The press, as usual, is getting it wrong.
But it is an important discovery anyway - that normal oxygen rich interactions have plants producing methane, rather than just as a bi-product of anoxic bacteria - shows that we have a great deal to learn about carbon's rich cycle, and cannot rely on hacks to simply sink the carbon. We have to face the fact that reducing carbon means reducing carbon input, not merely hoping to mop up.
So what does this study mean? Bottom line: we are going to have to remeasure how much carbon sinking trees do versus their methane output. We can't cut down tropical rain forests - they supply oxygen - but we can't simply believe that we can turn ashes back into wood to solve the Global Warming Problem.
[Yes, I'm an environmentalist, and global warming is a major deal. It's something that is in my writing, my economics and my music.]
What the Paper Contends
The Nature paper is entitled "Methane emissions from terrestrial plants under aerobic condition". Which to a biologist is shocking enough, look up methane and plants, and you will find that two areas of methane production are noted - one is decay, and the other is digestion by animals. One cow produces 25 moles of methane a day. At avagadro's number per mole, that's enough breaking wind jokes to power an entire episode of South Park.
The paper makes three arguments. First, by taking carbon isotopes, they found that methane coming out of plants in an oxygen environment contained methane fromed in the presence of oxygen. Basically, they fed the plants carbon that was slightly different from the ordinary kind, and measured it in the resulting gas. Second, they measured the same effect in the field - that is, this wasn't just a lab result. Third, they did some back of the envelop type calculations based on their few data points, and were shocked to discover that this source of methane may be a huge contributor to methane in the atmosphere.
What this doesn't do is say that "trees cause pollution" it is saying that the carbon dioxide sink effect of trees and other vegetation - which we still have problems measuring .
The Carbon Budget
From the beginning the question of global warming and controlling human impact was a question of the carbon budget - even in the 1990's the basic outline is visible - measure carbon in/carbon sunk/carbon out, and then look for feedback effects. It has been well known for some time that methane has been decreasing even though it was not clear whether this was from less methane being put into the atmosphere, or because of greater methane destruction through interaction with OH- radicals - that is a radical that is one hydrogen short of being water, and is called a "base", as opposed to "an acid" which is when there are free hydrogen (H+) radicals.
In the 1980's methane was estimated at 400-600 Tetragrams. The new process may add between 62 to 236 Tg to the total. That the range is so large should tell you how much error there is here, there is almost as much error in this first estimate than signal. But that isn't any reason to dismiss it, merely to take it with a cautious and preliminary eye.
For reference, the carbon sinking of forests is measured in PetaGrams - but given the estimated numbers - this effect could reduce the carbon sink capacity of forests by 10%. That is, they still sink lots of Carbon Dioxide, but they put out the longer lived Methane.
Carbon Stopping Rather than Carbon Sinking
What this means for Kyoto is that many nations which had hoped to reach their targets cheaply by planting trees, are going to find that this doesn't reduce global warming by as much as they thought it would. The next negotiation for Kyoto will have to have this new effect plugged in, so that the right choices can be made. The right wing is going to be all over this as an attack on Kyoto, when, in fact, because of the breaking down of the agreement into periods, Kyoto is better than setting emissions targets arbitrarily, and often with an eye to headlines. Harper's conservatives in Canada want out of Kyoto - and promise to cut Ozone, which is not a greenhouse gase of any significance.
In 2001 a better methodology has been proposed for Kyoto - instead of taking specific activities, it would measure the net change. How this would interact with the new discovery is this, we would measure the methane production for different kinds of land use, and over the course of a year count up the total changes, and calculate the difference between this year and last year, not calculate the total amount of creditted activities. This system would be more robust than counting credits, because there would be strong incentive to allow people to "cheat" - plant trees with one hand, and knock them down in some way that doesn't "count".
As people may know I am skeptical of carbon sinking the reason for this is complex, but the mathematics can be reduced to this: what matters, as the carbon budget papers show, is the flux, or change, in atmospheric carbon, not the activities that human beings "count" or not. Biological systems respond in complex ways. Increasing forests in one place uses up water, holds soil, and may well offset growth in other areas. What good does it do to pay one person to plant trees - if the result is to make trees less likely to grow someplace else? It is impossible to measure the net effect of carbon sinking because of the sensitivity to initial conditions. Cutting down rainforest trees reduces methane, but pours carbon dioxide from combustion, and methane from the bovines that are sent to graze in the resulting grasslands. Since we do not understand, and cannot measure, net effects of actions, it makes no sense to make actions the basis of our incentives.
In otherwords - planting a tree has complex effects in nature. While it is a good thing for all kinds of reasons - we should be "recanopying" our cities to reduce the need for air conditioning, reducing smog and so on - we can't expect this to dent the global warming problem in ways that we can pin down and measure, and thus put into treaties and charge people for.
Contrast this with preventing carbon from getting into the atmosphere in the first place. Almost none of the activities - combustion being numero uno - have the complex counterlevering results of planting vegetation or shifting land use. Instead, they kill plants where we mine, the kill them where we build developments, they kill them where we dump the waste. And they are simple to track and simple to deal with - cutting down methane and C02 emissions, as well as some aersols and other assorted trace gases - reduces the stress on the sink. The natural sink process will be better at finding ways of absorbing carbon - that is what life does, sink carbon in physical forms - than we are. We can certainly encourage the sink process, by giving over as much land and sea as possible to natural growth and perserving wilderness, but meddling to much in the selection is not going to produce demonstrable wins. Again, the reasons for this are complex, but they have to do with the feedback loop of life and the atmosphere.
Trees still sink more global warming than they cause, but they don't sink as much of it as we thought, for as long. So what this study means is that we are going to have to continue to improve our understanding of the carbon cycle. We are going to continue to have to improve the ways we credit and debit global warming credits. We are going to have to slash how much carbon we dump into the air, because plants, just like people, dump their wastes on to others if they can get away with it. The difference is that a tree only does as much of it as it needs to grow, as opposed to as much of it as it takes to be a billionaire several times over.