In the early morning on June 28, the hottest day in the brutal heat wave in the Pacific Northwest, climate scientists thousands of miles away were already deciding to run a “climate attribution” study that analyzed how much climate change supercharged the extreme temperatures. They’d seen the weather forecasts: It wasn’t an ordinary heat wave. “We just looked at each other on Zoom, and we said, ‘Well, there’s no way we cannot do this,'” says Geert Jan van Oldenborgh, a climate researcher at the Royal Netherlands Meteorological Institute and part of a coalition called World Weather Attribution.
For the next 10 days, an international team of 27 scientists worked nonstop to analyze what had happened during the heat wave, when some communities in the region broke heat records by margins of more than 10 degrees, and hundreds of people died. Seattle, normally in the 70s in June, hit 108 degrees. The village of Lytton, Canada, reached 121 degrees before a wildfire ignited and burned it down. It was clearly unusual—and when the researchers first started plugging numbers into models, the models said it was so extreme that it wasn’t physically possible.
“We first try to find observations that go back as far as we can find, and include the heat wave itself,” says van Oldenborgh. “We do a statistical analysis to see how much the probability of having an heat wave like this has changed in the observations. And we immediately ran into a problem because our standard statistical analysis says, ‘Oh, the heat wave you just observed was impossible.'”
The researchers were able to fine-tune the analysis until they found a statistical model that said that the heat wave could have occurred. When they downloaded climate models—computerized versions of the Earth that try to recreate the climate from the 19th century until the present—they calculated that climate change made the extreme heat 150 times more likely. But the analysis made it clear that scientists don’t understand how heat waves work as well as they thought they did. The impacts from climate change may be worse than models have predicted.
“There are basically two possibilities,” van Oldenborgh says. “Either, it was really, really bad luck—I mean, the world is a big place, and bad things happen somewhere, and the Northwest was just the place where it happens this year, and they were really unlucky. But the other option is that we’ve crossed some threshold that [made] a new mechanism kick in. And these heat waves are, in fact, a lot more likely now than we estimate from the history, because the history didn’t have this mechanism. And now we do.”
The study called the heat wave a “1 in 1,000-year” event, but noted that the temperatures were so far outside the normal range that it’s hard to say with certainty how rare it was. We don’t know yet if this is something that might start to happen every 20 years, or every decade, or even more often. We don’t know if the Pacific Northwest has unique conditions that made such extreme heat possible, or if the same thing could happen in Hong Kong or Norway. The scientists now plan to study the mechanisms that might have driven the heat wave, including drought and changes in the jet stream, and adjust predictive models.
There’s an unsettling possibility that current assumptions about how hot it can get—not just in the distant future, but now and next year—underestimate the problem. The same may be true for other impacts from climate change. The current flooding in Germany, for example, which has killed more than 170 people, left hundreds of others missing, and washed away some centuries-old houses, came after unprecedented rainfall. The World Weather Attribution scientists want to study the climate links to that disaster, too. “We are discussing whether we have the manpower to do the analysis,” says van Oldenborgh. “The problem is we’re a bit exhausted from the last few weeks.”