, causing warm anomalies in the central and eastern equatorial Atlantic basin and therefore triggering the Atlantic Niño. The Atlantic Niño in turn enhances seasonal rainfall over South America and Africa, and it can also weaken the Indian summer monsoon through atmospheric teleconnection. Projections of the IOD's teleconnection to the North Atlantic are themselves changing: one study finds that the influence of the IOD on the polar vortex roughly halves in the 21st century, which reduces the Indian Ocean pathway's role in shaping regional circulation.

Observation, Monitoring and Forecasting

ENSO forecasting depends on a sustained observing network, and its in-situ backbone is the TAO/TRITON mooring array in the equatorial Pacific. The array comprises 55 TAO moorings developed by NOAA's Pacific Marine Environmental Laboratory and maintained by NDBC, 12 TRITON moorings maintained by JAMSTEC, and 5 subsurface current-profiling moorings. The array was developed over the 1985–94 Tropical Ocean Global Atmosphere (TOGA) program for improved detection, understanding, and prediction of ENSO. Each mooring carries instruments that record temperature, salinity, velocity, and surface meteorological data, and the stations sit at roughly 15° longitude intervals across the Pacific. The spacing is designed to capture the equatorial waves that drive the cycle, and using these data generally improves the temperature and salinity reconstruction, especially in the eastern tropical Pacific. Continuity is not assured: JAMSTEC, NOAA's principal partner in maintaining TAO since 2000, announced a phased withdrawal of TRITON moorings from the western Pacific.

Moorings are complemented by satellite observations. Altimeters track sea-surface height, which reflects changes in the depth of the thermocline, while radiometers measure sea-surface temperature across the whole basin. Profiling floats drifting through the upper ocean fill the gaps between moorings with subsurface temperature and salinity profiles.

Forecast systems fall into several families. Statistical models extrapolate empirical relationships among observed indices, whereas dynamical coupled ocean–atmosphere models integrate the governing physics on a grid. Hindcast verification has favoured the latter in some periods: dynamical models exhibited a statistically significant advantage over statistical models when predicting ENSO events during the months of March to May. Operational outlooks combine these tools into ensembles. The CPC Official ENSO outlook, which synthesises multiple models from North America and abroad, indicates El Niño will continue to strengthen through the end of the year. Ensemble spread converts this into probabilities rather than a single trajectory. Data-driven approaches, such as neural-network methods, are still largely at the research stage; one convolutional LSTM study reported that the monthly mean Nino3.4 index can be skillfully predicted up to one year ahead, and the model was also able to predict strong El Niño cases, such as 1997–1998 and 2015–2016 a year ahead. Even so, prediction skill remains limited beyond 1 year.

The main obstacle to seasonal skill is the spring predictability barrier. Prediction accuracy sharply declines for predictions initialized from February to May, and for statistical forecasts correlation skill swiftly declines for longer lead forecasts, reaching values as low as 0.3 or even lower. The barrier is not uniform across event types: Central Pacific events have stronger spring forecast barriers than Eastern Pacific events. Forecasters try to reduce its effect by widening the set of predictors. Influences from other basins help here: in one deep-learning experiment, the exclusion of Indian Ocean effects reduces the effective lead time from 17 months to 12 months. Sea-surface salinity has also been tested, and in a linear dynamical model it plays an important role in improving the forecast skill of ENSO and weakening SPB.

Official products are issued by national centres. The Climate Prediction Center assigns an alert status, and the current one is El Niño Advisory. The monthly diagnostic discussion appears once a month (2nd Thursday of each month), and between releases the oceanic and atmospheric conditions are updated weekly on the Climate Prediction Center web site. Since February 1, 2026, the Relative Oceanic Niño index (RONI) is used for official monitoring and prediction. The World Meteorological Organization periodically consolidates national assessments into its own El Niño/La Niña updates.

