Climate Consensus: Where the 97 Percent Comes From

Almost everyone has heard that 97 percent of climate scientists agree humans are warming the planet. Fewer people know where the number comes from, and that gap leaves room for doubt: who was asked, what exactly did they agree with, and who decided what counts?

The figure is not a single poll. It is the result of several independent studies that measured agreement in different ways and kept arriving at similar answers. Looking at how each one worked is the best way to judge how solid the consensus is.

Counting papers: Oreskes, 2004

The first widely cited study came from the historian of science Naomi Oreskes, published in Science in 2004. She read 928 abstracts of papers on global climate change published between 1993 and 2003. About three quarters explicitly or implicitly accepted that recent warming was largely caused by humans, the rest took no position, and none of the abstracts explicitly rejected the consensus view.

The method had obvious limits. An abstract is a short summary, and many papers on narrow technical questions never state a view on the cause of warming. But the absence of a single explicit rejection was itself a finding.

Asking scientists: Doran and Zimmerman, 2009

In 2009 Peter Doran and Maggie Zimmerman took the opposite approach and asked people directly. They sent a short survey to 10,257 earth scientists and received 3,146 answers. Across everyone who replied, 82 percent said human activity was a significant factor in changing global temperatures.

The figure rose with expertise. Among the 79 respondents who were climatologists publishing mostly on climate, 76 agreed temperatures had risen and 75 named human activity as a significant factor. That subgroup is where an early version of the 97 percent comes from, and critics fairly pointed out that it is a small group.

Weighting expertise: Anderegg, 2010

A 2010 study in the Proceedings of the National Academy of Sciences, led by William Anderegg, looked at 908 researchers who had each published at least 20 climate papers. Between 97 and 98 percent of them supported the main conclusions on human-caused climate change set out by the Intergovernmental Panel on Climate Change. The pattern was the same as in the survey: the more a researcher had published on climate, the stronger the agreement.

The big one: Cook and colleagues, 2013

The best-known number comes from a 2013 study led by John Cook, which classified 11,944 abstracts published between 1991 and 2011. Two thirds, 66.4 percent, took no position on the cause of warming, which is normal for specialised research. Of the papers that did state a position, 97.1 percent endorsed human-caused warming.

This is the study critics target most. The economist Richard Tol argued that the classification lacked proper controls and checks on how consistently the raters scored abstracts. The authors defended their method, and later studies using different methods have produced similar or higher figures, which is the strongest answer to a methodological doubt.

Later checks: above 99 percent

Two 2021 studies brought the count up to date. Mark Lynas and colleagues took a random sample of 3,000 papers from about 80,000 climate studies published between 2012 and 2020 and found that 99.85 percent were not sceptical of the human cause. A survey by Krista Myers and colleagues of 2,780 earth scientists found that 98.7 percent of climate scientists agreed humans are causing climate change; agreement was lowest, at 84 percent, among economic geologists, a field closely tied to the mining and fossil fuel industries.

Reviews of the small number of papers that reject human-caused warming have also found that many contain errors or cannot be replicated, which is part of why they have not changed the overall picture.

What the scientific bodies say

The studies are backed by formal statements from scientific organisations. The American Association for the Advancement of Science said in 2014 that about 97 percent of climate scientists have concluded that human-caused climate change is happening. The American Geophysical Union and the American Meteorological Society both say it is extremely likely that human activity has been the dominant cause of warming since the mid-twentieth century, and the American Physical Society, the Geological Society of America and the US National Academy of Sciences have issued similar statements.

The Intergovernmental Panel on Climate Change goes furthest. Its 2021 assessment described human-caused warming of the atmosphere, ocean and land as unequivocal. The 2010s were about 1.1 degrees Celsius warmer than the late nineteenth century, and sea levels have risen 20 to 25 centimetres since 1900, half of that since 1980.

Consensus is not the same as proof

A consensus is not a vote that makes something true. Its value lies in what it summarises: many independent lines of evidence, examined by people who know the field, pointing the same way. Our explainer on what scientific consensus means looks at how such agreements form and at the times they have been wrong.

The useful questions are the ones these studies invite: how agreement was measured, by whom and among which experts. Asked that way, the climate consensus holds up well, because different methods, different samples and different decades keep giving the same answer. For a general method you can apply to any claim like this, see how to evaluate a claim.