¿ìè¶ÌÊÓƵ

Ice is a unique archive. The snow that falls over the Greenland icecap
carries gases, acids and dust added to the atmosphere from all over the
northern hemisphere. As it turns to ice, the chemical fingerprint of each
layer is buried intact, gradually moving down through the icecap as it is
covered by successive layers each year. If the layers can be dated, then
the ice provides a record of what was happening in the atmosphere thousands
of years ago.

For the past two summers researchers from the Greenland Icecore Project
(GRIP) have been drilling into the Greenland ice sheet from its highest
point, Summit Camp, 3200 metres high, at a latitude of roughly 73 degrees
North. Later this year the researchers, from Denmark, Switzerland, France,
Germany, Iceland, Britain, Belgium and Italy, will return for what may be
the most exciting field season yet. So far they have recovered 2321 metres
of ice core, reaching well back into the last ice age and containing a wealth
of information about the climate and atmospheric chemistry of the northern
hemisphere over the past 40 000 years. This summer, their drill will reach
bedrock more than three kilometres below the surface, cutting through ice
between 300 000 and 500 000 years old – the oldest and deepest ice recovered
from anywhere in the world.

The information frozen into the ice includes atmospheric composition
and chemistry, levels of dust and acids from volcanic eruptions. This record
of changes during both large and small variations in climate will reveal
more about what initiates climate change and how changes in the climate
system progress – vital topics in the debate on global climate change.

GRIP is already a success, according to Henrik Clausen, from the Department
of Glaciology at the Geophysical Institute at the University of Copenhagen,
who is in charge of field operations. ‘The drilling has so far exceeded
our best expectations, and we have recovered more than 99.9 per cent of
the core,’ he says. Clausen is confident of reaching the bedrock. ‘But of
course we are a little anxious to see if the drill can pass the final test
this summer.’ The slight doubts about the future success of the project
arise from the record depths that GRIP will reach. This year, GRIP’s drill
will be working under higher pressure than any other ice core project as
the drill bites deeper, the weight of ice above it increases.

GRIP will be the third core to reach the bottom of the Greenland ice
sheet. The first was a US project at Camp Century near Thule in northwest
Greenland, from 1963 to 1966. Its core was 1385 metres deep, reaching about
120 000 years back in time. The next core was at the radar station Dye-3
in southeast Greenland, between 1979 and 1981. This core was longer – at
2038 metres – and much better quality, because drilling techniques had improved.
But more ice survived each year, so the core represented only about 100
000 years.

Antarctica has had two deep cores, at Byrd and Vostok. The Byrd core,
drilled in west Antarctica by US scientists, spanned the past 70 000 years
in 2191 metres of ice. The core from Vostok, in east Antarctica, reached
a depth of 2521 metres and covered 160 000 years – setting the record for
time and depth so far. The Vostok core was drilled by Soviet engineers and
analysed jointly by Soviet and French scientists.

The GRIP core, together with another deep ice core, the American GISP-II
which is being drilled 30 kilometres away, will be superior to these valuable
records. The GRIP core is at the highest point of the ice sheet, at a site
where there is little sideways flow of the ice. When the flow of ice is
strong, older layers of ice are broken up, destroying the record. At Summit,
the deeper layers of the core are merely flattened, preserving more of the
valuable older ice. The core should give researchers a chance to track climate
and atmospheric chemistry further back in time than ever before.

The big question is how far back in time the GRIP core will go. That
depends mainly on whether or not the ice melted in the warm period that
ended about 120 000 years ago with the start of the last ice age. ‘Our models
indicate that the ice close to the bottom is between 300 000 and 500 000
years old. But nobody knows for sure until the drill reaches the bedrock,’
says Willi Dansgaard of the University of Copenhagen.

Models of what to expect in the ice sheet are based mainly on measurements
of ice flow from the two older boreholes. In general, flow drags layers
of ice into folds, thickening some layers and thinning others until they
can no longer be traced. Such distortion also degrades the information that
the layers hold – atmospheric gases trapped in bubbles in the ice leak away,
for example. But at the GRIP site there is no folding, just thinning, giving
a good correlation between the predictions of the models and the ice cored.

