The Strongest El Niño Since 1950 Is Coming. Cyprus's Dams Won't Notice.

The US climate agency gives a better than 90% chance of a very strong El Niño this winter, possibly the strongest since records began in 1950. Seventy-six winters of Cyprus rainfall and thirty-nine years of dam records say the same thing: it will make no difference here. The correlation is 0.05, the eight strongest El Niños averaged 82% of normal rain, and the two emptiest winters of the past decade were El Niño winters.

Every few years the Pacific Ocean warms along the equator, the trade winds slacken, and the world's weather desks start writing about El Niño. This year they have reason to. On 13 August NOAA's Climate Prediction Center declared an El Niño Advisory and said there was a greater than 90% chance of a very strong event through the coming winter, with a 69% chance that its October–December index exceeds +2.5 °C. No event since 1950 has managed that. The World Meteorological Organization's August update put the odds of El Niño persisting through the winter at near 100%. Australia's Bureau of Meteorology, which uses a stricter yardstick, measured the equatorial Pacific at +2.45 °C in the last week of August.

In Cyprus the question arrived on cue. Politis asked in May whether a strong El Niño would reach the island. Phileleftheros put it to the Department of Meteorology's director, Filippos Tymvios, in June. Both answers were careful. Tymvios said any effect would be "indirect" and pass through the large-scale circulation, that if it appeared at all it would be in autumn and winter, and that the department "does not treat it as an immediate forecast for Cyprus."

That caution is well placed. Because the intuitive story, that an ocean event big enough to flood Peru and parch Australia must do something to a Mediterranean island, has a data problem. The data exists, it is long, and it says no.

The 76-winter test

Cyprus's Department of Meteorology has published an island-wide rainfall figure for every hydrological year since October 1901. The series is easiest to get from Kairika, Fragmata's sister site for Cyprus weather, which serves it through an open API alongside a live rain-season tracker; we used that copy here, and extended it to this summer from the DoM's monthly reports. NOAA's Oceanic Niño Index, the standard measure of El Niño and La Niña, runs from 1950. Overlap the two and you get 76 winters, each with a Pacific temperature and a Cyprus rainfall total, from 1950–51 to the season that ended last week.

Plot them and this is what appears.

If El Niño winters were wet ones, the cloud of dots would lean up and to the right. It doesn't lean anywhere. The best-fit line through 76 winters is nearly horizontal and the correlation coefficient is 0.05, a number that in any other context would be called zero. Split the winters by phase and the averages sit within a few points of each other:

El Niño winters do come out about seven points wetter than La Niña winters on average. But the spread inside each group is enormous. El Niño winters have delivered anything from 42% of normal to 159%, and seven of the twenty-seven fell below 80%, the level at which Cyprus starts talking about drought. If you were told only that an El Niño was under way, the honest forecast for the island would be "somewhere between a drought and a flood," which is the forecast you had before you were told.

The strong ones

Perhaps weak El Niños are noise and only the big events matter. This winter's is expected to be a big one, so it is the right question to ask. Eight winters since 1950 have had a November–January index of +1.5 °C or more, the threshold climatologists use for "strong."

Six of the eight were dry. Together they averaged 82% of normal. The three that dominated the satellite era, the ones with their own Wikipedia pages, were all below normal in Cyprus: 1982–83 at 87%, 1997–98 at 77%, and 2015–16, the strongest event ever measured, at 61%. The most recent strong El Niño, 2023–24, brought 75% and opened the drought that emptied the island's dams by the start of this year.

And the winter of 1972–73, an El Niño with an index of +1.77 °C, is the driest hydrological year in the entire 125-year Cyprus record: 213 mm, 42% of normal.

There is one more way to see this, and it is the most persuasive. Lay the Pacific index and the Cyprus rainfall anomaly on the same time axis, year after year, and look for the pattern.

The red spikes in the top panel are El Niños. The bars in the bottom panel are Cyprus winters, coloured by phase. The wettest winter since 1950, 1968–69 at 159%, was indeed an El Niño, but a weak one. So was 2018–19, the season that refilled every dam on the island. And the bar for the winter that has just ended, 2025–26, with the island at 121% of normal and the dams recovering from their worst level since 2009, is blue. That was a weak La Niña, the phase that is supposed to bring drought.

What the dams saw

Rain is the input. What matters to a reader of this site is what reached the reservoirs, and for that there is a second, shorter record: Water Development Department storage readings for every dam since January 1988. From those we can measure each winter's refill, the rise from the autumn low to the following spring's peak across the 18 main dams, and tag each winter with its Pacific phase.

At first glance El Niño winters win: their median refill is 74 mln. m³ against 43 for La Niña and 34 for neutral winters. Look closer and the whole difference is one dot. The winter of 2018–19 took the dams from 32 mln. m³ on 1 December to 271 in mid-May, a refill of 238 mln. m³ that no other year comes close to. It was an El Niño winter, a weak one, with an index of +1.05. Remove it and the El Niño average drops to the middle of the pack. The correlation between the Pacific index and dam refill across all 39 winters is 0.12.