As of the 8 October 2026 discussion, most Niño indices increased in the past month, reaching +2.1°C in Niño-3.4, +3.0°C in Niño-3, and +3.9°C in Niño-1+2. The historic-event probabilities are 54% for September-November, 83% for October-December, and 70% for November-January, where a historic event is defined by a three-month RONI of +2.5 °C or more. The September discussion had given a 75% chance of a historic event that would exceed the strength of previous El Niño events dating back to 1950 for October–December. The October outlook also keeps a strong-to-very strong El Niño likely through January-March 2027 (remaining greater than an 83% chance). NOAA notes that the chances of experiencing impacts consistent with El Niño are larger, though not guaranteed, and the next discussion is scheduled for November 12.

Research History

The scientific understanding of ENSO was assembled in stages, beginning with statistical work in the 1920s. Gilbert Walker described a seesaw in surface pressure between the Indian Ocean and the tropical Pacific and named it the Southern Oscillation. Walker offered no physical mechanism, so many meteorologists treated the pattern with suspicion, and it could have been a statistical artefact of a short record. The gap was closed in 1969, when Jacob Bjerknes published a study that connected the Southern Oscillation to the equatorial Pacific. He identified wintertime atmospheric teleconnections over the North Pacific that accompanied warm episodes in the equatorial ocean, and he argued that these were part of Walker's global pattern. Bjerknes proposed that changes in the trade winds are tied to the east–west sea surface temperature gradient through the Walker circulation. A warmer eastern Pacific weakens the easterlies, which reduces upwelling and warms the surface further, so the feedback reinforces itself. This positive feedback is still known by his name. He could not explain, however, what turns a warm phase back into a cold one. Oceanographers and meteorologists then began to combine their efforts in the mid-1970s. Wyrtki's 1975 idea of sea-level "buildup" in the western Pacific warm pool treated El Niño as a more isolated event, and a first attempt to forecast an El Niño was made in 1974.

Forecasting progressed in the 1980s, when ENSO was studied far more intensively. The strong warm episode of 1982–83 was not recognized until it was well developed, which galvanized the research community. Mark Cane and Stephen Zebiak of Lamont-Doherty Earth Observatory built a simple coupled model, combining a steady-state linear atmosphere with a reduced-gravity ocean, to examine how the Pacific behaves. In 1986 Cane, Zebiak and Dolan reported experimental hindcasts of El Niño events since 1970, suggesting that events were generally predictable one to two years ahead. The same year the model was used in real time, and it predicted the onset of the 1986–87 event. A separate study by Barnett and colleagues in 1988 showed that three different classes of numerical models also predicted that event at lead times of three to nine months. The Zebiak–Cane model is often described as the first dynamical model to forecast an ENSO event successfully, and it became a benchmark for later work. Over the 1990s, more sophisticated models that included a full global atmosphere replaced it, and they laid the foundation for routine seasonal outlooks.

The Tropical Ocean Global Atmosphere (TOGA) programme was the main institutional response to the 1982–83 event. It ran as a ten-year international effort from 1985 to 1994 with the aim of studying and predicting ENSO. Its most lasting legacy was an observing system for the tropical Pacific, which included the moored buoy array known as TAO/TRITON. Before TOGA, measurements of subsurface ocean temperature were very sparse, even though thermocline changes were known to be central to ENSO. The array supplied the data needed to test the theories and to initialize forecast models. The strong 1997–98 event then tested these new tools and models, and it became another reference point for the field.

Standardized indices developed alongside these observations. For decades the Oceanic Niño Index (ONI) served as the main measure. It is a three-month running mean of Niño-3.4 sea surface temperature anomalies, referenced to a moving 30-year climatology. NOAA's classification used a threshold of ±0.5 °C and required anomalies to persist for at least five consecutive overlapping seasons. This made it possible to classify events consistently from 1950 onward. Because the ONI retains the long-term warming of the tropical ocean, researchers later argued that it blurs the seasonal variability that matters for climate impacts. The Relative Oceanic Niño Index (RONI) addresses this by subtracting the tropical-mean sea surface temperature anomaly, which damps the warming trend. Analyses have also shown that RONI-based classifications are less sensitive to the choice of climatology period than ONI-based ones. NOAA's Climate Prediction Center adopted RONI in February 2026. In most cases it weakens El Niño events and strengthens La Niña events compared with the ONI; for example, the 2023–24 El Niño measured about 0.6 °C cooler on RONI than on ONI.

