Biocycle - Global Carbon Cycle

Download a presentation summarizing the global carbon cycle and methods we are using to analyze it.

Read an article on the global carbon cycle in Physics Today (This article appeared in the August, 2002 issue).

Carbon is the natural currency of life itself: all organic compounds, including plants, animals, humans, food and fiber, and much of our energy sources are built of structures made of carbon atoms. Photosynthesis is the process by which plants convert inorganic CO₂ molecules in the air into organic molecules that form the base of the food chain and the source of energy for nearly all life on Earth. This process of producing reduced organic compounds in an oxidizing atmosphere requires energy, which plants obtain from sunlight. All other organisms, from bacteria to human beings, sustain themselves by recovering this stored energy through respiration by burning (oxidizing) these molecules back to CO₂ Globally, photosynthesis and respiration consume and replace about one fifth of all the CO₂ in the atmosphere every year in a huge natural carbon cycle which has been operating for over a billion years.

Atmospheric CO₂ has risen and fallen through geologic time, and these variations have been well-documented over the past 400,000 years by analyzing air trapped in glacial ice. Over the past four cycles of ice ages and interglacial periods, CO₂ concentrations in the atmosphere have followed a remarkable repeating pattern, decreasing dramatically as continental ice sheets spread across the northern hemisphere, then rising rapidly as the ice sheets melted. The carbon cycle clearly beats in time with the ice ages, probably because of changes in ocean circulation and biology. Whatever processes consume atmospheric CO₂ during ice ages have consistently stopped at concentration levels of about 180 ppm. During interglacial periods, the concentration repeatedly stabilized around 280 ppm. The mechanisms that determine these set points of the natural carbon cycle are not well understood, but probably involve transfers of large but consistent amounts of carbon between the atmosphere and the deep oceans.

Since the industrial revolution, human beings have been extracting fossil organic material from geologic reservoirs and burning it to extract energy stored by plants over many millions of years. As a result, CO₂ concentrations have now risen to about 380 ppm, compared with a very stable value of about 280 ppm for the previous 800 years. This has led to serious concerns about potential global warming, because CO₂ molecules in the atmosphere are strong absorbers and emitters of longwave radiation. The Intergovernmental Panel on Climate Change (IPCC) has projected that CO₂ concentrations will reach between 600 and 800 ppm by the end of this century, which is more than twice the preindustrial level.

In addition to the radiative impact on Earth's climate, elevated CO₂ levels and other human actions have led to substantial changes in the functioning of the natural carbon cycle. As CO₂ levels rose, the surface ocean became undersaturdated with respect to dissolved CO₂ so some of the extra CO₂ in the atmosphere has dissolved into the ocean. Because plants eat CO₂ the additional atmospheric carbon has led to enhanced plant productivity in some areas, which has removed some CO₂. These processes are known as carbon sinks (as opposed to sources). Other important sink processes inlcude the effects of nitrogen and other fertilizers, lengthening growing seasons in boreal and arctic ecosystems, and the uptake of CO₂ by regrowing fApril 28, 2006 3:42 PM in inorganic form in carbonate rocks (limestone and related minerals). The rest is labile, meaning that it can be transformed easily from one form to another. There are three important labile pools of carbon: in the oceans, the atmosphere, and the terrestrial biosphere. Of these, the oceans contain by far the most carbon, about 38,000 Gigatons. (1 gigaton of carbon, GtC is equal to 10¹²kg, which is the mass of one km³ of water). Most of this is in the deep ocean as dissolved bicarbonate ion (HCO3^-), and can only interact with the atmosphere on timescales of thousands of years required for deep ocean mixing and ventilation. Organic matter on land (plants, soils, microbes, animals, people) accounts for perhaps 2,000 GtC, and atmospheric CO₂ comprises the smallest labile reservoir at about 780 GtC.

The natural carbon cycle involves very large exchanges between the atmosphere and both the oceans and plants and soils on land. Terrestrial photosynthesis, for example, removes over 100 GtC/yr from the atmosphere to create new organic matter. This flux would completely consume atmospheric CO₂ in just a few years if it were not nearly balanced by emission of CO₂ by respiration and decomposition. Similarly, huge amounts of CO₂ are absorbed in the surface oceans every year by dissolution and biological fixation, but these large one-way fluxes are also nearly balanced by emissions. Small imbalances between the very large gross exchanges among the three labile carbon reservoirs in the Earth system currently produce a net uptake of about 3 GtC/yr, or roughly half of the rate of CO₂ emissions by combustion of fossil fuel.

The net effect of sink processes has been to substantially slow the rise of atmospheric CO₂ Carbon sinks have removed about half of the CO₂ emitted by fossil fuel combustion in the past century. This amounts to a free service by the oceans and atmosphere. The value of this service to our economy is enormous. Economists have studied the potential costs of CO₂ emissions reductions and mitigation methods, and have generally estimated such actions to cost between $10 and $100 per ton of carbon. The global carbon sink is currently removing about 3.5 billion tons of carbon from the atmosphere per year, and this service is therefore worth between $35B and $350B per year. Yet beyond outlining the basic mechanisms responsible, we know surprisingly little about these sinks. We lack the ability to identify where in the world the sinks operate, whether they will continue to operate, how long the stored carbon will stay out of the atmosphere, or what we might do to help sinks to operate more efficiently.

