Amazon during 2005 drought

The Amazon forest is widely considered by scientists to act as a ‘carbon sink’, absorbing from the atmosphere about 1.5 billion tonnes of carbon dioxide a year and offsetting much of the emissions from deforestation. This is a significant amount, equivalent to a quarter of the USA’s emissions, estimated at 5.4 billion tonnes. However, recent events suggest that the forest may be losing this useful function. In 2005 and 2010 the Amazon suffered from two devastating droughts, in which millions of tress died, resulting in major emissions from the rotting vegetation. Dr Simon Lewis, a forest scientist from the University of Leeds, was the lead author in an article published in the journal Science in February 2011, which made an initial assessment of the impact of the 2010 drought. Sue Branford interviewed Dr Lewis for LAB.*


S_LewisDr Simon LewisScientists used to think of the Amazon forest as carbon-neutral, emitting as much as carbon dioxide as it absorbed. Why has this view changed?

All plants in the Amazon forest take in carbon from photosynthesis and lose it through respiration and tree mortality. Scientists used to believe that these processes were in equilibrium, with a zero net change in carbon storage. But over recent decades there has been a steady rise in the amount of carbon stored in the Amazon forest. The most likely reason is the change in the composition of the earth’s atmosphere, with the increase in carbon dioxide. This carbon dioxide may be allowing plants to photosynthesise and grow more, and thus to store more carbon per unit area than in the past. After measuring 100,000 trees across the Amazon basin at regular intervals, we calculate that in a normal year about 1.5 billion tonnes of extra carbon dioxide are absorbed by the Amazon forest.

What happens if there is a drought in the Amazon?

If there is a drought and it is long enough for some plants and trees to die, then the dead vegetation returns carbon dioxide to the atmosphere and the forest changes from being a carbon sink to being closer to carbon neutral or a carbon source. Let’s take the last two severe droughts as an example. In 2005 there was an unusual drought affecting the south-western area of the Amazon basin, which was classified as a ‘one in one hundred years’ event. We went out into the forest and measured how many trees had died and which species they were, and then calculated the amount of carbon we can expect to go into the atmosphere when they rot. We calculated that about five billion tonnes of carbon would be released into the atmosphere, not all at once but stretched out over several years. Then in 2010 there was another, even more extreme, drought. It was extremely surprising for the 2010 drought to have come so soon after 2005 and to have been more severe. In 2005 some 37% of the Amazon basin that was drier than in an average year but in 2010 it was 67%. We made some first calculations, based on our earlier studies, and estimate that about 8 billion tonnes of carbon will eventually be released.

Have you any idea why the forest suffered another devastating drought so soon after the one in 2010?

The 2005 and 2010 droughts coincided with high Atlantic sea surface temperatures. You get rain where the northern hemisphere winds and the southern hemisphere winds meet. And if you have warmer temperatures, then this meeting can be displaced northward and you get droughts in the southern part of the Amazon. So these droughts are consistent with some of the most sophisticated climate change models, which, indeed, have been predicting droughts in the early 21st century.

Are these droughts drying out the forest?

There is no sign yet that the forest is drying out. Most scientists think that warmer temperatures will lead to more evaporation and therefore more rainfall. The important thing to realise about these events is that they are just periods of absence of rain. In 2005 and 2010 the total amount of rain, even in the drier parts of the forest, was about the same as the average over the whole year. It’s just that rain was missing at the driest time of the year. The problem was the distribution of rain across the year, not the total amount.

Is there any evidence that the damage done by a later drought is exacerbated by the damage already done to the forest by an earlier drought?

We don’t know yet. We have just got funding to look at the impact of the 2010 drought. It may be that more trees died in areas that were affected by the earlier drought, because they were weakened by the drought. Or it may be that fewer trees died in these areas because those susceptible to drought were killed earlier, leaving only drought-resistant trees. So there’s a lot of uncertainty but by the end of the year we should have a much better idea.

Could the Amazon forest go from being an absorber of carbon to an emitter of carbon?

This could happen, if the Amazon starts to suffer from three or more severe droughts per decade. Worse still, we could set into motion a ‘positive feedback’ mechanism, by which the droughts could lead to greater emissions of carbon dioxide, further exacerbating the problem. However, we can’t discount the possibility that 2005 and 2010 were merely caused by a decadal-scale natural cycle and that we may move to another phase, without these types of droughts. But, if we wait for the answer and find we have a positive feedback, then it will be too late to reduce fossil fuel emissions to deal the problem. Put another way, we risk playing Russian roulette with the world’s largest rainforest.

*A version of this interview has been published in the magazine, EcoAmericas.


Listen to the interview