Open Questions and Debates

Although the broad outline of the ENSO cycle is well established, several central questions remain open. The first concerns what starts an event. Westerly wind bursts are widely treated as important triggers of El Niño, and a forecast model that uses them as input can time strong events such as those of 1987–88 and 1997–98 more accurately The WWBs are known to be important triggering effects of the El Niño; the forecast with the additional input of the wind bursts can capture the timing of the strong El Niño (1987-1988 and 1997-1998) more accurately.. What remains disputed is whether these bursts are a cause, a response to the subsurface state of the ocean, or largely a stochastic forcing, and how much heat stored beneath the surface contributes to initiating an event.

A second group of problems lies in the climate models themselves. Coupled models have long shown two persistent tropical Pacific errors. One is an overly strong equatorial cold tongue: the cold bias is nearly spanning the equatorial Pacific Ocean, and it is in excess of 2°C over the central and western equatorial Pacific in one coupled model. The other is the double intertropical convergence zone, in which models place rainfall maxima on both sides of the equator while observations show a single maximum north of it. These biases also deteriorate El Niño–Southern Oscillation variability. Some studies go further, arguing that the cold tongue bias can spuriously shift the simulated ENSO mean state under warming, a finding that highlights the need for further progress to be made in reducing the double ITCZ problem. Because some machine-learning forecasts are pretrained on model output before being fine-tuned on reanalysis data, these errors can carry over into statistical forecasts as well.

The asymmetry between El Niño and La Niña is a third unresolved issue. Strong El Niño events reach larger amplitudes than strong La Niña events, while La Niña episodes tend to last longer. Extreme El Niño events have outsized impacts and strongly contribute to the warm/cold phase asymmetries of ENSO. Yet there is currently no consensus on the respective importance of oceanic and atmospheric nonlinearities for those asymmetries. One modelling study attributes about 40% of the peak amplitude of extreme El Niño events to a nonlinear response of deep atmospheric convection to sea surface temperature. Other work emphasizes the thermocline: because the surface bounds upward movement of the thermocline, larger subsurface anomalies can develop during El Niño than during La Niña, which in simple ocean models yields stronger warm events. Some analyses, by contrast, suggest that oceanic nonlinearities act as a weak damping during extreme El Niño events. Why extreme El Niño events are rarer but larger than their La Niña counterparts, and how these processes combine, is therefore still unsettled.

The fourth problem concerns the limits of predictability. Machine-learning methods have extended forecast range in some studies: one convolutional network reached skill above 0.5 for late-spring targets only up to about ten months ahead, while winter targets showed skill above 0.8 at short leads. Other frameworks, such as reservoir computing, have reported skill out to about 21 months, and a hybrid of deep-learning and dynamical forecasts has improved skill across many lead times. These figures depend on verification metrics, target seasons and hindcast periods, and many long-lead results rest on only a few decades of observations. A further debate concerns the sources of skill: one study found that near-surface zonal wind alone offered predictive skill comparable to sea surface temperature at lead times of 11 to 21 months, although it added nothing when combined with sea surface temperature. Whether such long-lead signals are robust, and how far the spring predictability barrier can be overcome, remain open.

Finally, the reasons why events differ so widely in location, timing and strength remain only partly explained. Whether these differences follow predictable patterns, or largely reflect irregular and nonlinear dynamics, is a question that links the mechanisms, model biases and forecasting debates discussed above.

