News•February 6, 2013

Study Downplays Risk of Catastrophic Amazon ‘Dieback’

By Lauren Morello

In a warming world, tropical forests may be hardier than previously thought.

For scientists who study the Amazon, the worst-case scenario has long been clear. As the planet warms, some models suggest, the rainforest will dry and die, sending a massive shot of carbon into the atmosphere to further warm the planet.

That risk now appears to be smaller than researchers feared, according to a study published Wednesday in Nature. It estimates that for every degree Celsius of warming, the Amazon and other tropical forests will release 53 billion metric tons of carbon.

The view from above the Amazon rainforest canopy at Cristalino Lodge, Alta Floresta, Mato Grosso, Brazil.
Credit: Unboxed Media

That’s a big number. But it’s well below the threshold of 100 billion tons carbon per degree warming that seems to pop up in models that project climate change will cause a catastrophic “dieback” in the Amazon.

“It’s a bit of good news, really,” said lead author Peter Cox, a climate modeler at the University of Exeter, though he emphasized that’s a relative term. “Even after our study, we think that climate change is quite damaging,” he said.

The new analysis places the risk of forest dieback this century at just 0.24 percent — with conditions. The projection depends on the notion that plant growth will increase as atmospheric carbon dioxide levels rise, which would suck up more of the greenhouse gas. And it means that effect would have to keep pace with the release of carbon from stressed trees and soils as the planet warms, drying tropical landscapes.

The conclusion surprised Cox, who in 2000 published one of the earliest papers sounding the alarm about the risk of dieback in the Amazon. The dramatic projections seemed prescient when the Amazon was hit by major droughts twice in the past decade, in 2005 and 2010.

“The concerns that we had, based on our work and other people’s work, is that the Amazon forest in particular could be vulnerable to climate change,” Cox said. “That’s what really motivated the study.”

The analysis draws on year-to-year observations of atmospheric CO2 levels as well as a full spread of model projections of tropical forests’ fate in a warming world.

Models that showed the most dramatic forest losses under climate change also projected large year-to-year swings in the amount of carbon dioxide the forests stored or absorbed — a consequence of temperature changes, drought and the presence of El Niño or its counterpart La Niña.

NASA satellite data show the extent of the 2005 megadrought in the western Amazon rainforest, with the hardest-hit areas in red and yellow in the map at left. The map on the right shows the forest's slow recovery in subsequent years, with red and yellow areas slowest to rebound.
Click to enlarge. Credit: NASA/Jet Propulsion Laboratory-Caltech/Goddard Space Flight Center

In contrast, models that showed smaller year-to-year swings in CO2 uptake — more in line with real-world measurements of forests’ carbon appetite — also forecast less forest loss.

With that knowledge in hand, the researchers were able to estimate future carbon loss per degree of warming.

“It’s a clever approach, and a useful thing to do,” said Yadvinder Malhi, a tropical ecologist at Oxford University who was not involved with the new study. “It’s nice to have another insight rather than just looking at the spread of models and saying, ‘They’re all disagreeing — what do they know?’”

But Malhi, whose own recent work has revealed that the 2005 drought caused damage to the Amazon forest canopy that lingered for years, leading into the 2010 drought, said he’s not sure the method accounts for all the factors that influence long-term forest health.

“There’s a need for more data on [trees’] physiologic response” to stress, he said, and to understand how the mix of species, the risk of fire, and the ability of the forest to rebound from repeated fire and drought may change as the temperature rises.

James Randerson, an Earth systems scientist at the University of California, Irvine, said the new paper represents “a very nice approach for limiting some of he uncertainty,” but cautioned that models can’t simulate important human responses to the changing forest.

His research suggests that in the rainforests of equatorial Asia, residents increase their use of fire to clear forests during dry years, driving up carbon loss. And other studies have shown that in the Amazon, developing road networks that reach into the forest can serve as ignition points for wildfires during times of drought.

Since 2004, deforestation has claimed nearly 43,000 square miles of Amazon rainforest, an area the size of Virginia, according to Brazilian government data, although the rate of loss has slowed dramatically in recent years.

Cox acknowledged that deforestation remains a major — and uncertain — influence in the future of tropical forests like the Amazon.

“It’s an open question whether the combination of deforestation and climate change could push the remaining forest over the brink,” he said. “Watch this space on both issues.”