Evening Weather Briefing, Saturday 18 July 2026

Evening Weather Briefing, Saturday 18 July 2026 ECMWF 12Z | Generated 19:30 UTC | For: Saturday 18 July --- 1. HEADLINE & KEY CHANGE A transient NAO flip toward neutral-negative, centred on days 3-5, delivers a frontal passage to Northwest Europe mid-week before a sharp Arctic Oscillation recovery reasserts zonal flow, the brief cold window is real but bounded. Since yesterday's 12Z run, the frontal system targeting the Low Countries around July 22-23 has strengthened marginally, the Northeast US Sunday maximum has been pulled back by 3.5°C, and southeast Australian wind speeds have been upgraded cohesively across the July 21-23 window. --- 2. SYNOPTIC SETUP & FORECAST EVOLUTION The Atlantic westerlies are currently doing what they are supposed to do in mid-July: pushing mild, moist air across the North Sea in a broadly zonal stream with the NAO at +0.58 and AO at +1.23. That arrangement is about to be interrupted, and the interruption is both well-signalled and time-limited. The GEFS ensemble mean traces the NAO declining from +0.15 on day one to -0.12 by days 4-5. The AO follows: a brief dip toward -0.08 before recovering to near +1.9 by day 7, one of the stronger AO recoveries in the near-term ensemble range. What this describes, in synoptic terms, is a transient Greenland blocking episode, or at minimum a westerly collapse, that allows a frontal system to deepen and stall across Northwest Europe around July 22-23. Then the polar vortex reasserts. The episode is not long enough to establish persistent blocking; most ensemble members bring the zonal regime back before the end of the month. The Amsterdam day-by-day data traces the passage cleanly: 21.9°C on Friday, a step down to 18.8°C on Saturday as the Atlantic flow weakens, relative stability through early next week, then the break, 9.9mm of rain and 23 km/h sustained winds on Wednesday, dropping to 15.8°C Thursday. Frankfurt already shows a 44% probability of temperatures more than one standard deviation below normal by day five. The ensemble is not evenly split; it leans cold for central Europe mid-week. The critical model disagreement is on duration. In the majority scenario, this is a 48-72 hour disruption: one frontal wave clears, the AO recovery is as advertised, and the zonal flow is restored before the end of July. In the minority scenario, approximately 25-30% of members, the Greenland blocking establishes briefly and the trough deepens over western Europe, extending the cold anomaly into the last week of July. The EC46 week-two spread for Frankfurt (17.9-25.0°C) marks the precise range of that disagreement. With a 7°C spread on week-two temperatures, the market needs to price that uncertainty rather than anchor on the ensemble centre. Beyond week two, the extended ensemble converges on a late-summer pattern of persistently above-normal warmth for Europe. The ECMWF seasonal bulletin, issued July 10, states that temperatures across virtually the entire European continent are predicted to exceed the 80th percentile of the historical distribution throughout the forecast period. The strengthening El Niño provides the backbone for that bias. --- 3. REGIONAL ANALYSIS a) NW Europe & Nordic Through the weekend and into early next week, conditions are unremarkable. Amsterdam and London hold in the 18-22°C band under southwesterlies; winds are light (ECMWF IFS averages 2.2 m/s in Amsterdam and London, 1.9 m/s in Frankfurt). These are below the wind speeds that deliver meaningful offshore North Sea capacity factors. This is not a high-output wind period for the first half of the week. The frontal passage changes that profile sharply around July 22-23. The 12Z run upgraded Amsterdam's Wednesday wind peak by 3.0 km/h to 20.0 km/h, consistent with a deepening frontal wave rather than a shallow coastal trough. The full ECMWF 7-day Amsterdam wind average sits at 18.3 km/h, healthy but concentrated in the back half of the week. For the German Bight and the key offshore zones feeding into German and Belgian grids, the trough deepening around days 4-5 implies a short window of elevated wind capacity factors before the zonal regime reasserts. The southern-track ensemble members, where the trough cuts further into France rather than passing through the Netherlands, would extend and intensify that wind generation window, while northern-track members see it clear more quickly. The Nordic hydro situation receives no material relief from this pattern. Oslo's EC46 shows week one at 16.3°C [14.9-17.6°C] and week two at almost identical 16.3°C [14.3-18.5°C]. Precipitation signals are weak. The same southwesterly setup