« Open Source Warfare | Main | I, For One, Welcome Our New Post-Singularity Overlords »

Amazon Carbon Cycling

At first blush, this doesn't seem like terribly good news, but it does give us a bit more information to use when thinking about our climate response options. Researchers studying the carbon cycle in the Amazon River basin (particularly the uptake and release of CO2 by the region's plants) have found that, rather than sequestering the CO2 for decades or centuries, the carbon is cycled out back into the atmosphere within five years. Among the implications of this discovery (assuming it's confirmed) is that the Amazon rain forests are not good candidates for long-term capture of CO2, and that we need to pay close attention to the actual carbon cycle for each biome before assigning sequestration targets.

Comments (10)

Reminds me of this:

Hydroelectric dam's fluctuating water levels = decomposing vegetation and CO2 release. Perhaps a similar mechanism in the amazon?


Sort of makes you wonder about the legitimacy of off-sets. Seems like we have to get the system worked out before we can proclaim climate neutrality.

Whoa! Let's be careful here folks.

(I work in the Amazon basin on climate and carbon cycling issues, so I happen to have some of the background information on this one. So let me try to clarify this point.)

These guys have basically measured the "age" (using radioisotope tracers that tell you when the carbon was fixed from the atmosphere and put into plant tissues through photosynthesis) of CO2 that is being released by the rivers of the Amazon basin. Interestingly, there is fair amount of carbon dioxide coming out of the Amazon river, into the atmosphere, that was previously assumed to be safely tucked away in the forests. And now we learn, surprisingly, that the average age of that carbon is about 5 years. That means that the carbon found in the rivers of the Amazon took about 5 years to get there, from the point it was first taken up by some plants in the basin.

That would seem to indicate that the system holds onto carbon for only a little while.

But that's not quite right. Essentially these results show that only *some* of the carbon fixed by plants in the Amazon (particularly in the cerrado, or savanna-like, regions) seems to be flushing into the Amazon river, and then back out to the atmosphere, quite rapidly.

But it does *not* necessarily mean that the whole carbon cycle of the basin is moving this fast, or that long-term carbon sinks aren't possible in the Amazon.

In fact, the vast majority of carbon fixed in the Amazon rainforests does not end up in the rivers. Only a small fraction of it does. And it's probably the organic matter leaching from grassland and cerrado soils (which also turnover pretty fast) that is the primary source of this river carbon.

Furthermore, we are finding that there are some rather old carbon stocks in the basin as well. Many trees, for example, can live there for decades -- if not a century or two. The soils can also hold onto the carbon for a while.

The bottom line is this: the carbon cycle of the Amazon has a whole bunch of different components -- some of them are fast (as shown in this study) and some are slow (as shown in others). The majority of them are far slower than 5 years, and that's where most of the carbon is going. Planting (or protecting) trees in the Amazon will still have a significant effect on the CO2 levels of the atmosphere.

But Jamais is completely right that we need to improve our understanding of these systems before prescribing their potential as carbon sinks. And it does appear that some of the carbon we thought might be locked-up in the Amazon rainforest is not. Instead, some (but only a small part) of it is leaching quickly into the rivers, where it is then burned off into the atmosphere.

Nature contiues to surprise us!

p.s. And the whole idea of using forests as carbon sinks continues to be a difficult science and public policy question. We may ultimately need every carbon sink we can get our hands on, but it is important to ask whether this is being done *in addition* to emissions reductions, or *instead of*.

Carbon sinks are only going to go so far, I'm afraid. To paraphrase Bill Clinton: It's the Emissions, Stupid.

Another thought.

This study is reporting the age (or turnover time) of carbon that is coming out of the *rivers* of the Amazon basin -- not the forests themselves.

Most of the carbon that goes into the Amazon basin (through photosynthesis) turns into trees. And most of that carbon stays in the Amazon for a long time.

The rivers in the Amazon basin appear to be a small, but important "short circuit" in the Amazon carbon cycle. That is, they quickly leach carbon from the land, mainly from the soil, and then immediately burn it off into the atmosphere.

But, again, this is only a fraction of the carbon fixed by trees in the Amazon. Most of the carbon does not leach into the river; instead, it turns into wood or other biomass. And most of that stuff stays put for a while.

This "river short-circuit" of the Amazon carbon cycle *is* pretty impressive, though. But it's unlikely to be such an important factor in other places. The Amazon is just amazingly wet! There is an astonishing amount of water (and dissolved and suspended stuff) running through it. In fact, the Amazon carries about 1/5 of all of the world's freshwater -- in single river! No other river comes close. (The other 9 largest of the 10 largest rivers *combined* still do not match the flow of the Amazon river!)

When you think Amazon, think water. Lots and lots of it.


I just read an article a few days ago about a similar and big Eurofunded research project which monitored carbon intake in the Siberian, Western Russian and Eastern European forests. It seems like these forests are disappointing as carbon storing mechanisms. Not only do they release the carbon very quickly, their total uptake is very low. The vast grasslands and steppes further down south did the work much much better, on both accounts.
It seems like we will have to revise our view on the value of forests as carbon sequestering mechanisms.

