Since the Intergovernmental Panel on Climate Change released its report summarizing six years of global warming science Friday, the mammoth document has been picked and prodded by just about every environmental journalist. Even before the report went public, much debate focused on how strongly the panel would weigh in on people’s role in climate change: specifically, if the IPCC would authoritatively attribute global warming to humans.
The panel did, with a little caveat: Humans are escalating global warming, the committee ruled—or at least, it’s “95 percent” certain they are. “It is extremely likely [95 percent confidence] more than half of the observed increase in global average surface temperature from 1951 to 2010 was caused by the anthropogenic increase in greenhouse gas concentrations and other anthropogenic forcings together,” read the report.
It’s not that scientists were uncertain prior to Friday that human activity is contributing to climate change—what the panel is expressing is a normal principle of probability called a confidence interval, which defines the likelihood that a set of calculations falls into expected parameters. In fact, 95 percent is pretty high; in the last IPCC report, released in 2007, scientists expressed the same statistic with only 90 percent certainty.
It’s almost impossible to express absolute certainty in any scientific finding—especially one that takes place on the massive and messy scale of planetary data, thus forgoing the neat, easily replicable qualities of a laboratory study. Still, it’s confusing to civilians accustomed to the colloquial use of “certain.” After all, if a witness to a hit and run is only 95-percent certain about the identity of the driver, there’s ample wiggle-room for the prosecution in that 5-percent uncertainty.
Maybe that’s why there’s been a stream of explainers on the IPCC’s idea of certainty—notably “What 95 percent certainty means to scientists,” from the Associated Press and Discovery News. But a growing body of research suggests that, even when the definition of scientific certainty is explained well, focusing on this linguistic gap confuses readers into thinking scientists haven’t reached a firm conclusion, when they most certainly have.
Still, according to “Climate Change and the Media,” a report released in late September by the Reuters Institute for the Study of Journalism at the University of Oxford, “certainty” still dominates coverage of climate change. The study group surveyed 350 articles in six countries for language expressing a paradigm and found over 80 percent of the pieces unpacked some idea of scientific certainty. (They also found that expressions of skepticism varied by nationality, with Australia presenting the highest number of inclusions from global warming skeptics, followed by the United States.)
The study’s lead author, James Painter, suggests a better way of framing climate change: risk. Using a risk paradigm won’t solve all the tricky negotiations between science-speak and human speak, but as Painter writes, it offers a “more helpful prism through which to analyse the challenge.” It’s a simple framing trick: Focusing on the positive of what we know (“Scientists say there is a 95-percent risk that humans are causing climate change”) versus the negative of our uncertainty (“Scientists are 95-percent certain that humans are causing climate change.”) It’s also a term that aligns the colloquial definition with the scientific definition of the word: 95 percent is a whole lot of risk.
As Painter points out in the full report, as climate models become increasingly more detailed in their reports of the future climate, it will become more necessary for journalists to be able to detail the findings by mastering the language of probability. It also allows for easy analogies, like this handy calculator, which compares the probability of predicted rises in global temperature to more concrete risks, such as a terrorist attack. 95 percent certainty that I will be struck by lightning? That’s a risk I’m not willing to take.
More in The ObservatoryRead More »