Odekon M. Encyclopedia of paleoclimatology and ancient environments

Подождите немного. Документ загружается.

the data in Tables C1 and C4, is about 11 years. Greater land

coverage by either forests or grasses would lead to consider-

ably longer or shorter, respectively, residence times of carbon

in land phytomass. Recycling time by oceanic phytomass is

about 0.06 yr or 20 days.

Volatilization of carbon from soil humus is due to respiration

or mineralization of the reactive fraction of humus (15–20% of

total). This implies a faster recycling of soil organic matter than

may be suggested by the size of the whole reservoir.

The ratio of mean carbon fluxes of carbonate and organic

carbon (carbonate/organic) in long-term storage or burial in

sediments is 3.6–5.6, comparable to the mass ratio of 5.2 in

sedimentary reservoirs (Tables C1 and C4). However, the rivers

transport much more organic carbon relative to DIC, with the

DIC to organic carbon ratio between about 1 and 1.8

(Table C4). This indicates that the rate of organic matter recy-

cling in sediments is considerably faster than that of carbonates

on a geologically long timescale.

Carbon trends through time

Sedimentary mass records

Preservation of sediments is subject to recycling, a process in

which materials are eroded and dissolved, the products trans-

ported to the ocean where deposition or new mineral formation

occurs, and ultimately the materials are reincorporated in the con-

tinental sediment cover or crystalline crust, through continental

accretion and/or the endogenic cycle. As a result, the older the

sediment, the less is preserved. Furthermore, most older sedi-

ments of the Early Proterozoic and Archean Eons, older than

about 2 billion years (2 Ga), have been metamorphosed, and it

is difficult to estimate the original proportions of the different

sediment types in metamorphic rocks.

The rates of carbonate storage during the last 550 Ma

(Figure C2b) vary by about 50% from the mean, depending

on the data source, but they show a decline from the higher

values of the past to recent values. Organic carbon shows an

increase in the storage rate around 300 Ma before present due

to massive accumulation of coal in the Carboniferous Period.

Variations in the rates of organic carbon storage result in varia-

tions of the rates of oxygen production, as shown in Reactions

(3b), (5), and (6), with more oxygen generated when more

photosynthetically produced organic carbon is sequestered.

A record of isotopic composition of organic carbon and car-

bonate carbon since the Late Proterozoic time, about 780 Ma

ago, is shown in Figures C2a and C3. In general, an increase

in the values of d

carb

is accounted for by a greater rate of

Figure C2 (a) Distribution of d

C in the shells of biogenic calcite

during the Phanerozoic, based on approximately 4,500 samples. The

black line is a running mean and the shaded area includes the 95%

confidence limit about the mean (from Veizer et al., 1999). The base of

the Phanerozoic Eon is the base of the Cambrian Period, 542 Ma BP, as

given by the International Commission on Stratigraphy 2004.

(b) Calculated CaCO

and organic carbon storage rates (accumulation

rates) during the Phanerozoic (from data of Berner and Canfield, 1989;

Mackenzie and Morse, 1992; Robert A. Berner, Yale University, personal

communication, 2004).

Table C4 Carbon fluxes between major reservoirs in pre-industrial time

(Figure C1). Data from Ver et al. (1999) unless indicated otherwise.

(10

g ¼1 Gt)

Flux Units

(10

gCyr

1

) (10

mol C yr

1

)

Net primary production (NPP) 113.5 9,450

Land 63.1 5,250

Ocean 50.4 4,200

Volatilization from soil organic

matter

62.5 5,200

Weathering consumption of CO

0.13 11

Net exchange atmosphere-ocean

(dissolution 8 10

, evasion

8.042 10

mol yr

1

)

0.51 42

River input of dissolved C

(DIC + DOC)

0.60 50

DIC 0.38 32

DOC 0.22 18

POC 0.19 16

PIC 0.18 15

Oceanic sediment long-term storage 0.28 23

Carbonates

0.22 to 0.34 18 to 28

Organic matter

0.06 5

Volcanism, metamorphism,

hydrothermal

0.22 18

Uplift

0.40 33

Phanerozoic averages, from Figure C2b.

