'You've got to move fast': Science learns to quickly link extreme weather and climate

Cabins destroyed in fire near Pemberton A number of cabins have been destroyed by a raging wildfire north of Pemberton that forced hundreds to evacuate.

As firefighters and other first responders battle an unprecedented summer of fires, floods, tornadoes and heat waves around the country, a group of Canadian scientists are asking why they're happening in the first place.

"May and June were record hot months in Canada and we've got the record wildfire season as well," said Nathan Gillett of Environment and Climate Change Canada. "Yes, it has been busy."

Gillett heads the Rapid Extreme Event Attribution Project, a new federal program that uses the growing field of attribution science to promptly establish to what extent -- if any -- a specific flood in British Columbia or wildfire in Quebec is due to climate change.

"The idea is to be able to make rapid extreme event attribution days or weeks after the extreme events occur," he said.

Twenty years ago, if you'd asked a scientist if climate change was linked to days of torrential rain or months of desiccating drought, you'd probably get an answer along the lines of "We can't say for sure but this event is consistent with the modelling."

But in 2003, a paper was published suggesting science could do better. Myles Allen of Oxford University borrowed a concept from epidemiology.

"You can say that smoking increases your risk of lung cancer by a certain amount," Gillett said. "In the same way, you can say human-induced climate change increased the risk of a certain event by a certain amount."

Since then, hundreds of attribution papers have been peer-reviewed and published. As well as Canada, governments including the United Kingdom, Australia, the Netherlands, South Korea, Japan and the United States are using attribution science.

Attribution science works by comparing climate models. One set of models will use data drawn from actual records while another, otherwise identical, set will be constructed with the influence of greenhouse gases removed.

Simulations will be run using those two sets and the difference in the results reveals the impact of climate change. It allows scientists to say to what extent the presence of greenhouse gases increased the likelihood of the event in question.

"It's probabilistic," Gillett said.

The process is now established enough, with peer-reviewed protocols and standards, that the calculations can be done quickly.

"Once you've got the method in place and it's validated, you really just have to get the observations from that event and you can provide a result," said Gillett.

Some events are easier to study than others. Gillett said his group hopes to be able to come to conclusions on heat waves in about a week, but wildfires, which involve more variables, will take longer.

Speed matters, said Clair Barnes, a researcher with the World Weather Attribution group in the U.K., which has studied the role of climate change since 2015 in more than 50 events around the world -- including the finding that the heat wave preceding the fire that levelled Lytton, B.C., was made 150 times more likely by climate change.

"Our aim is to look at high-impact events that are in the news," she said. "There was an appetite in the public and the media for more information about what's really happening now."

Promptly assessing the role of climate change after extreme events brings actual insight and information to the discussion, Barnes said.

"If you spend three years thinking about it, the media has already decided it was climate change or it wasn't climate change and has moved on. If you want to be involved in that discussion and bring some science to that discussion, you've got to move quickly."

But attribution science has more uses than just shaping public debate. Governments are using it inform their adaptation strategies. Financial institutions are using it to assess risk. It's come up in hundreds of court cases around the world attempting to attribute climate liability.

It does have its limitations.

Attribution science can only work where there's enough historical weather data to build an accurate climate model. That leaves out much of the global south, where some of the worst human impacts are occurring. As well, extremely local events are often beyond its resolving power.

"You do have to be careful to communicate the uncertainties," said Gillett. "We shouldn't be overconfident."

There's certainly no shortage of work. Barnes said her group has had to establish a strict protocol that weighs the magnitude of the event, the amount of damage it inflicts and its effect on human lives to weed out which events merit study.

"There are so many events that we just don't have the time to look at them all."

But World Weather Attribution has found the time to consider Canada's wildfires. It's a complex one, so results aren't expected for another month or so.

By then, chances are there will be a new extreme event to consider. When Barnes joined World Weather Awareness, she assumed winter and summer -- the times of peak temperature highs and lows -- would be the busiest. Not so.

"We've had temperature records set for the last few months and it's not even the peak of boreal summer," she said. "It's just been non-stop."

This report by The Canadian Press was first published July 22, 2023.