Why the ozone hole is healing, and still opens every spring
A reader asked: the ozone hole was fixed, so is it still healing? The short answer is that the cause was fixed and the damage was not. The chemicals that thin the ozone layer were phased out decades ago, but the ones already in the air break down over decades too, and they are still there. Measurements say the chlorine over Antarctica is falling and the layer is slowly recovering. A hole still opens over Antarctica every spring, and in a year with a cold stratosphere, such as 2026, it can open wide. Both things are true at once, and the reason is the most useful part of the answer.
What was broken
About 90% of the atmosphere's ozone is in the stratosphere, most of it between about 15 and 35 km up, the band called the ozone layer. It absorbs much of the sun's ultraviolet light before it reaches the ground. There is not much of it. Bring all the ozone in the atmosphere down to the surface and it would make a layer of pure ozone about 3 mm thick. Scientists measure that column in Dobson units, where 300 is about 3 mm, and an "ozone hole" is the area where the column falls below 220.
The damage came from gases made for fridges, air conditioners, foams, spray cans and fire extinguishers: the chlorofluorocarbons (CFCs), and their bromine-carrying cousins, the halons. They were prized because they are so unreactive. Nothing at ground level breaks them down, so over years they drift up into the stratosphere. There the hard ultraviolet light finally splits them, and the chlorine and bromine come free.
A free chlorine atom attacks ozone in a cycle that hands the chlorine back at the end, ready to go again. It is a catalyst, so one chlorine or bromine atom can destroy many thousands of ozone molecules before it is locked away. Most of the time it is locked away, in two fairly harmless "reservoir" gases, hydrogen chloride and chlorine nitrate.
Why the hole is over Antarctica, in spring
In the southern winter a ring of strong winds, the polar vortex, spins around Antarctica and walls off the air inside it. That trapped air gets colder than anywhere else in the stratosphere. Below about −78 °C, thin polar stratospheric clouds form. On the surfaces of their particles the two reservoir gases react with each other and turn almost entirely into reactive chlorine. In the dark of winter that chlorine has nothing to do. When sunlight returns at the end of winter, it starts the catalytic cycle all at once, and the ozone inside the vortex falls fast. When the vortex breaks up in late spring, ozone-rich air flows back in and the hole closes until the next year.
That gives every year's hole two ingredients. One is how much chlorine and bromine is up there, which changes slowly. The other is how cold the vortex gets and how long it holds together, which changes from one year to the next.
What was fixed, and what can't be hurried
The Montreal Protocol, adopted in 1987 and since ratified by every country, phased out production of these gases. Emissions fell. What no treaty can do is pull out the gas that was already in the air. Each gas leaves at its own rate, set by chemistry. The 2022 international assessment of the ozone layer gives an atmospheric lifetime, the time to remove about 63% of a gas, of 52 years for CFC-11 and 102 years for CFC-12. Old foam insulation and scrapped fridges in landfills also keep leaking small amounts of the banned gases. The widget below lets you drag that decay forward.
Interactive Suppose a gas stopped being released today. How much of it is still in the air some years later? Drag across the chart, or use the slider, to move forward in time. Pick a gas to follow it. The slider and the buttons work with Tab and the arrow keys.
| gas | lifetime, yr | half gone after, yr | still in the air |
|---|---|---|---|
| CFC-12 | 102 | 71 | 100% |
| CFC-113 | 93 | 64 | 100% |
| Halon-1301 | 72 | 50 | 100% |
| CFC-11 | 52 | 36 | 100% |
| Carbon tetrachloride | 30 | 21 | 100% |
| Halon-1211 | 16 | 11 | 100% |
| Methyl chloroform | 5 | 3 | 100% |
Nothing has broken down yet. Move the slider to go forward in time.
Is it healing?
Yes, slowly, and the measurements say so. The 2022 assessment found that by 2020 the effective chlorine over the poles had come down about 11% from its peak, a little under a quarter of the way back to its 1980 level. In late 2025, NOAA's Stephen Montzka put the Antarctic figure at about a third of the way back. The same assessment found that total ozone over Antarctica "continues to recover", even though the hole's size, strength and lifetime swing a lot from year to year. NASA's Paul Newman summed up the recent pattern in 2025: holes are smaller in area than in the early 2000s, form later in the season, and break up earlier.
