Dr Ashley Sparrow is wearing boardshorts and a T-shirt: beachy casual wear belying the Antarctic ice out of the window. His flight from Scott Base to Christchurch has been delayed due to weather, and he's killing time - working on his laptop, playing the piano in the bar with accomplished, booming grace.
Today, he and University of Waikato microbiologist Dr Ian McDonald are relaxed after their team completed two weeks' worth of field research in six days.
"We're finished!" exclaims McDonald, beaming. "We've got nothing to do."
"It's like waiting in an airport lounge," adds Sparrow, "except the food is better."
Sparrow has come from one extreme to another: he's an associate professor in environmental science at the University of Nevada in Reno, though he spent almost 12 years teaching at the University of Canterbury.
The biocomplexity project he's in Antarctica to work on is part of the International Polar Year, a scientific programme exploring the Arctic and Antarctic from March 2007 to March 2009.
Led by the University of Waikato's Professor Allan Green, and involving an international team of collaborators, this project's mission is ambitious: figuring out the entire ecosystem of the Dry Valleys.
"Never before has anyone tried to understand how all these organisms that you can't see with a naked eye vary across an entire landscape," says Sparrow.
"We're trying to understand how different organisms are being driven by very subtle variations in what looks like a uniformly polar desert environment."
From the air the valleys are vast, empty and silent, coloured in shades of brown, black, and the blue and white of ice, snow, glacial tongues.
The first humans to see them were Captain Robert Falcon Scott and his Discovery expedition team in 1903. Scott looked across the rugged peaks and troughs and called them "the valleys of the dead"; even the great glaciers that once carved them have now withered away.
But where there's a hint of water, there's life, even in the polar deserts of Antarctica. Though they cover a tiny percentage of the continent, the valleys play a significant part in our understanding of how an ecosystem works.
It was once thought that they were completely lifeless; instead, the gravel-strewn corridors are home to an array of organisms that fascinate the people who study them.
Dead seals, for example, create a microclimate beneath their rotting flesh. McDonald studies the soil beneath the carcases and says they act like "a big lump of fertiliser".
"You find [seals] all over the place. There's theories about them being shoved out the back of aircraft because they've been used in animal experiments. Another theory is that they start looking for open water, and above open water the sky turns a different colour, similar to what you see over the Dry Valleys as opposed to the ice. So they head towards the valleys, and they just keep on going, and then they just run out of water and food and die. There are some places where there are a lot of seals in a very small area and that's a huge amount of carbon and nitrogen. They don't decompose very quickly so it's very, very slow input. It's like Osmocote. And it's definitely a hotbed of activity underneath them. Activity rates are 100 times higher than you'd find in normal soil a metre or two away."
The dead seals create a tiny but significant variation in the desert environment, and it is these variations the project hopes to understand. Sparrow says the combination of the three valleys chosen for the survey - the Miers, Marshall and Garwood valleys - means the full spectrum of the ecosystem can be explored.
"By combining the three, we go from the most desert part of Antarctica to a place that looks like an oasis, when the ice melts and it gets quite wet. There are parts of the Garwood Valley that look tropical." Tropical?
"Well, by Antarctic standards. It has all the things that you'd really like in your garden - lots of liquid water in the summer, relatively fertile soil and warm soil temperatures. All of those things together make a garden grow, and it's exactly the same thing down here. You can look across the valley and it will appear green because all these mosses that have been shrivelled up and dried for years and years suddenly come to life."
The mosses can be thought of as resurrection plants, he says, that green up when the conditions are just right - for three days, about once a decade.
"They unfurl from around the bases of the rocks and so it looks like there's a moss carpet, the faintest hint of green. When you're in the Dry Valleys everything is rock-coloured, white or blue for the sky, so when you see green you're really struck by it because it's such a rare colour. That's a tropical resort in Antarctica."
Sparrow hopes the research will also shed light on one of the most fundamental questions in ecology - how to build life itself. "If we had to restore an area that humans have damaged in some way and left almost sterile, how would we start again? Or, if you go back 20 years when we were still mad about space travel, could we build an ecosystem so we could travel long term through space?"
He says the past ecological experiments, where scientists have tried to create mini-ecosystems in a lab, have never worked.
"There have been experiments in England - well, you can imagine things going mouldy easily in England. But in Antarctica we have a natural but simple ecosystem. It may be possible to use this data to go from the simplest community that we see, under the most stressful or disturbed conditions, and then put all the remaining samples in a line up to the most productive oasis and look at what the changes are."
In the Dry Valleys, change happens so slowly that scientists return year after year to see their old footprints still imprinted in the rocky soil. Because the ecological systems of the lake and valley floor are so simple, scientists are able to study each element individually. This is near-impossible in a more complex environment such as a forest, where millions of organisms interact in an impossibly tangled web.
The thought of being able to identify every organism in the Dry Valleys fascinates Sparrow.
"Down here it might be possible to actually get a pretty good idea of every type of organism - every bacterium, for example, which you can't do anywhere else in the world. And that may actually give us a handle on how to build or restore an ecosystem.
"If we can do it for the desert spectrum, we might be able to transfer those rules to somewhere more temperate, more like New Zealand. Because the more humans destroy or disturb the surface of the earth, the more restoration is going to become a big issue."
He looks pleased and pauses, leaning back on his chair. McDonald laughs: "I think we just solved the problem of life."
OUR OWN MARS
Threading through Victoria Land, about 90km west of the New Zealand and American research bases in McMurdo Sound, the Dry Valleys form the largest ice-free areas in Antarctica - covering about 0.03 per cent of the continent.
In summer, temperatures fluctuate between zero and minus 20C. In winter, temperatures have dropped to as low as -68C. Winds that rush downhill - known as katabatic winds - at speeds of 200km/h or more evaporate all moisture from the land and push pebbled ridges into wave-like forms.
Some seals in the Dry Valleys have lain there since the time of the Pharaohs; they have been radiocarbon dated to thousands of years old. The gritty, powerful winds grind the flesh and fur from their bones and over time reduce them to skeletons.
Some scientists consider the valleys' rocky, desolate surfaces to be the most Mars-like on Earth. During the 2008/2009 summer research season, Nasa tested duplicates of the Phoenix Mars Lander's instruments in Antarctica's Beacon Valley, and ran the last tests of ENDURANCE (Environmentally Non-Disturbing Under-ice Robotic Antarctic Explorer, the most over-wrought acronym ever fabricated).
ENDURANCE will leave Earth for good sometime in the next 20 to 30 years and fly to Europa, Jupiter's sixth moon, where scientists hope to use it to discover microscopic life-forms in suspected oceans under the planet's crust.