Rainforests: Carbon Sink or Carbon Source?

Mornings here at La Selva Biological Station, Terry McGlynn counts bugs. He heads into the forest, maps out two one-square-meter grids and sticks a wire into the damp coating of dead leaves that line the ground of this jungle at the northern end of a volcanic national park in Costa Rica. He pokes at the earth with the wire in several spots, measuring the depth of the dead leaves, then crams them into a plastic bag and weighs the bag on a portable digital scale. On his knees and sweating in the sticky jungle heat, McGlynn empties the bag back onto the ground and hacks the contents into bits with a machete, 20 or 30 whacks, to chop up the twigs and other big pieces. This extracts the bugs that live inside.

McGlynn, an assistant professor at the University of San Diego, and his lab technician, Eduardo Lopez, take measurements from dozens upon dozens of sites throughout the forest and bring carefully chosen collections of leaf litter—everything that falls from the trees—back to the lab. It is a closet of a room, lined with beetle carcasses, feathers and seed pods; the walls feature intricate pencil drawings of bugs seen under a microscope. Here, the two sift the litter through a curious device made of shiny black fabric, which looks like a cross between a butterfly net and a magician’s hat. Next, the litter goes into another shiny-fabric contraption, this time white, that’s known as a Winkler and is handmade by a woman in Brazil. Sifted litter goes into the Winkler, things happen, and unlucky bugs—mostly springtails, ants and bark beetles—fall out the bottom and into a cup containing a 70% ethanol solution. McGlynn sorts these insects, then weighs and analyzes them.

McGlynn’s routine is part of an attempt to understand how animals affect the decomposition of dead plant matter in a tropical rainforest, which, in turn, is part of an effort to understand the way carbon moves through the forest. The project is spearheaded by a husband-and-wife team of erstwhile botanists named David and Deborah Clark, who, much to their dismay, may have come up with some really bad news for life on Earth.

Even small changes in the way this type of forest functions can have big impacts on global climate. Tropical forests cover 17% of the Earth’s land mass, but account for more than a third of the world’s plant growth and store roughly 40% of all the carbon in terrestrial life, plus a third or more of all the carbon stored in soils. “Tropical forests move more carbon in and out of the atmosphere than any other ecosystem,” says Alan Townsend, an ecologist at the University of Colorado who studies carbon and nutrient cycling at another site in Costa Rica. “They’re the United States of carbon dioxide emissions and uptake.”

For years, scientists assumed these ecosystems would save us from ourselves: that they’d soak up some of the extra CO₂ we’re adding to the atmosphere by driving our cars and powering our houses and cutting down other parts of the forest for wood and crops. It makes sense: If plants use CO₂ to grow, and there’s more CO₂ available, then the plants will grow more. This has been demonstrated in short-term experiments. The concept is called CO₂ fertilization, and it was built into many early models that tried to predict the future of the Earth’s climate.

“The problem,” says Jay Gulledge, senior research fellow at the Pew Center on Global Climate Change, “is we don’t have any strong evidence that there really will be or is a significant CO₂ fertilization effect—it’s just assumed. There are certain situations where you can show this in a greenhouse, but whether real ecosystems do this is questionable.”

To make a model that can accurately predict the impacts of climate change, it’s crucial to know whether tropical forests will in fact be sinks for carbon dioxide—net reducers of CO₂ in the atmosphere—or whether they’ll be sources, releasing more CO₂ than they absorb. The Clarks’ research hints at the idea of rainforests as CO₂ sources, a scary proposition that could speed up the effects of global warming. Armed with decades of data, the Clarks are now showing that enough global warming will slow the growth of tropical trees, which could cause rainforests to emit more carbon dioxide than they soak up, and by century’s end, trees in the tropical forests could die. Instead of saving us from the greenhouse effect, the world’s rainforests could contribute to it.

