the observatory

Extreme Measures

Must reporters cite climate change in every article about severe weather?
February 24, 2011

Last week, the journal Nature made a big splash in the press with the publication of two studies which found that manmade climate change has contributed to the intensification of heavy rains and increased the likelihood of floods that have, collectively, affected millions of people.

After years of hearing the scientific refrain that no single weather event can be attributed to climate change, those papers—as well as a spate of abnormal events in the last year, including a heat wave in Russia, floods in Pakistan, drought in China, and intense snowfall in the United States—have many reporters wondering if they now have a green light to report at will that severe weather bears the hallmarks of global warming.

In column for the Seattle PostGlobe, for example, former Post-Intelligencer reporter Jake Ellison argued that The New York Times was remiss for not mentioning climate change in a February 8 article about one of China’s worst droughts in decades. “Reporters should include #climatechange angle in all weather-tragedy stories,” the PostGlobe wrote in a subsequent tweet. “What say u, @cjr?”

We say, “not necessarily.” When talking about a heavy storm, many scientists—seemingly more every day—will say, “This is what we expect to see more of in a warmer world.” It is still impossible to attribute any single weather event to manmade climate change, however, and the connections between severe weather and climate change remain highly complex, nuanced, and uncertain. Andrew Freedman, the managing editor for online content at Climate Central, summed up the situation perfectly in an excellent post for The Washington Post’s Capital Weather Gang blog on Wednesday:

Although the ties between climate change and extreme weather events often elicit absolute statements from advocates on various sides of the climate change issue, the reality is that while much can be said, much remains unknown. The challenge for reporters as well as scientists is to accurately convey both the scientific findings and the uncertainties surrounding them.

Many journalists, politicians, and climate scientists have run into trouble by portraying the links between climate change and extreme weather in stark terms, rather than shades of gray.

Sign up for CJR's daily email

In one of the two studies published in Nature last week, researchers compared records of precipitation in the Northern Hemisphere from 1951 to 1999—which show an intensification of the heaviest rain and snow events—to dozens of model simulations that either accounted for human greenhouse-gas emissions or didn’t. They found that only when the gases were included did the models match what actually happened. In the other study, researchers harnessed the power of myriad personal computers through the project in order to run thousands of model simulations of a major flood in England and Wales in 2000. Again, some simulations included human greenhouse-gas emissions and some didn’t, and the researchers found that including the gases significantly increased the odds that the flood would occur.

The physical explanation for the connection between heavy precipitation and climate change is fairly straightforward—warmer air can hold more water vapor. It is important to note, however, that while the first study found that climate change had contributed to the intensification of heavy rains, it did not find that warming had caused any particular storm. Likewise, while the second study found that climate change had increased the likelihood of a particular flood, it did not find that warming has caused the flood. Moreover, both papers pointed out that, like all scientific studies, they are subject to a range of uncertainties. This doesn’t mean the studies aren’t robust and important, it just means that they cannot be used to generalize about weather-climate connections. (In a companion piece for Nature, climatologist Richard Allen, who wasn’t involved with either study, has a great rundown of their limitations and promise.)

As NASA climate modeler Gavin Schmidt explained in a post at, “This is a complex issue, and one not well-suited to soundbite quotes and headlines… The two new papers deal with the attribution of a single flood event, and the attribution of increased intensity of rainfall across the Northern Hemisphere. While these issues are linked, they are quite distinct, and the two approaches are very different too.” With that in mind, Schmidt presented “some very basic, but oft-confused points”:

Not all extremes are the same. Discussions of ‘changes in extremes’ in general without specifying exactly what is being discussed are meaningless. A tornado is an extreme event, but one whose causes, sensitivity to change and impacts have nothing to do with those related to an ice storm, or a heat wave or cold air outbreak or a drought.

There is no theory or result that indicates that climate change increases extremes in general. This is a corollary of the previous statement – each kind of extreme needs to be looked at specifically – and often regionally as well.

Some extremes will become more common in future (and some less so). We will discuss the specifics below.

Attribution of extremes is hard. There are limited observational data to start with, insufficient testing of climate model simulations of extremes, and (so far) limited assessment of model projections.

New York Times reporter-turned-blogger Andrew Revkin felt that the two studies published in Nature were not as “definitive” as they seemed and that their authors should have done more to highlight caveats in the research. “Did the authors stress the uncertainties in discussions with journalists?” he asked in a post at his blog, Dot Earth. “It sure doesn’t look that way. Should the journalists have pushed harder when confronted with definitive language? To my mind, yes.”