References

1. Understanding ENSO Indices in a Warming Climate ONI vs. RONI indiceswmo.int
2. El Niño Index Dashboard: NOAA Physical Sciences Laboratorypsl.noaa.gov
3. Relative Oceanic Nino Index and Nino3.4 Data - Brian McNoldybmcnoldy.earth.miami.edu
4. A Relative Sea Surface Temperature Index for Classifying ENSO Events in a Changing Climate in: Journal of Climate Volume 37 Issue 4 (2024)journals.ametsoc.org
5. Climate variability: Relative Oceanic Niño Indexclimate.us
6. The Relative Oceanic Niño Index (RONI) Becomes NOAA’s Official ENSO Classification Index for Tracking El Niño and La Niñagujaratweather.com
7. ONI vs RONI: how NOAA's two ENSO indices differ · ENSO Dashboardoni.bureau7.com
8. RONI vs ONI: Why NOAA Changed How El Niño Is Measurediamelnino.com
9. El Niño-Southern Oscillation (ENSO)cgd.ucar.edu
10. Monthly Oceanic Nino Index (ONI)data.humdata.org
11. El Niñobritannica.com
12. El Niño - Studentskids.britannica.com
13. The origin of the El Niño and La Niña weather pattern nameswgrz.com
14. An El Niño Fish Tale - UCAR Center for Science Educationscied.ucar.edu
15. El Niñormets.org
16. Southern Oscillation - Glossary of Meteorologyglossary.ametsoc.org
17. Southern Oscillationbritannica.com
18. Walker, Gilbert Thomasencyclopedia.com
19. Climate - Southern Oscillation, El Niño, La Niñabritannica.com
20. History of the SOI - Physical Sciences Laboratory - NOAApsl.noaa.gov
21. Southern Oscillation - an overviewsciencedirect.com
22. The Walker Circulation: ENSO's atmospheric buddyclimate.gov
23. El Niño - NASA Sciencescience.nasa.gov
24. Nonlinear Dimensionality Reduction Methods in Climate Data Analysisarxiv.org
25. Local and Global Effects of El Niño and La Niñacourses.ems.psu.edu
26. The Walker Circulation: The Atmospheric Engine Behind ENSOelninoguide.com
27. How Trade Winds Drive El Niño and La Niñathe-weather.com
28. On the ENSO Mechanisms By Chunzai Wang Physical Oceanography Divisionaoml.noaa.gov
29. review of ENSO theoriesacademic.oup.com
30. The El Niño Southern Oscillation (ENSO) Recharge Oscillator Conceptual Model: Achievements and Future Prospects - Vialard - 2025 - Reviews of Geophysics - Wiley Online Libraryagupubs.onlinelibrary.wiley.com
31. (PDF) The El Niño Southern Oscillation (ENSO) Recharge Oscillator Conceptual Model: Achievements and Future Prospectsresearchgate.net
32. The El Niño Southern Oscillation (ENSO) Recharge ... - Archimerarchimer.ifremer.fr
33. Realistic ENSO Dynamics Requires a Damped Nonlinear Recharge Oscillatorarxiv.org
34. What's a 'super El Niño'? And other El Niño questions, answered » Yale Climate Connectionsyaleclimateconnections.org
35. December 2023 El Niño update: adventure!climate.gov
36. A Super El Niño Is Very Strong, But Even Weaker El Niños Affect the Weatherscied.ucar.edu
37. Could Upcoming Super El Niño Be Strongest Ever?opensnow.com
38. The Impacts of the Eastern Pacific and Central Pacific El Niño on North American Winter Climateescholarship.org
39. Contrasting the eastern Pacific El Niño and the central Pacific El Niño: process-based feedback attributionlink.springer.com
40. Emergence of changing Central-Pacific and Eastern-Pacific El Niño-Southern Oscillation in a warming climatenature.com
41. Eastern Pacific El Niño - an overviewsciencedirect.com
42. One forecaster’s view on extreme El Niño in the eastern Pacificclimate.gov
43. Strongest El Niño Events in History: The Complete Recordthe-weather.com
44. United States El Niño Impactsclimate.gov
45. El Niño and La Niña: Frequently asked questionsclimate.gov
46. La Niña Impacts and Outlook Western Region December 2020drought.gov