that is mildly warm and dry for the UK and Benelux is also keeping Norway under a blocking-adjacent ridge with below-normal runoff conditions. The hydro reservoir deficit does not narrow this fortnight. The AO recovery to near +1.9 by day 7, if it validates, is the structural support for elevated North Sea wind output in the latter part of the forecast window. Strongly positive AO states are associated with a more consolidated polar vortex and stronger, more persistent westerlies. The week-two EC46 warmth for Northwest Europe (Amsterdam 19.4°C [16.9-22.3°C]) would also be consistent with a well-established Atlantic ridge delivering warm southwesterlies rather than cold northwesterlies, the two scenarios have very different implications for wind generation even if the temperature centres land similarly. b) Southern & Eastern Europe The Mediterranean pattern is straightforward and high-confidence through both forecast weeks. Rome posts 28.0°C [27.1-29.0°C] in week one, the EC46 spread is just 1.9°C, indicating strong model agreement on summer heat persistence. Week two eases to 26.3°C [24.6-27.9°C], a modest cooling that likely reflects downstream influence from the NW European trough in some ensemble members. Madrid week one averages 30.1°C [28.7-31.3°C], week two 28.7°C [25.9-31.5°C]. Solar irradiance across southern Europe should remain high throughout. The ensemble does not suggest persistent cloud or meaningful precipitation for the Mediterranean basin in either week, which supports strong solar generation from the Iberian peninsula through Italy. The summer high pressure that characterises Iberian conditions in July-August is not challenged in any plausible ensemble scenario. The EU gas storage context sits quietly in the background: Belgium at 28.0% full, Netherlands at 31.4%, Germany at 44.9%. Italy, at 71.6%, is the outlier. These levels, particularly for the northwest, are below typical seasonal expectations for mid-July. The mild-to-warm weather pattern suppresses heating demand, which helps ongoing injection, but the relatively low starting point for Belgium and the Netherlands means the injection season has limited buffer against any early autumn demand pull. c) East Asia Japan is in sustained summer heat with high near-term confidence. Tokyo's 15-day CDD accumulation of 152.7 and Nagoya's 172.4 reflect a sustained cooling demand well above any comfort threshold. The EC46 week-one ensemble for Tokyo centres at 29.1°C [27.4-30.5°C], a tight range indicating robust model agreement. Week two eases modestly to 28.7°C [26.2-30.9°C], with wider spread, but the ensemble centre remains firmly in cooling territory. Seoul warms from 25.4°C in week one to 26.8°C [25.6-27.8°C] in week two. Shanghai holds near 29.6-30.0°C across both weeks with tight ensemble spread. The overall East Asian picture is one of a subtropical ridge maintaining its grip on the region, consistent with El Niño's suppression of the northward monsoon retreat that would normally bring relief. The MJO is in Phase 4 at amplitude 0.7, barely above the threshold that normally generates meaningful teleconnections. Phase 4 places the convective envelope over the Maritime Continent to western Pacific sector, which tends to reinforce the subtropical ridge over Japan rather than weaken it. At this amplitude, no significant MJO-driven modification to the Japan heat pattern is expected in weeks one or two. If the MJO strengthens and progresses toward Phase 6-7 over the coming two weeks, a scenario consistent with the El Niño-enhanced western Pacific convective environment, it would become a potential moderating influence on Japanese demand in weeks 3-4. The current amplitude does not yet support that call. The sustained Japan heat signal, holding with moderate confidence through week two, is the most consistent demand signal in the Pacific basin this cycle. d) Americas The US pattern is unusually well-defined and internally consistent across all three guidance sets. A strong anomalous 500-hPa ridge covers the western and southeastern United States; a persistent trough is anchored over the northeast. CPC's 6-10 day and 8-14 day discussions both point toward the same structure, with the western ridge gradually de-amplifying in week two but not collapsing, and the northeastern trough retrograding toward the Ohio Valley before weakening toward the end of the period. The practical consequence across the southern tier is extreme and persistent heat. Dallas's 15-day average temperature is 34.3°C with a CDD accumulation of 240.6; Houston at 32.2°C averages CDD of 207.5; Phoenix at 36.7°C reaches 275.2 CDD. The NOAA seasonal outlook for August-September-October