CO2 and the End of the Last Ice Age: Lag Redux

by Bill Chameides | April 12th, 2012
posted by Erica Rowell (Editor)

Permalink | Comments Off on CO2 and the End of the Last Ice Age: Lag Redux

That infamous lag between temperature change and carbon dioxide (CO2) questioned in new paper.

Way back when, in the days when Al Gore’s An Inconvenient Truth was a box office hit, the climate denier community was all over an apparent Achilles heel in the global warming story: the lag.

CO2 is a greenhouse gas, no question. We know this from so many lines of evidence — from laboratory measurements to satellite observations of radiation leaving the Earth’s atmosphere — that there is no legitimate grounds for argument. What is less certain is how much global temperatures change as a result of specific changes in CO2 — the so-called climate sensitivity. (See here, here, here and here.)

The Ice Age Record and CO2

Climate scientists have used a variety of methods to get at the climate sensitivity, and together those methods point to a warming of 5.4 degrees Fahrenheit (3 degrees Celsius) for a doubling of CO2 concentrations with a probable range of 3.6 to 8.1 degrees Fahrenheit (2 to 4.5 degrees Celsius).

One of the methods used to assess the current climate sensitivity has been looking at paleoclimatic variations: correlating global temperature and CO2 changes in the Earth’s past. Among the most compelling paleoclimatic evidence of a climate-CO2 link comes from information gleaned about the climate during the epoch known as the Pleistocene, running from about 2.5 million to 12,000 years ago.

During the most recent half of the Pleistocene, the climate was dominated by lengthy periods of cold conditions (lasting roughly 100,000 years) known as glacials, followed by shorter warm periods (lasting in the tens of thousands of years) known as interglacials. The end of the Pleistocene came with the end of the last glacial period, which marked the beginning of the current epoch, the Holocene.

Probably our most extensive, detailed record of this period has come from data gathered from bubbles trapped in ice cores over the past 800,000 in the Antarctic. Importantly, with regard to the climate sensitivity question, these data provide a contemporaneous record of both temperature and atmospheric CO2.

The striking aspect of the dataset is the close correspondence of the temperature changes with CO2 concentrations.

Analysis of the composition of bubbles trapped in ice cores from Antarctic show a close correlation between temperatures (indicated by the variables δD and benthic δ18O) and greenhouse gas concentrations. But a careful inspection of this data indicates that changes in CO2 lagged behind the temperature changes. The work of Shakun et al suggest that this lag is a local phenomenon and not reflective of the global trend. (Figure 6.3, IPCC Fourth Assessment Report: Climate Change 2007: Working Group I: The Physical Science Basis)

When temperatures were cold, CO2 concentrations were low (generally about 180 parts per million); when temperatures were warm, CO2 was high (around 300 parts per million). (Note: current CO2 concentrations have reached 394 parts per million, far larger than the atmosphere has seen for some 2 million years.)

If there was any doubt that CO2 plays a controlling factor in the climate, this evidence would surely seem to dispel it, right?

Not So Fast, Said the Skeptics.

Upon closer inspection, it turns out that that there is a lag between the record’s temperature changes and CO2 concentrations. The increase in CO2 generally lags behind both the temperature rise during deglaciations (the transition periods between glacials and interglacials) and the temperature drop that occurs at the onset of a glacial. Aha!, claimed the skeptics, CO2 is not driving the climate — the climate is driving CO2.

There is a strong and compelling answer to this criticism. While CO2 isn’t the catalyst that forces temperature trends to reverse direction, it does amplify the trend. We’ve known for some time that what initially drives the climate to shift between glacials and interglacials and back again is astronomical. Slight variations in the Earth’s position and orbit around the Sun lead to subtle changes in the amount and distribution of sunlight absorbed by the planet. And yet, these changes in sunlight that cause small temperature changes cannot by themselves have led to the large temperature fluctuations that are seen in the ice core record. The explanation: The small temperature changes set off a chain of feedbacks that drove the climate even further. One of those feedbacks, the ice core data suggest, involved CO2, and climate scientists argue that it was the subsequent changes in CO2 that were responsible for most of the actual temperature changes observed in the record.