The idea that carbon sinks might saturate or stop efficiently removing CO₂ from the atmosphere is no mere conjecture. Nearly all the major sink mechanisms are expected to decline in strength over time, or even to reverse themselves and become sources of carbon to the atmosphere. Forest regrowth only works as a sink until the forest has grown back from previous cutting. Nitrogen fertilization works only on nitrogen-limited ecosystems: just as in a vegetable garden, further additions of fertilizer are ineffective in enhancing growth. Boreal warming may allow shrub growth to store carbon in previously tundra ecosystems, but may also dry out wetlands and melt permafrost that have preserved organic matter for millennia. Even CO₂ fertilization is expected to saturate at the levels projected by 2100.

Even more troublesome is the possibility that climate change and the carbon cycle might interact to release large amounts of CO₂ into the atmosphere. A recent study by Peter Cox and his colleagues at the Hadley Centre in the UK explored this possibility. They found that as a result of CO₂-induced warming of the ocean surface in the tropical Pacific, that a sort of perpetual El Nino condition could occur by mid-century. In their simulation, this El Nino-like warming led to severe drought over the South America which caused most of the trees in the Amazon rainforest to be replaced by grassland. The carbon stored in trees and soils was released to the atmosphere. As a result, the simulated CO₂ concentration in this model rose to nearly 1000 ppm by 2100, compared to about 700 ppm without the carbon cycle feedback. This in turn led to nearly twice the global warming. Carbon cycle-climate feedbacks like such as that studied by the Hadley Centre are among the primary sources of uncertainty in predictions of global change, and have profound implications for management of energy and carbon sinks. Even if the outcome simulated by Cox and his colleagues has a very low probability of being realized, the potential for major impacts on the people and economy of the USA justify a major research effort to evaluate global models against available data. On the other hand, a similar calculation made with a different numerical model by Pierre Friedlingstein and his colleagues at the Institut Pierre-Simon Laplace in Paris found much weaker feedback between the climate and carbon cycle. In their simulation, both land and ocean sinks continued to strengthen throughout the 21st century, resulting in CO₂ concentrations below 800 ppm in 2100, and only about half the climate warming simulated by the Hadley Centre.

Given the value of carbon sinks, their potential for saturation, and the very expensive consequences if they should stop operating, it is imperative that research efforts be implemented to provide policymakers with adequate information to address these issues. We must be able effectively evaluate the major hypotheses regarding sink processes and sink saturation in order to predict changes and evaluate risks.

Studies of global carbon cycling have been conducted at local scales, focused on elucidating processes in the field, and at global scales, focused on understanding and predicting changes in the atmosphere and oceans. Each kind of inquiry has been very valuable in developing our understanding, yet each has inherent limits that must be traded off against one another. Local studies, for example of the effects of nitrogen deposition on regrowing forests, are crucial to establish patterns of ecosystem response to environmental forcing. Yet it is difficult to know how representative a particular study area is of all the vast areas being subjected to global change. Sources and sinks can also be inferred from changes in the amount of CO₂ and other gases in airmasses as they pass over a study region, using a technique called inverse modeling. This approach has the advantage of providing area-averaged sources and sinks over huge regions, but provides no information about the causes or mechanisms involved.

Recently, technological advances have provided carbon cycle scientists with new tools to bridge the wide conceptual gap between local process studies and global inverse studies. Micrometeorological techniques have allowed carbon, water, and energy fluxes to be estimated over entire ecosystems, and have been applied at over 100 sites around the world over periods of several years. New instrumentation for estimating atmospheric CO₂ concentrations from satellites has been developed, which will enable a revolution in the study of regional sources and sinks using inverse models over the next decade. Finally, powerful computers and numerical models have been developed to make use of these new data.

The carbon cycle science community now faces a new challenge: the synthesis of many disparate data streams from local to regional to global scales; the development of rigorous methods of evaluating the new generation of models; and ultimately, the prediction of carbon-climate interactions over the coming decades. One idea that has been advanced toward achieving the required synthesis is to adapt mathematical methods that have long been used in weather forecasting, seismology, and medical imaging to the analysis of the global carbon cycle. These tools are collectively known as model-data fusion or data assimilation methods. The idea is to construct mechanistic models of carbon cycle processes that predict observable features at many scales in space and time. These predictions are compared to measurements made in-situ and by remote sensing, and used to adjust aspects of the models that are the most uncertain. Thus a mechanistic description of the function of the global carbon cycle would be created which is optimally consistent with many types of data, and yet which is defined at all points in the global atmosphere, oceans, and land surface. Such a tool would be used to evaluate major hypotheses about carbon sources and sinks, and thus to develop credible predictive models of future carbon-climate interactions needed to inform rational public policy.