New research suggesting that the planet might already be committed to vastly greater warming than previously thought is being dismissed as deeply flawed by prominent climate scientists.
A study published today in one of the world's top science journals, Nature, offers the most complete reconstruction to date of global sea-surface temperatures for the past two million years—a valuable addition to the climate record, scientists say.
But the conclusions the study's author drew from that research—that even preventing any further increase of greenhouse gases in the atmosphere could still leave the Earth doomed to a catastrophic temperature rise of up to 7 degrees Celsius (about 13 degrees Fahrenheit)—isn't supported by the data, several top scientists said.
"This is simply wrong," said Gavin Schmidt, chief of NASA's Goddard Institute for Space Studies.
Jeffrey Severinghaus, a paleoclimatologist at Scripps Institution for Oceanography in San Diego, was equally vehement, arguing that the study's result isn't logical: "It's based on a fundamental mistake," he said. "The problematic conclusion doesn't flow from the main meat of the paper."
Two other scientists reached by National Geographic shared that blunt assessment.
The study's author, a former postdoctoral researcher at Stanford University, said she wasn't attempting to offer a detailed climate forecast. But she examined the past's tight link between sea-surface temperature changes and natural releases of carbon dioxide and tried to show what that might imply for the future. Her result: an alarming 3 to 7 degree Celsius temperature rise by several thousand years from now, even if fossil fuel emissions were capped today.
"We do find this close relationship between temperature and greenhouse gases that is remarkably stable, and what the study is developing is the coupling factor between the two," said Carolyn Snyder, who now works on climate issues for the Environmental Protection Agency.
Other scientists said, however, that Snyder has drastically overstated what that relationship tells us about today's situation because of the way she extrapolates from past ice ages to the very different climate of the modern era.
"The number she gets is huge," said Eric Steig, an earth sciences professor at the University of Washington. "She's making a statement about the future, but I don't find anything in the paper that explains why she thinks she can do that."
In reality, NASA's Schmidt said, if greenhouse gas levels in the atmosphere stay the same as today, temperatures may rise an additional 0.5 degrees Celsius to 1 degree Celsius. Other studies suggest they could hit 2 degrees. But, Schmidt said, no credible evidence suggests they could rise 7 degrees—"and nothing in the study changes that."
Bad Deduction From a Good Record
For decades scientists have used ocean sediment cores to reconstruct past sea-surface temperatures at particular times and locations. Snyder cataloged and organized 20,000 such reconstructions from 59 sediment cores. Then she went through the painstaking process of weighting them to build a time line of global average surface temperatures.
Previous research had only organized these temperature records into short snippets of history. Snyder’s critics applauded her for creating the first temperature time line that spans the globe and extends back two million years.
"That's great," said Jeremy Shakun, a geologist and climate scientist at Boston College. "It seems like an absolute must-do kind of study. It had to be done."
During the past two million years, Earth has cycled in and out of prolonged ice ages; the most recent one ended 11,000 years ago. Snyder’s new time line reveals that temperatures gradually cooled to a nadir around 1.2 million years ago. She also found that the phenomenon known as polar amplification—the tendency of temperatures at the poles to fluctuate more strongly than the global average—has held steady for the past 800,000 years.
Today the whole planet is warming, but the Arctic is warming much faster than average. Scientists have no doubt that the current warming is caused primarily by human emissions of carbon dioxide and other greenhouse gases.
It has also been known for a long time, from analysis of ancient air bubbles trapped in Antarctic and Greenland ice cores, that atmospheric levels of CO2 fluctuated naturally during the ice ages—and that they were tightly correlated with local air temperature above the ice. Snyder’s time line confirms that correlation between CO2 and temperature at the global scale and over two million years.
Her mistake, the critics said, was in the simplistic way she used that correlation to infer the Earth’s sensitivity to CO2—not only in the past but also in the future. Her conclusions simply didn't make sense, they said. The reason is that much of the variation in temperature during the ice ages was caused by factors other than CO2.
Over the past 800,000 years, massive ice sheets have expanded across the northern continents and then retreated again every 100,00 years or so. The cycle is driven by variations in Earth’s orbit about the sun, which determine whether enough sunlight falls on the Northern Hemisphere in summer to melt ice.
The presence of ice itself has a dramatic cooling effect on the planet, because it reflects sunlight back to space. Conversely, when ice melts at the end of a glacial period, the increased absorption of sunlight warms the whole planet—triggering feedback effects, including a rise in CO2, that warm it further.
"Fundamentally, the orbit of the Earth causes ice ages to end, and the ocean warms in response," Severinghaus said. "This warming causes some CO2 to outgas from the ocean, raising atmospheric CO2 concentrations."
Whereas today the rise in CO2 is the prime driver of global warming, during the ice-age period it was merely an important feedback. Snyder insisted she wasn't trying to suggest that greenhouse gases caused all the temperature changes back then. But the paper's critics said her work had in effect done just that, and as a result produced what seems an exaggerated estimate of the current sensitivity to CO2.
An Open Question
The question of just how sensitive Earth’s climate will be over the long term to our CO2 emissions remains open. There is evidence that over millennia, once all the feedbacks have had a chance to kick in, the warming could be indeed be substantially larger than in the short term.
But Snyder’s critics agree that her method is not the way to resolve the question. Over the weekend Schmidt wrote a blog post to explain why the relationship between CO2 and temperature during the ice ages isn't enough by itself to describe our potential future.
"There's a lot of good work in here," he said of Snyder's paper. "But one small error has been made that has been translated into something with a huge implication for the future. And unfortunately, that's just wrong."
Or perhaps he should have said “fortunately.” If Snyder’s analysis were correct, our descendants might already be doomed to a catastrophically warmer world, with sea levels hundreds of feet higher than they are today. As it stands, that outcome appears preventable.
Follow Craig Welch on Twitter.
Editor's Note: An earlier version of this story said temperatures would rise between .5 and 1 degree Celsius "if humans halt greenhouse gas emissions today." It has been corrected to say "if greenhouse gas levels in the atmosphere stay the same as today."