I'm searching the link for the article.


Ok, found it. The European research project is called Terrestrial Carbon Observation System Siberia I, headed by the Max Planck institute.

Project page:

An article about the disappointing sequestering capacity of the forest:


(registration required)

Copy translated from Dutch (babelfish):

Siberian forest takes less carbondioxide than thought. (CO2) are a gas that promote the greenhouse impact - the reheating of the planet -. Of the forests in Siberia it was assumed that they set out much CO2 in plant material. Thus the forest would compensate partly the industrial emission. But the results of the European TCOS-Siberia-project draws a line through that account. From airplanes and on observation towers scientists have measured the gas exchange of the forests and grasslands of Siberia. The forest appears to take less than twenty per cent of the carbondioxide which is expelled by the Russian federation, and this is much less than previously thought. Howevern, when the climate warms up and the growth season becomes longer, Co2-intake increases. It is not known whether this effect will last long and sustains itself.
From the research, which was conducted by Martin Heimann of the Max-Planck-Instituut for Biogeochemie, it became clear further that the grasslands and steppes south of the siberian forests take in much more CO2.

(Sorry for the bad translation).

Jon Foley:

Yes, there probably needs to be a big adjustment of our ideas of carbon sequestration in terrestrial ecosystems. We have focused a lot of attention on forests, but some recent results (like those mentioned above) suggest that grasslands may be a good candidate as well.

In my area (Madison, Wisconsin), there are studies underway to determine how prairie restoration and gentler farming techniques (no-till) may help restore the organic matter of the soil, and thereby store carbon out of the atmosphere. Remnant prairies in our region store a *lot* of carbon -- but, unfortunately, most of the restored ones do not seem to be doing as well. Perhaps we need more restoration experiments to figure out how to do restore the species mix, and also the soil carbon levels? Stay tuned.

Thanks, Jon, for helping us lay persons better understand the report. I agree that we'll need every carbon sink we can reasonably use, but the *best* ones will be those associated with things we want to do anyway - such as eat, preserve habitat, protect watersheds, etc. So restoring prairies is good - restoring and pasturing ruminants on them is better. Trees as carbon sinks is good - but riparian buffers and food forests are better. Increasing soil carbon is good - increasing organic matter in agricultural soils is better. One action always has many outcomes - we need to get really good at finding actions with multiple benefits.

Jon Foley:


I'd love to see us find ways to do farming that improves long-term productivity, sequesters more carbon from the atmosphere, stabilizes soil and reduces erosion, reduces flooding, and increases biodiversity. Restoring soil organic matter does *all* of these things, amazingly enough!

I'm also inspired by the wind farms being built in Wisconsin. Often the turbines are simply put on top of a farm field: harvesting the wind above the harvested crop. Not a bad way to go, especially when the turbines provide a steady income to the farmer (through rent from the utility company) that doesn't depend on commodity prices or the weather.

Emilio Mayorga:

Hi Jon,

I'd seen your contributions at this site before; great work.

But more to the point, thank you for your clarifications about our paper. You're completely correct in highlighting that the export of carbon to rivers is only one component of Amazonian forests' carbon budget; and a small (but important!) one at that. This is a complicated system with many pieces, and as far as I know (but I'm less up to date on this than you) we still don't have well constrained estimates of carbon sinks in old-growth forests that nicely resolve methodological issues or include all important fluxes.

Your clarification is a very important one. As it turns out, The Economist (July 29th issue) briefly cited our work in an unrelated article about the hydrological role of trees in arid and semi-arid lands. They never contacted us, but in one short paragraph the reporter managed to completely misinterpret the scope of our work, suggesting that Amazon forests don't store carbon for a long time, after all -- ignoring carbon in wood and soils, as you pointed out.

I totally agree with you on nearly all of your points, but I wanted to add a couple of clarifications. First, while leakage of carbon from land (forests) to rivers is a pretty small fraction of total forest photosynthesis or ecosystem respiration, the more important point is that this leakage appears to be significant when compared to the range of estimates of CO2 *sinks* in old-growth Amazon forests. It's one of several important pieces of the complicated puzzle that is estimating the magnitude of this sink. Second, we do highlight (based on the stable carbon isotope, 13C) that when grasses are present (natural ones or introduced pastures), they tend to have a "disproportionate influence" on respiration fluxes. But our results in rivers where grasses are absent, like the Rio Negro and its tributaries, are nearly identical. We really need more data to fine-tune the role of respiration stemming from decomposition of tree vs. "macrophyte/grass" material, but river respiration in fully forested watersheds appears to be young and vigorous as well!

Take care, and I hope to run into you at a conference.

-Emilio Mayorga


This page contains a single entry from the blog posted on July 27, 2005 2:05 PM.

The previous post in this blog was Open Source Warfare.

The next post in this blog is I, For One, Welcome Our New Post-Singularity Overlords.

Many more can be found on the main index page or by looking through the archives.

Powered by
Movable Type 3.34