The GRIP core will provide climate researchers with the most comprehensive
and detailed body of data ever on the climate and atmospheric chemistry
of the past. In particular, the core could be the first to contain ice from
several glacials and interglacials – ice ages and the warmer periods between
them. If it does, researchers will have a chance to investigate a vital
but poorly understood phenomenon: how the climate changes when glaciations
start and end. ‘Speaking in general terms, the new deep core will provide
us with a very long climate record, which contains an unsurpassed level
of detail. This will make it possible to develop more reliable climate models,
concerning both the large climatic changes and the minor variations,’ says
Dorthe Dahl Jensen, from the Geophysical Institute at the University of
Copenhagen.

Understanding the climate of the past is a fundamental part of predicting
the progress and consequences of climatic changes in the future – both natural
changes and the global warming that is foreseen as a result of human pollution
of the atmosphere. One of the pioneers of ice core drilling, Chester Langway
from the State University of New York at Buffalo, puts it this way: ‘The
climate of the past, the present and the future are all interrelated. The
new ice core will be a cornerstone in the new International Geophysical
and Biological Programme, particularly as related to past climate, because
it will provide us with such a detailed chronology for the past several
hundred thousand years. And as a scientific enterprise, GRIP is state of
the art.’

For ice cores, state-of-the-art science does not mean just the best
laboratories, equipment and personnel. To make the most of these valuable
samples, high-quality research must begin as soon as the core leaves the
borehole. As they emerge from the ground, sections of the core are analysed
by researchers working in a 100 metre trench cut 5 metres into the ice.

The first measurements – of electrical conductivity – take place on
intact cores. The conductivity of the core depends on how acid the ice is,
layer by layer. If there was a major volcanic eruption at around the time
the ice fell as snow, that snow tends to be more acid. Gases and particles
from an eruption are blown high into the stratosphere: sulphur dioxide,
for example, spreads throughout the hemisphere and changes into tiny droplets
of acid. Eventually the droplets sink back into the troposphere, the lowest
12 kilometres or so of the atmosphere, where they are washed out by rain
and snow.

The traces left by these big eruptions give the researchers the first
clues to when a particular section of the core was laid down as ice, providing
the basis for more detailed investigations. ‘We measure the conductivity
of the whole core every two centimetres,’ says Eric Wolff from the British
Antarctic Survey in Cambridge. This immediately identifies parts of the
core that formed during major volcanic eruptions and other large natural
events. ‘Then we beat the drum to colleagues working with the more detailed
measurements further down the trench. This keeps up the enthusiasm.’ The
British Antarctic Survey, which is a full member of GRIP, sees the project
as important not only in its own right, but also as a precursor to future
European drilling projects in Antarctica.

In the next stage, a team of Danish researchers make more detailed measurements
of the electrical conductivity of the ice core – taking a reading every
two millimetres. These measurements also give a quick date for that segment
of the whole ice core, because the acid content of the snow that falls varies
with the seasons. For instance, there is most acid in the summer snow on
the icecap. Glaciologists detect the peak of acidity every summer and can
count the annual layers in the core in the same way that archaeologists
count growth rings in a tree. An extra refinement comes from the amount
of dust in the ice. In early spring, dust from the continents spreads towards
to poles, falling with the snow to form layers in the ice.

Farther down the laboratory trench the ice core is sliced in half lengthways.
One half is sent to Copenhagen to be stored at the Geophysical Institute.
The other is cut into smaller pieces, which are either analysed immediately
or distributed to other laboratories participating in GRIP. At the end of
the trench, Katrin Fuhrer from the Physical Institute at the University
of Bern in Switzerland analyses the ice chemistry.

‘The most exiting thing that we have discovered are some peaks of ammonium
ion, which are most probably caused by biomass burning in high northern
latitudes. Some of the peaks reach 20 times the normal background level
in the core,’ she says. They appear in ice from times such as the Younger
Dryas, a cool time about 12 000 years ago that was followed by rapid warming,
and the present interglacial. Fuhrer says that this is the first time ever
that signs of biomass burning have been found in ice cores from anywhere
in the world.

When the chemical analysis in Greenland has picked up such an event,
Michel Legrand from the Laboratory of Glaciology and Geophysics of the Environment
in Grenoble, France, makes a more detailed analysis of that part of the
core. ‘Biomass burning contributes to the input of trace gases into the
troposphere: carbon dioxide and nitrous oxides, for example. But until now,
we didn’t know much about biomass burning in the past,’ says Fuhrer. ‘First
we will compare our results with other ice cores to see how representative
they are. Then we will try to find out if there is a correlation between
temperature and biomass burning through the ages. Our goal is to try to
establish a record of big natural fires similar to the existing record of
volcanic eruptions. We hope to draw some conclusions on the influence of
biomass burning on climate.’

