麻豆传媒

鈥淭HE first time I came here was in October, getting close to polar sunset. It
was dark, and as our plane broke through the clouds there was suddenly ice
stretching out for ever and these huge dark cracks of open water. Then there was
the tiny glow from the camp and the ship, like some last vestige of
civilisation. It made shivers go up and down my spine. It was like landing on
some other planet.鈥�

Taneil Uttal, a meteorologist from the Environmental Technology Laboratory in
Boulder, Colorado, is sitting in her cabin on the Des Groseilliers, a Canadian
icebreaker that has been frozen into the pack ice hundreds of kilometres north
of Alaska since October 1997. From the ship鈥檚 bridge, you can see what she
means: through the huge panoramic windows, scattered huts, tents and scaffolding
towers are silhouetted against an alien landscape of ice and sky. It鈥檚 an
awesome sight, but the first officer, here on watch duty, seems bemused. 鈥淚t鈥檚
better when we鈥檙e moving,鈥� he shrugs.

Actually, the icebreaker is moving, but with the ice rather than through it.
The ship is the heart of a $19 million experiment to piece together the
mysterious workings of the Arctic Ocean. The year-long project, dubbed SHEBA,
for 鈥渟urface heat budget of the Arctic Ocean鈥�, has brought together an
unprecedented collection of atmospheric scientists, oceanographers and
glaciologists. Together they are studying the interactions between the three key
components of the Arctic climate: air, sea and, above all, ice, which forms,
breaks, melts and re-forms as the seasons change.

鈥淚t鈥檚 not a smoking-gun, global warming-type experiment,鈥� says Dick Moritz,
head of the SHEBA project office in Seattle. 鈥淲e鈥檙e not trying to figure out
what the climate has done. We鈥檙e there to figure out how it works.鈥� If it
succeeds, SHEBA will give climate modellers information that鈥檚 badly needed.
According to the global climate models (GCMs) that simulate how increasing
levels of greenhouse gases will alter the Earth鈥檚 climate, the polar regions are
highly sensitive, and any temperature changes could be up to five times as large
as those at the tropics.

Cool reflection

The main reason for this, says Andrew Weaver, a climate modeller at the
University of Victoria in British Columbia, is a mechanism known as the
ice-albedo feedback. White sea-ice reflects plenty of sunlight, keeping the
polar regions cool. If the ice melts, it exposes dark seawater, which absorbs
sunlight and causes yet more ice to melt in a positive feedback loop.

Unfortunately, modellers don鈥檛 know exactly how this works so each climate
model tries to account for it in a slightly different way鈥攁nd each comes
up with a different answer. Worse still, says Weaver, no one knows how to
incorporate clouds into the models. Clouds could make a significant difference
to the way that heat passes through the system, and to the interactions between
temperature change and the pack ice.

If the Arctic really is as sensitive as the models predict, the whole
ecosystem could be threatened, to say nothing of the effects large-scale melting
could have on the weather patterns in the northern hemisphere. To know for sure,
the modellers need a better understanding of the way the whole system works. But
the Arctic is almost entirely ocean, and relatively little is known about it.
鈥淭here could be whole new processes involved in the Arctic that we鈥檙e just not
aware of,鈥� says Weaver.

This is where SHEBA comes in. Funded mainly by the US National Science
Foundation, with support from the US Office of Naval Research, the Canadian
government, the US Department of Energy and NASA, the idea is to measure all the
different components simultaneously under a wide range of conditions, and then
to piece together this information to work out what modellers need to include in
their models. It鈥檚 no easy task when the place you are studying is literally
shifting under your feet.

Not that you鈥檇 know it when you arrive. Though the Des Groseilliers is
surrounded by ice, the ship seems strangely land-bound. On the short hike to the
main parts of the camp, the ground feels as solid and unshakable as anything
you鈥檙e ever likely to walk on, and the life belt hanging on a pole seems
incongruous. Peter Guest, a meteorologist from the Naval Postgraduate School in
Monterey, California, stops and stamps on the ground a few times. 鈥淒on鈥檛 forget
the ocean is right down there. You鈥檙e walking on water now.鈥�

It鈥檚 May, and the ice floe is holding up well. But the past months have
brought plenty of reminders that the camp rests on a thin skin of ice, not much
more than a metre thick. Most dramatic was the event of 7 February, when the
whole camp came close to a watery grave (see 鈥淐racking up鈥�). Guest points
out one of the casualties of the near break-up鈥攖he mangled remains of a
Skidoo repair hut that was thrust to the top of a pressure ridge when the
separated edges of the ice crashed back together again.

