VRTINEC·News

Event analyses

19 November 1999: a Mediterranean cyclone over cold air – snow across inland Slovenia, the deepest November snow in decades

Event analyses · 19 November 2026

  • snow
  • ciklon

In mid-November 1999 autumn was still mild, but in less than a week everything turned into proper winter. On Friday, 19 November, a deep Mediterranean cyclone crossed the northern Adriatic just south of Slovenia. Because the whole air column over the country was already below freezing, all precipitation inland fell as snow, even in the lowest valleys of Dolenjska, Bela krajina and Štajerska. In Kočevje, Postojna, Celje and Črnomelj there had never before been so much snow in November since measurements began. Unlike many other Mediterranean cyclones, this time no milder air from the south reached us, so the snow did not turn into rain and it stayed on the ground almost until the end of the month.

The bigger picture: a block over the British Isles and cold air from the north

Early November had been very mild for the season, then the flow over Europe rearranged. Around 10 November a strong anticyclone settled over the British Isles and the North Sea, shutting off the westerly flow. On its eastern side the airflow turned to come from the north and north-east, while a cold trough deepened over Scandinavia and stretched down to the Mediterranean.

Tuesday, 16 Nov: anticyclone over the Atlantic, cold trough from Scandinavia to the Mediterranean. Source: ERA5, plot VRTINEC.

In Slovenia it turned much colder on 15 November; in places snow already reached the lowlands, and light snow fell again the next day. Bora set in on the coast and in the Vipava valley. Cold air filled the basins and valleys and daytime temperatures stayed around freezing. As in many big lowland snowfalls, this was the foundation: when moisture arrived later, there was no warm layer left to melt the snow into rain.

Night to 16 Nov: cold air (blue) spreads from the north-east across central Europe to the Alps and the Adriatic. Source: ERA5, plot VRTINEC.

17 and 18 November: the first heavy snow

The Atlantic ridge rebuilt, and on its eastern flank a new, even deeper trough dropped from the North Sea towards Germany and the Alps. Ahead of it, pressure began to fall over northern Italy. On the evening of Wednesday, 17 November, heavier snow reached Slovenia, heaviest overnight in the south and south-east. By morning around 30 cm had fallen in the Kočevje area, Notranjska and Dolenjska. It partly cleared during the day, but the snow stayed, as even in the lowlands the temperature barely rose above zero.

Night to 18 Nov: snow over all of Slovenia (dots: mostly snow). The freezing level is at the ground everywhere inland. Source: ERA5, plot VRTINEC.

Friday, 19 November: the cyclone crosses the northern Adriatic

The decisive part came on the night to Friday. The upper-level cold core slid across the Alps and a deep cyclone developed over the Gulf of Genoa within a few hours. This is typical for the Gulf of Genoa: the Alps block and deflect the cold air from the north, pressure falls quickly in their lee, and the warm sea gives the cyclone extra energy. By Friday morning the cyclone had moved over the northern Adriatic and Istria, where ERA5 shows the lowest pressure, and during the day it moved east across Croatia towards Serbia and Hungary.

Friday, 19 Nov, night: upper low over central Europe, new cyclone over the Gulf of Genoa and the Tyrrhenian Sea. Source: ERA5, plot VRTINEC.
Animation: sea-level pressure and 500 hPa height, Europe, every 12 hours from 15 to 22 Nov 1999. Source: ERA5, plot VRTINEC.

Slovenia was on the northern side of the cyclone. There, moist Mediterranean air rises over the cold layer near the ground, cools, and the moisture falls out as precipitation. Near the ground a light north-easterly to easterly wind kept feeding cold air. The freezing level therefore stayed at the surface everywhere inland and it snowed down to the lowest places – even where Mediterranean cyclones usually bring rain. The snow was dense, persistent and heavy. The heaviest precipitation fell over the Kočevje area, Gorski kotar and Notranjska, where the airflow ran into the Dinaric mountains. On the coast it rained with bora, and there was no snow by the sea.

Friday morning: cyclone over Istria and Kvarner, most precipitation over the Kočevje area and Gorski kotar. Source: ERA5, plot VRTINEC.