 Vinicio Paniagua Alfaro assembles, disassembles, moves and maintains the modular towers that are critical to measuring the nighttime exchange of CO₂ between the forest and the air. Credit: Beth Perkins

In 1979, recently out of grad school and looking for a real-world opportunity, the Clarks saw a job posting in the back of Science. They’d met as undergrads at the University of North Carolina, traveled together to the Galapagos with the Peace Corps, and completed PhDs, with matching National Science Foundation Fellowships, in the same lab at the University of Wisconsin. Intrigued by the ad, they offered themselves jointly for the position: director of La Selva Biological Station, a research facility in a 16-square-kilometer rainforest in Costa Rica, run by a consortium called the Organization for Tropical Studies. Shortly thereafter, the Clarks packed their bags for Central America.

At La Selva, the husband-and-wife ecologists shared the administrative tasks of running the station—an outpost of lime-green buildings straddling the Puerto Viejo River, midway between San Jose and the Caribbean coast. La Selva had been operating for just over 20 years, producing dissertations on leaf-cutter ants and neotropical frogs. The Clarks settled into life at the station, eating rice and beans twice a day in the cafeteria, and mingling with a constantly-rotating mix of biologists and grad students from around the world. In their spare time, they studied trees. Their goal was relatively simple, if long-term: to learn how different, ecologically speaking, six species of trees were from one another.

The Clarks began their landmark study, called “Trees,” in 1983, measuring more than 2000 individual trees once a year, keeping track of information like the tree’s physical condition, how much light its crown was getting and the status of the immediately surrounding forest. “It’s very hard to be a tree here in the understory,” says David Clark. He and his wife speak of La Selva’s particular trees as though they’re old friends. “They’re being plastered by things falling on them all the time.”

By the early 1990s, their data began to tell them something strange. “In a totally separate finding we weren’t even particularly interested in at the time,” recalls Deborah Clark, sitting in her office on the edge of the rainforest, “we discovered that adult trees showed big differences in how much they grow from one year to the next. And because they’re different species, what really stood out was they were telling us the whole forest was growing at different rates from one year to the next.”

Several years later, the Clarks’ 16 years’ worth of data—initially intended simply to help shed light on the form and function of trees—yielded even more surprising results, showing a clear relationship between the temperature in a given year and how much the trees grew. Even modest changes made a difference: Although there was only a 2° C difference in average temperature over the decade and a half the Clarks had been taking measurements, the connection between tree growth and temperature was clear. In the hottest year, the trees grew 34% less than the long-term average, and in the coolest year, they grew 33% more. The difference in the amount they grew in the warmest and coolest years was 90%.

The Clarks were floored. “Our mental state when we started was, one year is the same as the next,” Deborah says. “So this was a real shock.”

The disturbing implication is that global warming caused by increasing levels of atmospheric CO₂ might be impeding the growth of rainforests and initiating a “positive feedback loop”: CO₂ causes warming, which slows growth, which reduces CO₂ uptake, which causes warming. Even more worrisome, when the Clarks compared their results with data on carbon dioxide levels in the tropics as a whole—the research of a legendary climate scientist named Charles Keeling, who died last year—the graphs showed a clear correlation. The years the tropics emitted the most CO₂ were the hottest years for the tropical regions overall. And, the Clarks discovered, those were the same years in which the temperatures were highest at La Selva—and the same years in which tree growth at La Selva slowed down.

The conventional wisdom on tropical forests had been that because their climate can vary considerably from year to year, small changes in the average temperature shouldn’t make for a noticeable difference in the forest’s cycles. But the Clarks found otherwise. “We are now coming to understand that tropical forests may be being stressed by this ongoing climate change we’re causing,” Deborah says, “and if some predictions about this turn out to be true, the stress on tropical forests could, in fact, speed up global warming.” Tropical forests, she says, “may be a very major source of carbon dioxide emissions to the atmosphere. It could be a very big positive feedback. We may be at a point now where this has gone too far.”