Uplift assumed as balanced by subduction of ocean floor over geologically

long periods (Mackenzie, 2002, p. 178).

CARBON CYCLE 115

storage of organic matter that incorporates

C in preference to

C. The fluctuating increase in d

carb

during the earlier part

of the Paleozoic Era, from 500 to 300 Ma ago, probably indi-

cates long-term expansion of marine biological production

and emergence of higher land plants in the Devonian Period,

between about 400 and 360 Ma ago (see also Carbon isotope

variations over geologic time). The magnitude of isotopic frac-

tionation due to biological production and storage of organic

carbon is the difference between the d

C values in carbonates

(taken as representing the DIC isotopic composition in ocean

water, Figure C2a) and organic carbon (Figure C3c; see preced-

ing section on Isotopic fractionation).

The fractionation factor (e,in%) varied within a range of

28–33% during the Paleozoic Era, from about 540 to 245 Ma.

Departures outside this range, in geologically earlier and later

periods may represent, at least to some extent, evolutionary

changes in the oceanic and terrestrial biotas. In Late Proterozoic

time, older than 540 Ma, large variations in the isotopic values of

organic and carbonate carbon (Figure C3a, b) may correspond to

the times of three major glaciations that occurred between 800

and 600 Ma ago, when rates of storage of organic carbon might

have been relatively low (Hayes et al., 1999, p. 111). A decline

in d

carb

since the beginning of the Cenozoic Era about

65 Ma ago (Figures C2a and 3b) is usually attributed to the rise

in the elevation of land and emergence of the major mountain

terrains, such as the Alps and the Himalayas, which exposed

more of the old organic carbon from shales to weathering,

thereby adding more

C-depleted carbon to the oceans. How-

ever, the reasons behind a decrease in biological fractionation

during the same period are not altogether clear.

As sedimentary carbonates and organic matter account for

most carbon on the Earth’s surface (Table C1), the isotopic

composition of these two reservoirs has been used to estimate

their mass fractions in the sedimentary record. Because carbon

on the Earth’s surface ultimately originated in the mantle, d

of the mantle should equal the mean d

C of the two major sur-

face reservoirs (sedimentary carbonates and organic carbon).

Thus, the isotopic composition balance requires that

mean

¼ f d

carb

þð1  f Þd

org

ð11Þ

where f is the mass fraction of carbonate carbon and 1 – f is the

mass fraction of organic carbon in sediments. The mean d

for mantle and surface carbon lies in the range of 5  2%,

the organic carbon reservoir in the Phanerozoic is about

28%, and the carbonate carbon is between +1 and +2%

(Figure C3a,b). These values of d

C, used in Equation (11), give

the fraction of organic carbon in sediments, 1 – f = 0.15 to 0.29.

The value given in Table C1 is 0.16.

Atmospheric CO

The latest major period of the Earth’s history, the Phanerozoic

Eon, covers approximately the last 550 Ma when such major

evolutionary events occurred as the emergence of large vascu-

lar land plants and different groups of CaCO

-secreting organ-

isms in the ocean. The biosphere and the global carbon cycle

most likely affected one another interactively throughout the

Earth’s history, when atmospheric CO

undoubtedly played a

major role in both because of the properties of the atmosphere.

Apart from its interactions with the biosphere, atmospheric

is variably controlled by the major tectonic processes of

the formation of the new ocean floor at the oceanic spread-

ing zones, subduction of oceanic sediments, and emission

of CO

from the deeper parts of the Earth by volcanism and

Figure C3 Carbon isotopic composition (d

C, in %) in the Late Proterozoic (age > 570 Ma) and Phanerozoic time of (a) organic matter and

(b) carbonates. (c) Fractionation between carbonate and organic carbon, e in % (from Hayes et al., 1999, and data provided by John M.

Hayes, Woods Hole Oceanographic Institution, June 2003).

116 CARBON CYCLE

metamorphic processes. Furthermore, the rates of weathering on

land, transport of weathered materials to the ocean, and sedi-

mentary storage of organic matter and carbonates are all parts

of the carbon cycle that involve the atmospheric CO

reservoir.