layers on top of this with above-normal temperature probabilities across the West and South, highest over the Pacific Northwest and southwestern Texas. There is no credible cooling scenario for the Gulf Coast and Texas demand corridor within any planning horizon covered by this briefing. The northeast is a different story. Boston is running a sequence of frontal events: 10.8mm of rain on Sunday July 19, 8.9mm Tuesday, 8.0mm Wednesday. Average temperatures remain mild, well below cooling thresholds. The July 19 downgrade of 3.5°C for New York's Sunday maximum is this run's most consequential US change, it pulled the afternoon peak to 29.3°C, consistent with the leading edge of the trough arriving ahead of the weekend temperature peak that yesterday's run had suggested. The CPC 8-14 day describes the trough retrograding from southeastern Canada toward the Ohio Valley, meaning the mid-Atlantic states eventually come under ridge influence from the west, but the transition timeline sits near the boundary of reliable guidance. For the Atlantic hurricane season, no storm data entered today's feed. The El Niño background is the dominant context: strong El Niño events suppress Atlantic tropical activity through increased upper-level wind shear, and the 81% probability of a very strong El Niño by October-December is a meaningful structural dampener on Gulf energy infrastructure risk through the peak of the season. This does not eliminate storm risk, but it reduces the probability density of major Gulf events. Brazil's São Paulo EC46 data shows a week-two spread of 14.2-23.1°C around a centre of 19.3°C, a 9°C range that is exceptional even by extended-range standards. This represents genuine model divergence on whether a cold frontal outbreak penetrates into southeastern Brazil in the week-two window. In the cold scenario, southern Brazil precipitation could be depressed as the cold air is dry; in the warm scenario, monsoon moisture maintains higher runoff for the Paraná basin. The ECMWF seasonal bulletin identifies above-normal precipitation for parts of southern South America in October-December, but the week-two uncertainty is a live model disagreement, not a climatological expression. e) India & Australia Mumbai holds at 26.1°C in both week one [25.2-27.0°C] and week two [25.3-27.0°C] with remarkably tight ensemble spread across both periods. The Indian summer monsoon is imposing its characteristically narrow temperature corridor. No anomalous heat or cold signal is present in the South Asian data. The ECMWF seasonal signal of below-normal precipitation for southern Asia in August-October, a canonical El Niño pattern, is an emerging tail-risk for sub-continental hydro power systems toward the fourth quarter, but lies outside the reliable range here. Australia is in its heating season with meaningful demand loading across southeast cities. Melbourne's 15-day HDD accumulation of 119.7 (10.0°C average) and Adelaide's 105.3 (11.0°C) represent sustained gas and electric heating demand for the NEM winter. The EC46 for Melbourne confirms a cold bias in the near term, week one at 8.6°C [7.3-9.8°C], gradually recovering through weeks three to six toward 11.0°C [8.5-13.9°C]. The most operationally relevant development from today's run in the Australian domain is a cohesive upgrade to southeast wind speeds across the July 21-23 window. Melbourne's July 22 wind peak was raised 6.4 km/h (now 17.6 km/h), July 23 by 4.4 km/h, July 21 by 3.9 km/h. Adelaide mirrors the pattern: +5.4 km/h July 22 (now 17.7 km/h), +3.4 km/h July 21, +3.1 km/h July 23. This is a coherent upgrade across adjacent stations, consistent with the 12Z run tracking a more robust cold front or trough crossing southern Australia around mid-week. The NEM South Australia and Victoria wind capacity factors for July 22-23 are likely to exceed yesterday's expectations by a material margin. The upgrade persists into the sub-seasonal window as well: Adelaide August 24-25 wind peaks are also raised (+4.5 and +3.3 km/h respectively), suggesting a more active late-July to August westerly regime than the previous run captured. --- 4. EXTENDED RANGE & REGIME (WEEKS 3-6) The EC46 extended ensemble for Europe through weeks 3-6 tells one dominant story, warm, with uncertainty growing quickly in amplitude rather than direction. Paris weeks 3-4 centres around 22.6-23.1°C with upper-end ensemble members reaching 26.9-27.3°C. Frankfurt week four averages 21.8°C [18.0-26.3°C]. The 8°C spread in week four for Frankfurt is not model noise; it reflects genuine uncertainty about whether the July blocking episode described above sets up a more persistently amplified pattern or