What is left when you take away 2018–19 is not flattering to El Niño. The dams refilled by less than 10 mln. m³ in five winters since 1988. Two of them, 2015–16 (+2.6 mln. m³) and 2023–24 (+7.6 mln. m³), were the two strongest El Niños in the whole storage record. The third-strongest, 1997–98, managed 12.9. In the WDD's seasonal inflow accounts, 2015–16 (17.9 mln. m³) and 2023–24 (24.7) are two of the three worst seasons of the past eleven.

Meanwhile the three biggest refills after 2018–19 were 2011–12 (+152 mln. m³, La Niña), 2001–02 (+145, neutral) and 2021–22 (+124, La Niña). The phase that is supposed to be dry filled the dams; the phase that is supposed to be wet emptied them. The truth is neither. The Pacific is simply not in charge here.

Why not

None of this should surprise the people who study it. Cyprus's winter rain comes from a specific machine: mid-latitude depressions that form or deepen over the warm eastern Mediterranean, the "Cyprus lows," steered by the position of the jet stream and by pressure patterns over the North Atlantic and Russia. The indices that track those patterns, the North Atlantic Oscillation and the East Atlantic–West Russia pattern, are what correlate with rain in the Levant. A 2025 study of 33 stations in the occupied north of the island found the western Mediterranean and North Atlantic oscillations to be the leading controls on the island's extreme rainfall, with the Pacific nowhere on the list. The Department of Meteorology's own published research on Cyprus rainfall extremes looks at the eastern Mediterranean pattern and the Siberian high, not at El Niño.

The best-known paper claiming a link, a 1998 study of Israel, found that El Niño years tended to be wetter, but only in the last quarter-century of its record, not before, and its authors called it a "possible link." A wider 2002 analysis of Mediterranean rainfall found the El Niño signal concentrated in the western basin, and in autumn actually drier conditions in the east. Later work showed the relationship switching on and off over decades. The IPCC's latest assessment lists the North Atlantic Oscillation as the driver of Mediterranean winter rain and assigns no Mediterranean rainfall teleconnection to El Niño at all.

Which is exactly what the Cyprus data shows: a weak, unreliable, sign-flipping influence buried under much larger ones.

What to watch instead

The winter now beginning will be an El Niño winter, and a big one. If it turns out wet, El Niño will get the credit; if dry, it will be quietly forgotten, as it was in 2015 and 2023. Neither would be evidence of anything. Seventy-six winters already are.

For anyone trying to guess what the dams will do by spring, there are better places to look:

• The North Atlantic Oscillation. A negative phase pushes the storm track south into the Mediterranean and is the single best-documented predictor of a wet Cyprus winter. Seasonal centres publish NAO outlooks from October.
• The DoM's seasonal outlook. The department issues a three-month forecast each month, produced with the regional SEECOF consortium. It will say "no clear signal" more often than not, which is itself information.
• The rain as it falls. Half of Cyprus's annual rain arrives in December, January and February. By mid-January the season's character is usually set. The Kairika season tracker plots the island's running total against the 1961–90 normal, week by week, with every season since 1940 as a backdrop, and Kairika's daily brief will say when the first real Cyprus low is on its way. This site's inflow chart then shows what reached the dams.

And, as ever, the level of the dams themselves. Cyprus enters this winter with 111 mln. m³ in storage, the most for the time of year since 2022, and a decade-long lesson in how little the Pacific has to do with it.

Author: Vladimir Bugay, Fragmata developer Data: El Niño index from NOAA Climate Prediction Center's Oceanic Niño Index (3-month running mean of Niño 3.4 sea-surface temperature anomaly), retrieved 6 September 2026; winters are classed El Niño at +0.5 °C or above and La Niña at −0.5 °C or below on the November–January value, the CPC threshold applied to a single season. Cyprus rainfall from the Department of Meteorology's island area-average series (data.gov.cy dataset 1042, 1901–2022), obtained via the open API at kairika.info, extended from the DoM's monthly area-average precipitation reports for 2022–23 to 2024–25 and the DoM figure of 608.3 mm to 31 August for 2025–26; all expressed as % of the 1961–90 normal of 503 mm. Dam storage 1988–2026 from the Water Development Department via Fragmata's dataset; refill is the spring peak minus the lowest reading between October and January across the 18 main dams, and is capped by capacity in years when the dams spilled. Forecast statements from the CPC ENSO Diagnostic Discussion of 13 August 2026 and the WMO El Niño/La Niña Update of August 2026. Studies cited: Price et al. 1998 (Geophysical Research Letters), Mariotti et al. 2002 (GRL), Krichak et al. 2002 and 2014, López-Parages & Rodríguez-Fonseca 2012, Zaifoglu & Brocca 2025 (International Journal of Climatology), IPCC AR6 WGI Annex IV. Dam levels updated daily at fragmata.info; Cyprus weather, rainfall history and the rain-season tracker at kairika.info.