Atmospheric CO

underwent great changes in concentration

during the Phanerozoic, as shown in Figure C4. Relative to

its pre-industrial concentration of about 300 ppmv, CO

atmospheric levels might have been 15–25 times higher in

the Early Paleozoic time, before a decline began about

380 Ma ago due to emergence of large vascular land plants.

The high atmospheric CO

concentrations at that time might

have been comparable to those of modern soils that contain

about 30 times more CO

in their pore space than the atmo-

sphere, due to the respiration or oxidation of organic matter.

The calculated CO

curve, and its upper and lower limits of esti-

mated error (e.g., Berner and Kothavala, 2001 ), is based on a

geochemical model of the carbon cycle in a system of land-

ocean-atmosphere reservoirs, and it includes numerous pro-

cesses that affect weathering, biological production, and storage

of reduced and oxidized carbon in sediments.

Among such processes are the increase in solar luminosity

during the Phanerozoic, changes in the area and elevation of

the global land mass, changes in the abundance of gymnosperm

and angiosperm plants, water runoff from land, temperature,

volcanic activity, and interactions among these processes. The

low CO

values between about 380 and 280 Ma reflect the

emergence of land plants in the Devonian and storage of coal

in the Carboniferous Period. The higher values in the Mesozoic

Era, from about 240 to 100 Ma, reflect a low relief of land and

relatively low rate of utilization of atmospheric CO

in the

weathering of continental rocks and old sediments. Thereafter,

from the beginning of the Tertiary, the tectonically-driven rise

of land resulted in higher weathering rates and a decline of

atmospheric CO

The variations in shallow ocean temperature and atmo-

spheric CO

content during Phanerozoic time may be con-

trasted with those of the last 18,000 years, since the Last

Glacial Maximum to the beginning of industrial time about

200 years ago and to the present. Since the Last Glacial Max-

imum, global mean temperature increased from about 9



Cto



C and CO

from about 185 to 280 ppmv at the beginning

of industrial time, indicating a change in R

CO2

from 0.66 to 1.

Since 1861 to 2000, mean global temperature of the Earth’s

surface has increased further by about 0.7



C and atmospheric

has risen by a factor of 1.33 since pre-industrial time.

Concern is focused on the continued increase of atmospheric

due to burning of fossil fuels and human practices of land

use, and the possible release of methane from the methane-

hydrates due to warming (Figure C1). The rise in atmospheric

carbon dioxide since the Last Glacial Maximum, and especially

the last century, has not been countered by a greater biological

production of organic carbon and, importantly, its sequestration

in sediments that provide, together with mineral weathering, a

powerful geological buffer of the carbon cycle.

Abraham Lerman

Bibliography

Arrhenius, S., 1896. On the influence of carbonic acid in the air upon the

temperature of the ground. Phil. Mag., 5th ser., 41, 237–276.

Barker, D.S., 1997. Carbonatites. McGraw-Hill Encycl. Sci. Technol.,

8th ed., vol. 3, New York, NY: McGraw-Hill, pp. 239–240.

Bassham, J.A., 1974. Photosynthesis. Encycl. Brit., Macropaedia, vol. 14,

Chicago, IL: University of Chicago Press, pp. 365–373.

Berner, R.A., and Canfield, D.L., 1989. A new model for atmospheric oxy-

gen over Phanerozoic time. Am. J. Sci., 289, 333–361.

Berner, R.A., and Kothavala, Z., 2001. Geocarb III: A revised model of

atmospheric CO

over Phanerozoic time. Am. J. Sci., 301, 182–204.

Berner, R.A., and Maasch, K.A., 1996. Chemical weathering and controls

on atmospheric O

and CO

: Fundamental principles were enunciated

by J.J., Ebelmen, in 1845. Geochim. Cosmochim. Acta, 60, 1633–1637.

Clark, I.D., and Fritz, P., 1997. Environmental Isotopes in Hydrology.

New York, NY: Lewis Publishers, 328pp.

Dumas, J., 1842. Essai de Statique Chimique des Êtres Organisés. 2ème

édit. Fortin, Masson, Paris, 4 þ88pp.

