Trees use water more efficiently as atmospheric carbon dioxide rises

Studies have long predicted that plants would begin to use water more efficiently as atmospheric carbon dioxide levels rose. A research team led by Research Associate Trevor Keenan and Assistant Professor of Organismic and Evolutionary Biology Andrew Richardson, however, has found that forests across the globe are becoming more efficient than expected.

Using data collected from forests in the northeastern US and elsewhere around the world, Keenan and Richardson, found increases in efficiency larger than those predicted by even the most state-of-the-art computer models. The research, which was done in collaboration with researchers from the Harvard's Department of Earth and Planetary Sciences, the USDA Forest Service, Ohio State University, Indiana University, and the Karlsruhe Institute of Technology in Germany, is described in a July 10 paper in Nature.

"This could be considered a beneficial effect of increased atmospheric carbon dioxide," said Keenan, the first author of the paper. "What's surprising is we didn't expect the effect to be this big. A large proportion of the ecosystems in the world are limited by water -- they don't have enough water during the year to reach their maximum potential growth. If they become more efficient at using water, they should be able to take more carbon out of the atmosphere due to higher growth rates."

While increased atmospheric carbon dioxide may benefit forests in the short term, Richardson emphasized that the overall climate picture would remain grim if levels continue to rise.

"We're still very concerned about what rising levels of atmospheric carbon dioxide mean for the planet," Richardson cautioned. "There is little doubt that as carbon dioxide continues to rise -- and last month we just passed a critical milestone, 400 ppm, for the first time in human history -- rising global temperatures and changes in rainfall patterns will, in coming decades, have very negative consequences for plant growth in many ecosystems around the world.."

How do increasing carbon dioxide levels lead to more efficient water use?

The answer, Keenan said, is in the way photosynthesis works. To take in the carbon dioxide they need, plants open tiny pores, called stomata, on their leaves. As carbon dioxide enters, however, water vapor is able to escape.

Higher levels of carbon dioxide, however, mean the stomata don't need to open as wide, or for as long, meaning the plants lose less water and grow faster. To take advantage of that fact, Keenan said, commercial growers have for years pumped carbon dioxide into greenhouses to promote plant growth.

To test whether such a "carbon dioxide fertilization effect" was taking place in forests, Keenan, Richardson and others turned to long-term data measured using a technique called eddy covariance. This method, which relies on sophisticated instruments mounted on tall towers extending above the forest canopy, allows researchers to determine how much carbon dioxide and water are going into or out of the ecosystem.

With more than 20 years of data, the towers in the Harvard Forest -- which have the longest continuous record in the world -- are an invaluable resource for studying how forests have responded to changes in atmospheric carbon dioxide levels. Though more than 300 towers have since sprung up around the globe, many of the earliest -- and hence the longest data record -- are in the northeast U.S.

When Keenan, Richardson and colleagues began to examine those records, they found that forests were storing more carbon and becoming more efficient in how they used water. The phenomenon, however, wasn't limited to a single region -- when they examined long-term data sets from all over the world, the same trend was evident.

"We went through every possible hypothesis of what could be going on, and ultimately what we were left with is that the only phenomenon that could cause this type of shift in water-use efficiency is rising atmospheric carbon dioxide," Keenan said.

Going forward, Keenan (now based at Macquarie University in Sydney, Australia) said he is working on a proposal to get access to data collected from even more sites around the world, including several that monitor tropical and Arctic systems.

"This larger dataset will help us to better understand the extent of the response we observed," he said. "That in turn will help us to build better models, and improve predictions of the future of the Earth's climate. Right now, all the models we have under-represent this effect by as much as an order of magnitude, so the question is: What are the models not getting? What do they need to incorporate to capture this effect, and how will that affect their projections for climate change?"