The ozone hole should heal around 2066

The ozone hole over Antarctica should be back to its 1980 state around 2066, about forty years from now, according to the latest international assessment of the ozone layer. The rest of the layer heals sooner: around 2045 over the Arctic and around 2040 across most of the world. The models behind the Antarctic date spread from 2049 to 2077, so “the mid-2060s” is as precise as anyone can honestly be².

“Recovered” here means back to 1980, not back to an untouched sky. 1980 is about when the Antarctic hole became readily detectable¹⁴, and NASA’s records already show a small one that year⁶. A recent study suggests human-caused ozone loss could first have been detected as early as 1957, high in the stratosphere over the tropics²⁴. Until about mid-century, a hole will keep opening over Antarctica each spring².

A layer 3 millimetres thick that blocks the sun’s harshest light

The ozone layer sits in the stratosphere, between about 15 and 35 kilometres up, where about 90% of the atmosphere’s ozone is found. It absorbs all of the sun’s most dangerous ultraviolet light (UV-C) and most of the next most harmful kind (UV-B), the kind that causes sunburn⁵. Squeezed down to the pressure at ground level, the whole layer would be about 3 millimetres thick¹⁵.

Scientists measure it in Dobson units: 100 would make a layer of pure ozone 1 millimetre thick¹⁰. Over Antarctica in the early 1970s, spring ozone was about 350 Dobson units. The “hole” is the area where it falls below 220, and in its core it drops to about 120, a third of what was there before⁵.

This is a different problem from the ozone in smog. Ozone near the ground, formed when traffic and industrial pollution react in sunlight, is a pollutant that harms lungs and crops. The same molecule high up protects life⁵.

A hole the size of North America, most springs

In a typical recent year the ozone hole often reaches about 25 million square kilometres, about the size of North America⁵. NASA ranks years by the hole’s average size from 7 September to 13 October, and on that measure the worst was 2006, at 26.6 million square kilometres. The largest single day on NASA’s record was 29.9 million, on 9 September 2000⁶.

Size swings a lot from year to year. In 2019 a burst of warm air in the stratosphere kept the hole under 10 million square kilometres for most of the season, the smallest since 1982. NASA called it great news, then added in the next breath that it was “not a sign that atmospheric ozone is suddenly on a fast track to recovery”¹¹. The 2025 hole averaged 18.7 million square kilometres, the fifth smallest since 1992¹⁰.

The 2026 hole is a big one. On NASA’s preliminary daily data it peaked at about 27 million square kilometres on 18 September⁷, large but well short of that 2000 record. On the records of Copernicus, the European monitoring service, its peak of 27.4 million on 20 September was the third-largest reading since 1979, behind 2000 and 2015⁸. The final rankings come later in the year.

Why the hole opens over Antarctica

The hole opens over Antarctica because four things meet there every spring: the coldest stratosphere on Earth, thin clouds that form in that cold, a ring of winds that seals the air in, and the returning sun. The chlorine and bromine that destroy ozone come from human-made gases, mainly chlorofluorocarbons (CFCs) and halons, and they are spread through the whole stratosphere, but it takes all four to set them loose⁵.

First, the cold. In the polar winter the stratosphere drops below about −78 °C, cold enough for thin clouds to form in air that is normally bone dry. Over Antarctica that happens for about five months a year, over the Arctic for one to four. Second, those polar stratospheric clouds change the chemistry. On the surface of their tiny particles, chlorine locked in harmless compounds is turned into forms that attack ozone⁵.

Third, a ring of strong winter winds, the polar vortex, seals the air in, so the reactive chlorine builds up undisturbed. The southern vortex is stronger because the Southern Hemisphere has fewer mountains and less contrast between land and ocean to disturb it⁵.

The last ingredient is sunlight. The destruction needs light, so it starts when the sun returns to Antarctica in late winter and can eat 2 to 3% of the ozone a day. A single chlorine atom can destroy many thousands of ozone molecules before it is locked away again. The hole deepens through September and October. The vortex breaks up in late November or early December, and the hole is usually gone by mid-December⁵.

The Arctic rarely gets a true hole. It starts the winter with more ozone and its vortex is warmer and leakier, so losses are usually 10 to 20%. In unusually cold winters such as 2011 and 2020 they reached about 30%, and in March 2020 the chemical loss matched what Antarctica sees routinely⁵.

The chlorine will take decades to clear

Under the Montreal Protocol, the world has phased out 99% of the chemicals that cause the hole³, and the hole is still here because they last so long. CFC-11, once used in foam and aerosol cans, survives about 52 years in the air, and CFC-12, the old fridge coolant, about 102. Air takes about five and a half years to travel from the ground to the stratosphere over Antarctica⁵¹⁴, and old fridges, foams and fire extinguishers are still leaking what they contain⁵.

At the start of 2024, the ozone-destroying chlorine and bromine over Antarctica were 28% of the way back from their peak to the 1980 level. NOAA expects them to get all the way back around 2070¹⁴. The ozone itself should recover a few years before that, around 2066², partly because greenhouse gases cool the upper stratosphere, which slows ozone loss⁵.

