The Changing Carbon Cycle

Frontmatter

1. The Global Atmospheric CO2 Distribution 1968–1983: Interpretation of the Results of the NOAA/GMCC Measurement Program

Abstract

The modern period of precise atmospheric CO₂ measurements began with Keeling’s pioneering determinations at Mauna Loa, Hawaii, and the South Pole during the International Geophysical Year. The Mauna Loa record (e.g., Keeling 1983) remains the single most valuable CO₂ time series for carbon cycle model verification. For very recent interpretations of atmospheric CO₂ measurements and carbon cycle relationships, the reader is referred to Cleveland et al. (1983), Keeling (1983), Machta (1983), Mook et al. (1983), Pearman et al. (1983), Keeling et al. (1984), Wong et al. (1984), Bacastow et al. (1985), Komhyr et al. (1985), and Fraser et al. (this volume). In this chapter, the global atmospheric CO₂ records, particularly of the NOAA/GMCC (National Oceanic and Atmospheric Administration/Geophysical Monitoring for Climatic Change) program, are summarized for the period 1968 through 1983, with emphasis on the mean properties of the global carbon cycle as viewed from the atmosphere (i.e., global mean CO₂ concentration, latitude dependence of concentration and seasonal amplitude, airborne fraction, etc.).

Richard H. Gammon, Walter D. Komhyr, James T. Peterson

2. Simulating the Atmospheric Carbon Dioxide Distribution with a Three-Dimensional Tracer Model

Abstract

Atmospheric CO₂ concentrations exhibit spatial and temporal variations reflecting the distribution and time dependence of various natural and anthropogenic CO₂ sources and sinks at the earth’s surface. The location, magnitude, and time history of these sources and sinks are of paramount importance for global carbon cycle studies. From data obtained in a global network of suitably located CO₂ monitoring stations, it appears feasible to infer at least some of the characteristics of the sources from measured atmospheric CO₂ concentration data. To do so, the following two main problems have to be dealt with:

Martin Heimann, Charles D. Keeling, Inez Y. Fung

3. Presentation of the 20th Century Atmospheric CO2 Record in Smithsonian Spectrographic Plates

Abstract

The history of 20th century scientific measurement programs does not contain many examples of long-running studies in which an important property of the atmosphere is reliably measured by the same person at the same site with precisions of >1%. One outstanding example is C. D. Keeling’s Mauna Loa record of CO₂ measurements (Keeling et al. 1982). A program that ran for an even longer period of time in the early and mid-20th century is the Smithsonian Solar Constant Program, directed by Charles G. Abbott. Over the past several years, we have been studying whether a portion of the Smithsonian data set can be used to extend the record of atmospheric CO₂ concentrations back to the early 20th century.

Gerald M. Stokes, James C. Barnard

4. Atmospheric CO2 Record from Direct Chemical Measurements During the 19th Century

Abstract

In a paper published almost 50 years ago Callendar (1938) suggested that, during the period 1900 to 1935, a 6% increase in atmospheric CO₂ occurred as a result of the combustion of fossil fuels. He calculated that approximately three-quarters of the anthropogenically produced CO₂ remained in the atmosphere, which induced a global warming of 0.1°C and accounted for most of the increase observed in the available long-term temperature records. The calculated atmospheric CO₂ increase was based on a few observations (Brown and Escombe 1905) that, Callendar suggested, indicated a background CO₂ level in 1900 of 274 ± 5 ppmv (parts per 10⁶ by volume). As pointed out by Keeling (1978a), it is difficult to see how Callendar arrived at such a low concentration from the Brown and Es-combe data, and, from a later analysis of additional data sets (Callendar 1940, 1958), he revised his estimate of a late 19th century background CO₂ level to be 288 to 290 (±3) ppmv. From a more rigorous statistical analysis of 19th and 20th century CO₂ data, Slocum (1955) questioned whether concentrations had increased, and Bray (1959) showed that a significant increase is observed only when highly variable data are removed from the observations.