47. How Does El Niño Impact the Weather?scripps.ucsd.edu
48. Orographic amplification of El Niño teleconnections on winter precipitation across the Intermountain West of North Americancbi.nlm.nih.gov
49. A new interhemispheric teleconnection increases predictability of winter precipitation in southwestern USncbi.nlm.nih.gov
50. A distinct and reproducible teleconnection pattern over North America during extreme El Niño events - PMCpmc.ncbi.nlm.nih.gov
51. Possible shift in the ENSO-Indian monsoon rainfall relationship under future global warmingncbi.nlm.nih.gov
52. Role of the North Atlantic in Indian Monsoon Droughtsarxiv.org
53. Redefining climate index with sea surface temperature of key area can better express atmospheric physical activityarxiv.org
54. NWS Jackson, MS: El Nino and La Ninaweather.gov
55. Impacts of El Niño and La Niña on the hurricane seasonclimate.gov
56. <center>El Niño-Southern Oscillation and the seasonal predictability of tropical cyclones</center>aoml.noaa.gov
57. Regional Effects Of ENSO On U.S. Hurricane Landfalls - Florida Climate Centerclimatecenter.fsu.edu
58. Influence of Mean Flow on the ENSO–Vertical Wind Shear Relationship over the Northern Tropical Atlantic in: Journal of Climate Volume 25 Issue 3 (2012)journals.ametsoc.org
59. Impacts of El Niño and La Niña (ENSO) on Hurricane Seasontrackthetropics.com
60. The role of ENSO in shaping global rainfall, storms and temperature patterns - The Watcherswatchers.news
61. The Impact of El Niño and La Niña in Each Region of Brazilczapp.com
62. El Niño 2026 raises concerns in Brazilsciencearena.org
63. A change in coral extension rates and stable isotopes after El Niño-induced coral bleaching and regional stress eventsncbi.nlm.nih.gov
64. A ‘super’ El Niño has the power to devastate fishingpreventionweb.net
65. 1972 Peruvian anchoveta crisisen.wikipedia.org
66. A ‘super’ El Niño has the power to devastate fishingtheconversation.com
67. El Nino and Indonesia's 1997-98 Forest Firesscholarspace.manoa.hawaii.edu
68. The amount of carbon released from peat and forest fires in Indonesia during 1997nature.com
69. Assessing costs of Indonesian fires and the benefits of restoring peatland - PMCpmc.ncbi.nlm.nih.gov
70. How climate change affects mosquito-borne infectious disease: The different trajectories of malaria and denguescience.org
71. Global Disease Outbreaks Associated with the 2015–2016 El Niño Eventnature.com
72. Effects of the El Niño-Southern Oscillation and seasonal weather conditions on Aedes aegypti infestation in the State of São Paulo (Brazil): A Bayesian spatio-temporal studyjournals.plos.org
73. [PICTURE HERE] Public Health Situation Analysis El Niño Global Climate Eventcdn.who.int
74. El Niño Southern Oscillation (ENSO) effects on fisheries and ...openknowledge.fao.org
75. El Niño: The Not So Little Global Eventoceanfutures.org
76. El Niño-driven phase shift to algal dominance on Isla del Caño’s coral reefs: implications for urgent restorationncbi.nlm.nih.gov
77. El Niño fuels wildfires in Indonesia, sickening some 100,000 peoplescientificamerican.com
78. Indonesia’s Parched Peatlands Burn Under El Niñoiri.columbia.edu
79. El Niño and its health impacts - Worldreliefweb.int
80. Effects of temperature, rainfall, and El Niño Southern Oscillations on dengue-like-illness incidence in Solomon Islandsncbi.nlm.nih.gov
81. enso-li2011.txtncei.noaa.gov
82. A Review of Paleo El Niño-Southern Oscillationmdpi.com
83. Dampened El Niño in the Early and Mid‐Holocene Due To Insolation‐Forced Warming/Deepening of the Thermoclinepar.nsf.gov