Guest has almost finished his stint at the SHEBA ice station and is handing
over to Ed Andreas, who has just arrived from the Cold Regions Research and
Engineering Laboratory (CRREL) in New Hampshire. We are on our way to 鈥淢et
City鈥�, site of much of the meteorology research.

At the heart of Met City is a 20-metre metal tower. Around the tower
radiometers face up and down, recording all the radiation striking the
surface and reflecting off it. The whole of SHEBA is ultimately focused on the
surface, says Andreas, because that is where the three components of the
system鈥攁ir, ice and ocean鈥攎eet. 鈥淚t鈥檚 where all the action is.鈥�

He and Guest are particularly interested in how the wind pushes the ice
around, making it splinter and re-form. From automatic weather stations which
are dotted throughout the Arctic and report their measurements through satellite
links, researchers can determine the speed and direction of the winds. But this
can give a misleading picture of the ice dynamics, especially during the long
polar night, when stable layers form near the surface. Unless turbulence carries
the wind鈥檚 momentum downwards, the ice remains untouched. 鈥淲e鈥檙e trying to
understand how the wind can get down to the surface and move the ice around,鈥�
says Guest.

Near the tower is a sodar鈥攕ound detection and ranging鈥攊nstrument,
which constantly chirps as it uses sound pulses to measure tiny turbulent eddies
in the air above. Every few metres up the tower itself, instruments measure the
wind speed in three dimensions, and the flow of heat to and from the surface. At
the top of the tower, the wind is biting. Off in the distance, a tethered
balloon, like a miniature white airship, is exploring the next layer up. Higher
still, weather balloons climb to an altitude of 20 kilometres or more, measuring
the clouds they pass through, while as part of other programmes timed to
coincide with SHEBA, planes fly through the clouds over the site. And back on
the ship, Uttal鈥檚 lidar and radar instruments are constantly watching the sky,
firing light and radio waves upwards to locate cloud droplets of water or
particles of ice.

Invisible clouds

Clouds are particularly important, says Uttal, because they make a huge
difference to the amount of radiation that reaches the surface, and yet the
models handle them very poorly. Part of the problem is that it can be hard to
tell from satellite pictures if the clouds are there. 鈥淪ometimes when the radar
is showing that there are ice particles going from the surface up to 5 or 6
kilometres, you can go outside and see blue sky all around,鈥� says Uttal. Add to
that the long months of Arctic darkness, and it can become very difficult to
know exactly how much of the time the Arctic spends under cloud. 鈥淚f you don鈥檛
know they鈥檙e there then a GCM is not going to put them in. [The models] could be
expecting clouds 20 or 30 per cent of the time when in fact they鈥檙e there 80 or
90 per cent of the time.鈥�

Should you care about clouds you can鈥檛 even see? Yes, says Uttal, if they
affect the radiation balance. In principle, the presence of clouds could change
everything about the heat balance in the Arctic. Depending on their height,
structure and whether they are made of liquid or ice particles, clouds can warm
or cool the ice below, and the SHEBA researchers hope to pin down which clouds
do what. They also hope to confirm satellite data by comparing them with the
SHEBA cloud measurements. Without satellites, researchers can鈥檛 hope to cover
enough of the Arctic to monitor how the region is changing.

Next to Met City lies Ocean City, a motley collection of tents where the
focus is not up but down. Through holes drilled in the ice, oceanographers are
looking at how heat moves between the upper ocean and the ice. Roger Anderson,
an engineer working for a consortium of scientists, is using a winch to lower
instruments for measuring temperature and salinity down to 150 metres. He
watches his cable anxiously as the instruments descend.

