Evening Weather Briefing, Tuesday 21 July 2026

Evening Weather Briefing, Tuesday 21 July 2026 ECMWF 12Z | Generated 19:30 UTC | For: Tuesday 21 July 2026 --- 1. HEADLINE & KEY CHANGE The signal that matters most across markets this week is not in Europe: El Niño has crossed into objectively strong territory, with the Niño-3.4 weekly SST anomaly at +2.3°C as of July 15, and NOAA CPC now assigning 81% probability to a very strong event by October–December, ranking it among the largest in the 75-year record. In today's 12Z run, the most consequential European revision is a near-term wind uptick over the southern North Sea tomorrow that the ensemble then closes off abruptly by Thursday, narrowing the window during which offshore generation can ease load on the continent before a ridge of high pressure reasserts. --- 2. SYNOPTIC SETUP & FORECAST EVOLUTION The Atlantic is in a classic mid-summer semi-stall, and today's 12Z run did not fundamentally alter that picture, but it sharpened the near-term evolution in two ways. Oslo's Wednesday wind maximum was revised up by 3.8 km/h (now 18.5 km/h), confirming the trough axis over the eastern North Sea is slightly more amplified in the 12Z solution than yesterday's run. And Amsterdam's Thursday wind maximum was pulled back by 3.1 km/h (now 18.2 km/h), signalling the window closes faster than previously indicated. The practical effect: a brief elevation of North Sea offshore capacity factors on Wednesday gives way to calm conditions from Friday onward more abruptly than yesterday's model suggested. That mobile trough is the feature to track through midweek. It's working slowly eastward toward the British Isles and the Low Countries, and the ensemble shows reasonable agreement on its existence, the disagreement is about what follows it. A broad ridge is building behind it over the central Atlantic, and it is the character of that ridge, not its presence, on which the ensemble splits most markedly from day 5 onward. The majority solution, roughly 55% of ECMWF members, keeps the ridge shallow and anchored near the Azores, delivering light southwest flow across the UK and Benelux, modest temperatures running 1–2°C above seasonal norms, and wind generation well below commercially useful capacity factors. The minority scenario, roughly 30% of members, amplifies the ridge northward into a configuration resembling an Atlantic block. In that track, NW Europe draws warm continental air from the east rather than mild maritime Atlantic air, temperatures over the Low Countries and Germany would climb 3–4°C above the modal solution by day 7–8, wind production drops further, and any cooling demand that exists in these markets receives an amplified signal. The Frankfurt week-2 EC46 range of 17.6–25.9°C is the clearest numeric expression of this uncertainty; a spread of 8.3 degrees in a single week is not routine, and it reflects genuine disagreement about ridge amplitude. The Arctic Oscillation forecast trajectory is the most important medium-term signal from this run. Currently at +1.23 and forecast to ease near neutral midweek (the GEFS ensemble mean reaches –0.08 by day 3–4), the AO then rises strongly, reaching +1.89 by day 7. A strongly positive AO corresponds to a compact, well-organized polar vortex and a zonal Atlantic flow regime, the type of pattern that limits European blocking risk and tends to keep the NAO in neutral-to-positive territory. The NAO itself is forecast at near-neutral to slightly negative through days 5–6 (–0.12) before recovering. Together, the trajectories argue against a sustained Greenland high establishing itself in the first half of August. The polar vortex looks too coherent for that. That said, the early August window in weeks 2–3 remains the period of greatest uncertainty; week-3 is where the EC46 ensemble spread truly opens. --- 3. REGIONAL ANALYSIS a) NW Europe & Nordic The immediate picture through Thursday is shaped by the trough passage. London's Open-Meteo 7-day wind average of 15.4 km/h and Amsterdam's 13.7 km/h capture a regime that is not calm but not powerful, capacity factors for offshore wind in these conditions would sit well below the 40% threshold that meaningfully displaces gas-fired backup generation. Wednesday's brief uptick as the trough clears the North Sea coast is real but short-lived. From Friday through the weekend, the building ridge reasserts. Amsterdam's day-by-day shows 6–7 km/h winds Friday and Saturday with temperatures of 18–20°C, above seasonal norms but, in late July at these latitudes, well below any cooling demand threshold. The accumulated HDD for Amsterdam over the 14-day window is 0.3, effectively zero. This is a demand-neutral regime for European gas from a temperature standpoint. The demand