Correction Issued for Alarming Ocean Temperature Paper

Published January 2, 2019

Researchers with the University of California at San Diego’s Scripps Institution of Oceanography and Princeton University issued a correction to their widely publicized study which had raised fears of a rapid increase in the Earth’s temperature.

The study published in Nature on October 31 claimed ocean temperatures have risen roughly 60 percent more than the United Nations’ Intergovernmental Panel on Climate Change (IPCC) has estimated.

Independent climate researcher Nicholas Lewis discovered significant methodological flaws in the paper. The authors admitted their error after Lewis and others publicized his findings.

Real Data vs. Proxy Estimates

Ocean temperature data comes largely from the Argo array of robotic devices that record actual temperatures at different depths in the world’s oceans. Argo was started in 2000.

Rejecting the Argo data as covering too short a time, Ralph Keeling, a climate scientist at the Scripps Institution of Oceanography, and Laure Resplandy of Princeton calculated temperatures based on the amount of oxygen and carbon dioxide rising off the ocean, by filling round glass flasks with air collected at research stations around the globe.

A recent IPCC report estimated net greenhouse gas emissions must be reduced 45 percent below 2010 levels by 2030 and zeroed out by 2050 to keep warming from exceeding 2 degrees Celsius above preindustrial levels, which it claims would result in dangerous climate changes. Based on their proxy temperature calculations, Keeling and Resplandy estimated emissions cuts would have to be 25 percent steeper than IPCC estimated in the coming decades to avoid exceeding the two-degree cap.

Large Math Errors

Lewis found the authors made mathematical errors overstated the amount of ocean warming and understated the uncertainty of their projections.

“[They used a] novel method to estimate heat uptake by the ocean over the period 1991 to 2016 and came up with an atypically high value,” Lewis wrote in his analysis.

Keeling and Resplandy quickly acknowledged making the errors Lewis identified.

“When we were confronted with his insight it became immediately clear there was an issue there,” Keeling told various news outlets.

After correcting their mistake, Keeling and Resplandy found their margin of error was between 10 and 70 percent, making their ocean temperature estimates virtually worthless.

“Our error margins are too big now to really weigh in on the precise amount of warming that’s going on in the ocean,” Keeling told the media. “We really muffed the error margins.”

‘They Got the Headlines’

The large discrepancy between the IPCC’s ocean temperature estimates and those posited by Keeling and Resplandy should have called the study’s claims into question with Nature’s peer reviewers well before publication, says climatologist Tim Ball, Ph.D., chairman of the Natural Resources Stewardship Project and a policy advisor to The Heartland Institute, which publishes Environment & Climate News.

“The 60 percent difference from all previous measures should scream problems to the authors and the reviewers,” said Ball. “The case parallels another finding well outside the normal in climate studies, namely the data for the hockey stick.”

The hockey stick was a graphic representation of a reconstruction of the earth’s temperature record over the past 1,000 years which incorrectly indicated neither the widely recognized Medieval Warm Period nor the Little Ice Age occurred.

“Why didn’t they get an independent statistician to examine the work, as the Wegman Report [requested by the U.S. House of Representatives’ Committee on Energy and Commerce following the hockey stick controversy] recommended?” Ball asked. “The reality is, in both cases they got the result they wanted, they got the headlines they sought, which is all the media and the public remember, so they didn’t question the findings.”

Sticky Errors

This paper would almost certainly not have been published had it said IPCC significantly overestimated global ocean temperatures, says David Legates, Ph.D., a professor of climatology at the University of Delaware and a policy advisor to The Heartland Institute.

“I am confident a paper that calculated the ocean warming was 60 percent less than the IPCC and other research had estimated would not even have been published,” said Legates. “The fact they had a paper arguing the IPCC was substantially underestimating ocean temperature—in other words, the IPCC’s projections were ‘conservative,’ if you will, and the real figure was really 60 percent higher—was the impetus behind the paper being published.

“Now, as it turns out, the results were flawed and the figures are only ‘in line’ with IPCC’s estimates, but the original stories are out there and no doubt journalists will cite the ‘first-run’ articles without the caveat the paper was flawed,” Legates said.

Calls for Greater Scrutiny

Extraordinary claims of climate disaster do not get the scrutiny they deserve before being published or covered by the media, Legates says.

“There was a time when extreme claims required more scrutiny,” said Legates. “To say the values heretofore calculated were 60 percent too low demands a response as to why the old values are to be discarded in favor of the new ones.

“This event underscores the issue that claims of an increased impact of anthropogenic climate change are more readily believed, particularly if they are ‘worse than previously thought,’ and the media are quick to hype the ‘latest and greatest’ finding when it fits the disaster narrative,” Legates said. “Correcting themselves and pointing out the mundane fact the findings were flawed is all too often shunned or not covered because ‘it will only be used by skeptics to discredit the science,’ but when science discredits itself, that must be part of the narrative as well.”

H. Sterling Burnett, Ph.D. ([email protected]) is a senior fellow at The Heartland Institute.