TOOLIK LAKE, Alaska -- The last days of summer are rapidly fading here in the land of the midnight sun, 160 miles above the Arctic Circle, spurring a frenzy by animals and plants alike to prepare for the hyperborean winter's nine-month onslaught.
Also racing against the clock to finish their field work are dozens of international scientists at this remote research site, searching for small pieces that could fit into the puzzle of global warming.
With an urgency that drives 18-hour workdays (much less tiring under the 24-hour sun), these ecologists tediously dissect blocks of earth cut from test plots to measure the tiniest mosses and green sprigs.
They wade up to their chests in cold rushing streams to install test equipment, spend hours sitting in wetlands collecting minute insects from river grass. They dive to the bottom of frigid, murky lakes to collect soil core samples. They weed and pick apart vegetation on small hillside plots.
And for everyone, the long treks to these unspoiled test sites are invariably over soggy tundra, a swamp dotted by hard tussocks of cotton grass that erupt from the ground, making footing as perilous as slogging over a field of wet sponges on a bed of bowling balls.
Everywhere, whining clouds of the notorious Alaskan mosquitoes attack the researchers, the voracious insects desperately seeking mammal blood so they can reproduce before the freeze.
After all the summer hardships and long days, most of the scientific study of Toolik will be done back home in laboratories and on powerful computers. The wealth of samples and readings collected during this short Arctic growing season, or summer, will provide raw material for months of analysis and comparison and, perhaps, a key discovery about the complex ecological dynamics near the top of the world.
Toolik Field Station, a former staging post for building the "haul road" along the Trans-Alaska Pipeline, looks much like a setting for "M*A*S*H" -- a dusty spread of old green trailers, Quonset huts and Army-style latrines.
It's 130 miles along the rutted, rocky highway to the nearest town, Deadhorse. Communications depend on satellite telephone relays; mail delivery depends on the good will of supply-truck drivers acting as volunteer postal workers.
The scenery is stunning: The translucent purple, snow-streaked Brooks Mountains stand guard to the south. Placid Toolik Lake glistens like a jewel in the sun, home to loons and gulls and terns and lake trout. The summer weather is ever-changing: One day might be hot enough for T-shirt and shorts; the next brings a sleety, penetrating July snow, the transitions often decorated by rainbows.
The ecology field station was the dream of John Hobbie, a veteran Arctic scientist with the Marine Biological Laboratory in Woods Hole, Mass.
The pipeline and the rough highway provided access to this remote, relatively unspoiled tundra environment; acquiring the abandoned trailers, Hobbie and colleagues set up shop in 1975. The station is now operated by the University of Alaska Fairbanks, on a lease from the federal Bureau of Land Management.
The primary goal of the Toolik station is to provide long-term data on changes in the Arctic environment -- air, water, land, flora and fauna.
These days, global warming is the hot-button (and best-funded) science topic. Most of the Toolik experiments focus in some manner on the natural cycle of carbon use and how changes can affect this fragile environment.
Worldwide, larger quantities of carbon dioxide (and related methane) gas are being released from Earth's surface into the atmosphere, trapping the planet's heat and promoting global warming.
The planet has warmed by an average of little more than 1 degree Fahrenheit over the past century.
At Toolik Lake, the weather record shows a rise of about 6 degrees over the past two decades.
That magnitude of difference is not altogether surprising. Scientists expect, because of a variety of factors, that the Arctic (and Antarctic) regions will experience the first, and greatest, effects of increased global warming. But the warming effects on plants and animals are far from clear.
"This is where the weather begins," says Gaius Shaver, a botanist from the Marine Biological Laboratory.
Of greatest concern is the immense mass of carbon locked in the frigid peat layer beneath the surface soil, slowly building up over 10,000 years, since the end of the last Ice Age. One-third of Earth's estimated total carbon is stored in this partially decomposed vegetable matter.
Higher temperatures could thaw layers of peat, some scientists warn, releasing the carbon as carbon dioxide gas into the atmosphere, reinforcing and accelerating global greenhouse warming through a feedback effect.
But there are complicating factors.
Warmer soils can also increase the growth rate of Arctic vegetation. More carbon dioxide released from the ground can fuel greater plant growth, resulting in a larger uptake of carbon dioxide from the air and preventing its release into the upper atmosphere.
Temperature and moisture variations also dramatically affect any predicted carbon cycle.
So scientists at Toolik Field Station are manipulating the tundra environment to see how it could change as climate changes.
Test plots on the gentle hills around Toolik Lake are dotted with small greenhouses, shade covers, ground-warming devices. Some parcels are fertilized with nitrogen and phosphorus, enriched with lime and sulfur, fenced off from root-eating animals such as voles.
Some plots are treated with water containing a rare carbon isotope, enabling ecologists to trace the course of carbon cycling through land and water into plants and animals and the air.
Greg Starr of Florida International University is warming the soil electrically in early and late season to see whether heat can lengthen the limited Arctic growing season. (It does for some plants, but those that thrive may crowd out such desirable species as the snowflake-shaped reindeer lichen, the favorite winter food of caribou.)
On the Kuparuk River, Karie Slavic of the Marine Biological Laboratory strings a pump across the roaring waters to feed nitrogen fertilizer into the current, to determine how the nutrient is dispersed and which aquatic plants benefit from it.
Responses are often delayed and unpredictable. It took nine years of experimental fertilizing before river moss erupted over the streambed rocks; but it was all washed away by one violent summer flood.
Understanding how this sensitive ecosystem reacts to measured experimental alterations will help scientists cope with the vagaries of natural change -- and of human-caused greenhouse gases from the burning of petroleum, coal and other fossil fuels.