Watching Forests Take Up Carbon

Because of the present and future importance of forests affecting the time course of the rise in atmospheric C02, there is presently a lot of work going on to understand whether—and how fast—they are taking up carbon in particular parts of the world, and how they respond to climate fluctuation. When modern ecosystem ecology first began in the 1960s, studies of forest growth concentrated on estimating the amount of wood added to all the trees throughout the forest, from the width of tree rings or the increased girth of the trunks. From this it was possible to infer the approximate amount of carbon by which a growing young forest increased its carbon storage each year, or in an old forest in equilibrium that rate at which carbon was flowing through the ecosystem (balanced by death of old trees and loss of branches). This approach has been used to estimate that the eastern USA forests are taking up carbon at the rate of several tens of millions of tonnes per year. However, it is a very broad-brush approach. It would be better if we had more details about exactly how much carbon is being taken up where, as this would enable us to predict better what will happen in the future.

In the 1980s ecologists began to consider a more ambitious and detailed approach to understanding where and how fast forests take up carbon. This relied on taking very comprehensive and precise measurements of the C02 concentration around the trees. The idea is, that if a forest is photosynthesizing and sucking up carbon from the atmosphere around it, this should show up as a localized depletion of C02 in the air just above, around and inside the forest canopy (Figure 7.19). Using

Photosynthesis dominates over

Figure 7.19. The eddy flux covariance method picks up C02 (a) blown towards the forest during the day and (b) compares it with C02 moving out of the forest at night.

Photosynthesis dominates over

Sensors' measure C02 concentration

Respiration releases CO,

Respiration releases CO,

Figure 7.19. The eddy flux covariance method picks up C02 (a) blown towards the forest during the day and (b) compares it with C02 moving out of the forest at night.

large numbers of well-placed sensors to measure C02 concentration, it is in theory possible to estimate just how much net photosynthesis is going on during the day, and thus how fast the forest is accumulating carbon. Even though the forest ecosystem is also respiring during the day, in daytime there will normally be more photosynthetic uptake of carbon than carbon released from respiration. This estimate has to be balanced against the amount of carbon lost from the forest at night, when there is only respiration and no photosynthesis. Again this night-time assessment can be done using the sensors to measure how much the C02 concentration around the forest has been raised at night relative to the background level in the atmosphere. To make these estimates properly, it is necessary to estimate how fast more C02 is getting delivered (during the day) or taken away (during the night) by air movement. This involves a lot of complex physics and calculation. If the measurements are continued month after month, year after year, then it may be possible to infer how the balance of carbon in a sample patch of forest is changing over time.

This approach, known as the eddy flux covariance method, is compelling but also very ambitious. It requires a huge investment of labor and money to put in place the complex measuring equipment and maintain it, and analyze the data that comes out. At an intuitive level, it is easy to see that if a slight portion of the carbon loss or gain each day was not included in the accounting (e.g., because a sensor missed it) this error would accumulate over many months and might give a totally misleading picture of the direction in which the forest's carbon balance was changing. 0ne big problem that this method has run into is that it is difficult to summarize the amount and direction of air movement over the forest during day and night. At night, especially, air movement from the forest canopy is very sensitive to local conditions and it may stay stable (stratify), or may continue to be turbulent. Turbulence in the air is generally a good thing for this methodology, because it carries extra C02 coming from the plants and soil out towards the sensors up on the pole, where they can detect it and correctly add it to the respiration side of the budget. Sometimes at night the cooling C02-rich air from within the canopy at night sinks right to the forest floor, underneath the sensors, and flows away undetected. This too gives the impression of less respiration actually occurring in the ecosystem.

Perhaps because of these problems, eddy flux covariance has sometimes given strange results that do not seem to tally with previous knowledge of ecosystem processes built up during the 1960s and 1970s. A study site in pristine Amazon rainforest seemed to be inexplicably gaining carbon so rapidly that it was set to double in carbon mass within 60 years. In Europe, forests and their soils in the north seemed to be accumulating carbon more slowly than those in the south, not because trees were growing slower in the cooler climate but because the northerly soils were breaking down carbon more rapidly. This seemed to contradict decades of knowledge about how soil respiration responds to temperature. In many other areas such as the eastern USA and Japan, the measurements seem to make much more sense in terms of previous knowledge of how climate and forest age affect carbon balance. However, there is the nagging question of whether these studies too might contain errors which are too small to clash with previous understanding of ecosystems, but scientifically important nevertheless.

Some ecologists such as Christian Koerner of Bern University have pointed out that, in addition to the measurement problems, the fact that important processes which affect forests such as tree falls, land slides, droughts and fires occur in an occasional and unpredictable way means that intensive measurements of small patches of forests may not give a particularly relevant picture of long-term trends on a broad scale, which is the sort of question these studies are basically attempting to answer.

While the eddy flux covariance method is a considerable achievement of engineering and scientific collaboration, it remains an open question as to how much it can really teach us, compared with more old-fashioned methods of looking at the carbon balance of ecosystems.

7.9.1 Predicting changes in global carbon balance under global warming

From the year-to-year variability in the amount by which CO2 builds up in the atmosphere, it looks as if the amount of CO2 taken up or released by the world's vegetation responds quite a lot to changes in the climate from one year to the next. Such small responses to year-to-year climate variation might give us clues to longer term trends that will emerge as the global climate warms due to the greenhouse effect.

Since CO2 responds to climate, and climate responds to CO2, there is the potential here for some important feedbacks. It could be that as the world gets warmer it will favor more carbon being stored in vegetation and soils, slowing the warming by taking CO2 out of the atmosphere. This would be a negative feedback loop, tending to act against the main cause of the warming. On the other hand, in a warmer world, vegetation and soils might actually respond by releasing CO2, adding further to the warming in a positive feedback loop.

