Despite lots of year-to-year variability, trends are now becoming clear.

The Montreal Protocol, which went into effect in 1989, is a rare instance of a global agreement to solve a global problem: the release of vast quantities of ozone-destroying chemicals into the atmosphere. In the decades since, however, changes in ozone have been small and variable, making it hard to tell whether the protocol is making any difference.

But evidence has been building that the ozone layer is recovering, and a new paper claims to have directly measured the ozone hole gradually filling back in.

CFCs and ozone

During the 1970s and '80s, evidence had been building that a class of industrial chemicals, the chloro-flurocarbons (CFCs), were damaging the ozone layer, a region of the stratosphere rich in this reactive form of oxygen. Ozone is able to absorb UV light that would otherwise reach the Earth's surface, where it's capable of damaging DNA. But the levels of ozone had been dropping, which ultimately resulted in a nearly ozone-free "hole" above the Antarctic.

The ozone hole spurred countries and companies into action. As companies developed replacements for CFCs, countries negotiated an international agreement that would limit and phase out their use. The Montreal Protocol codified that agreement, and it is widely credited with reducing (though not eliminating) the CFCs in our atmosphere.

But determining whether the protocol is having the desired effect on the ozone layer has been challenging. Ozone is naturally generated in the stratosphere at a very slow rate, and the amount of destruction that takes place over the Antarctic varies from year to year. Hints of a recovery have often been followed by years in which ozone levels drop again. Recovery has been so slow, in fact, that it's possible to find people who claim the whole thing was a scam—and even a conspiracy designed to test whether it was possible to create a similar agreement for greenhouse gasses.

The challenge with getting definitive measurements are legion. To begin with, CFCs aren't the only chemicals that can react with ozone, so tying things to them is tricky. In addition, the weather has a big influence on ozone's destruction. The atmosphere over Antarctica picks up the CFCs during the Southern Hemisphere summer as they're brought down by winds from the mid-latitudes. Over the winter, strong winds block off a variable-sized region above the pole where the hole develops.

In addition to winds, temperature also has a large effect on the rate of the ozone-destroying reactions, which take place on ice crystals.

As a result of all these factors, the amount of ozone destroyed and the physical size of the hole vary from year to year. While there have been some strong hints of the ozone hole's recovery, scientists have ended up arguing over whether they're statistically significant.

Tracking chemicals over the Antarctic

Which brings us to the new research. Early last decade, NASA launched the Aurora satellite, which was specifically designed to track the chemical composition of the atmosphere. Two researchers from NASA's Goddard Space Flight Center, Susan Strahan and Anne Douglass, have now analyzed a dozen years of Aurora data to directly tie ozone recovery to the drop in CFCs.

To begin with, the team found a proxy for the amount of chemicals that arrived over Antarctica during the Southern Hemisphere summer. Nitrous oxide is also brought in on the winds, and unlike CFCs, it doesn't undergo reactions over the course of the winter. Thus, it can serve as a trace for the amount of CFCs that are available each winter.

Strahan and Douglass use some additional chemistry to demonstrate that nitrous oxide levels correlate with the amount of CFC present. They focus on measuring chlorine levels at the end of the winter season. By this point, in addition to destroying ozone, the chlorine in the CFCs would have reacted with methane and produced hydrochloric acid. Thus, the amount of hydrochloric acid present should also relate to the starting concentration of CFCs.

This data showed that the overall levels of CFCs and nitrous oxide were highly correlated but that the ratio of the two was slowly shifting as the total amount of chlorine was trending downward. That's a key piece of data, as it shows the Montreal Protocol is working as intended—fewer CFCs are being delivered to Antarctica each year.

The next step is to compare those numbers to ozone concentrations within the hole. This was done simply by comparing the concentrations of ozone over 10-day windows in the early and late winter. Rather than providing an absolute measure of the amount of ozone present, the comparison gives an indication of what percentage of ozone is being destroyed each winter.

Overall, the researchers show that the amount of chlorine over the Antarctic is declining at a rate of 25 parts-per-trillion each year (that's 0.8 percent), although the weather-driven variability means that there are some years it increases. This resulted in a 20-percent drop in ozone destruction during this period.

The good news is that the variability in chlorine present paralleled that of the annual ozone destruction, validating both the overall science behind ozone destruction and nicely matching detailed chemical models of ozone's sensitivity to CFCs.

It's hard to summarize these findings any better than the authors do: "All of this is evidence that the Montreal Protocol is working—the chlorine is decreasing in the Antarctic stratosphere, and the ozone destruction is decreasing along with it."