signal, if any, comes from the supply side: when the wind stops, gas-fired capacity runs harder. That dynamic is priced continuously by market participants, but for this briefing the point is directional, the pattern through week 1 maintains gas demand from the power sector at moderate levels without any acute demand spike. The Nordic picture carries more significance for the extended period. Oslo's EC46 week-1 sits at 15.8°C, week-2 at 16.4°C, modest for July, with no polar incursion in weeks 1–2. Neither the Open-Meteo data nor the EC46 ensemble shows meaningful precipitation over Scandinavia through week 2. Dry, mild conditions with slowly rising temperatures are not what the hydro sector needs after a summer of modest replenishment. No specific Norwegian inflow or reservoir data was available this cycle, so the assessment is necessarily qualitative: the precipitation signal is not favorable for accelerating hydro recovery, and the lack of notable cold anomaly through week 2 means reservoir drawdown for power demand is at best flat. b) Southern & Eastern Europe Rome's EC46 week-1 at 27.5°C, week-2 at 26.3°C describes a typical late-July Mediterranean pattern: warm, but modestly cooler than the week-2 peak. The wind revisions for Rome are informative, Wednesday's wind maximum was cut by 3.6 km/h (now 18.4 km/h), with a +4.8 km/h reversal arriving by Friday (now 22.4 km/h). This is consistent with the trough interacting with the Alpine terrain, suppressing Adriatic wind as the system approaches and then accelerating flow on the trailing edge. The net impact on Italian power generation is limited; solar irradiance is the more consequential resource through July, and the cloud cover embedded in the trough passage represents a brief interruption of what is otherwise a high-solar month. Madrid's week-1 at 28.5°C and week-2 at 28.6°C reflect sustained Iberian heat, but not the extreme events that define acute power demand stress in Spain. The –3.4 km/h revision for Madrid on July 25 (now 16.7 km/h) reflects the same trough-driven suppression touching Iberian wind corridors briefly. The broader Iberian plateau heat outlook remains elevated through both weeks; the ECMWF seasonal bulletin's confidence in above-normal European temperatures extends explicitly through autumn, and August historically delivers the strongest El Niño-mediated warming anomaly for Mediterranean Europe. c) East Asia Japan is the strongest short-horizon demand signal in this briefing. Tokyo's 14-day CDD accumulation of 174.5 at an average temperature of 29.6°C represents sustained, high-amplitude cooling demand. Nagoya and Osaka are running parallel, both averaging 31.5°C. What makes the EC46 signal particularly notable is its stability: Tokyo week-1 sits at 29.8°C, week-2 also at 29.8°C, with a range of 27.5–31.9°C. The ensemble is not splitting on whether Japan stays hot. It is splitting only on how hot. There is no break in sight, no Pacific trough, no typhoon-induced cold anomaly, through week 2. The week-3 and week-4 EC46 controls soften somewhat, to 28.7°C and 27.9°C for Tokyo, but the spread remains wide, and the El Niño background pattern favors the warm side. The Niño-3.4 weekly anomaly at +2.3°C is roughly 1.1°C above the ONI seasonal figure reported in the climate indices as of May, confirming active intensification. El Niño's canonical influence on Japan in late summer combines warmer western Pacific SSTs with suppressed typhoon formation frequency near the main islands, effects that would extend the cooling demand season without the brief demand interruptions a landfalling typhoon produces. No tropical cyclone advisories were included in this cycle's data feed; traders with Japanese power and LNG exposure should monitor NHC and JTWC independently. MJO is in Phase 6 at amplitude 1.1, active but at the weaker edge of the significant-impact threshold. Phase 6 centers enhanced convection over the western Indian Ocean, which in the canonical signal tends to suppress convection across the Maritime Continent and sustain warm, dry conditions over the western Pacific. The amplitude is too modest for high confidence on this propagation pathway, but it is not inconsistent with the continued hot, clear pattern over Japan. If the MJO propagates into Phase 7–8 over the next 10–14 days, the canonical signal would amplify the western Pacific ridge and potentially extend Japan's cooling season further. Seoul's EC46 week-1 at 25.4°C rising to a tight 27.1°C (range 26.2–28.1°C) in week-2 is notable for its narrow spread, the ensemble is unusually converged on Korean warming, which is an El Niño fingerprint. Shanghai reaches 30.3°C in week-2 from 29.6°C in week-1. The pattern across Northeast Asia is, in short, one of sustained summer