As the cyclone moved away to the east, snow spread over Štajerska, Kozjansko and Prekmurje too. By Saturday morning 40 cm of fresh snow had fallen in Kočevje, and around 35 cm in Kozjansko and around Laško, Šentjur and Maribor. The same cyclone brought unexpectedly heavy snow to Croatia; in Bosnia up to a metre of snow fell in places, and a storm hit the Dalmatian coast.

Night to 20 Nov: the cyclone moves east, snow spreads over eastern Slovenia. Source: ERA5, plot VRTINEC.
Animation: 12-hour precipitation, snow and freezing level, Alps and Slovenia, every 12 hours from 15 to 22 Nov 1999. Source: ERA5, plot VRTINEC.
Ljubljana according to ERA5: from 16 Nov all precipitation as snow, freezing level (pink) almost always at the ground. Pressure was lowest on Friday, 19 Nov, in the morning. Source: ERA5, plot VRTINEC.

Saturday, 20 November: a white landscape and frost

On Saturday it cleared. Above the fresh snow the air cooled strongly overnight; Celje and Maribor dropped below –10 °C. The snow cover was deepest in the south and south-east: that is where most snow fell, and also where the cold air held on most firmly.

Snow depth on Saturday morning: most snow in the Kočevje area, Notranjska, the Idrija hills and Kozjansko. No snow on the coast and in the Vipava valley. Data: ARSO, plot VRTINEC.

Sunday, 21 November: another round of snow

The upper trough stayed over central Europe, and a new, weaker cyclone formed over the Gulf of Genoa. On 21 November it also crossed the northern Adriatic. Snow fell again across the interior, this time more in the west and centre, and the snow cover grew further. Bora blew again on the coast. On Monday the snow cover was at its deepest in most of the country; clear and very cold weather followed, with morning temperatures around –15 °C in frost hollows.

Night to 22 Nov: the second cyclone over the northern Adriatic, snow again across the interior. Source: ERA5, plot VRTINEC.
Snow fell in three waves: on the night to 18 November, from 18 to the morning of 20 November, and on the night to 22 November. Data: ARSO, plot VRTINEC.
Monday, 22 Nov, morning: snow cover at its deepest in most of the country. Data: ARSO, plot VRTINEC.
Where ERA5 put the most snow: southern and central Slovenia and Gorski kotar. Source: ERA5, plot VRTINEC.

What the records show

For November, the lowland snow cover was unusually deep: more than half a metre in the Kočevje area and Notranjska, around half a metre in Bela krajina and lowland Dolenjska, and about 30 cm in Prekmurje. Celje, Postojna, Kočevje and Črnomelj had never measured so much snow in November before. In most of the interior the snow stayed until the end of the month, with a lake of cold air over the lowlands. Bora caused additional trouble in Primorska and Notranjska.

Measured: maximum snow depth 15–23 Nov 1999 – Kočevje 65 cm, Postojna 60 cm, Črnomelj (Dobliče) 60 cm, Novo mesto 50 cm, Rateče 45 cm, Maribor 35 cm, Celje 32 cm, Ljubljana 30 cm, Murska Sobota 30 cm; Turški Vrh in Haloze (282 m) 79 cm; Trava and Mrzla Rupa 105 cm; Kredarica 100 cm. Lowest temperatures after the snow: Nova vas na Blokah –16.6 °C, Postojna –15.6 °C, Maribor (airport) –14.4 °C.

Addition 1: Why it happened – an Atlantic ridge and a breaking wave

The event started with a block. Around 10 November a strong anticyclone settled over the British Isles. Such high pressure stops the westerly flow that in winter usually brings milder Atlantic weather. The flow then has to meander around the obstacle: on the eastern side of the block it turns to come from the north and cold air spills southwards.

From the block over the British Isles to the upper cold core over the Alps. Colours: 500 hPa height anomaly. Source: ERA5, plot VRTINEC.

The first block weakened around 15 November and cold air pushed from the north towards the Alps. Then the ridge over the eastern Atlantic rebuilt, this time even more steeply, towards Iceland. Warm air flowed north on its western side, while on its eastern side Arctic air slid straight south. Between 17 and 19 November a deep, narrow trough formed over central Europe.

Warm air (red) rises along the western side of the ridge towards Iceland; cold air (blue) spreads over central Europe on its eastern side. Source: ERA5, plot VRTINEC.