Friedman, I., and O’Neil, J.R., 1977. Compilation of stable isotope fractio-

nation factors of geochemical interest. In Fleischer, M. (ed.), Data of

Geochemistry. 6th ed, U.S. Geol. Survey Prof. Pap. 440–KK.

Gaffron, H., 1964. Photosynthesis. Encycl. Brit., vol. 17, Chicago, IL:

University of Chicago Press, pp. 855–856B.

Graham, L., 1974. Heat. In Encyclopedia Britannica, Macropaedia, vol. 8,

pp. 700–706; Micropaedia, vol. 4, p. 1007. Chicago, IL: University of

Chicago Press.

Hayes, J.M., Strauss, H., and Kaufman, A.J., 1999. The abundance of

in marine organic matter and isotopic fractionation in the global biogeo-

chemical cycle of carbon during the past 800 Ma, Chem. Geol., 161,

103–125.

Kvenvolden, K.A., 1988. Methane — a major reservoir of carbon on a

shallow geosphere? Chem. Geol., 71,41–51.

Kvenvolden, K.A., and Lorenson, T.D., 2001. Global occurrences of nat-

ural gas hydrates. In Paull, C.E., and Dillon, W.P. (eds.), Natural Gas

Hydrates: Occurrence, Distribution, and Detection. AGU Geophys.

Monograph Series, vol. 124, pp. 3– 18.

Li, Y.-H., 2000. A Compendium of Geochemistry: From Solar Nebula

to the Human Brain. Princeton, NJ: Princeton University Press,

xiv þ475pp.

Locklair, R.E., and Lerman, A., 2005. A model of Phanerozoic cycles of

carbon and calcium in the global ocean: Evaluation and constraints on

ocean chemistry and input fluxes. Chem. Geol., 217,113–126.

Lotka, A.J., 1925. Elements of Physical Biology. Baltimore, MD: Williams &

Wilkins, xxx þ460pp. Also published as Elements of Mathematical

Biology, 1956. New York, NY: Dover, pp. xxx þ465.

Mackenzie, F.T., 2002. Our Changing Planet. Upper Saddle River, NJ:

Prentice Hall, xii þ580pp.

Mackenzie, F.T., and Lerman, A., 2006. Carbon in the Geobiosphere.

Dordrecht, The Netherlands: Springer xxii þ402pp.

Mackenzie, F.T., and Morse, J.W., 1992. Sedimentary carbonates through

Phanerozoic time. Geochim. Cosmochim. Acta, 56, 3281–3295.

Meyer, B.S., 1964. Plant Physiology. Encycl. Brit., vol. 18, Chicago, IL:

University of Chicago Press, pp. 16–31.

Figure C4 Calculated ratio of atmospheric CO

to the Holocene

Epoch value taken as 300 ppmv (R

CO2

) and shallow ocean temperatures

(from data of Berner and Kothavala, 2001, and Wallmann, 2004,as

plotted in Locklair and Lerman, 2005).

CARBON CYCLE 117

Mohr, F., 1875. Geschichte der Erde. 2. Aufl. Bonn, Germany: Verlag Max

Cohen & Sohn, xx þ554pp.

Mojzsis, S.J., Arrhenius G., McKeegan, K.D., Harrison, T.M., Nutman, A.P.,

and Friend, C.R.L., 1996. Evidence for life on Earth by 3800 Myr.

Nature, 384,55– 59.

Mook, W.G., Bommerson, J.C., and Staverman, W.H., 1974. Carbon iso-

tope fractionation between dissolved bicarbonate and gaseous carbon

dioxide. Earth Planet. Sci. Lett., 22, 169–176.

O’Leary, M.H., 1988. Carbon isotopes in photosynthesis. Bioscience, 38,

328–335.

Poldervaart, A., 1955. Chemistry of the Earth’s surface. In Poldervaart, A.

(ed.), Crust of the Earth, Geol. Soc. Am. Spec. Pap., 62, pp. 119–144.

Rankama, K., and Sahama, Th.G., 1950. Geochemistry. Chicago, IL: Univ.

Chicago Press, xvi þ912pp.

Redfield, A.C., Ketchum, B.H., and Richards, F.A., 1963. The influence of

organisms on the composition of seawater. In Hill, M.N. (ed.), The Sea,

vol. 2, New York, NY: Wiley, pp. 26–77.