The oceanography instruments are performing well, recording turbulent eddies
at 4 metres and 8 metres, the wet analogy of the met tower work, and making
sonar measurements in the depths. But fate has delivered an unkind blow. The
experiment started over the Arctic abyssal plain, which lies 3.5 kilometres
below the ice. The water in the upper ocean there was unexpectedly fresh. The
researchers were intrigued: did this mean that the past few summers had seen
unusual amounts of ice melting? But a steady westward drift has taken SHEBA over
the Chukchi Cap to a different body of water, as little as 500 metres deep. No
one knows how much this will affect the results, but everyone would have
preferred to stay with the same patch for the whole experiment. Still, Anderson
is sanguine. 鈥淵ou have to play the hand you鈥檙e dealt if you鈥檙e drifting with the
颈肠别.鈥�

Back on the surface, Guest and Andreas are more concerned about ensuring
their measurements encompass the broadest possible range of conditions. As well
as the met tower, and two other smaller towers, a handful of semiautomated sites
are scattered farther from the ship, each named after one of the teams in last
year鈥檚 baseball playoffs.

Engineer Jeff Otten from the Environmental Technology Laboratory is on his
way to the site called Atlanta, to perform the last service visit of his 7-month
stint. The temperature is a balmy 鈥�7 掳C, the Sun is shining and the
wind is down to a gentle breeze. It鈥檚 a far cry from the miserable weather Otten
encountered in the depths of winter. Day after day, he says, the temperature
stuck at 鈥�35 掳C, with blowing snow and a howling, chilling wind. 鈥淲e
had three chemical hand warmers in each boot.鈥� What made him choose to work for
so long in these conditions? 鈥淚f I could find the same kind of job in a warmer
place, I鈥檇 look into it,鈥� he says.

In the winter darkness, everything had to be done by flashlight, and extra
people were drafted in to stand polar bear watch. The bears are a real danger.
Curious and often hungry, they are supreme killing machines. No one is allowed
to leave the ship without a shotgun, although unless things get really serious
the gun is for warning shots only. On the Skidoos, we pass the place where just
a few days ago a 鈥渂ig 10-footer鈥� arrived to dig out a seal hole. 鈥淛umper鈥�
Bitters, the project鈥檚 Mr Fixit, was sent to scare the beast away. In spite of
his 23 years鈥� experience in the Arctic, he still found this one too close for
comfort. Only reluctantly did it yield to warning shots and leave the camp.

The Atlanta site consists of little more than a silver suitcase with an
antenna, solar panel and metal tripod, and looks for all the world like a
miniature lunar lander. Nearby, a crack in the ice鈥攌nown as a
lead鈥攈as opened and is beginning to refreeze. Here, away from the ship and
the huts, the remoteness begins to hit home. At the edge of the lead, blocks of
ice gleam an extraordinary aqua blue against the white snow. The water is almost
black in contrast with the bright ice, while on the water鈥檚 surface, streaks of
newly formed ice crystals are dusted with snow. Above, the dense white stratus
cloud seems low enough to touch. The silence is eerie.

The instruments here were perfectly placed to record the effects of the lead
as it formed. 鈥淟eads are like open windows into the ocean,鈥� says Moritz. In the
winter, with the air temperature 鈥�35 掳C or lower, and the freezing
seawater a mere 鈥�1.7 掳C, heat pours out of the ocean. In the summer,
the dark water absorbs much more sunlight than the white snow or grey ice around
it. So although leads occupy a relatively small area, they could be
significant.

But it鈥檚 not just through cracks that heat can pass between the ocean and
atmosphere. The ice and snow play a vital role in trapping and releasing heat.
Back at the ship, Terry Tucker from CRREL, who is presently chief scientist for
the project, is preparing for a trip to one of the ice research sites about 5
kilometres away. Careful preparations are made. At this distance, we could
easily become stranded behind a lead, and if the weather suddenly turns bad, the
helicopter will be unable to lift us to safety. A sledge must be loaded with
survival gear: tents, stoves, clothing, sleeping bags, flares, shotgun
cartridges and emergency rations of chocolate.

