Three of six rings north of Manaus will bathe Amazon trees in 620 ppm of CO2 for a decade

Free-air enrichment means there is no chamber around the trees. Sensors on a central tower in each ring will read wind direction and CO2. In three of the rings, tubes on the outer towers will release CO2-enriched air at levels set by those readings ^[3]. The other three rings get no added gas and are the controls ^[4]. Both groups share the site, so a difference between them can be put down to the gas.

"Our atmosphere has about 420 parts per million (p.p.m.) of CO2. What we do is increase it to about 620 p.p.m. and see what happens," said David Lapola, the climatologist at the State University of Campinas whose team runs the experiment ^[18][6]. The added dose is 200 ppm, about 48 percent above today's air ^[20].

The team's main question comes after photosynthesis. They want to know whether the extra carbon is actually stored in trees, as the IPCC predicts ^[19]. Photosynthesis is the main route for carbon into the ecosystem, according to Lapola, but not the only one ^[19]. "It could be that [carbon] enters tree leaves and escapes through the soil," Lapola said. "If that's the case, we won't have the response the world expects." ^[7]

The form the carbon takes matters too. When soil nutrients are balanced, carbon taken from the air mostly becomes trunk tissue, a long-lived store ^[12]. "Phosphorus is essential in the construction of cell membranes, genetic material, energy molecules and complex carbohydrates that are structural to tree trunks," Lapola said ^[13]. Low soil phosphorus was cited as a possible reason the Sydney forest stored no extra carbon, and AmazonFACE is set up to test that factor as well ^[10].

Water is the second thread. With more CO2 in the air, stomata, the tiny pores on the backs of leaves, do not need to stay open as long to collect the carbon photosynthesis needs ^[14]. Lapola said that could protect trees from drought. It might also dry the local atmosphere, because the trees have less time to transpire ^[15]. "Multiply that process for a whole forest, and you disturb the hydrological cycle locally and beyond," he said ^[16]. An electrical device attached to a trunk measures water moving from the roots to the canopy ^[17].

The thing this does not tell you yet is the answer. CO2 release is due to begin within the next few months, and the project runs for a decade ^[2]. When results come, they will compare three treated rings with three controls ^[4]. They will come from a forest that Lapola describes as far richer in tree species than the temperate sites of earlier experiments ^[8]. I think three plots per group is a small denominator for a forest that varied, and early growth differences deserve caution until several years of data are in. Scaling a result from one site north of Manaus ^[1] up to the whole basin will be a job for climate models; the rings measure only the trees inside them.