Rosing, M.T., 1999.

C-depleted carbon microparticles in >3700-Ma sea-

floor sedimentary rocks from West Greenland. Science, 283, 674–676.

Rubey, W.W., 1951. Geologic history of seawater, an attempt to state the

problem. Geol. Soc. Am. Bull., 62, 1111–1147.

Rubey, W.W., 1955. Development of the hydrosphere and atmosphere, with

special reference to probable composition of the early atmosphere. In

Poldervaart, A. (ed.), Crust of the Earth, Geological Society of America,

Special Paper, vol 62, pp. 631–650.

Rubinson, M., and Clayton, R.N., 1969. Carbon-13 fractionation between

aragonite and calcite. Geochim. Cosmochim. Acta, 33, 997–1002.

Salomons, W., and Mook, W.G., 1986. Isotope geochemistry of carbonates

in the weathering zone. In Fritz, P., and Fontes, J. Ch. (eds.), Handbook

of Environmental Isotope Geochemistry, vol. 2, The Terrestrial Environ-

ment, Amsterdam, The Netherlands: Elsevier, pp. 239–269.

Thode, H.G., Shima, M., Rees, C.E., and Krishnamurty, K.V., 1965.

Carbon-13 isotope effects in systems containing carbon dioxide, bicar-

bonate, carbonate, and metal ions. Can. J. Chem., 43, 582–595.

Urey, H.C., 1952. The Planets: Their Origin and Development. New

Haven, Conn.: Yale University Press, xvii þ245pp.

Veizer, J., Ala, D., Azmy, K., Bruckschen, P., Buhl, D., Bruhn, F., Carden,

G.A.F., Diener, A., Ebneth, S., Godderis, Y., Jasper, T., Korte, C.,

Pawellek, F., Podlaha, O.G., and Strauss, H., 1999.

Sr/

Sr, d

and d

O evolution of Phanerozoic seawater. Chem. Geol., 161,59–88.

Ver, L.M.B., Mackenzie, F.T., and Lerman, A., 1999. Biogeochemical

responses of the carbon cycle to natural and human perturbation: Past,

present, and future. Am. J. Sci., 299, 762–801.

Vogel, J.C., Grootes, P.M., and Mook, W.G., 1970. Isotopic fractionation

between gaseous and dissolved carbon dioxide. Zeitschr. Physik, 230,

225–238.

Walker, J.C.G., 1977. Evolution of the Atmosphere. New York, NY:

Macmillan Publishing, xiv þ318pp.

Wallmann, K., 2004. Impact of atmospheric CO2 and galactic cosmic

radiation on Phanerozoic climate change and the marine d

O record.

Geochem., Geophys. Geosyst. doi: 10.1029/ 2003GC000683.

Whitmarsh, J., and Govindjee, 1995. The photosynthetic process. In

Singhal, G.S., Renger, G., Soppory, S.K., Irrgang, K.-D., and Govindjee

(eds.), Concepts in Photobiology: Photosynthesis and Photomorpho-

genesis. Dordrecht, The Netherlands: Kluwer Academic, pp. 11–51.

Wilkinson, B.H., and Algeo, T.J., 1989. Sedimentary carbonates record of

calcium and magnesium cycling. Am. J. Sci., 289, 1158–1194.

Cross-references

Archean Environments

Atmospheric Evolution, Earth

Carbon Dioxide, Dissolved (Ocean)

Carbon Isotope Variations Over Geologic Time

Carbon Isotopes, Stable

Faint Young Sun Paradox

Isotope Fractionation

Marine Biogenic Sediments

Marine Carbon Geochemistry

Methane Hydrates, Carbon Cycling, and Environmental Change

Ocean Paleoproductivity

Phosphorus Cycle

Weathering and Climate

CARBON DIOXIDE AND METHANE, QUATERNARY

VARIATIONS

Introduction

A straightforward method to reconstruct variations of atmo-

spheric greenhouse gas concentrations during the past few

hundred thousand years is the analysis of air enclosed in

bubbles of suitable polar ice. For polar ice to be suitable, the

possibility of any influence of melt water must be excluded,

which is the case for very cold regions in the center of the Ant-

arctic and the Greenland Ice Sheet, in part. Such ice is formed

by the compaction of snow and firn by a dry sintering process.

