Ozone Layer Depletion
This is the one global air pollution problem the world has clearly turned around. Scientists predicted it in 1974. Ground stations proved it in 1985. Countries agreed to fix it in 1987. It is now healing. This slate covers how that happened, and the new problems the fix created on the way.
Status
mostly solved. Recovery is now measured, not just predicted. But the finish line is still about 40 years away, and several harmful gases are still not controlled.
About 99% of the controlled ozone-damaging chemicals have been phased out. [8]
2066Antarctic Return to 1980 Levels [8]
The Problem
CFCs were cheap, safe to handle, and used everywhere in fridges, foam and spray cans. That was the trouble. They do not break down at ground level, so they last 40 to 150 years and float up to the stratosphere, where sunlight splits them and releases chlorine that destroys ozone. [1] One chlorine atom destroys many ozone molecules before it is removed, so small leaks caused big losses. The ozone layer blocks the ultraviolet light that causes skin cancer, cataracts and weakened immune systems, and that harms crops and ocean plankton. At its worst the hole over Antarctica covered 29.9 million square kilometres, which is bigger than North America. [21] If nothing had been done, models say about two thirds of the world's ozone would have been gone by 2065, and you could have burned your skin in five minutes. [20]
Key Numbers
The figures this slate rests on. Each has a value, a unit, a year and a source. Correcting one of these is the smallest and most welcome contribution you can make.
How long CFCs last, 1974 [1]
40-150 yr
Largest hole on record, 2000 [21]
29.9 M km2
Countries signed at the start, 1987 [4]
46
Controlled chemicals phased out, 2022 [8]
99%
World return to 1980 levels, 2022 [8]
2040
Antarctic return to 1980 levels, 2022 [8]
2066
CFC-11 emissions drop, 2018 to 2019 [13]
26%
Warming avoided by Kigali [9]
0.5 C
Delay caused by feedstock leaks [17]
7 yr
UV rise in a world with no treaty, by 2065 [20]
650%
Recovery dates differ by region and by scenario, and the reports give them as estimates. The 2066 date in [8] and the 2073 date in [17] do not contradict each other. They are different scenarios, so both belong here. Where two good sources disagree, both stay on the slate with their citations.