Mountains of ice

No one has visited the site for two weeks, and snow has already covered the
previous tracks. Lines of flags are meant to show the way, but they are little
help thanks to the shifting of the ice. We plough through the snow, over ridges
and down ice gullies, stopping to scan the horizon for the orange flag that
marks the site. Out here, evidence of the ice鈥檚 movements is everywhere. The
plains of ice resemble miniature rift valleys, encircled by the tumbled mountain
ranges of the pressure ridges. 鈥淚t鈥檚 plate tectonics fast-forwarded,鈥� says
Tucker.

Once we reach the site, he scrabbles in the snow and pulls out a chunk that
holds together unexpectedly well, unlike the powdery surface. This is the hoar
layer, snow made of long, thin crystals, which was uncovered a few weeks ago by
snow scientist Matthew Sturm from the CRREL in Fairbanks. He found a layer of
hoar just above the ice virtually everywhere he dug. It is filled with closed
cells of air, like loft insulation, and could make a difference to the movement
of heat from the ocean up through the ice and snow. 鈥淢atthew believes that the
snow out here is three times less conductive than people had been using in their
models,鈥� says Tucker. 鈥淭hat would have a big impact on how much sea ice growth
you鈥檇 get during the winter.鈥�

Another surprise, says Tucker, has been the whiteness of the ice under the
snow. The snow is very reflective. Models tend to assume that it will reflect
around 80 per cent of the incoming light at the start of the winter, and that
reflectivity falls to around 40 per cent when the snow melts in the summer,
exposing greyer ice. But beneath the snow, Tucker has found badly deteriorated
ice, full of pinnacles and bubbles, which is almost as white as the snow. 鈥淚t鈥檚
almost like a self-defence mechanism,鈥� he says. Usually, when snow melts, the
surface below can absorb more heat, causing further melting. This white ice
could help to retard the feedback, another mechanism that the models don鈥檛 take
into account.

Results like these are tantalising, but everyone agrees that the significance
of SHEBA will not be clear until all the different interactions are extracted
from the data and plugged into the models. This will happen in about a year鈥檚
time.

Meanwhile, although the ice has been steadily melting through the summer, the
camp is holding together well. For those who have endured ice camps before, this
is Easy Street. 鈥淚 wouldn鈥檛 consider it doing luxurious science,鈥� says Tucker,
鈥渂ecause we still go out and freeze our butts off every day. You get to the
point where you get so damned cold you want to cry sometimes. But still you can
come back to this wonderful hotel in the middle of the Arctic Ocean, with a crew
that bends over backwards to help you. They have an excellent electronics
technician, they have an excellent machinist. The food is gourmet. You can take
a hot shower every morning鈥攊t鈥檚 just fantastic.鈥�

The worst of the trouble began in January, when a 鈥渓ead鈥� of open water
separated the ship from the main camp. Buster Welch, a biologist from the
Canadian Department of Fisheries and Oceans in Winnipeg, who is head of the
Canadian ancillary research programme, was working in his hut. Suddenly he heard
a loud crack. Two metres away, the ice had split. Welch and his colleagues
hastily bulldozed snow to stop water flooding his storage hut, and the crisis
seemed averted. By the next day, the lead had opened to 60 metres wide. Then, to
everyone鈥檚 relief, it closed again.

But at noon on 7 February, another lead opened, this time right by the ship.
Worried about his hut, Welch began to 鈥渨interise鈥� his equipment, removing
everything that couldn鈥檛 stand freezing. When he emerged 45 minutes later he
couldn鈥檛 believe his eyes. 鈥淲hen I went in I was 120 metres off the starboard
side. When I came out there was the ship aimed right at me, about 60 metres
away.鈥� Fortunately for Welch, the ship nosed up against the Blue Bio and then
stopped.

But the drama was not over. While much of the water refroze, the main crack
separating the ship from most of the camp shifted fitfully day by day. Then one
night, while the researchers were in bed, most of their tents and equipment slid
half a kilometre north before coming to rest.

Nothing this drastic has happened since, although there have been several
alarms. 鈥淚鈥檝e been woken up a few times,鈥� says Welch. 鈥淚 was standing up here on
the bridge one night and the whole ship lurched. Everyone comes up to see what鈥檚
happening, and check if their buildings are still there. You get to have a
辫补谤迟测.鈥