whether the zonal regime fully reasserts. But across all spread scenarios, the ensemble centre never drops below seasonally near-normal levels for NW Europe in this window. The dominant large-scale driver is the strengthening El Niño. The Nino 3.4 SST anomaly of +2.3°C on July 8 was already elevated; a downwelling Kelvin wave has deepened the eastern Pacific thermocline further since then. The NOAA NMME ensemble forecasts intensification through 2026, with 81% probability of a very strong El Niño event by October-December. At those magnitudes, this would rank among the largest events in the post-1950 record. For European extended-range weather, strong El Niño events drive above-normal temperatures through a combination of tropical heating effects on the stratospheric polar vortex (strengthening it, reducing blocking tendency) and direct sea-surface temperature influences on European climate through the North Atlantic warming pattern. The ECMWF seasonal confidence in well-above-normal European temperatures through the forecast period rests primarily on this coupling. The QBO adds nuance. The 50hPa QBO is currently in its easterly phase at -1.5 m/s. The Holton-Tan relationship associates easterly QBO with a weaker polar vortex and increased blocking risk at high latitudes, particularly during Northern Hemisphere winter. In mid-summer, this signal is substantially attenuated, but it provides a background bias toward NAO variability, which is exactly what the GEFS is showing in the days 3-5 NAO dip. The tension between the El Niño vortex-strengthening signal and the QBO vortex-weakening tendency is visible in the post-day-5 AO trajectory: the AO recovers strongly toward +1.9 by day 7, suggesting El Niño wins in the near term. But the easterly QBO seeds higher blocking variability into autumn, and the combination of strongly negative PDO (-9.90 as of December 2025) and strongly positive ENSO creates a complex North Pacific background state that historically generates wider spread in NAO forecasts beyond week four. The MJO at Phase 4, amplitude 0.7, is marginally active. Phase 4-5 propagation enhances the Pacific subtropical ridge, which supports East Asian summer heat and is consistent with the current Japan pattern. If the MJO strengthens, plausible in the El Niño-active western Pacific, Phase 6-7 propagation in weeks 3-4 could moderate Japan demand slightly and deliver enhanced precipitation to the western Pacific typhoon track. This is a scenario to watch in subsequent briefings rather than a confirmed signal today. --- 5. DATA FRESHNESS & CONFIDENCE - Open-Meteo 16-day: Current as of 2026-07-17. Fresh; used for day-by-day city data and 15-day HDD/CDD accumulations. - ECMWF IFS 12Z: Initialized 2026-07-17. Fresh; primary driver for the NW Europe frontal analysis and regional confidence assessments. - EC46 46-day ensemble: Current as of 2026-07-17. Fresh; used for weeks 1-6 temperature ranges and regime probabilities. Point forecasts beyond day 14 should be read as probabilistic envelopes, not predictions. - GEFS NAO/AO forecast sequence: Current but carries the standard GEFS day 5-8 degradation. The day 7 AO recovery toward +1.9 is an ensemble mean; individual members show considerable spread. The directional signal (recovery from the dip) is robust; the magnitude is not. - Climate indices (NAO/AO/MJO/QBO/ENSO): The observed NAO (+0.58) and AO (+1.23) values are dated June 30; MJO to July 5; QBO and ONI to May-July. These are structural baselines, not real-time readings. The GEFS forecast sequence provides the current trajectory. - NOAA CPC outlooks: Updated 2026-07-16, one day old, valid for the 6-14 day US analysis. No material degradation expected. - ENSO: Nino 3.4 weekly reading as of July 8. The structural El Niño is well-established and the nine-day lag does not materially affect the seasonal analysis. The NOAA ENSO discussion is current as of July 16. - ECMWF Seasonal Bulletin: Dated July 10. Valid for the extended-range regime discussion; seven days old but seasonal-scale signals are stable. - West Pacific tropical storms: No JTWC advisory data entered today's feed. The storm watch desk should be consulted for any active systems in the Philippine Sea, South China Sea, or northwest Pacific before Asian markets open. - Brazil hydro: No direct reservoir or precipitation data available. The São Paulo EC46 spread is the only proxy for southeastern Brazil hydro conditions in this briefing; the uncertainty range is genuinely wide. - India monsoon precipitation distribution: Only Mumbai temperature data available. Sub-continental rainfall distribution and reservoir status are not represented in today's feed.