One of the most important of the analyses is carried out in the laboratory
of the Geophysical Institute at the University of Copenhagen, using the
oxygen isotope method developed by Dansgaard to measure the temperature
of the snow as it fell.

There are two natural isotopes of oxygen: the usual form, with 8 protons
and 8 neutrons in the atomic nucleus and the rare heavier isotope, oxygen-18,
containing two extra neutrons. The ratio of these two isotopes in ice cores
is related to temperature because water containing different proportions
of the isotopes melts, freezes and evaporates at slightly different rates.
At each stage in the transformation from sea water to ice, the proportion
of the two isotopes changes in a way that depends on the air temperature,
among other things. The net effect is that when the Arctic air becomes colder,
the concentration of oxygen-18 in the snow decreases, and when the climate
gets warmer, it increases.

Subtle shifts in the snow

The relative concentrations of the two isotopes are measured by mass
spectrometry. ‘In this way we are able to trace both the large and minor
climatic variations as far back in time as the ice core takes us,’ says
Dansgaard. Taking the temperature together with the dating and chemical
analysis of the ice core, the GRIP scientists hope to understand more of
the relationship between the climate and atmospheric chemistry.

The GRIP researchers are also looking at more subtle climate variations.
Jorgen Peder Steffensen from the Geophysical Institute in Copenhagen, is
analysing how the proportion of atmospheric aerosols coming from the continents
and the oceans changes when the climate shifts. He uses chromatography to
find the proportions of ions such as sodium, potassium, magnesium, calcium,
chloride, sulphate, ammonium and nitrate. Most of the water in ice comes
from the sea, so Steffensen relates all the samples to the average composition
of sea water, with its known proportions of each ion. Any departure from
this composition shows that some other processes must be operating.

‘Except for a few known volcanic eruptions, sea water is by far the
dominant source of chloride in the ice core. So we normally start with the
chloride-sodium ratio, and it often fits perfectly with the ratio in sea
salt. If there is an excess of sodium, it is likely to come from the continents,’
says Steffensen. The researchers have to consider all the processes that
could contribute to the chemical composition of the ice. For example, nitrate
ions are produced mainly in the atmosphere and sulphate is produced in swamps,
by fires and volcanic eruptions and by biological activity in the oceans.

The amount of dust in the atmosphere also changes. There was far more
dust in the atmosphere of the last ice age than there is now, mainly because
the seas were lower and previously submerged land was exposed to very strong
winds. The glaciologists have also shown that the amount of dust in the
atmosphere peaks every spring and winter. Tracing the dust in the core identifies
annual layers – in fact it is one of the most sensitive methods of dating
these cores. Researchers compare the dust dating with the other techniques.

Taken together, these diverse dating methods are far more powerful than
each one would be alone, because they support and complement each other.
If one of them fails to give a clear stratigraphy, another may be useful.
For example, counting annual layers back from the surface is valid only
if each year’s ice layer has survived. The signatures of events such as
major volcanic eruptions acts as a check that can be used to show where
layers are missing. ‘So far we have dated the GRIP core some 40 000 years
back in time by combining the dating methods,’ says Clausen. According to
these results, the last ice age ended about 11 500 years ago. ‘In the Dye-3
core, we found that the last glaciation ended 10 700 years ago, but the
new results are more reliable because we this time have used more dating
methods simultaneously,’ says Clausen.

There was more carbon dioxide in the atmosphere at warmer times in the
past, but so far there is no proof that this gas drives climate change.
A team of Swiss and French researchers led by Bernhard Stauffer of the University
of Bern are tackling this problem by measuring the carbon dioxide and methane
in bubbles of air trapped from the atmosphere when the snow fell. The team
hopes for a correlation between greenhouse gases and the other indicators
of climate change.

Claus Hammer from the Geophysical Institute in Copenhagen measures dust
and nitrate levels, working in a small laboratory on the ice. ‘Our first
aim is to date and study the large climatic changes like the termination
of the last glaciation,’ he says. After several thousand years of climate
variations between cold and milder periods, the temperature in the northern
hemisphere suddenly rose 7 °C in just 50 years, and after that the climate
has stayed warm. ‘This shows that climatic change may happen very fast,
especially changes from cold to warmer climate,’ says Hammer. ‘The dust
measurements are so sensitive that we can trace the precipitation month
by month in the transition period. We will also be able to show how the
seasons changed during the shift from glacial to interglacial.’

