A Historic El Niño Tightening Its Grip: Position for Supply Stress and Demand Divergence This Winter
---
ENSO & Teleconnections
The dominant story for the next three months has a name: a potential once-in-a-generation El Niño. The weekly Niño-3.4 index reached +3.4°C as of August 26, and NOAA's CPC now puts a 69% probability on OND 2026 exceeding +2.5°C on the RONI scale, a threshold that would make this the strongest event on record since modern monitoring began in 1950. Subsurface thermal anomalies have reached +10°C at depth across the eastern equatorial Pacific, and the thermocline sits well above normal. The ocean is loaded. The atmosphere is responding: westerly wind anomalies dominate the west-central equatorial Pacific, convection is suppressed over Indonesia, and the Southern Oscillation Index sat near –2 standard deviations through July. NOAA places the probability of a very strong event through NH fall and winter at above 90%.
The MJO is active at Phase 5 with amplitude 1.1, placing enhanced convection over the Maritime Continent and western Pacific. Over the next two to four weeks, MJO propagation into Phases 6–7 would typically enhance the El Niño atmospheric coupling already in place, reinforcing suppressed rainfall over Indonesia and boosting precipitation over the eastern Pacific corridor. The QBO is in its easterly phase at 50hPa (–0.7 m/s), which in combination with the negative PDO (–9.90 as of December 2025) tilts the NH winter stratospheric pattern toward weaker polar vortex conditions, a setup that statistically increases the probability of cold air outbreaks over the eastern US and northern Europe in December–February.
The NAO ended July at –1.28 but the GEFS 16-day ensemble has it oscillating near neutral through mid-September. The AO, neutral at –0.35 in July, is trending clearly positive in the GEFS forecast, reaching +1.89 by day seven, suggesting a reinforced polar vortex in the short term. That near-term positive AO may suppress European heating demand through September, but traders should not anchor to it; the QBO-easterly/El Niño combination has historically broken the polar vortex with greater frequency in mid-winter, and the stratospheric setup will warrant close monitoring from November onward.
---
6-Week Temperature Trajectory
The EC46 ensemble through mid-October tells a consistent story for Western Europe: a gradual but unambiguous slide into seasonal norms and below. London goes from 16.1°C in week one to 11.7°C by week six, with the ensemble range compressing to 8.8–14.6°C, relatively narrow for a six-week lead, indicating above-average confidence that the pattern will deliver below-normal temperatures in the back half of October. Frankfurt mirrors this: 16.6°C now, 11.1°C by week six, with the cold tail extending to 7.1°C. Amsterdam's EC46 cold probability exceeds 34% at one standard deviation below normal, with warm probability barely at 3.5%. Madrid drops sharply from 27°C to 14.6°C, a 12-degree fall in six weeks, well outpacing climatological autumn cooling, though ensemble spread (11.5–18.1°C by week six) keeps the magnitude uncertain.
By contrast, ensemble spread widens dramatically for southern Australia. Adelaide moves from 12.8°C to 16.1°C as Southern Hemisphere spring advances, but the week-five range spans 11.5–21.3°C. Melbourne carries a similar story: 12.2°C to 15.0°C with 10.8–20.4°C spread. El Niño spring warmth bias for Australia is the background signal, but confidence degrades quickly beyond week three.
Japan cools steadily and with relatively narrow spread: Nagoya from 24.4°C to 17.9°C (range 15.3–20.3°C by week six), Osaka from 23.5°C to 16.5°C. Houston begins a slow seasonal retreat from 27.7°C to 23.5°C, residual late-season cooling demand fading but not yet gone. The wide New York spread by week six (10.7–19.6°C) flags genuine uncertainty about whether October brings a cold snap or lingering warmth, a key variable for NYMEX prompt gas.
---
Regional Seasonal Outlooks
Europe. ECMWF's August seasonal bulletin calls for temperatures above the 80th percentile climatological threshold across most of continental Europe through autumn, with only the far northeast and far north showing a marginally weaker warm signal. NOAA's multi-model synthesis reinforces this. The ECMWF also flags wetter-than-average conditions for western and southern Europe through autumn, though the agency explicitly flags higher forecast uncertainty and lower skill here than in the tropics. DWD's seasonal and Météo-France have not broken from the ECMWF consensus on temperature. The wrinkle: despite the warm seasonal mean signal, the EC46 ensemble structure for NW Europe in weeks three through six now shows cold probability materially exceeding warm probability at the city level, a tension between seasonal climatology and the ensemble's medium-range dynamical evolution. This kind of "warm mean, cold tail" structure is worth watching; it can disguise demand spikes.
