From the University of Wisconsin:
Greenland ice may exaggerate magnitude of 13,000-year-old deep freeze
Ice samples pulled from nearly a mile below the surface of Greenland glaciers have long served as a historical thermometer, adding temperature data to studies of the local conditions up to the Northern Hemisphere’s climate.
But the method — comparing the ratio of oxygen isotopes buried as snow fell over millennia — may not be such a straightforward indicator of air temperature.
“We don’t believe the ice cores can be interpreted purely as a signal of temperature,” says Anders Carlson, a University of Wisconsin–Madison geosciences professor. “You have to consider where the precipitation that formed the ice came from.”
According to a study published today by the Proceedings of the National Academy of Sciences, the Greenland ice core drifts notably from other records of Northern Hemisphere temperatures during the Younger Dryas, a period beginning nearly 13,000 years ago of cooling so abrupt it’s believed to be unmatched since.
Such periods of speedy cooling and warming are of special interest to climate scientists, who are teasing out the mechanisms of high-speed change to better understand and predict the changes occurring in our own time.
In the case of the Younger Dryas, average temperatures — based on the Greenland ice — plummeted as much as 15 degrees Celsius in a few centuries, and then shot back up nearly as much (over just decades) about 1,000 years later.
“In terms of temperature during the Younger Dryas, the only thing that looks like Greenland ice cores are Greenland ice cores,” Carlson says. “They are supposed to be iconic for the Northern Hemisphere, but we have four other records that do not agree with the Greenland ice cores for that time. That abrupt cooling is there, just not to the same degree.”
Working with UW–Madison climatologist Zhengyu Liu, collaborators at the National Center for Atmospheric Research and others, Carlson found their computer climate model breaking down on the Younger Dryas.
While it could reliably recreate temperatures in the Oldest Dryas — a similar cooling period about 18,000 years ago — they just couldn’t find a lever in the model that would simulate a Younger Dryas that matched the Greenland ice cores.
“You can totally turn off ocean circulation, have Arctic sea ice advance all the way across the North Atlantic, and you still will have a warmer climate during the Younger Dryas than the Oldest Dryas because of the carbon dioxide,” Carlson says.
By the time the Younger Dryas rolled around, there was more carbon dioxide in the air — about 50 parts per million more. The warming effects of that much CO2 overwhelmed the rest of the conditions that make the Oldest and Younger Dryas so alike, and demonstrates a heightened sensitivity for Arctic temperatures to rising greenhouse gases in the atmosphere.
The researchers zeroed in on the Northern Hemisphere’s temperature outlier, Greenland ice cores, and found that the conversion of oxygen isotope ratio to temperature typically used on the ice cores did not account for the sort of crash climate change occurring during the Younger Dryas. It assumes prevailing winds and jet streams and storm tracks are providing the moisture for Greenland precipitation from the Atlantic Ocean.
“The Laurentide ice sheet, which covered much of North America down into the northern United States, is getting smaller as the Younger Dryas approaches,” Carlson says. “That’s like taking out a mountain of ice three kilometers high. As that melts, it allows more Pacific Ocean moisture to cross the continent and hit the Greenland ice sheet.”
The two oceans have distinctly different ratios of oxygen isotopes, allowing for a different isotope ratio where the water falls as snow.
“We ran an oxygen isotope-enabled atmosphere model, so we could simulate what these ice cores are actually recording, and it can match the actual oxygen isotopes in the ice core even though the temperature doesn’t cool as much,” Carlson says. “That, to us, means the source of precipitation has changed in Greenland across the last deglatiation. And therefore that the strict interpretation of this iconic record as purely temperature of snowfall above this ice sheet is wrong.”
By the study’s findings, Greenland temperatures may not have cooled as significantly as climate headed into the Younger Dryas relative to the Oldest Dryas, because of the rise in atmospheric carbon dioxide that had occurred since the Oldest Dryas.
“You can say at the end of the Younger Dryas it warmed 10, plus or minus five, degrees Celsius. But what happened on this pathway into the event, you can’t see,” Carlson says.
It’s a fresh reminder from an ancient ice core that climate science is full of nuance, according to Carlson.
“Abrupt climate changes have happened, but they come with complex shifts in the way climate inputs like moisture moved around,” he says. “You can’t take one difference and interpret it solely as changes in temperature, and that’s what we’re seeing here in the Greenland ice cores.”
The National Science Foundation and Department of Energy funded the research.
Note: as of this writing, the article is not yet online at PNAS – I’ll update this post when it becomes available – Anthony
Commenter and Contributor Bill Illis submitted this comment to Tips and Notes with the request that it be moved here. Since we cannot move comments from one thread to another and since the comment is quite relevant to this thread, I decided to post this as an update. (Bill if you had trouble posting the comment, please let us know what the trouble was.) -REP
(A long post meant to explain the problem with the Greenland ice core temperature reconstructions thread posted earlier today. If this could be moved there, I consider it an important problem which is not noted that often).
The problem with the Greenland temperature reconstructions is that, starting about 20 years ago, they began calibrating the isotope data with “borehole temperature” reconstructions rather than the typical measured formula.
This has overstated the temperature change by more than twice what the normal isotope formula for Greenland’s circumstances would have indicated and set-back the science 20 years.
The Borehole temperature profile for the Greenland ice cores at this link (just slightly different in each location). The ice sheet is melting at the bottom since it is -2.4C which is just enough to melt the ice at the pressure levels at the bottom of the ice-sheet (the new recently completed Neem ice core, however, is at -3.0C and appears to be unmelted).
So they use a model about how temperatures at the surface and temperatures at bedrock migrate through the 3 km deep ice-sheet and they can guesstimate what the surface temperatures were in the past. They then recalibrate the dO18 isotopes to those guesstimates rather than the known conversion formula. The problem is that the borehole models appear to be wrong by a factor of 2.
If one used a more typical formula for the isotopes in Greenland’s ice cores, we get temperatures that are more consistent with what we know about the North Atlantic, Northern Hemisphere, Antarctica and Europe (during the ice ages and during the Younger Dryas). In addition, now that the Greenland ice cores go all the way back to the Eemian interglacial, 123,000 years ago, we can say the borehole temperature conversion formula is wrong because it has temperatures in Greenland at +9.0C for the Eemian. This would have melted out most of the glaciers given a few thousand years and this did not happen. The typical isotope formula provides a number of +4.0C which is more consistent with Antarctica, sea level and global conditions during the Eemian.
The Younger Dryas temperatures only declined by about half what is typically quoted.
- Younger Dryas -The Rest of the Story! (wattsupwiththat.com)
- New evidence of Younger Dryas extraterrestrial impact (wattsupwiththat.com)