Paul J. Fraser, William P. Elliott, L. S. Waterman

5. Review of the History of Atmospheric CO2 Recorded in Ice Cores

Abstract

Since the pioneering attempts by Scholander et al. (1961), the observation that the porous spaces in natural ice contain samples of ancient air, the study of ice cores for potential insights into the history of the atmospheric CO₂ concentration has received great attention from scientists interested in the reconstruction of environmental parameters. Progress, however, was made possible only because deep ice cores from Greenland and Antarctica, which are continuous sequences of generally high-quality samples formed during the last 100,000 and 50,000 years, respectively, were available for study. Research has led to new techniques for extracting gases from ice and to recent developments of sensitive and accurate techniques for the analysis of gas. Today, analysis of gas concentrations in air entrapped in natural ice is considered to be the most promising method for reconstructing the history of the atmospheric CO₂ concentration (WMO 1983).

Hans Oeschger, B. Stauffer

6. Ancient Carbon Cycle Changes Derived from Tree-Ring 13C and 14C

Abstract

Current research places major emphasis on atmospheric CO₂ concentration because this variable is important in modeling climatic change. The history of atmospheric CO₂, however, is described not only by concentration change but also by changes in its isotopic composition. Two stable isotopes, ¹²C and ¹³C, exist for carbon, whereas ¹⁴C, a radioactive isotope, is also present in the natural carbon cycle.

Minze Stuiver

7. Interpretation of the Northern Hemispheric Record of 13C/12C Trends of Atmospheric CO2 in Tree Rings

Abstract

The present long-term increase of atmospheric CO₂ has been accompanied by changes in the isotopic composition of its carbon atoms. Both the ratios of stable ¹³CO₂/¹²CO₂ and radioactive ¹⁴CO₂/¹²CO₂ have decreased as a result of anthropogenic activity. The decrease of the latter, until the beginning of nuclear testing caused an increase in the mid-1950s, is almost exclusively due to dilution of atmospheric CO₂ resulting from fossil fuel burning, from which the ¹⁴C isotope had already decayed. The decrease of the ¹³C/¹²C ratio is a better measure of total anthropogenic activity because it reflects not only fossil fuel burning but also CO₂ released as a result of deforestation and agricultural manipulation of the soil. The CO₂ from both of these sources is marked by a δ¹³C deficit of about 18 to 20‰ (PDB scale; e.g., Freyer 1979c).

Hans D. Freyer

8. Revised Estimates of Atmospheric CO2 Variations Based on the Tree-Ring 13C Record

Abstract

Since the publication of a paper by Peng et al. (1983) regarding the analysis of the tree-ring-based δ¹³C record,* a number of reports giving new measurements of δ¹³C in tree rings have emerged (Leavitt and Long 1983; Stuiver et al. 1984; Freyer, this volume; Stuiver, this volume). The results of the original analysis were based on a global ¹³C/¹²C trend compiled by Freyer and Belacy (1983). Freyer (this volume) recalculated the ¹³C/¹²C trend of the Northern Hemisphere using measurements on 65 trees, including those presented in the publication of Leavitt and Long (1983), and he concluded that no long-term trend exists during the period of a few centuries before 1800 AD. By contrast, a clear trend of decreasing ¹³C was observed after 1800 AD. Freyer estimated the overall decrease of δ¹³C from 1800 to 1980 AD to be about −1.5‰, which is 0.5‰ less than that obtained from the composite ¹³C trend used by Peng et al. (1983). Therefore, a reevaluation of the terrestrial biosphere contribution to the lowering of δ¹³C in the atmospheric CO₂ seems appropriate. This is the main objective of this chapter.

Tsung-Hung Peng, Hans D. Freyer

9. Carbon Isotope Measurements in Baseline Air, Forest Canopy Air, and Plants

Abstract

This chapter reviews recent experimental results from a variety of Australian programs, each having potential relevance to the direct or proxy measurement of atmospheric ¹³C/¹²C variation, in particular the global variation resulting from the combustion of isotopically light carbonaceous material in fossil fuel.