Given what we know of the responses of forest carbon balance to year-to-year climate fluctuation, the effect that global warming might have on CO2 uptake or release by forests is complex. It depends on the particular region, and the detailed nature of the climate shift: how big it is, and whether rainfall changes as well as temperature. The whole task of predicting what will happen tests the limits of understanding of both climate and the global carbon cycle. Attempting to model the whole system over the coming centuries requires inter-disciplinary teams of experts using some of the fastest computers available. One study by Peter Cox and his colleagues based at the Hadley Center in the UK predicted that, as the world warms, carbon will gush out of the world's ecosystems into the atmosphere, amplifying the warming in a positive feedback (Figure 7.20). The effect, then, is for an initial push of CO2 to give warming that is amplified further by more CO2 coming out of the world's forests. In the model, a large part of the positive feedback occurs due to more frequent and more severe El Nino events affecting the carbon balance in the tropical forests of South America, South-East Asia and Africa. Drying in the Amazon region is also predicted to occur as a result of increased temperatures in the Atlantic due to global warming.

Though such models are impressive, there are many uncertainties, and a slight error in the parameters of a model could throw the predictions way off from what will actually happen. Not all climate models predict more El Ninos under global

% carbon loss % carbon gain

Figure 7.20. Model results with and without the "gushing out'' of carbon that would result from warming affecting the carbon balance of forests. In the lower scenario, the extra C02 that the forests give out (in response to warming) warms the climate further and this in turn promotes more forest growth in the high latitudes. (a) Without C02 feedbacks. (b) With C02 feedbacks involving the extra C02 released by die-back of vegetation and decay of soil carbon, adding further to the warming and affecting vegetation in a positive feedback cycle. Source: Cox et al. (2001).

warming, and there is even some doubt whether most of the El Nino events over the past few decades have pushed more CO2 into the atmosphere. The "blip" in CO2 that tends to be associated with an El Nino could actually be due to the opposite "La Nina'' climate episode that tends to follow soon afterwards.

There is in particular a need for more fieldwork and basic observations of how plants and soils respond to changing climate conditions. A big problem for models to predict is exactly how vegetation will respond through the direct CO2 effect (Chapter 8). The global model for the future carbon cycle that Cox and colleagues have put together also included the influence of the direct CO2 fertilization effect for the future, and even though they have tried to bring this in to the model, it is hard to say how this factor would modify the overall outcome in terms of carbon storage. In a more recent version of the same model, Richard Betts and his colleagues have attempted to look in more detail at this direct CO2 effect on global carbon balance in the future: but more about that in the next chapter (Chapter 8). It is important to bear in mind, however, that the Cox et al. and Betts et al. scenarios are using just one general circulation model, the Hadley Centre Model, when there are several other ones out there. The Hadley Centre model happens to be the only one that forecasts such a widespread "Amazon Meltdown'' scenario (others have the Amazon rainforest surviving the next century mostly intact, unless of course humans cut it down), so perhaps we should not take it so seriously as to exclude everything else. Another thing to bear in mind is that even quite small details of how the model of Cox and colleagues works have a big effect on the outcome of global warming in its predictions. For example, how sensitive tropical forest trees are to temperature is key to their response to warming; yet we actually lack the key data that are needed. The model itself assumes that tropical forests respond to temperature just as temperate-zone forests do. There is an urgent need for more data from experiments and field observations to show how tropical forests will actually respond to temperature increases.

Various other feedback mechanisms involving CO2 and ecosystem carbon reservoirs are also thought possible, and not all of them have been modeled yet. One major worry for the future is how the very large reservoir of carbon in peatlands in Siberia and Canada will respond to global warming. At present, both regions are extensively blanketed in a layer of peat. This has built up from dead plant material over thousands of years under waterlogged soil conditions and cool temperatures that have slowed down decay. If temperatures increase, the slow breakdown of the peat might accelerate and result in rapid addition of CO2 (and also the greenhouse gas methane, CH4) to the atmosphere. CO2 output is especially likely if the climate gets drier overall so that water tables fall, allowing oxygen to get into the peat. Under these conditions microbes, fires and also a gradual chemical reaction with the air can oxidize the peat. Digging down within the peat itself, one can often see "oxidized layers'' from the past when drought caused water tables to fall and some of the peat broke down to give CO2 gas. The fear is that under global warming this could occur on a massive scale, pushing billions of tonnes of CO2 into the already-warming atmosphere. If all the peat in the northern latitudes were to oxidize, it would push up the CO2 level by at least 50% beyond the present level. This is the sort of problem that models of the carbon cycle under global warming need to try to tackle. Especially important will be the field data that the models are based on, yet such data are often in short supply. A lot of the sort of basic science that enables such understanding is unglamorous and does not attract the same sort of funding as, for example, the eddy flux covariance studies mentioned above, or the FACE experiments discussed in Chapter 8. Much of what we do know is based on a long and rather esoteric tradition of the study of nature—the sort of thing that until recently funding agencies were keen to cut back on. All that we can say at present is that it is quite possible that vegetation and the soils underneath it will have a major influence on C02 levels once global warming gets fully under way.

Was this article helpful?

0 0
Renewable Energy Eco Friendly

Renewable Energy Eco Friendly

Renewable energy is energy that is generated from sunlight, rain, tides, geothermal heat and wind. These sources are naturally and constantly replenished, which is why they are deemed as renewable.

Get My Free Ebook


Post a comment