Published online 5 March 2009 | Nature | doi:10.1038/news.2009.136

News: Briefing

Climate change crisis for rainforests

Drought could turn carbon sinks into sources.

rain forestWill rainforests still absorb CO2 in future?Punchstock

The tropical forests of South America, Africa and Asia take up and release huge amounts of carbon each year. On the whole, they are a significant 'sink' for atmospheric carbon dioxide, but their future role in sequestering the greenhouse gas is uncertain. If rainforests are hit by serious drought, as they were in the Amazon basin during 2005, they could turn into a carbon 'source' sooner than we thought. So, are we in danger of losing our closest allies in the fight against climate change?

How is climate change affecting the growth of the forests?

Atmospheric CO2 levels are now 40% above what plants were experiencing just a century or so ago. Plants may have benefited from the availability of extra CO2, which they convert, through photosynthesis, into biomass. But plant growth benefits from elevated CO2 levels only up to a point, and more negative aspects of climate warming may still be ahead.

The biggest worry is drought. Scientists who have for the first time determined the drought sensitivity of a tropical forest report in Science1 today alarming results from the Amazon basin: the unusual 2005 drought there has apparently turned some of the affected areas of the Amazon from a carbon sink to a carbon source.

A comparison of plots that were monitored regularly before and after the drought revealed that forest patches subjected to a 100-milimetre decrease in rainfall released on average 5.3 tonnes of carbon per hectare as trees in the area died.

Basin-wide, between 1.2 billion and 1.6 billion tonnes of carbon were released as a result of the intense dry season and weakened wet season during 2005, the team estimates. The exceptional growth in atmospheric CO2 concentrations in 2005 may actually have been caused by these releases.

So does climate change mean that rainforests will not be carbon sinks in the future?

That's not clear, because current climate models are not very good at simulating rainfall. The formation and distribution of clouds and precipitation are controlled by atmospheric processes that occur on smaller scales than existing climate models can resolve. As a result, climate models reproduce observed temperatures reasonably well but diverge rather wildly when it comes to rainfall, and particularly so in the tropics. Projections of rainfall must therefore be taken with a pinch of salt.

Nonetheless, many scientists do strongly suspect that, in a warmer climate, dry conditions such as those of 2005 will become more frequent in the Amazon region and around the tropics. If they are right, tropical forests could gradually cease to act as a solid buffer against climate change.

How large a carbon sink are the world's tropical forests at the moment?

Scientists estimate that mature tropical forests, which cover about 10% of Earth's land, take up as much as 1.3 billion tonnes of carbon per year. This is a substantial amount, equivalent to almost 20% of carbon emissions from fossil-fuel burning. Tropical forest thus accounts for around 40% of the global terrestrial carbon sink.

Leaf from a young sapling, dying after droughtLeaf from a young sapling, dying after the 2005 drought in Columbian Amazonia.Peter Vitzthum

The good news is that undisturbed old forests keep getting better at sequestering carbon from the atmosphere. Over the past couple of decades, mature tropical forests in Africa and South America seem to have taken up an extra 0.6 tonnes of carbon per hectare each year on average2,3. Tropical forests in Asia are likely to have improved their carbon uptake as well, although probably at a lower rate.

How reliable are these figures?

Measuring tree growth is notoriously difficult, not least because tropical observation networks are pitifully few, particularly in Africa. Problems related to plot selection, comparability and converting tree-diameter measurements to carbon content have led to an intense debate about the size and fate of the tropical (and global) terrestrial carbon sink. Given the many uncertainties, forests have been excluded from national carbon budgets under the 1997 Kyoto Protocol on Climate Change.

However, data gathered over the past decade suggest that undisturbed old-growth forests — in and outside the tropics — do indeed continue to grow and accumulate carbon. There is little doubt that tropical forests have acted as a substantial carbon sink for at least the past couple of decades. Old-growth boreal forests, which were long suspected to be carbon-neutral, have recently been found to keep accumulating carbon as well4.

How long will old forests continue to get better at taking up CO2?

That is a key question. Deforestation and forest degradation, through logging, clearing and fire, are only the most obvious problems. Between 2000 and 2005, South America and Africa have each lost around 4,000 square kilometres of forest annually. But even undisturbed forests cannot continue to grow for ever. Their accelerated growth in recent decades is probably a temporary phenomenon, explained either by the fertilization effect of elevated CO2 levels or by the fact that they are still in the process of growing back from major disturbances in past centuries.

What does all this mean for forest management and the politics of climate change?

Climate change and deforestation pose a double threat to rainforests. Keeping alive large amounts of forest will require big areas to remain undisturbed from logging and clearing. Fragmented forest areas are more vulnerable and more likely to be overrun by climate change.

"These forests have given us a subsidy for a long time, but this cannot be taken for granted," says Oliver Phillips, an ecologist at the University of Leeds, UK, who coordinates the Amazon Forest Inventory Network, which was responsible for the latest study. "So when putting a carbon value on them we'd rather be conservative." 

  • References

    1. Phillips, O. L. et al. Science 323, 1344-1347 (2009).
    2. Lewis, S. L. et al. Nature 457, 1003-1006 (2009).
    3. Phillips, O. L. et al. Science 282, 439-442 (1998).
    4. Luyssaert, S. et al. Nature 455, 213-215 (2008).
Commenting is now closed.