The Clarks, says Gulledge of the Pew Center, “are actually digging into the black box a bit and pulling out some of the mechanisms.” By asking first what is happening, and second how it’s happening, he says, “the Clarks are refining our understanding of the carbon cycle, in order to see how it’s going to change.”

The Clarks’ research goals—which have turned La Selva into a Mecca for rainforest ecologists—have now widened into a grand scheme: to understand how tropical rainforests function and how they will behave as the climate changes. David is scrutinizing how trees in the forest reproduce, grow and die—and what affects these processes—with a series of projects that run from individual trees all the way up to the entire tropical landscape. They will help predict how tropical forests will behave in the future, under the impacts of anthropogenic global warming. Meanwhile, Deborah is honing in on carbon, coordinating the growing gaggle of scientists coming to La Selva to continue her research: to learn how carbon moves through a tropical forest, how this system is influenced by the availability of water and nutrients, and what this may ultimately tell us about how these forests will respond to a changing climate.

In the grass outside the La Selva dining hall, an iguana soaks up the sunlight after two days of unrelenting rain. Fist-sized songbirds in startlingly bright shades of blue, green, yellow and red flit about in the flowering bushes. Daytripping birdwatchers train their binoculars on a pair of chestnut-mandibled toucans perched in a treetop near the riverbank, as a turkey vulture scopes it all from on high.

Across the swaying suspension bridge and not far from the path leading into the forest, several peccaries, a particularly pungent type of wild pig, munch on leaves outside the Clarks’ cabin. “I love the pigs,” says David, who has just returned from a meeting in London. “I never get tired of watching them.”

“I love when they’re sunning themselves and they all pile on top of each other,” adds Deborah.

The Clarks’ travel schedule is often hectic—two days after David returned, Deborah was off to a meeting in the Southeast US, and then he was headed to a field site in Braulio Carillo National Park, on La Selva’s southern border, for a week. Perhaps as a result, the rainforest hasn’t lost its allure, even after 20-plus years; almost every afternoon, the couple walk in the forest together, talking—or arguing—about science.

“We try to be collegial,” says David, “but we’ve been married for 35 years, so you don’t have to be as careful as you would with someone who’s just a professional colleague.” Nevertheless, the couple has an unspoken rule against talking science at dinner. “That wouldn’t be good for your digestion,” David says.

At home in their cabin, David cooks dinner and Deborah does the dishes. In the mid-90s, the Clarks stepped down from their directorship of La Selva to devote all their time to research. Giving up the director’s house, they moved to a standard-issue researcher cabin—no kitchen, bathroom out the door and down the stairs. Three years ago, the Clarks paid for $30,000 of renovations to the house, which now has tile floors, an indoor bathroom, his-and-hers offices on separate floors and a small but functional kitchen. “We’d been out of the directorship almost 10 years, and it somehow occurred to me, we should cook!” says David. He subsequently stocked up on cookbooks, and is re-acquainting himself with the use of pots and pans.

One of many outbuildings on the jungle side of La Selva, this one stores equipment for climbing and maintaining the 40-meter towers. Credit: Beth Perkins

Upstairs, Deborah’s office houses the wine collection and an electronic keyboard and other instruments. Downstairs, David keeps the DVD collection, the stringed instruments—he plays guitar, mandolin, and fiddle, and she plays banjo—and some Costa Rican rocking chairs, one of the few tropical souvenirs found in the Clarks’ house. “When you live here,” says David, “you don’t vacation here. You wouldn’t have knick-knacks from your home town.” Deborah’s favorite vacation was a canoeing trip in southern Ontario—where, she laughs, “there are like 12 tree species.” (La Selva has more than 300.)

The Clarks have crafted a cozy world of domestic harmony in this tangled tropical landscape. They periodically host jam sessions at which scientists rock out on jungle evenings. The science is still paramount, of course, and despite—or because of—the couple’s important findings, there is much more work to be done.