Roger J. Francey

10. Estimating Changes in the Carbon Content of Terrestrial Ecosystems from Historical Data

Abstract

When forests are cleared for agricultural crops, the carbon stored originally in trees is oxidized and released to the atmosphere, either rapidly if the trees are burned or slowly if they are left on the ground to decay. Similarly, the organic matter of soil is reduced through cultivation. Such reductions in the carbon stocks of terrestrial systems occur with the harvest of forests for wood and with the clearing of forests for cropland, pasture, or other uses. On the other hand, the regrowth of forests following harvest, the abandonment of agriculture, or the establishment of plantations increases the storage of carbon on land, both in vegetation and in soils. The balance between the clearing and regrowth of forests is the major factor in determining changes in the net storage of carbon in terrestrial systems. Non-forested systems can also lose or accumulate carbon, such as when grasslands are converted to agriculture; however, the changes in carbon per unit area are much smaller than for forests.

Richard A. Houghton

11. Changes in Soil Carbon Storage and Associated Properties with Disturbance and Recovery

Abstract

Organic matter in the world’s soils contains about three times as much carbon as the land vegetation. Soil organic matter is labile and is likely to change as a result of human activities. Agricultural clearing, for example, results in a decline in soil organic matter. At the present time, there may be a net release of 0.85 × 10¹⁵ g C • yr⁻¹ from soils of the world due to agricultural clearing (Houghton et al. 1983; Schlesinger 1984), or about 15% of the annual release from fossil fuels. The release of carbon may have been greater near the turn of the century as a result of more rapid agricultural expansion into virgin areas (Stuiver 1978, Wilson 1978). It is the purpose of this chapter (1) to review briefly the present estimates of the size of the pool of carbon in world soils and (2) to offer a review and analysis of what is known about the effects of agriculture on soil carbon storage.

William H. Schlesinger

12. Continental and Global Scale Remote Sensing of Land Cover

Abstract

In recent years, a number of investigations have indicated that the sensors aboard meteorological satellites have potential for land-cover monitoring at regional, continental, and global scales. The outstanding characteristic of data from such satellites relates to their high temporal resolution; imagery is available for the whole globe on a near-daily basis. Thus the possibilities of obtaining cloud-free imagery are greatly enhanced, and the temporal dynamics of land cover can be observed. The Advanced Very High Resolution Radiometer (AVHRR) of the National Oceanic and Atmospheric Administration (NOAA) series of sun-synchronous, polar-orbiting, operational satellites has been identified as having particular potential in this context (Gray and McCrary 1981, Schneider et al. 1981, Townshend and Tucker 1981, Cicone and Metzler 1982, Ormsby 1982, Schneider and McGinnis 1982, Tucker et al. 1982). This is because the radiometer’s first band in the visible-red part of the spectrum and the second band in the near-infrared (Table 12.1) are two bands of particular use in vegetation mapping of green leaf area, green leaf biomass, or the intercepted photosynthetically active radiation (Tucker 1979, Curran 1980, Kumar and Monteith 1982). These bands correspond approximately to bands 5 and 7 of the Multispectral Scanner System (MSS) of the LAND-SAT series of satellites.

Compton J. Tucker, J. R. G. Townshend, T. E. Goff, B. N. Holben

13. Changes in the Area of Forests in Rondonia, Amazon Basin, Measured by Satellite Imagery

Abstract

The transformation of tropical forests to pasture, crops, and barren land, now rapidly underway wherever tropical forests occur (Myers 1980; Fearnside 1982; Lanly 1982), contributes to the increase in CO₂ in the atmosphere (Woodwell and Houghton 1977; Woodwell et al. 1978, 1983a,b; Houghton et al. 1983). The greatest promise in measurement of the rate of change in area of forests lies in use of satellite imagery, available since 1972 in the LANDSAT series with a resolution of about 80 m [NASA 1983, Woodwell et al. 1983a, Klemas and Hardisky 1983, Woodwell 19841. We have developed a special technique using LANDSAT imagery from different times to make direct measurements of changes in the area of forests. The technique was developed for Maine and tested in Washington (Woodwell et al. 1983a). We report here an application of the technique in the tropical moist forests of the state of Rondonia in the Brazilian Amazon (Fig. 13.1).