It’s a somewhat complicated but nevertheless compelling explanation for what we see in the ice core data and it clearly places CO2 at the center of the climate swings over the period of data record. But questions have still persisted about the length of the lag and how data gathered at specific ice core sites (mainly those in Antarctica) relate to climate conditions around the globe.

A Detailed Look at the Last Deglaciation: What Lag?

To address these issues, Jeremy Shakun of Harvard University and co-authors carried out a detailed analysis of temperature and CO2 records during the last deglaciation. Their results, which were published in last week’s issue of the journal Nature, are quite startling.

Before we get into the paper’s findings, a bit of review. The last glacial began about 116,000 years ago and reached its maximum extent around the globe between 27,000 and 19,000 years ago — a period known as the Last Glacial Maximum. Temperatures then began a warming trend (with some small excursions, such as this one) that brought us to our present warm period called the Holocene, a period that began about 12,000 years ago. And so the transition from the last glacial to the current interglacial lasted from about 19,000 to 12,000 years ago. It is this period of deglaciation that Shakun et al focused on.

However, and this is important, Shakun et al didn’t just consider data from ice cores; they included other paleotemperature proxy records distributed across the globe. And it was the use of this larger global dataset (80 sites in all) that led to an important insight.

A key finding: The temperature rise during deglaciation was not uniform throughout globe. Around 19,000 years ago, temperatures began to increase in the Southern Hemisphere but not in the Northern Hemisphere. It took another 2,000 years for Northern Hemisphere temperatures to begin rising.

Shakun et al argue that the so-called lag in CO2 behind temperature is not real but an artifact resulting from extrapolating local temperatures obtained from Antarctic ice cores to the globe. Looking at just Antarctica data gives the appearance of CO2 lagging behind temperatures. But, examine a larger global temperature record, and the lag essentially disappears.* The rapid rise in global temperatures according to Shakun et al’s analysis began about 17,000 years ago when global atmospheric CO2 concentrations began their increase.

Further analysis of these data suggests that what was behind the Northern and Southern Hemisphere’s asynchronicity of temperatures were oscillations in the so-called Atlantic Meridional Overturning Current (often shortened to AMOC and popularly known as part of the conveyor belt circulation), which transports heat from the Southern to the Northern Hemisphere via the Atlantic Ocean. Shakun et al propose that changes in the AMOC initially caused a small amount of warming in the Southern Hemisphere, which led to a release of CO2 from the oceans, which in turn drove the larger global temperature increase that began about 17,000 years ago. Which begs the question of what triggered the variations in the AMOC. A likely candidate: orbital changes we have long believed to be the first mover in the glacial-interglacial cycle.

Does this study mark the end of the lag debate? We’ll have to wait and see — it will be interesting to find out what similar analyses uncover for earlier deglaciations and glaciations. But lag or no lag, it is hard to place CO2 anywhere but at the center of the Earth’s climate system. A sobering thought as atmospheric CO2 levels approach 400 parts per million.


End Note

* The exception: A brief period at the onset of the deglaciation (between ~19,000 and 17,500 years ago) saw a very modest initial warming of about 0.5 degrees Fahrenheit (0.3 degrees Celsius). This initial warming precedes the rise in CO2 indicating that a different mechanism, most likely astronomical, was responsible for the onset of warming.

filed under: Antarctica, carbon dioxide emissions, climate change, faculty, global warming, temperatures
and: , , , , , , , , , , , , , , , ,

comments disabled after 30 Days

©2015 Nicholas School of the Environment at Duke University | Box 90328 | Durham, NC 27708
how to contact us > | login to the site > | site disclaimers >

footer nav stuff