Step six meters into La Selva’s forest, and you’re immediately confronted with the tangled, chirping, intertwined mess of life that is the rainforest. Howler monkeys whoop, woodpeckers knock at the trees so loudly it sounds like they’re using hammers, and the wings of hummingbirds whir like propellers. The air is thick, and tiny patches of blue sky and dappled sunlight barely make their way through the thick forest canopy.

La Selva and the surrounding forest boast more than 5,000 species of vascular plants, more than 400 kinds of resident and migratory birds, 56 species of snakes and 116 different mammals—among them tapirs, agoutis, capuchin monkeys, jaguars, the pigs and 65 different kinds of bats. There are thousands of species of arthropods: spiders, flying insects, 450 ant species. And countless other life forms call this lowland tropical rainforest home—all of them composed of carbon and taking in or emitting carbon dioxide.

Rainforests are biologically constrained: the vastly diverse species that live there have adapted to a narrow range of environmental conditions, like how much overall moisture is available. At La Selva, which gets more than four meters of rainfall per year, scientists have found that there’s less biomass when there’s less water, even though rainforests are always relatively wet. “It could be that some plants adapted to very wet conditions, and small variation makes a big difference,” says Javier Espeleta, a Costa Rican ecologist who works with the Clarks, studying carbon cycling in roots.

Other ecologists believe changes in rainfall patterns in the tropics could have as big an impact as temperature, if not bigger. “A lot of the precipitation that falls in those areas is generated internally,” says Cory Cleveland, an ecologist at the University of Colorado who works with Alan Townsend. “If those areas dry out because of land use change”—deforestation, which leads to changes in the way water moves around—“that leads to changes in precipitation on big scales.” To study this, Cleveland and Townsend are beginning a project in a national park in southern Costa Rica, building roofs over small forest plots in order to decrease the annual rainfall by 60%, to observe the effect on carbon cycling in the soil.

Figuring out how carbon moves into, through, and back out of a place like La Selva involves an almost infinite number of things that must be measured. There’s carbon below ground, in roots, soil and microbes; on the ground, in both the leaf litter and the detritivores, the animals that eat the litter; aboveground, in all the forest’s plants and the herbivores that consume them; and in the air. Each of these components is a macrocosm of other things that bear examination.

Take trees. Trees absorb CO₂ from the atmosphere during photosynthesis. They also respire, releasing carbon dioxide back to the air. Photosynthesis occurs in the leafy green parts of trees but respiration occurs all over: in the leaves but also in the stems, the branches, the roots. To understand the net balance of carbon in a forest, you need to measure or calculate all of the gas exchanges within it, and figure out the processes by which trees exchange nutrients with mycorrhizae, a type of fungus that lives symbiotically on roots. You need to discern the speed at which a particular tree grows, because trees use more carbon dioxide from the atmosphere when they’re growing faster; the way sunlight availability, temperature, moisture and soil nutrients affect photosynthesis and respiration; what role animals play; and more.

“There are three things I’m trying to find out,” says Deborah. “How much carbon is there in this forest? We’ve got a pretty good sense of that right now. How does it move around on a day-to-day and year-to-year basis, and how is that cycling affected by climate variation? And how can we learn enough about that to adequately build a model that will predict changing climate?”

To get answers to these questions, the Clarks have brought in an international team of ecologists, biologists, ecosystem modelers, botanists and entomologists. “It’s kind of like a religion,” Deborah jokes. McGlynn describes the project, gathered loosely under the umbrella name Carbono, as a “dreadnought,” the name of a hulking, 10-gunned British battleship. The Clarks, particularly Deborah, thrive on the collaboration. “I like working with lots of different people, with what [H.G. Wells] called ‘the world mind,’” she says. “That excites the heck out of me.”

