Cheaper than printing it out: buy the paperback book.

"I feel I am far out in space," said Roy Walford, one the people who lived inside Biosphere 2. Walford was speaking to a reporter via a video hookup during the first two-year closure of the ark, from September 26, 1991, to September 26, 1993. During that time eight humans, or biospherians as they are called, dramatically removed themselves from the direct touch of all other life on the planet, and from all the affirming flows of materials propelled by life, and lived instead in the tiny autonomous backwater swirl of life they had conjured up in a miniature surrogate Gaia. They could have been in space.

Walford was healthy but extremely skinny and underfed. For two years, all the biospherians were hungry. Their pocket-size farm had been plagued with insect infestations. Because they couldn't spray the beasties with poisons -- since they would be drinking the evaporated runoff later in the week -- they ate less. At one point the desperate biospherians crept down their rows of potato plants with portable hair dryers to drive the mites off the leaves, but without success. Altogether they lost five staple crops. One of the biospherians plummeted from 208 to 156 pounds. But he was prepared for this. He brought in clothes several sizes too small at the start.

Some scientists felt starting the Bio2 project with humans inside was not the most productive way. Peter Warshall, their consulting naturalist, said, "As a scientist, I would have preferred that we closed the whole thing up for one year with only the first two or three kingdoms in it: unicellular organisms and below. We could have seen how much the microbial cosmos controls the atmosphere. Then later we'd put everything in, close it up for the next year and compare the fluctuations." A few scientists felt the troublesome and difficult-to-support species of Homo sapiens shouldn't be in there at all, and that the humans became a mere entertainment factor. But many were sure the ecological study was pointless compared to the practical goal of developing technologies of human survival away from the Earth. To review the conflicting views of the scientific import and agenda of the project, an independent Scientific Advisory Committee was commissioned by Bio2's financier Ed Bass. They issued a report in July 1992 which acknowledged the dual nature of the experiment. It stated:

The committee recognizes that there are at least two major areas of science to which Biosphere 2 can contribute significantly. One is the understanding of biogeochemical cycles of closed systems. From this perspective Biosphere 2 represents a much larger and more complex closed system than has ever been studied. For these studies the presence of human beings in the system is not essential except that they provide the capacity to make observations and measurements not initially regarded as important. The second is to gain the knowledge and experience to maintain humans within equilibrium in a closed ecological system. For these the presence of people is central to the experiment.

As an example of the latter case, within the first year people living inside the closed system yielded a completely unexpected medical result. Regular blood tests of the sequestered biospherians showed increased levels of pesticides and herbicides in their blood. Since every aspect of the environment within Bio2 was monitored constantly and precisely -- it was the most monitored environment of all time -- scientists knew that there were no pesticides or herbicides anywhere inside. One biospherian who had previously lived in third world countries had traces in her blood of a pesticide banned in the U.S. twenty years ago. What the medics guessed was that as the biospherians lost significant weight due to their restricted diet, they burnt up fat reserves stored in the past and flushed out toxins deposited in them decades ago. Until Bio2 was built, there was no scientific reason to precisely test people for internal toxins because there was no way to rigorously control what they ate, drank, breathed, or touched. Now there was. Just as Bio2 provided an experimental lab for meticulously tracking the flow of pollutants through an ecosystem, it also provided a lab for meticulously tracking the flow of pollutants through a human body.

Human bodies themselves are a vast complex system -- despite our advanced medical knowledge, still unmapped -- which can only be properly studied by isolating them from the greater complexity of life. Bio2 was an elegant way to do this. But the Science Advisory Committee missed another reason to have humans aboard, perhaps one of equal importance to getting ready for space. This justification was matter of control and scaffolding. Humans were to serve as the "thumb on the way to thought," the chaperone present at the introduction, but not needed past that. People were not necessary for a closed ecosystem to run once stable, but they might be helpful in stabilizing it.

For instance, there was the practical matter of time. No scientist could afford to run the emerging ecosystem for years and let it crash whenever it wanted, only to have to start over. As long as the humans inside measured and recorded what they did, they could steer the closed system away from the precipices of disaster and still be scientific about it. Within great latitudes, the artificial ecosystem of Bio2 ran its own course, but when it veered toward a runaway state, or stalled, the biospherians nudged it. They shared control with the emergent system itself. They were copilots.

One of the ways the biospherians shared control was by acting as "keystone predators" -- biological checks of last resort. Populations of plants or animals that outran their niches were kept in reasonable range by human "arbitration." If the lavender shrub began to take over, the biospherians hacked it back. When the savanna grass shouldered out cactus, they weeded fiercely. In fact the Biospherians spent several hours per day weeding in the wilderness areas (not counting the weeding they did on their crop plots). Adey said, "You can build synthetic ecosystems as small as you want. But the smaller you make it, the greater role human operators play because they must act out the larger forces of nature beyond the ecological community. The subsidy we get from nature is incredible."

Again and again, this was the message from the naturalists who assembled Bio2: The subsidy we get from nature is incredible. The ecological subsidy most missing from Bio2 was turbulence. Sudden, unseasonable rainfall. Wind. Lightning. A big tree falling over. Unexpected events. Just as in a miniature Ecosphere, nature both mild and wild demands variance. Turbulence is crucial to recycle nutrients. The explosive imbalance of fire feeds a prairie or starts a forest. Peter Warshall said, "Everything is controlled in Bio2, but nature needs wildness, a bit of chaos. Turbulence is an expensive resource to generate artificially. But turbulence is also a mode of communication, how different species and niches inform each other. Turbulence, such as wave action, is also needed to maximize the productivity of a niche. And we ain't got any turbulence here."

Humans in Bio2 were the gods of turbulence and the deputies of chaos. As pilots responsible for co-controlling the ark, they paradoxically were also agents provocateurs responsible for staging a certain amount of out-of-controllness.

Warshall was in charge of creating the minisavanna within Bio2 and its miniturbulence. Savannas, said Warshall, have evolved in conditions of periodic disturbance and require a natural kick every now and then. Any savanna's plants need a jolt by being burnt to the ground by fire or grazed by antelope. He said, "The savanna is so adapted to disturbances that it can not sustain itself without it," and then joked about putting a sign in the Bio2 savanna that says "Please Disturb."

Turbulence is an essential catalyst in ecology, but it was not cheap to replicate in a man-made environment like Bio2. The wave machine that sloshed the lagoon water was complicated, noisy, expensive, endlessly breaking down, and after all that, only made tiny highly regular waves -- minimal turbulence. Huge fans in the basement of Bio2 pushed the air around for some semblance of wind, but it hardly moved pollen. Pollen-moving wind would have been prohibitively expensive to manufacture. And fire would have smothered the humans with captive smoke.

"If we were really doing this right, we would be piping in thunder for the frogs, who are stimulated to reproduce by rain splatters and thunder," said Warshall. "But we are not really modeling the Earth, we are modeling Noah. In reality the question we are asking is, How many links can we break and still have a species survive?"

"Well, we haven't had a crash yet!" Walter Adey chuckled. Both his analog coral reef in Bio2 and his analog swamp at the Smithsonian (which gets a thunderstorm when someone turns a gushing water hose onto it) thrived despite the sustained shock of being isolated and closed off from the big subsidies of nature. "They are hard to kill, given reasonable treatment. Or even occasional unreasonable treatment," Adey said. "One of my students forgot to remove a certain plug from the [Smithsonian] swamp one night, which flooded the main electrical panel with saltwater, which blew up the whole damn thing at 2 A.M. It wasn't until the next afternoon that we got its pumps running again, but it survived. We don't know how long we could have been down and still lived."