84. Coral geochemical records: An overview of their use as climate proxies and of available databasesclimatedataguide.ucar.edu
85. skip to main contentpar.nsf.gov
86. Holocene changes in eastern tropical Pacific climate inferred from a Galápagos lake sediment record - ScienceDirectsciencedirect.com
87. Holocene ENSO variability in the South China Sea recorded by high-resolution oxygen isotope records from the shells of Tridacna spp.nature.com
88. Past ENSO Variability: Reconstructions, Models, and Implicationsresearchgate.net
89. Koutavas Joanides 2012 282704whoi.edu
90. Reduced El Niño-Southern Oscillation during the Last Glacial Maximum - PubMedpubmed.ncbi.nlm.nih.gov
91. ENSO variability and teleconnections during glacial timesresearchgate.net
92. Robust expression of ENSO throughout the Last Interglacial - ADSui.adsabs.harvard.edu
93. Multiproxy El Nio-Southern Oscillation (ENSO) Reconstructionsncei.noaa.gov
94. High-resolution marine data and transient simulations support orbital forcing of ENSO amplitude since the mid-Holocene - ScienceDirectsciencedirect.com
95. The ENSO–Precipitation Teleconnection and Its Modulation by the Interdecadaljournals.ametsoc.org
96. Eastern Pacific Corals Track Robust ENSO Variability and Stronger La Niña Events 4,000 Years Before Present - Tripp - 2026 - Geophysical Research Letters - Wiley Online Libraryagupubs.onlinelibrary.wiley.com
97. CP - Comparison of past and future simulations of ENSO in CMIP5/PMIP3 and CMIP6/PMIP4 modelscp.copernicus.org
98. A mystery in the Pacific is complicating climate projections » Yale Climate Connectionsyaleclimateconnections.org
99. How the pattern of trends across the tropical Pacific Ocean is critical for understanding the future climateclimate.gov
100. 4 El Niño and Southern Oscillation (ENSO): A Reviewstaff.cgd.ucar.edu
101. Central-Pacific El Niño-Southern Oscillation less predictable under greenhouse warmingncbi.nlm.nih.gov
102. Uncertainty of ENSO-amplitude projections in CMIP5 and CMIP6 modelslink.springer.com
103. Intensity-dependent changes in ENSO frequency and lifecycle characteristics under future climate changelink.springer.com
104. ESD - The future of the El Niño–Southern Oscillation: using large ensembles to illuminate time-varying responses and inter-model differencesesd.copernicus.org
105. On the role of ozone feedback in the ENSO amplitude response under global warmingncbi.nlm.nih.gov
106. Contributions of Climate Change and ENSO Variability to Future Precipitation Extremes Over California - Huang - 2023 - Geophysical Research Letters - Wiley Online Libraryagupubs.onlinelibrary.wiley.com
107. Climate change alters the Indian Ocean Dipole and weakens its North Atlantic teleconnectionnature.com
108. Indian Ocean Dipole leads to Atlantic Niñopar.nsf.gov
109. Indian Ocean Dipole leads to Atlantic Niño - PubMedpubmed.ncbi.nlm.nih.gov
110. Indian Ocean Dipole leads to Atlantic Niñoncbi.nlm.nih.gov
111. Historical and Future Asymmetry of ENSO Teleconnections with Extremes in: Journal of Climate Volume 37 Issue 22 (2024)journals.ametsoc.org
112. Indian Ocean Dipole leads to Atlantic Niñonature.com
113. Refreshed Data System for Tropical Atmosphere Ocean (TAO) Buoy Arrayieeexplore.ieee.org
114. THE GLOBAL TROPICAL MOORED BUOY ARRAYaoml.noaa.gov
115. Bolstered by Buoys: Predicting El Niño - Eoseos.org
116. Impact of the TAO/TRITON Array on Reanalyses and Predictions of the 2015 El Niño - Hackert - 2023 - Journal of Geophysical Research: Oceans - Wiley Online Libraryagupubs.onlinelibrary.wiley.com
117. Tropical Ocean Observations for Weather and Climate: A Decadal Overview of the Global Tropical Moored Buoy Arraytos.org