The Carbono team has run experiments ranging from the mundane to the lofty. In 1997 they installed an “eddy flux tower,” an enormous aluminum contraption that reaches 42 meters to the rainforest canopy, 22 separate sections of two-meter-tall platforms and vertiginous stairs. Instruments on top measure the flow of gases between the forest and the air above. The tower, part of a network of towers around the world tracking this interaction in all types of different landscapes, measures a square kilometer of forest, and the results are extrapolated to the whole ecosystem. Frustrated that the forest’s dynamism was preventing the tower from producing complete data, David Clark hit on the idea of a mobile, modular tower that can show what static ones can’t; this initiative takes forest measurements in a way that’s unique in all of rainforest eco-science.

The Clarks and Steve Oberbauer, an ecologist at Florida International University in Miami, set up a series of 18 plots in the forest, where they began tracking carbon flow. The 50-by-100 meter plots are dotted throughout the forest to account for varying types of soil, plants and terrain geography. At each plot, chambers sunk in the earth help measure gas exchange between the soil and the air. Three-meter pits dug into the soil show how gases move around deeper underground. Mesh baskets on the jungle floor collect litterfall, which field technicians sort and weigh to see how much the forest is growing. Minirhizotrons, cameras sheathed in long plastic tubes and stuck into the dirt, photograph the roots and the fungi that grow on them—another snapshot of carbon moving through the system. The tower allows access to the rainforest canopy, where precisely calibrated instruments measure the exchange of carbon between the atmosphere and the forest.

Meanwhile, the insects Terry McGlynn collects are a window onto how carbon flows through litter-eaters. His role in the network of Carbono scientists is to visit La Selva several times a year to gather specimens, then return to San Diego to carry out his research on detritivores. (Lopez, his lab tech, is one of many native Costa Ricans who keep the various experiments running year-round, maintaining continuous field data.) Since they’re eating what falls from the trees, he says, “The carbon that makes up the bodies of the animals in the litter, comes from the litter.”

“You say litterfall and people roll their eyes, especially population people who study things like tapirs and jaguars,” says David Clark. “But the forest is built around the trees—it starts with the trees. One of the first things trees do when stressed is lose their leaves.” So understanding litter, he says, is “like having your finger on the most basic pulse of the forest.”

Despite the significance of tropical forests, they are so complex that scientists are only just beginning to understand them. There’s nothing to do but keep plowing forward with the research, and while the Clarks’ students, Deborah says, tend to worry that there’s too much work to do and the clock is ticking, she is, ultimately, optimistic. “I have many students who say, ‘We’re fried because nobody else has been doing these long-term observations, so we don’t know how generalizable these observations from La Selva are,’” says Deborah. “And I say, ‘We’re not fried. We have to be somewhat Zen about it.’”

It’s a slow process—figuring out the inner workings of a rainforest often involves devising experiments just to fill the holes in other experiments—and the forest may already be reaching a threshold. “Our working hypothesis right now,” says Oberbauer, who set up the original Carbono study with the Clarks and now runs the tower research, “is that trees are getting too warm to photosynthesize. The temperature optimum for these things is not very high.”

But there’s not enough data yet to know for certain. “There is so much that we don’t know,” says Deborah, navigating Costa Rica’s chaotic roads in the couple’s SUV on the way to their pied-à-terre in the capital. “We will never have perfect knowledge,” she continues. “But I feel that the science is getting stronger and stronger.”

Many plants have been shown to acclimate to gradually higher temperatures, continuing to photosynthesize and grow as the temperature rises. But at some point, Deborah believes, the temperature will reach a critical limit and photosynthesis rates will crash. Still, respiration could continue at its normal rate, meaning plants would be emitting stored carbon—not a desirable situation for the world’s climate.

“It seems really clear to me, when I read about modeling of what’s going to happen in the tropics,” she says, “that the focus is too heavily weighted on drought and not taking into account the huge impact that is likely to come from the rate at which temperature is increasing. One reason is that we don’t yet have a clear enough sense of the threshold at which global warming is going to hit like a sledgehammer on tropical forests.”