Authors of the survey, published on Thursday in the journal Environmental Research Letters, said the finding of near unanimity provided a powerful rebuttal to climate contrarians who insist the science of climate change remains unsettled.
The survey considered the work of some 29,000 scientists published in 11,994 academic papers. Of the 4,000-plus papers that took a position on the causes of climate change only 0.7% or 83 of those thousands of academic articles, disputed the scientific consensus that climate change is the result of human activity, with the view of the remaining 2.2% unclear.
The study described the dissent as a “vanishingly small proportion” of published research.
“Our findings prove that there is a strong scientific agreement about the cause of climate change, despite public perceptions to the contrary,” said John Cook of the University of Queensland, who led the survey.
Public opinion continues to lag behind the science. Though a majority of Americans accept the climate is changing, just 42% believed human activity was the main driver, in a poll conducted by the Pew Research Centre last October.
“There is a gaping chasm between the actual consensus and the public perception,” Cook said in a statement.
The study blamed strenuous lobbying efforts by industry to undermine the science behind climate change for the gap in perception. The resulting confusion has blocked efforts to act on climate change.
The survey was the most ambitious effort to date to demonstrate the broad agreement on the causes of climate change, covering 20 years of academic publications from 1991-2011.
In 2004, Naomi Oreskes, an historian at the University of California, San Diego,surveyed published literature, releasing her results in the journal Science. She too came up with a similar finding that 97% of climate scientists agreed on the causes of climate change.
She wrote of the new survey in an email: “It is a nice, independent confirmation, using a somewhat different methodology than I used, that comes to the same result. It also refutes the claim, sometimes made by contrarians, that the consensus has broken down, much less ‘shattered’.”
The Cook survey was broader in its scope, deploying volunteers from the SkepticalScience.com website to review scientific abstracts. The volunteers also asked authors to rate their own views on the causes of climate change, in another departure from Oreskes’s methods.
The authors said the findings could help close the gap between scientific opinion and the public on the causes of climate change, or anthropogenic global warming, and so create favourable conditions for political action on climate.
“The public perception of a scientific consensus on AGW [anthropogenic, ie man-made, global warming] is a necessary element in public support for climate policy,” the study said.
However, Prof Robert Brulle, a sociologist at Drexel University who studies the forces underlying attitudes towards climate change, disputed the idea that educating the public about the broad scientific agreement on the causes of climate change would have an effect on public opinion – or on the political conditions for climate action.
He said he was doubtful that convincing the public of a scientific consensus on climate change would help advance the prospects for political action. Having elite leaders call for climate action would be far more powerful, he said.
“I don’t think people really want to come around to grips with the fact that climate change is a highly ideological issue and it is not amenable to the information deficit model,” he said.
“The information deficit model, this idea that if you just pile on more information people will get convinced, is just completely inadequate, he said. “It strengthens the people who actually read and pay attention but it is certainly not going to change or shift the opinions of others.”
Jon Krosnick, professor in humanities and social sciences at Stanford university and an expert on public opinion on climate change, said: “I assume that sceptics would say that there is bias in the editorial process so that the papers ultimately published are not an accurate reflection of the opinions of scientists.”