At the tropopause, the boundary between troposphere and stratosphere, this showed up as a breaking wave. A tongue of cold lower-stratospheric air stretched across Germany all the way to the Alps and the Gulf of Genoa. When such a tongue reaches a warm sea, a cyclone forms beneath it very quickly. An exceptionally strong jet was blowing over the Atlantic at the same time. The Genoa cyclone was therefore short-lived but very sharp.

Blue and green: lower-stratospheric air reaching low down. The tongue extends from the North Sea to the Alps. White lines: strongest upper-level wind. Source: ERA5, plot VRTINEC.

Did the stratosphere play a role? Before the event the polar vortex was slightly stronger than normal for the season, and it weakened only afterwards. The NAO index was only briefly and weakly negative, and already positive again during the snowfall. So the event was not a consequence of a weakened polar vortex. It was a tropospheric process: a block, a breaking wave, and the Alps that triggered the cyclone.

Addition 2: The tropical background – MJO, La Niña and Hurricane Lenny

The tropical background was striking. A pronounced La Niña held over the Pacific. The Madden–Julian Oscillation (MJO), which slowly travels along the equator, was very strong in the first half of November. Thunderstorm clouds lingered for weeks over the Maritime Continent between the Indian and Pacific Oceans (phases 5 and 6), and in mid-month the MJO moved into phase 7.

Green: more tropical thunderstorms than normal. Convection stays over the Maritime Continent for weeks; the MJO is very strong in early November. Source: NOAA, BoM, plot VRTINEC.

Such tropical convection launches Rossby waves aloft that travel across the Pacific and North America towards the Atlantic. The sequence of 15–19 November fits that picture well: a ridge over western North America, a trough over the east, a ridge over the Atlantic and a deep trough over Europe, where our cyclone formed.

Ridges (red) and troughs (blue) follow each other from the Pacific across North America and the Atlantic to Europe. Source: ERA5, plot VRTINEC.

At the same time Hurricane Lenny was raging in the Caribbean, a very strong late-season hurricane. Unusually, it moved from west to east and reached category 4 on 17 November. It stayed in the tropics, so it had no direct effect on Europe and its moisture never reached us. The snow clouds over Slovenia drew their moisture mainly from the Mediterranean.

Lenny stayed in the Caribbean; moist tropical air did not connect with the cyclone over the Adriatic. Source: ERA5, NOAA, plot VRTINEC.
Hurricane Lenny on a satellite image on 17 Nov 1999, near its peak. Image: NOAA, Wikimedia Commons, public domain.

An honest caveat is needed here. Research shows that after MJO phases 6–8, blocked patterns and a negative NAO over the Atlantic become more frequent on average over the following one to two weeks. But this is a statistical link that holds across many cases together, not a rule for each individual event. In November 1999 the timing fits and the wave train across North America is visible, but the NAO did not turn clearly negative. From one case we cannot prove that the MJO caused the snowfall – at most it raised the chances of such a pattern.

Why this event is instructive

A Mediterranean cyclone over the Adriatic brings the most precipitation, but what decides the outcome is the air beneath it. With many cyclones from the south, milder air flows in and lowland snow turns to rain – as at the end of November 2005, for example. In November 1999, however, the cyclone passed south of us, Slovenia stayed on its cold northern side, and a cold north-easterly blew near the ground the whole time. So it snowed down to the lowest places, and the snow cover grew where it is usually thinnest in November: in the south and south-east of the country.

When forecasting such situations, the track of the cyclone is crucial. Just a few tens of kilometres further north and milder air would have reached Bela krajina and Posavje, turning the snow into rain there. A little further south and the precipitation would have been weaker.

Maps: ERA5 reanalysis (Copernicus C3S/ECMWF), plotted by VRTINEC. The reanalysis grid is about 25 km, so mountains are smoothed and its precipitation is not a measurement. Measured data: ARSO, observation archive; T. Cegnar, Climate characteristics of Slovenia in 1999 and 2000, Ujma 14–15. Background: I. H. Bulić, Il Nuovo Cimento C 2005; NOAA (OLR, HURDAT2, NAO), BoM (MJO).

Gregor Skok