George M. Woodwell, Richard A. Houghton, Thomas A. Stone, Archibald B. Park

14. One-Dimensional and Two-Dimensional Ocean Models for Predicting the Distribution of CO2 Between the Ocean and the Atmosphere

Abstract

To assert anything about what the concentration of CO₂ in the atmosphere will be in the future requires that one understand the oceanic-atmospheric distribution of carbon and the way it may change when excess CO₂ is introduced into the atmosphere. The span of time during which the emission occurs is short compared to the slow oceanic circulation; therefore, it becomes natural to separate the mixing process into two parts: (1) the transport from the atmosphere into a thin surface layer of the ocean and (2) the exchange of water between the surface layer and deeper down.

Anders Björkström

15. Three-Dimensional Ocean Models for Predicting the Distribution of CO2 Between the Ocean and Atmosphere

Abstract

The basic ingredients necessary to predict fossil fuel CO₂ uptake with a three-dimensional model of the oceans are threefold.

Jorge L. Sarmiento

16. Calibrating Ocean Models by the Constrained Inverse Method

Abstract

From the results of several investigations (Keeling 1973; Björkström 1980; Killough and Emanuel 1981; Björkström, this volume; Bolin 1983; Bolin et al. 1983, p. 231; Fiadeiro 1983; Bolin, this volume) the approach of using highly aggregated box models (less than 15 boxes for the world oceans) appears to be inadequate for the task of estimating accurately the current rate at which the ocean is absorbing excess atmospheric CO₂. However, general circulation models of the entire ocean are not at hand; therefore, what is needed is a new generation of models that can serve usefully in the interim to study the important question of the current rate of oceanic CO₂ uptake. Further, such models may have other applications, not the least of which may be their use as diagnostic tools for general circulation models of the ocean (see also Bryan et al. 1975; Sarmiento, this volume).

Berrien Moore III, Anders Björkström

17. Chemical and Biological Processes in CO2-Ocean Models

Abstract

The amount of CO₂ that will ultimately be released to the atmosphere by mankind’s use of fossil fuels will doubtless be several times the amount of CO₂ contained in the preindustrial atmosphere and equivalent to many times the amount of carbon that can be taken up by the terrestrial biosphere. Thus the oceans will become the sink for most of the excess carbon that does not remain in the atmosphere. To avoid too rapid or too great a rise in the atmospheric concentrations of CO₂ during this massive transfer of fossil carbon to the oceans, it will be necessary to predict reliably the rate at which the oceans will take up excess CO₂ from the atmosphere.

Charles F. Baes Jr., George G. Killough

18. Measurements of Total Carbon Dioxide and Alkalinity in the North Atlantic Ocean in 1981

Abstract

The ocean uptake of fossil fuel CO, has long been recognized as the principal modulator of the rising atmospheric CO, level. If we are to observe and understand this effect, then an essential step is the accurate measurement of the CO, properties of the ocean. Historically, this has been quite difficult to achieve. Although measurements of some kind date back to the late 19th century, complete, documented, and verifiable measurements are scarce indeed. This chapter describes and documents the series of total CO, and alkalinity measurements of seawater made on the North Atlantic Ocean during the Transient Tracers in the Ocean (TTO) expedition in 1981, and presents briefly the signals these data reveal.

