The Biosphere 2 rainforest greenhouse, built in Arizona in the late 1980s, contains 90 plant species.

ALPHA STOCK/ALAMY STOCK PHOTO

Unprecedented drought in an artificial ecosystem may reveal how rainforests will cope with climate change

Earlier this month, the doors to the tropical rainforest, enclosed under a ziggurat of glass, were sealed shut. Christiane Werner turned a valve to release about $12,000 worth of carbon dioxide (CO2) spiked with carbon-13, an isotope that is normally scarce in the atmosphere. The luxuriant plants inside Biosphere 2, a 30-year-old set of greenhouses and artificial ecosystems in the Arizona desert, soaked up the isotopic tracer, enabling investigators to follow the flows of carbon through the healthy forest. Werner, an ecosystem physiologist at the University of Freiburg in Germany, and her team gathered these baseline data for the harsh test to come: the largest forest drought experiment ever monitored with isotopes. "It will be amazing to see the results," says Tamir Klein, a plant ecophysiologist at the Weizmann Institute of Science in Rehovot, Israel, who is not involved.

On 7 October, the researchers shut off the sprinklers that irrigate the rainforest, beginning a 6-week drought. Next month, they will inject another pulse of isotopically enriched CO2 into the densely instrumented ecosystem, and apply other tracers. A forest's consumption of CO2 slows during drought, but scientists haven't pinned down how thirsty rainforest plants—especially large trees—use and release their stored carbon. The answers are important for the global climate cycle, Klein says. Droughts, expected to become more severe as the climate warms, could turn tropical forests from sinks of greenhouse gases into sources that accelerate climate change.

Field experiments in the Amazon, in which plastic panels intercept rain to keep large swaths of forest dry, have sketched out how drought kills trees of different sizes. Smaller studies targeting individual plants with isotopic tracers have revealed some of the impacts on plant function. But the Biosphere 2 experiment will do both by applying tracers across an entire forest. "We have an ecosystem in a lab," Werner says.

The $150 million Biosphere 2 was built in the late 1980s as a kind of spaceship on Earth, in which humans would attempt to survive inside a sealed ecosystem. That mission flopped, but the University of Arizona now operates the facility for research, education, and tourism. It has hosted large ecology studies and an ongoing $3 million experiment in landscape evolution. Biosphere 2's original funder, financier Edward Bass, helped support that earlier work, but much of the new experiment is funded by part of a €1.9 million grant Werner won from the European Research Council. About 50 researchers from 13 institutions are contributing equipment and expertise.

The focus is Biosphere 2's tropical forest, which includes some 90 plant species across an area the size of seven tennis courts. All summer, the team prepared by building canopy platforms where they could enclose dozens of leaves and stems in small chambers to capture their emissions. They drilled into tree trunks to insert probes, and dug observation pits to measure emissions from soil and roots. Four kilometers of tubing carry gases from the probes to a room full of instruments. "The scale of measurements on this drought is completely unparalleled," says co-leader Laura Meredith, a biogeochemist at the University of Arizona in Tucson and director of rainforest research at Biosphere 2.

By tracking the carbon-13, the researchers will learn how quickly carbon is taken up during photosynthesis and then moves through the forest. They will compare those rates before and during the drought across six tree species that differ in their drought resistance. And they will learn how the trees apportion stored carbon in their leaves, trunks, and roots. It's a "huge black box," and crucial for predicting how plants respond to stresses like drought, says plant physiologist William Anderegg of the University of Utah in Salt Lake City.

Another set of tracers will show in finer detail how particular metabolic pathways use carbon. During the past month, the researchers have supplied a solution of isotopically enriched pyruvate, a chemical building block used in many biological processes, to leaves, roots, and clumps of soil. One type of pyruvate tracer reveals how much carbon is given off during daytime respiration—a key part of the carbon cycle that needs to be better quantified, Werner says.

Another pyruvate tracer, taken up into a different pathway, shows how much carbon the plants and soil microbes use to synthesize volatile organic compounds (VOCs). When plants are stressed, these chemicals make up a considerable fraction of their carbon emissions. They can warm the atmosphere or turn into aerosols that cool it, but their overall climatic effect is unknown. Plants use VOCs for many purposes, including as a homing signal for a vast web of soil fungi that provide water and nutrients to roots during drought. The researchers hope to quantify rates and amounts of VOCs exchanged between the microbes and plants and whether they change during drought.

At the end of the drought, the researchers will perform one last tracer experiment, irrigating the deep soil with water enriched in an isotope of hydrogen. They expect large trees to take up most of the water, and they hope to learn whether their deep root systems will leak some of the water into the shallow soil, helping smaller plants recover.

Finally, the sprinklers will turn on and return the ecosystem to normal. When parched soil and fallen leaves are rewetted, microbes go into metabolic overdrive and churn out CO2 and VOCs. Meredith and her colleagues will measure emissions and link them to patterns in microbial genes.

Ultimately, results from the drought test will improve the way global climate models account for vegetation. "You need these experiments to unlock the physiology and add it into the models," Anderegg says. "It gets us much more mechanistic and rigorous projections of how tropical trees and forests might respond to climate change."

After the experiment wraps up, tourists will be let into the rainforest again. But the canopy platforms will remain for future research, and some of the carbon tracers will also stick around. "We can look for the signal for years to come," Meredith says.