The 2025 hole is a good example. NASA and NOAA ranked it the fifth smallest since 1992, averaging 18.71 million km² between 7 September and 13 October. Newman estimated that with the chlorine of 25 years earlier it would have been more than a million square miles, about 2.6 million km², larger. The assessment's projection is that ozone over Antarctica returns to its 1980 level around 2066, over the Arctic around 2045, and averaged between 60°N and 60°S around 2040.
The recovery is not even everywhere. Ozone in the upper stratosphere is recovering, but the 2022 assessment found that ozone in the lower stratosphere outside the polar regions had not shown signs of recovery, while models said it should have a little. That gap was still open in that assessment.
Then why was 2026's hole so big?
Because the slow ingredient sets the trend, and the fast one sets the year. In 2026 the vortex got strong and cold early. The Copernicus monitoring service reported that the hole's area jumped in early September as temperatures about 20 km above the South Pole dropped. Annika Seppälä, Dan Smale and Hannah Kessenich report that it peaked at about 27.4 million km² on 20 September, the second largest ever seen at that time of year. They also point out that it was not as deep as the big holes of the past, so less ozone was missing overall. When this page was written in mid-October the season was not over and no final ranking existed.
The weather can push the other way too. In 2002 the southern vortex split in a sudden stratospheric warming and the hole broke up far earlier than usual. In early September 2019 a strong warming of the stratosphere cut the area to 12 to 16 million km², where 20 to 25 had been typical for that date over the previous two decades. Neither of those small years meant the problem was solved, and a large year does not mean recovery has stopped. Smoke from Australia's 2019–20 wildfires has been linked to worse ozone loss in the years that followed, and the 2022 Hunga Tonga eruption to an early start of the 2023 hole. The 2022 assessment had not yet assessed those events, and the next one is due at the end of 2026.
What could still slow it
The chemistry only heals if the gases stay banned. Between 2012 and 2019 there were unexpected emissions of CFC-11, traced to unreported new production, and a large fraction of them came from eastern China. Once the source was found, global CFC-11 emissions fell, to about 45 thousand tonnes a year in 2019 and 2020. The 2022 assessment estimated that this episode could delay the polar recovery by up to three years. It also said the uncertainties, and the gaps in the monitoring network, were too large to be sure every such source had stopped. Looking further ahead, the same report expects nitrous oxide, which the treaty does not control, to matter more for ozone as the old gases fade. Carbon dioxide cools the stratosphere, which in some cold Arctic winters could make more of the clouds that set chlorine loose.
The short version
The fix was turning off the tap. The healing is the slow draining of what was already in the tank, at rates set by chemistry, over most of this century. Each spring's hole is that slowly falling chlorine meeting that year's weather. Judge the recovery by the chlorine and the trend of many years, not by the size of one hole.
Where this comes from
- Early indications of anomalous behavior in the 2019 spring ozone hole over Antarctica linked only, not reproduced
arxiv.org/abs/1909.07574 - Scientific Assessment of Ozone Depletion: 2022 — Executive Summary linked only, not reproduced
csl.noaa.gov/assessments/ozone/2022/downloads/executivesummary.pdf - Twenty Questions and Answers About the Ozone Layer: 2022 Update linked only, not reproduced
csl.noaa.gov/assessments/ozone/2022/downloads/twentyquestions.pdf - NOAA, NASA: 2025 ozone hole is 5th smallest since 1992 linked only, not reproduced
www.noaa.gov/news-release/noaa-nasa-2025-ozone-hole-is-5th-smallest-since-1992 - Small and short-lived 2025 ozone hole confirms long-term recovery trend linked only, not reproduced
wmo.int/media/news/small-and-short-lived-2025-ozone-hole-confirms-long-term-recovery-trend - Ozone hole reaches notable threshold early again as area increases above average in 2026 linked only, not reproduced
atmosphere.copernicus.eu/copernicus-ozone-hole-reaches-notable-threshold-early-again-area-increases-above-average-2026 - September's ozone hole was the biggest in 20 years. The warning signs appeared months ago linked only, not reproduced
theconversation.com/septembers-ozone-hole-was-the-biggest-in-20-years-the-warning-signs-appeared-months-ago-292457 - Scientific Assessment of Ozone Depletion: 2026 (in preparation; release expected late 2026) linked only, not reproduced
csl.noaa.gov/assessments/ozone/2026/