Peter G. Brewer, A. L. Bradshaw, R. T. Williams

19. Geologic Analogs: Their Value and Limitations in Carbon Dioxide Research

Abstract

The CO₂ research community has recently shown much interest in the use of geologic analogs to verify climate model predictions. This application rests, of course, on the premise that both climate and atmospheric CO₂ have varied in the geologic past. Climate variability is widely documented in the geologic record, and CO₂ changes have recently been documented in ice cores. Although there is no other direct evidence for CO₂ variations in the geologic past, the relatively small size and short residence time of the atmospheric CO₂ reservoir suggest strongly that it must have been sensitive to perturbations in the larger reservoirs with which it exchanges.

Eric T. Sundquist

20. Requirements for a Satisfactory Model of the Global Carbon Cycle and Current Status of Modeling Efforts

Abstract

The concentration of CO₂ in the atmosphere is rising as a result of emissions into the atmosphere by fossil fuel combustion, deforestation, and expanding agriculture. Experiments with numerical models of the earth’s climatic system indicate that a further increase of the atmospheric CO₂ concentration might change climate on earth significantly. To understand how rapidly such possible changes might occur, we must be able to project the likely rate of change of atmospheric CO₂ due to future CO₂ emissions. For this purpose a better knowledge of the global carbon cycle is required, particularly of how atmospheric CO₂ is exchanging with the terrestrial ecosystems and the oceans. It has been maintained that about half of the CO₂ emitted into the atmosphere has stayed there; that is, the “airborne fraction” of the emissions has been about 50%. The emissions due to deforestation and changing land use need more careful consideration in this context, and the airborne fraction of the total emissions has not been determined accurately as yet. We need to know more precisely what has happened in the past to be able to validate models of the global carbon cycle.

Bert Bolin

21. The Use of Observations in Calibrating and Validating Carbon Cycle Models

Abstract

To determine the most appropriate data for calibrating and validating carbon cycle models, it is first necessary to determine the aims of the modeling study. Among the main uses of such studies, we can identify in particular:

1.

Prediction of future atmospheric CO₂ concentrations (to assess climatic and biological impacts)

 

2.

Reconstruction of past atmospheric CO₂ concentrations (to determine the driving force when attempting to analyze past climatic records in the search for a response to changes in CO₂)

 

3.

Interpretation of current measurements involving the carbon cycle (to determine the major carbon fluxes)

 

4.

Cross-comparisons of models with different degrees of resolution, including different dimensionality (to determine which models are most appropriate for particular studies).

 

Ian G. Enting, Graeme I. Pearman

22. Analysis of the Seasonal and Geographical Patterns of Atmospheric CO2 Distributions with a Three-Dimensional Tracer Model

Abstract

A great amount of information on the sources and sinks of atmospheric CO₂ is contained in the geographical, seasonal, and interannual variations of the global atmospheric CO₂ distribution. The measured concentrations of CO₂ at several locations illustrate large variations in the amplitude and phase of the seasonal cycle superimposed on an increasing long-term trend (see e.g., Gammon, this volume; Keeling 1983). Recent analysis of CO₂ records by Keeling and his collaborators (1983) reveals that the amplitude of seasonal cycle has detectable interannual variations and may be increasing with time.

Inez Y. Fung

23. Fossil Fuel Combustion: Recent Amounts, Patterns, and Trends of CO2

Abstract

Several types of human activity have introduced perturbations that impinge on the natural global carbon cycle. During the past century or so, one of the major perturbations has been the release of carbon from long-term storage through the combustion of fossil fuels. During the next 100 years fossil fuel use will almost certainly be the major source causing increased levels of CO₂ in the atmosphere.

Ralph M. Rotty, Gregg Marland

24. Distribution and Quantitative Assessment of World Crude-Oil Reserves and Resources

Abstract

The enumeration of world crude-oil reserves and resources is a difficult undertaking. One must rely on data gathered by others and, through various spot-check and cross-check mechanisms, distinguish the reasonable from the unreasonable. Estimating techniques vary; standards of measurement vary; classification of the reported elements differ; and, in some cases, even the recognition of what is being reported is obfuscated by language and nomenclatural ambiguities. Therefore, to improve understanding of world petroleum resources, it is prudent to describe the methodologies and measurement standards explicitly, to disaggregate the data as much as possible to gain insight into the source of the differences, and to standardize the classification and nomenclature scheme to establish unambiguous communication.