East Asia. JMA's 3-month outlook aligns with El Niño composite patterns: above-normal temperatures through autumn over Honshu and Korea, consistent with the Niño-3.4 reading. The KMA seasonal supports above-normal temperature probability for October–December across South Korea. Japan's LNG demand may remain suppressed relative to prior-year through October but the onset of below-normal temperatures in November–December under a strong El Niño is not uniform: there is historical precedent for cold Decembers in Japan during strong events, and JMA's winter outlook will be the critical update to watch on September 20.
Russia. Roshydromet's El Niño analogue composites suggest a warmer-than-normal winter for western Siberia but with elevated uncertainty for the Urals and western Russia. A weakened polar vortex scenario, which the QBO-easterly phase supports, increases cold risk for European Russia in January–February. This is the key swing factor for European piped gas demand that no seasonal agency has yet resolved with confidence.
South and Southeast Asia. ECMWF's suppressed Maritime Continent precipitation signal aligns with Indonesia's deepening dry season, LNG export logistics at Bontang and Tangguh face water and operational risk through October. India's monsoon is in its final weeks; the El Niño suppression signal, while partially offset by a positive Indian Ocean Dipole predicted by ECMWF with high model agreement, points to a drier-than-normal finish to the monsoon season across central and peninsular India. CMA's seasonal model is consistent with this read.
Americas. NOAA CPC's SON 2026 outlook is straightforward: above-normal temperatures dominate the western and central CONUS with probabilities exceeding 50% across the Pacific Northwest and Northern Rockies, while above-normal precipitation covers the Southwest through the central High Plains. This is a textbook El Niño SON pattern. The active Atlantic hurricane season interacts here, La Niña would normally favor high hurricane activity but a strong El Niño suppresses Atlantic shear, reducing the Gulf production disruption tail risk that was elevated in prior years.
---
Hydro & Storage
Brazil's hydro situation warrants close attention. ECMWF and NOAA both flag below-normal precipitation for northern South America under El Niño, directly threatening reservoir inflows across the Paraná and São Francisco basins entering the wet season. The EC46 shows Sao Paulo running +2.4–2.7°C warmer than the prior ensemble run over mid-September, which amplifies near-term cooling demand and accelerates drawdown. ONS has been managing dispatch conservatively, but if the wet season onset is delayed by four to six weeks, the El Niño-typical outcome, Brazil's thermal power substitution need and LNG import demand will rise materially through Q1 2027.
Nordic reservoir levels enter autumn from a moderate deficit position relative to the 10-year median, after a drier-than-normal summer across Scandinavia. The EC46 Oslo trajectory (11.8°C to 6.0°C over six weeks, with the cold end reaching 2.7°C by week six) points to heating season arriving early in Scandinavia. EU gas storage is tracking close to the 5-year average at this writing, but the pace of injections will need to sustain into late October to buffer what increasingly looks like a below-normal temperature December–February tail risk, particularly if the polar vortex weakens mid-winter.
---
Strategic Positioning
- TTF Q1 2027 and NBP Win 2026 carry underpriced cold-tail risk. The QBO-easterly/El Niño polar vortex disruption signal does not appear fully reflected in the Q1 forward curve; the window to build length in TTF Q1 before seasonal models start flagging vortex disruption probabilities in October is narrowing.
- Japan LNG spot (November–December) may see demand overshoot consensus if JMA's September winter outlook aligns with cold analogue years for strong El Niños. Watch for the September 20 JMA release as a potential catalyst.
- Brazilian power and LNG spot imports are a buy on weakness into September–October; the El Niño wet-season delay risk is not yet priced into Q4 thermal dispatch assumptions.
- European gas Cal-27 spreads warrant attention if October EC46 updates confirm the cold ensemble tail. The warm seasonal mean masks the cold-probability structure already visible in six-week EC46 city-level data.
- Indonesian LNG supply risk is underappreciated. Dry season intensity, consistent with MJO Phase 5–7 propagation and El Niño suppression over the Maritime Continent, raises operational risk at key facilities through October; the production tail risk supports a long bias in spot Asian LNG.
- US Henry Hub Nov–Dec: the NOAA CPC warm SON signal for the western and central CONUS suppresses early-season gas demand, but watch December closely. The QBO-easterly setup increases the probability of a cold December for the eastern US, which could sharply reverse the bearish autumn bias in prompt-winter contracts.