IT’S IN THERE The concentrations of carbon dioxide and methane found in bubbles of ancient air (dark spots) trapped in Antarctic ice provide clues to ancient climate. AWI Bremerhaven, University of Bern
A kilometers-long ice core from Antarctica
has recorded climate information for the past 800,000 years and has revealed a three
millennia–long period when carbon dioxide levels in the air were lower than any
previously measured.
The longest detailed records of atmospheric gases previously
reported, from the uppermost sections of a 3.2 kilometer–long ice core drilled
in eastern Antarctica, go back 650,000 years, says Thomas Stocker, a climate
physicist at the University of Bern in Switzerland. Isotopic analyses of
the ice in the deepest portions of that sample — at depths between 3,060 meters
and 3,190 meters — have revealed how temperature in the region varied between
650,000 and 800,000 years ago. But researchers previously hadn’t assayed the
gases trapped in bubbles in that portion of the core, Stocker notes. He and his
colleagues have now performed those analyses and report their findings in the
May 15 Nature.
Once snow piles up more than 80 meters or so deep, the
pressure at the bottom of the heap converts the densely packed, somewhat porous
snow into impermeable ice, thereby locking bubbles of air in place. As snow
continues to accumulate, the mass of ice — whether a mountain glacier or a
continent-wide ice sheet — becomes a chronicle of long-term variations in the
atmospheric concentrations of various gases, including those such as carbon
dioxide and methane that are linked to climate change.
GOING WAY BACK Researchers drilled a 3.2 kilometer–long, 98 millimeter–diameter ice core from East Antarctica, one portion shown here, that includes precipitation that fell during the past 800,000 years.L. Augustin/LGGE
In many aspects, the new results provide no surprises,
Stocker says. Earth still plunged into an ice age every 100,000 years or so,
punctuated with warm spells, or interglacials, that lasted about 10,000 years.
And, as found in previous studies of this core’s shallower ice, the rises and
falls of the region’s temperature are closely linked to increases and decreases
in levels of carbon dioxide and methane trapped in the ice’s bubbles. In other
aspects, however, the samples provide new clues about ancient climate.
Previously, ice core studies have found that natural levels
of atmospheric carbon dioxide varied between 180 and 300 parts per million. However,
during a 3,000-year period that began about 670,000 years ago, CO2
levels dropped to a minimum of 172 ppm, a low level unseen in other, more recent
samples. Since the Industrial Revolution began, CO2 concentrations
have been on the rise because of the burning of fossil fuels; today, levels of
that greenhouse gas exceed 380 ppm and are increasing, on average, about 2 ppm
each year (SN: 5/10/08, p. 18).
The geologic record, including seafloor sediments, suggests
that the long-term average amount of CO2 in the air has been
declining for at least 50 million years. The new findings, however, hint that atmospheric
levels of CO2 in general rose from 800,000 to 400,000 years ago and
then began to decline again. Also, the researchers report, interglacial periods
between 800,000 and 400,000 years ago weren’t as warm as those that have
occurred more recently. These variations, although small, may reveal previously
unrecognized cycles in climate that scientists don’t yet understand, Stocker says.
Determining the duration and magnitude of these cycles, if
indeed they are real, may require scientists to discover Antarctic locales that
harbor ice older than 800,000 years, says Ed Brook, a paleoclimatologist at
Oregon State University in Corvallis. The coldest and thickest parts of the ice
sheet in East Antarctica, for example, may
retain deep ice that fell as precipitation more than 1.5 million years ago, he
notes. Samples that old, he adds, could help solve a long-standing mystery
posed by the geologic record: Why does a 100,000 year–long climate cycle
recorded in recently deposited ocean sediments disappear in rocks laid down as
sediments more than 900,000 years ago?
Even if scientists never find ice more than 800,000 years
old, the new findings confirm that Earth’s atmosphere today is unusual, Brook says.
“Modern levels of greenhouse gases have no natural analogue in the ice record,”
he notes.
Found in: Climate Change, Earth Science and Environment
Text Size
Global climate depends on many factors other than the CO2 level, but what theories account for this mix of counterintuitive findings about how climate change has operated over time?
Several articles within the last handful of years in your magazine have noted that ice core sampling has only recently gone back to close to 1 million years in time. See the 650,000 figure in the February 7, 2007 article From Bad to Worse: Earth's Warming to Accelerate. I recall earlier articles on annual ice core sampling from the Greenland ice sheet which seemed to stop in the 350,000 years ago range.
As a reader, I would like to know more about the sampling methodology and the statistical methods used to analyze these ice core results. In the press, positions have been taking around global warming and single numbers have been widely used to impress the point (for the simple-minded public, perhaps, to be able to grasp at least one aspect of the enormity of the issue). I could trust these numbers more if I knew how they were arrived at. This is particularly the case when the record consists of analysis of just one 3.2 kilometer long ice core in one location.
Pardon my ignorance, but here are some of the gaps.
1. How is the present-day sampling done? My understanding is that the longest, continuous modern record comes from Hawaii. Are CO2 levels affected by that latitude, atmospheric weather patterns, altitude, surrounding ocean versus mid-continent? What are the variables in CO2 concentration in the atmosphere?
2. How are the ice core samples evaluated? The historic CO2 level is determined by analyzing air bubbles in ice (which forms at approximately 80 meters snow depth). During the entrapment process are there variations that occur in CO2 concentration? How many air bubbles are tested from a sample to determine the average? What is the testing technique? Are there any corruption concerns for samples exceeding a certain age? Since ice has plastic motion, doesn't the accuracy of the CO2 level for specific spans of time deteriorate with the depth of the core? The June 7, 2008 article mentions a 175 ppm CO2 level for a 3,000 year interval. How long ago was that interval? As with present-day sampling, what are the factors affecting CO2 levels at specific locations/latitudes?
3. What methods are there for comparing the ice core CO2 levels with other climate records? The June 7, 2008 article mentions a 100,000 year climate pattern in sea floor sediments, which peters out at 800,000 years. Any others?