The rapid end of the last ice age seen in the ice core correlates well
with other records, such as those from lake and bog sediments in western
Europe. But the ice core record will be the most detailed so far. Recently
a group led by Wallace Broecker from Columbia University, New York, has
found that the oscillations of climate from cold to warm, glacial to interglacial,
seen in the Greenland ice cores are matched by cycles in marine sediments
from the North Atlantic. Broecker proposes that these events are a result
of an oscillation in the salt content of Atlantic sea water. And he suggests
that the changing salinity is driven by fluctuations in the amount and path
of meltwater coming from the shrinking ice sheets.

Hammer sees the excellent resolution of the ice core record as a way
to investigate ideas of this kind in more detail. ‘The ocean is undoubtedly
involved in rapid climatic change, but more research is needed before it
is possible to make a firm conclusion about what initiates the process,’
he says. ‘By analysing the transition period between glacial and interglacial
in detail in the GRIP core, we are trying to find out what makes the climate
system shift so quickly to new states and how the atmosphere and the climate
react to astronomical changes, like changes in the solar output, changes
in the Earth’s rotation and the angle of the axis of the Earth.’

The resolution and length of the GRIP record make it unique. One of
the biggest questions facing climatologists today is how fast the Earth’s
climate can change, and whether it might change rapidly in the future. Could
a sudden switch in the pattern of ocean circulation have triggered rapid
warming at the end of the last ice age? If the climate system changed quickly,
how soon did the ice sheets shrink or grow?

The detailed data from GRIP should help to answer some of the questions
raised by climatological research so far. But if existing ice cores are
anything to go by, the project will also raise enough questions to fuel
research for years to come.

Rolf Haugaard Nielsen is a freelance science journalist based in Copenhagen.
He visited the GRIP camp last summer.

* * *

THE DRILL FOR DRILLING INTO THE ICY PAST

During the short summer season, the camp at Summit hums with activity
round the clock. Between 30 and 50 people live and work around the drill
site. Their lifeline to civilisation is a three kilometre-long skiway, where
Hercules transport aircraft from the US Air National Guard based in New
York, and a small Twin Otter plane supplied by the British Antarctic Survey,
ferry people and cargo.

The camp itself consists of three dome-shaped buildings and a few rows
of tents. One dome shelters the drill trench, 5 metres below the surface,
where the temperature never exceeds -15 °C. This is where most of the
work goes on.

The focus of the camp is the drill hall, at one end of the trench. Drilling
teams work night and day in three shifts. Instruments and computers keep
the drillers constantly informed about the position of the drill bit, the
temperature and the pressure in the borehole, the pull on the cable, the
number of turns of the bit and other relevant variables.

The drilling itself is done in bursts of about nine minutes, which produce
a piece of core 2.5 metres long at most. By the end of the 1991 season,
when the drill had penetrated more than 2300 metres of ice, it took an hour
to raise the drill with each section of core and lower it again.

The project uses a drill that was developed and built in 1978 at the
University of Copenhagen primarily by the director of the GRIP Operational
Centre, Niels Gundestrup, together with Sigfus Johnsen, who is now professor
of Geophysics at the University of Reykjavik in Iceland. The drill is 11
metres long and consists of two parts. Three bow-shaped pieces of metal
mounted at the top of the drill keep the assembly centred in the borehole
while it ascends and descends. When the drill is working, these components
engage the ice walls of the borehole preventing the upper part of the drill
from turning. The lower part of the drill is a spinning cylinder which cuts
around the core. The core itself is held inside the drill.

The borehole is filled with a hydrocarbon fluid, similar to diesel fuel,
with a density close to that of ice. This ensures that the enormous pressure
of the surrounding ice does not squeeze the borehole out of shape.

Drilling poses endless technical challenges; it is by no means a routine
operation. For example, the borehole must be kept clear of ice chips, which
can damage the mechanisms of the drill.

‘We have solved the problem by placing a long screw between the two
parts of the drill. During drilling the screw is driven out making the drill
1 metre longer,’ says Johnsen, who leads the drill teams. ‘This creates
an underpressure at the cutters, and the ice chips are sucked up through
three thin channels on the outside of the drill.

Unexpected problems can strike at any time. On one occasion, in spite
of all precautionary measures, an oilcan fell into the hole and slid the
2 kilometres to the bottom. A magnet was specially attached to the drill
and successfully retrieved the can. The team avoided disaster, drilling
went on and GRIP continued its journey into the unknown climatological past.