118. Real-time ENSO forecast skill evaluated over the last two decades, with focus on the onset of ENSO eventsnature.com
119. Aus, US agencies point to El Niño strengthening - Grain Centralgraincentral.com
120. Prediction of ENSO Beyond Spring Predictability Barrier Using Deep Convolutional LSTM Networksieeexplore.ieee.org
121. Tropical basin interactions reduce spring predictability barrier of ENSO in a deep learning modelscience.org
122. The Indian Ocean Weakens the ENSO Spring Predictability Barrier: Role of the Indian Ocean Basin and Dipole Modes in: Journal of Climate Volume 36 Issue 24 (2023)journals.ametsoc.org
123. Sea Surface Salinity Strongly Weakens ENSO Spring Predictability Barrier - Pang - 2023 - Geophysical Research Letters - Wiley Online Libraryagupubs.onlinelibrary.wiley.com
124. ENSO: Recent Evolution, Current Status and Predictions Update prepared by:cpc.ncep.noaa.gov
125. Climate Prediction Center: ENSO Diagnostic Discussion - NOAAcpc.ncep.noaa.gov
126. ENSO Conditions and Coral Bleachingcoralreefwatch.noaa.gov
127. El Niño/Southern Oscillation (ENSO) Diagnostic Discussioncpc.ncep.noaa.gov
128. Abstract: Milestones on the Road toTOGA (95th American Meteorological Society Annual Meeting)ams.confex.com
129. El Niño and Southern Oscillation (ENSO): A Review Chunzai Wang 1aoml.noaa.gov
130. An Overview of El Niño-Southern Oscillation Understanding By Chunzai Wang 1aoml.noaa.gov
131. Experimental forecasts of El Niñonature.com
132. On the Prediction of the El Nino of 1986-1987ocp.ldeo.columbia.edu
133. Predicting El Niño Then and Nowclimate.gov
134. El Niño modeling improvements introduced to Zebiak-Cane model - Climate Program Officecpo.noaa.gov
135. Revealing the most disturbing tendency error of Zebiak–Cane model associated with El Niño predictions by nonlinear forcing singular vector approachlink.springer.com
136. The El Niño factorcarlzimmer.com
137. Forecasts of Tropical Pacific SST Using a Simple Coupled ...cpc.ncep.noaa.gov
138. ENSO Prediction - L'Heureux - AGU Journals - Wileyagupubs.onlinelibrary.wiley.com
139. New NOAA El Niño-Southern Oscillation Index Supports Drought Early Warningdrought.gov
140. (PDF) A Relative Sea Surface Temperature Index for Classifying ENSO Events in a Changing Climateresearchgate.net
141. A Bayesian Machine Learning Algorithm for Predicting ENSO Using Short Observational Time Seriesarxiv.org
142. The Equatorial Pacific Cold Tongue Bias in a Coupled Climate Model in: Journal of Climate Volume 21 Issue 22 (2008)journals.ametsoc.org
143. Cold Tongue Bias in CMIP6 Spuriously Causes Mean State to Move Towards El Nino in Response to Warming over the Historical Period - ADSui.adsabs.harvard.edu
144. Dominant contribution of atmospheric nonlinearities to ENSO asymmetry and extreme El Niño eventsncbi.nlm.nih.gov
145. Dominant contribution of atmospheric nonlinearities to ENSO asymmetry and extreme El Niño eventsnature.com
146. Nonlinearity and asymmetry of ENSOresearchgate.net
147. El Nino-Southern Oscillation - an overviewsciencedirect.com
148. A Machine Learning‐Based Approach to Quantify ENSO Sources of Predictabilityopensky.ucar.edu
149. A Machine Learning‐Based Approach to Quantify ENSO Sources of Predictability - Colfescu - 2024 - Geophysical Research Letters - Wiley Online Libraryagupubs.onlinelibrary.wiley.com
150. Machine Learning for the Physics of Climatearxiv.org
151. (PDF) DLENSO: A Deep Learning ENSO Forecasting Modelresearchgate.net
152. Combined dynamical-deep learning ENSO forecastsnature.com
153. pacific cold tongue: Topics by Science.govscience.gov