Charles D. Masters, David H. Root, William D. Dietzman

25. Long-Term Energy Projections and Novel Energy Systems

Abstract

Energy demand and supply projections are bound inextricably with external realities. Consequently, they reflect the dominant views on the development of those factors that primarily determine energy consumption. Some of these factors are highly uncertain or so difficult to quantify that it is impossible to model their effects. Other factors, such as economic activity or population development, can serve as variables that, to a large extent, determine energy consumption. Hence, parallel to the prospect of global economic activity slowly shifting from exponential growth (until the 1970s considered the norm) to a prolonged period of low growth, there have been similar downward revisions in the corresponding energy consumption. But, there are other reasons for the recent downward revisions of energy projections. In contrast to energy forecasters of the 1970s, today’s analysts can base their projections on empirical evidence about the response of industries and households to two unprecedented oil price increases of the last decade. Adjustments to the realities of high energy costs have resulted in energy efficiency improvements along the entire energy chain from resource extraction to consumption, as well as interfuel substitution, structural economic changes, and changes in individual attitudes.

Hans-Holger Rogner

26. Atmospheric CO2 Projections with Globally Averaged Carbon Cycle Models

Abstract

The principal objective of this analysis is to attempt to project the level of CO₂ in the atmosphere over the next century (to the year 2075). Knowledge of future atmospheric CO₂ concentrations is necessary to establish whether there are likely to be significant climatic and biological consequences resulting from continued fossil fuel combustion. To estimate the effects of projected fossil fuel emissions on future atmospheric CO₂ levels, it is necessary to know how this anthropogenic source impacts the natural biogeochemical cycle of carbon—a complex and dynamic set of processes linking the atmosphere, the oceans, and the terrestrial environment. Because of the complicated and changing nature of the anthropogenic source terms and the carbon cycle, simulation models that deal with relationships between economic patterns, energy use, and CO₂ emissions and with the major components of the global biogeochemical cycle are needed. These models appear to offer the only practical means to integrate the vast array of detailed quantitative data needed to provide projections of future atmospheric change. Such modeling efforts should be internally consistent with the natural carbon fluxes among atmospheric, oceanic, and terrestrial pools; they should accurately reflect the impact of fossil and biospheric releases over past centuries; and, of course, they should satisfactorily account for empirical measurements of recent decades (e.g., 316 ppmv in 1959 rising to 345 ppmv at present).

John R. Trabalka, James A. Edmonds, John M. Reilly, Robert H. Gardner, David E. Reichle

27. Possible Changes in Future Use of Fossil Fuels to Limit Environmental Effects

Abstract

Full use of the world’s recoverable fossil fuels could lead to atmospheric CO₂, concentrations up to 10 times the preindustrial concentration (Keeling and Bacastow 1977; Siegenthaler and Oeschger 1978; Perry et al. 1982) (See Fig. 27.1). It is not possible at present to specify a maximum allowable CO₂, concentration, i.e., a level at which the incremental losses would begin to exceed the incremental benefits, in some aggregate sense. Nevertheless, it is often assumed that a concentration of 600 parts per million by volume (ppm) would induce significant climatic and other changes that should perhaps be avoided if possible. Calculations of future CO₂, concentrations are usually based on estimates for future worldwide consumption of fossil fuels that do not incorporate any restrictions on CO₂, emissions. Some of us at the Oak Ridge National Laboratory, along with colleagues at the Institute for Energy Analysis and at the Massachusetts Institute of Technology, have asked how difficult it might be (and when it might be necessary to begin) to modify the future use of fossil fuels so that the atmospheric CO₂, concentration would never exceed some specified level, such as 500, 600, or 800 ppm (Perry et al. 1982).

Alfred M. Perry

Backmatter