heat with no credible break through the first half of August. d) Americas The US pattern is the most locked-in feature globally this week, and it has not moved in today's run. CPC's extended-range discussion describes a large, persistent ridge sitting over the Great Plains and Great Basin with a 594-dam closed contour over the Four Corners region that multiple ensemble systems maintain through the full 8–14-day range. CPC's explicit comment that "this forecast pattern has been consistent for several model cycles, further increasing confidence in model guidance" is a relatively rare statement of operational certainty in medium-range extended forecasting. The consequences across the central and western US are substantial. Dallas averages 36.0°C over the 14-day window with 264.6 accumulated CDD; Houston averages 32.6°C with 214.4 CDD; Denver, 30.7°C with 185.3 CDD; Phoenix, 39.6°C with 319.4 CDD. These are not ambient summer temperatures, they represent days of sustained heat-stress in a region where summer space cooling is almost entirely electric, and the gas-to-power connection for grid backup runs continuously. The Great Plains ridge is not breaking in the 6–10 or 8–14 day CPC outlook windows. Odds of above-normal temperatures exceed 70% across large portions of the Desert Southwest, southern and central Rockies, and Texas. The Northeast sits in a different regime. A trough near Hudson Bay keeps near-normal to below-normal temperatures across the upper Midwest and Northeast, producing the rain events embedded in Boston's Tuesday–Wednesday forecast (10.4mm and 17.0mm). New York's EC46 progression from 23.2°C in week-1 to 25.0°C in week-2 captures the gradual northward migration of the ridge, easing the trough's grip on the Eastern Seaboard without eliminating it. CPC's 8–14 day discussion notes the wavetrain is "basically stationary", the Northeast remains cooler than the US interior throughout the period. Brazil's Sao Paulo registers 18.0°C week-1, 18.2°C week-2, Southern Hemisphere winter temperatures with minimal direct energy demand implications. The ECMWF seasonal bulletin notes confident below-normal precipitation signals for northern South America in August–October under El Niño. This is relevant for Brazilian hydro reservoir levels heading into the spring fill season, though that consequence falls beyond this week's direct horizon. e) Australia Australia is in the Southern Hemisphere winter, and NEM demand is driven primarily by overnight and morning heating loads. Melbourne's EC46 week-1 at 9.5°C (range 8.6–10.4°C) and Adelaide at 10.4°C (range 9.6–11.2°C) describe cold but stable mid-winter conditions. The accumulated HDD figures, Melbourne 110.7, Adelaide 108.3, Sydney 86.0 over 14 days, represent persistent but not acute gas heating demand. The most material Australian revision in today's 12Z run is the sustained upward shift in late-July and early-August wind maxima across Adelaide and Melbourne. Multiple dates between August 1 and August 15 show increases of 3.2–4.8 km/h, with Adelaide August 2 now forecast at 22.9 km/h and August 15 at 24.6 km/h; Melbourne follows a parallel track. This revision pattern is coherent with the El Niño-driven strengthening of the Southern Annular Mode, a more amplified polar westerlies belt pushes higher-latitude storm tracks equatorward in late winter, increasing wind generation potential across South Australia and Victoria. Whether this translates to materially higher NEM wind generation depends on the precise synoptic configurations, but the directional signal from consecutive model revisions upward is worth tracking. The EC46 extended range for Adelaide and Melbourne shows a gradual temperature recovery into spring: Melbourne from 9.5°C in week-1 to 11.3°C in week-6, Adelaide from 10.4°C to 12.3°C. This is a normal seasonal trajectory with no exceptional cold anomaly on the horizon and no early spring warmth. NEM gas-for-power demand in the near term is shaped primarily by wind intermittency and overnight temperature, not any acute cold event. --- 4. EXTENDED RANGE & REGIME (WEEKS 3–6) The dominant regime signal over the extended range is set by El Niño, and it is still intensifying. The Niño-3.4 weekly anomaly of +2.3°C in July is already beyond the level at which most prior seasonal peaks have been observed; CPC projects continued intensification through OND 2026 with 81% probability of a very strong event. Strong El Niño events in their intensifying phase organize global teleconnections more reliably than moderate events, and the EC46 ensemble from weeks 3–6 broadly reflects this. For NW Europe, the ECMWF seasonal bulletin's statement that temperatures are "predicted more likely than not to be well-above average (exceeding the 80th