To some degree, Revkin has a point. Referring to the study that looked at the flood in England and Wales in 2000, an Associated Press article reported, “Researchers found that global warming more than doubled the likelihood of that flood occurring.” This seems to be what the paper’s authors told other reporters as well, but it doesn’t accurately reflect what they found. Ninety percent of the time, models showed that manmade warming increased the risk of floods by 20 percent; only two-thirds of the time did they show that warming increased the risk by more than 90 percent. In other words, a doubling of the odds was not the most likely scenario, even though that was the impression delivered by multiple articles. (Though, to be fair, the authors suspect that their models may underestimate the effects of manmade warming on the intensification of rainfall.)

On the other hand, the news coverage did, on balance, convey many of the caveats related to the Nature studies. The AP article cited above included a comment from climate scientist Jerry North, who expressed reservations about some of the data the researchers used (conversely, it was the only article to dig deeper into the “fingerprinting” technique used in the climate models, relying on NOAA’s Climate Indicators website to point out that it has been used to link greenhouse gases to more than a dozen other ecological changes). Richard Black reported for the BBC that “Both research groups were at pains to emphasize that these two papers are not the end of the road.”

Indeed, articles from The News York Times, The Washington Post, and NPR included sentences, respectively, stressing that models “cannot fully capture the complexity of the real world,” that “there is uncertainty in this work,” and that “not all extreme weather events can be blamed on climate change.” (Ironically, Nature News was the only one to employ the verb “causing” in relation to climate change and weather.)

Many articles (such as a good one from Scientific American) also did excellent work explaining why these types of modeling experiments are so difficult, and why results pertaining to particular phenomena cannot be used to extrapolate conclusions about other weather events, or weather in general. According to the Times, for instance:

That it took a decade to come to that conclusion [about the flood in the U.K. 2000] illustrates one of the major problems of climate science at the moment. Researchers are barraged with questions about weather extremes like the recent winters in Europe and the United States and the heat waves and droughts of last summer.

Yet, even when adequate weather statistics are available for an affected region, the scientists need years to run computer analyses of any specific event and calculate whether it was made more or less likely by global warming.

As many articles noted, one of climate modelers’ central goals is to improve their research methodologies to the point where they can analyze severe weather in real time. But that ability is still a ways off, which bring us back to the Seattle PostGlobe’s question about whether or not the Times was remiss to neglect a mention of climate change in its article about the severe drought currently afflicting China. The fact is, reporters simply cannot make such connections without evidence, and (as far as I can tell) nobody has performed a truly robust analysis of the Chinese drought.

Peer-reviewed analyses are not the only kind of evidence available to journalists, however. If a qualified scientist tells a reporter that he or she feels there is connection between a weather event and climate change, the reporter is perfectly free to quote the scientist saying that. For example, Xu Yinlong, a researcher at the Chinese Academy of Agricultural Sciences, told the Los Angeles Times, “This drought is occurring in front of the big backdrop of global warming and is part of the phenomenon of extreme weather events. The direct cause is months of lack of rainfall, but it definitely is connected with climate change.”

Such opinions are perfectly valid. Indeed, following last summer’s devastating heat wave in Russian and flood in Pakistan, CJR observed that news outlets avoided the temptation to pin the events on climate change, limiting themselves to the statement that scientists merely expect their frequency and intensity to increase in a warmer world. In a round of day-two stories, however, some journalists dug deeper, finding a number of scientists who were willing to go out on a limb.

“If you ask me as a person, do I think the Russian heat wave has to do with climate change, the answer is yes,” NASA’s Schmidt told The New York Times. “If you ask me as a scientist whether I have proved it, the answer is no — at least not yet.”

Such reporting should be commended, as long as it puts these quotes in their proper context, laying out the nuance, complexity, and uncertainty that surround them. In a roundup of coverage of the two Nature studies, the Knight Science Journalism Tracker’s Charlie Petit observed that “An official shift may just have occurred not only in news coverage of climate change, but the way that careful scientists talk about it.” He’s probably right, and while that shift doesn’t mean that reporters have carte blanche to link global warming to extreme weather without any supporting evidence, it does mean, as Freedman put it in his post for the Capital Weather Gang, that:

At the same time, ignoring the growing evidence that certain types of extreme events are already shifting, or claiming that a single study proves there is no climate change link with extreme weather – as a Wall Street Journal op-ed writer did last week – is also flat out wrong, and not just because of the two studies published last week (don’t just take my word for it, the authors of the study published a letter to the editor late yesterday correcting some of the writer’s misconception).

Indeed, there is an increasingly large body of research demonstrating that climate change is not, as the author of one of last week’s studies characterized it, a “victimless crime.” But as reporters plumb the depths of weather-climate connections, they should repeat this mantra: evidence, nuance, complexity, uncertainty; evidence, nuance, complexity, uncertainty.

Curtis Brainard writes on science and environment reporting. Follow him on Twitter @cbrainard.