Atmospheric air is gradually enclosed in bubbles at the transi-

tion from firn to ice, typically 70–120 m below the surface. If

the age of air in ice is defined as the time since its separation

from the free atmosphere, it is clear that it is younger than

the surrounding ice (defined as the time since deposition as

snow). The age difference varies between a few decades for

high accumulation rate sites and up to 5 millennia or more

for low accumulation rate sites. For a given drilling site, the

age difference generally varies with depth due to variations of

temperature and accumulation rates in the past.

The gradual enclosure of air in bubbles also implies that air

of an ice sample does not have a well-defined age but has a cer-

tain age distribution, depending on temperature and accumula-

tion rate. This limits the possible time resolution of greenhouse

gas records to a few years for high accumulation rate sites and

more than a century for low accumulation rate sites. The air is

well preserved in the bubbles. Diffusion of gases in cold ice is

low. However, chemical reactions with impurities cannot be

excluded, even in very cold ice.

The main foci of analyses of air samples extracted from air

bubbles are the concentrations of the greenhouse gases CO

and CH

, together with N

O and the O

ratio. The isotopic

composition of the gases is also of interest, especially in finding

possible causes of observed variations. In the following descrip-

tion, only variations of the atmospheric concentration of the

important greenhouse gases CO

and CH

will be discussed.

Records measured along the Antarctic Vostok (78



106



E) core go back to 420,000 yrBP (Petit et al., 1999)

and those from the recently drilled Dome C (75



S, 123



E) core date back to 740,000 YrBP (EPICA members,

2004). More detailed records are available for the last glacial

epoch, the transition from the last glacial epoch to the Holo-

cene, the Holocene and the last millennium. All these time

intervals are discussed in separate sections.

Atmospheric CO

and CH

concentrations are relatively

constant globally at any given time due to a short global mix-

ing time. However, small CH

concentration differences bet-

ween Greenland and Antarctica allow a rough estimate of the

latitudinal distribution of CH

sources during different climatic

epochs.

Analytical methods

To measure the CO

and CH

concentration of air enclosed in

air bubbles of polar ice samples, the air has first to be extracted

and then measured in a second step. For the extraction of air for

measurements, most laboratories use a dry extraction

method to avoid any contamination with CO

that could

be produced by the interaction of carbonates with water. Air

118 CARBON DIOXIDE AND METHANE, QUATERNARY VARIATIONS

bubbles are opened mechanically by cracking, milling or grat-

ing ice samples at low temperature. For CH

measurements,

air is removed by wet or dry extraction. In a typical wet extrac-

tion, ice is melted in an evacuated container and after all ice has

melted, the sample is frozen very slowly from the bottom again

to expel any air dissolved in the meltwater.

The CO

concentration in the air is measured by gas chro-

matography or by laser absorption spectroscopy. For the second

method, a tunable infrared laser (wavelength about 4.23 mm)

is tuned several times over the absorption line of a vibration-

rotation transition of the CO

molecule. The CH

concentration

is generally measured by gas chromatography. Measurement by

laser absorption spectroscopy is possible but has not been used

for routine measurements yet.

The minimal sample size for a single CO

measurement is

12 g ice; for a single CH

measurement, about 40 g ice is

needed. Certain laboratories are using larger sample sizes but

perform multiple analyses on the extracted air sample.

Reliability of CO

and CH

data from ice cores

The reproducibility of CO

measurements in air of polar ice

samples is on the order of 1.5 ppmv. Results from direct atmo-

spheric measurements of several ice cores that have an overlap-

ping time period are in good agreement. However, this is no

guarantee that the air in bubbles of older ice has exactly the

composition of atmospheric air at the time of ice formation.

could be produced very slowly by carbonate-acid reac-

tions or oxidation of organic material, or CO

dissolved or

enclosed in micro bubbles in snowflakes could diffuse slowly

into air bubbles. Only air from normal size bubbles and clath-

rates are extracted by dry extraction methods. The agreement

of results for ice cores from various drill sites with different

impurity concentrations show that the effect would be small,

but cannot be excluded. For inland sites from Antarctica, it

can be concluded that such artifacts are smaller than 5 ppmv.