There was also a hidden setback. Unreported production of CFC-11 between 2012 and 2019 is estimated to delay the polar recovery by up to three years². How it was caught is told on the Montreal Protocol page.

Over Antarctica the ozone layer is recovering, slowly

The ozone layer over Antarctica is recovering, and the evidence now points to the ban on ozone-destroying chemicals as the reason. Elsewhere the evidence is weaker: outside Antarctica the 2022 assessment had only low confidence that ozone has started to return², and over the Arctic there is little sign of it yet⁵.

Watching the hole’s size alone would never have shown the Antarctic recovery. From the 1990s the core of the hole held more chlorine than there was ozone to destroy, so its size came to depend on the weather rather than on how much chlorine was left¹³. Scientists had to measure the chemistry instead. NASA’s Aura satellite found about 20% less ozone destruction over Antarctica in the winter of 2016 than in 2005, in step with the falling chlorine¹².

In 2025, a study that searched the ozone record for the treaty’s signature found “robust” statistical and physical evidence that the chemical cuts are causing the start of the recovery¹⁶. NASA’s ozone team estimates that with the chlorine levels of 25 years ago, the 2025 hole would have been more than a million square miles (about 2.6 million square kilometres) larger¹⁰. The holes of 2024 and 2025 were both well below the 1990–2010 average in depth and size, the World Meteorological Organization reported¹⁵.

Then 2026 brought a giant one. It grew fast when the stratosphere above the South Pole turned unusually cold⁹. Researchers at New Zealand’s University of Otago had predicted a big hole back in May, from satellite signs of a very strong polar vortex²⁰. The hole is wide but not especially deep: the lowest ozone and the total amount missing were close to average⁹, and no rise in reactive chlorine has been detected¹⁹.

Some scientists see no recovery in the core of the hole

A 2023 study by the University of Otago group found that in October, ozone in the middle of the stratosphere at the heart of the hole had fallen by 26% since 2004, and linked the loss to changes in the winds, not only to chemicals¹⁷. If they are right, falling chlorine alone would not bring the core of the hole back on schedule.

Other scientists were not persuaded. Martin Jucker, of the University of New South Wales, argued that the large holes of 2020 to 2022 already had known explanations, such as bushfire smoke¹⁸.

Some of those explanations hold up better than others. Smoke from Australia’s 2019–2020 bushfires did make the 2020 hole larger, though it does not explain why that hole lasted so long¹. Water vapour thrown into the stratosphere by the Hunga Tonga volcano in 2022 had only a minor chemical effect on the 2023 hole²¹.

The last international assessment was written in 2022, before this study appeared. The 2026 assessment, due at the end of the year, should be the first to weigh the dispute⁴.

Antarctic spring sun now beats San Diego’s

The hole is over one of the emptiest places on Earth, but its effects reach further. At Palmer Station on the Antarctic Peninsula, the strongest ultraviolet light of the year is now about two and a half times what it was before the hole formed, and in spring it beats what San Diego gets⁵. In December 2020, when the hole lingered unusually late, King George Island off the Peninsula recorded a UV index of 14.3, with a daily dose comparable to Chile’s Atacama Desert²³.

That timing matters for wildlife. The hole used to close before summer, but in several recent years it has lasted into December, just as snow melts, plants emerge and many animals are breeding²².

The hole has changed the weather, too. Since the early 1980s it has pulled the Southern Hemisphere’s main band of westerly winds towards the pole, shifting where summer rain falls, a change that reconstructions of past climate suggest is unprecedented in 600 years⁵. It has brought more rain to parts of Australia and less to parts of South America²³.

Not every rise in UV is the ozone layer’s doing. In parts of Europe UV has risen by as much as 20% since the mid-1990s, mostly because skies have grown less cloudy¹⁵.

Climate change will move the finish line

The 2066 date assumes a middle path for greenhouse gases, and the climate will change it. Carbon dioxide warms the air near the ground but cools the upper stratosphere, and colder air there slows the reactions that destroy ozone, so more CO2 means faster recovery over most of the world. That may sound backwards after the Antarctic story, where cold is the problem. Over the poles in spring, cold matters because it makes the clouds that free chlorine; in the upper stratosphere it simply slows the chemistry that destroys ozone. Under high emissions, Antarctic ozone would be back at its 1980 level around mid-century⁵.

The reverse also holds. Under low emissions, ozone between 60°N and 60°S might not reach its 1980 level by 2100, partly because of nitrous oxide, whose human emissions now do more than twice the ozone damage of all the CFCs released each year. In the tropics, changes in how air circulates could keep ozone below its 1980 level until at least 2100 under some scenarios⁵².

Put together, the evidence suggests the chemistry is on track and the doubt now lies in weather and climate. The new assessment at the end of 2026 will update every date on this page⁴, and cold years like 2026 will keep producing large holes for decades. Each one will look like a setback without being one.