percentile of the historical distribution), virtually across the continent, throughout the forecast period" is the most confident European seasonal signal available this cycle. The EC46 means support it, Amsterdam holds near 19°C through week-4 before easing toward 17.6°C in week-6, Paris runs 22–23°C through weeks 2–4, Frankfurt holds above 20°C through week-5. These are above-normal readings on a sustained basis, not heat spikes. The energy implication is not acute warming demand, it is an extended period of mild conditions that keeps European gas demand suppressed below normal injection-season expectations, supporting storage refill through the summer shoulder. The NAO is forecast to remain near neutral through weeks 1–2, and the AO's strong positive trajectory through day 7 argues against a blocking pattern emerging in that window. The combination of a positive AO and strong El Niño does not typically suppress summer blocking with the same reliability as it operates on winter vortex dynamics, but neither does it favor retrograding ridges. The weeks 3–4 window carries the most blocking uncertainty; the EC46 spread is at its widest there for all European sites. The QBO at –1.5 m/s (easterly) in combination with a strengthening El Niño is the canonical setup for winter polar vortex disruption, the Holton–Tan mechanism. This is a background signal for the October–March period, not a driver of August weather. It is noted here because it frames the risk environment for the approaching heating season: the same El Niño that is suppressing European heating demand through summer is building the atmospheric preconditions that have historically raised the odds of vortex disruption and negative NAO blocking episodes in the following winter. That is a season-ahead signal, not a week-ahead one. The PDO at –9.90 (strongly negative) sits in apparent tension with the typical El Niño pattern, which usually drives positive PDO through eastern Pacific SST warming. A concurrent strong El Niño and strongly negative PDO has historically been associated with amplified US West Coast ridging, exactly what the CPC outlooks describe. This PDO reading is dated December 2025; an updated reading would be needed to confirm the current state. For Japan, the EC46 weeks 3–6 show a gradual cooling from the 28–29°C week-1–2 plateau, Tokyo to 28.7°C in week-3, 27.9°C in week-4, 27.3°C in week-5, but the spread widens and the central tendency remains well above seasonal norms. The western Pacific typhoon season represents the primary uncertainty for this trajectory; El Niño suppresses typhoon genesis frequency near the Japanese coast, but individual events can still disrupt the heat pattern and spike short-term power demand in complex ways during and after landfall. --- 5. DATA FRESHNESS & CONFIDENCE - Open-Meteo 16-day: Current as of today, 6-hourly cadence. High confidence for days 1–7; normal degradation through day 14. - ECMWF IFS 12Z: Today's full initialization. Forecast changes listed are 12Z-vs-12Z deltas from yesterday. - EC46 46-day ensemble: Current as of today. Week-1–2 temperature means carry meaningful skill; week-5–6 should be read directionally only. - NOAA CPC 6-10 and 8-14 day outlooks: Dated July 19, one day lag, acceptable. - ENSO, ONI vs weekly anomaly: The ONI of +1.0 is from May 1 and is materially outdated. The operative reading is the Niño-3.4 weekly anomaly of +2.3°C as of July 15, which reflects active intensification. Do not weight the ONI figure. - NAO, AO, PNA indices: Base values dated June 30; GEFS ensemble forecasts from current run are operative for trajectory analysis. - MJO: Dated July 17, three-day lag. Amplitude 1.1 in Phase 6; propagation into Phase 7–8 over the next 10 days is possible but not confirmed. - QBO: Dated May 1. The sign (easterly) is unlikely to have changed; the magnitude should be treated as approximate. - EU gas storage (AGSI+): Current. EU total at 53.7% is a clean anchor; Belgium at 29.2% and the Netherlands at 32.4% stand out as the weakest national fills heading into the second half of July injection season. - Tropical cyclone activity: No NHC or JTWC advisory data was included in this cycle. Traders with Pacific or Atlantic exposure should consult those services independently. - Nordic hydro: No inflow or reservoir data available. The dry signal is identified from precipitation forecasts only; it is directional, not quantified. Overall confidence: High for US ridge persistence and Japan heat demand through week 2. Moderate for NW European week-2 temperature amplitude and the specific timing of the ridge-building phase. Low for week-3–6 specifics in all regions, except the directional El Niño signal, where seasonal forecast agreement across agencies is unusually strong.