The reproducibility of CH

measurements is about 10 ppbv.

is much less soluble in water than CO

, but still twice as

soluble as N

. Therefore, it cannot be expected that analyses

of ice from locations with a mean temperature above 20



will provide reliable results. Methane results are certainly less

affected by chemical reaction between impurities in the ice,

but a small production of CH

by chemical reactions or by bio-

logical activity cannot be excluded (Sowers, 2001).

Variations of atmospheric CO

and CH

concentrations

during the past 650,000 years

Figure C5 shows variations of atmospheric CO

and CH

con-

centrations during the past 650,000 years compared with the

Antarctic temperature derived from dD records. The CO

and

concentrations are based on analyses performed along

the Vostok and the Dome C ice cores (Petit et al., 1999;

Siegenthaler et al., 2005; Spahni et al., 2005). At Vostok, an

ice core reached a depth of 3,623 m in 1998. The stratigraphy

is undisturbed down to a depth of 3,310 m, and the ice at this

depth is about 420,000 years old. More recently, an ice core

was drilled at EPICA Dome C down to a depth of 3,270 m.

The stratigraphy is probably undisturbed down to a depth

of 3,200 m where the ice has an estimated age of about

800,000 years. Gas analyses have hitherto been published for

ages up to 650,000 years. A very important result is that the

atmospheric CO

and CH

concentrations never reached the

present levels (about 370 ppm for CO

and 1,700 ppb for

in 2002) during this entire period.

The delta-deuterium (dD) record is a measure of regional

temperature (see Deuterium, deuterium excess, this volume).

A rather close correlation between variations of the two green-

house gas concentrations and Antarctic temperature is obvious.

variations are more or less caused by climatic variations

but there is also an important interplay between climatic and

variations, as will be discussed later.

Temperature during interglacial epochs before 430,000 yr

was lower than in the subsequent interglacials but the duration

of warm phases was longer. CO

concentrations were also

lower in the interglacial epochs before 430,000 yr

BP. At the

end of glacial epochs, temperature started to increase between

600 and 2,800 years before the beginning of the increase in

Figure C5 CO

and CH

records from Dome C and Vostok compared with the dD record from Dome C, which are characteristic for the local

temperature (Petit et al., 1999; EPICA members, 2004; Siegenthaler et al., 2005; Spahni et al., 2005). Black lines are from Dome C, gray lines from

Vostok. The shaded areas characterize interstadials.

CARBON DIOXIDE AND METHANE, QUATERNARY VARIATIONS 119

the CO

concentration. This time lag does not call into question

the interplay between temperature and CO

increase at the end

of glacial epochs, which take place over several millennia.

The last glacial epoch

The last glacial epoch was characterized by large and fast tem-

perature variations in Greenland, so called Dansgaard-Oeschger

events, as documented by the d

O profile shown in Figure C6

(Dansgaard et al., 1993). The CH

concentration measured

along the GRIP (Greenland Ice core Project) ice core shows

variations parallel to the temperature variations, with higher

concentrations during mild interstadials and lower concentra-

tions in cold phases (Chappellaz et al., 1993). The variations

are mainly caused by changes in CH

sources due to the pro-

ductivity and the extension of wetlands in low and mid lati-

tudes. The CH

variations are, therefore, a very convincing

indication that Dansgaard-Oeschger events were not limited

to Greenland but were climatic events of global significance.

The deuterium record from Antarctica and the d

O record

from Greenland are scaled so that equal amplitudes correspond

approximately to equal regional temperature variations. While

these two records are quite different, the CH

records from

Greenland and Antarctica are practically identical and show the

same characteristics. Fast variations can, therefore, be used to

synchronize age scales between Antarctic and Greenland ice core

records (Blunier and Brook, 2001). This synchronization, how-

ever, shows an asynchrony in temperature variations between

Greenland and Antarctica. As soon as a fast temperature increase

in Greenland occurred, the temperature in Antarctica decreased.

The CO

concentrations measured along the Taylor Dome

ice core (Indermühle et al., 2000) also show significant fluctua-

tions but these are smoother and smaller than for CH

. The var-

iations are parallel to Antarctic temperature. Temperature

variations in Antarctica were smaller than in Greenland, but

were accompanied by large variations in non-sea salt calcium

deposition (nssCa) in Antarctica, indicating significant varia-

tions of terrestrial dust. A possible cause of the CO

variations

with an amplitude of about 20 ppmv could, therefore, have

been a variation of iron fertilization by eolian dust in the

Southern Ocean (Röthlisberger et al., 2004).

The transition from the last glacial epoch to the

Holocene

The transition from the last glacial epoch to the Holocene was

the last large global temperature increase and was accompa-

nied by a substantial increase in atmospheric CO

and CH

It is therefore a key period for investigating the interactions

between global temperature and greenhouse gas concentrations.

Figure C7 shows a section of Figure C6 with greater resolution

(Monnin et al., 2001). The CO

concentration increased by 76

ppmv from 189 to 265 ppmv between about 17,000 and 11,100

BP in Antarctica. The increase occurred in four clearly distin-

guishable intervals. After a first, relatively fast (20 ppmv per

thousand years) linear increase (I), a reduced linear increase fol-

lowed in interval II, with a very fast increase of about 8 ppmv

at its end. In interval III, characterized by a temperature dec-

rease in Antarctica (Antarctic Cold Reversal), the CO

con-

centration also decreased slightly. In interval IV the CO

concentration increased at about the same rate as in interval I.

This interval was concluded by another fast increase of

about 6 ppmv, rising to 265 ppmv at the beginning of the

Holocene.

The increase of the CO

concentration parallels the tempera-

ture increase in Antarctica, confirming that the Southern Ocean

played a key role in causing the CO

increase. The rapid con-

centration increases at the end of interval II and IV occurred

during rapid temperature increases in the Northern Hemisphere,

indicating that it also had a significant global influence. Several

Figure C6 CO

and CH

records representing variations of the past 80,000 years compared with stable isotope records from Antarctica and

Greenland, characterizing the corresponding local temperature (Chappellaz et al., 1993; Dansgaard et al., 1993; Indermu

hle et al., 2000; Blunier and

Brook, 2001). Open circles of the CO

record are from Taylor Dome and diagonal crosses from Dome C ice cores (Antarctica). The dD record is also

from Antarctica. The CH

record is from the Greenland GRIP ice core. Vertical lines highlight some important, fast temperature increases in

Greenland (d

O curve).

120 CARBON DIOXIDE AND METHANE, QUATERNARY VARIATIONS

mechanisms have been suggested as causes for the atmospheric

increase: Reduced solubility of CO

in the Southern

Ocean because of temperature increase, enhanced air-sea

exchange due to reduced sea-ice extent, decreasing bioactivity

due to reduced iron fertilization by eolian dust and changes in

deepwater formation parallel to the fast temperature increases

in the North Atlantic region.

The evolution of the atmospheric CH

concentration during

the transitional periods is, as during Dansgaard-Oeschger

events in the glacial epoch, approximately parallel to that of

the Greenland d

O record. However, the beginning of the

increase lags the beginning of the temperature increase

by 800  600 years (Monnin et al., 2001). This time lag is

small compared with the duration of 6,000 years for the entire

increase. It does not call into question the important amplifica-

tion effect of the CO

increase with regards to the global tem-

perature increase during the transition from the glacial epoch to

the Holocene. The increase of atmospheric CO

concentration

from 189 to 265 ppmv corresponds to a direct radiative forcing

of about 2 W m

2

. The radiative forcing caused by CH

inc-

rease is about 10 times smaller (Joos, 2005). CH

increases par-

allel to fast temperature increases in Greenland are typically

delayed by a few decades giving rise to the conclusion that

variations are mainly a response to abrupt northern

climatic changes rather than a driver.

The Holocene

Ice cores from Taylor Dome and Dome C allow reconstruction

of detailed records of the CO

and CH

concentration during

the Holocene (Blunier et al., 1995; Flückiger et al., 2002;

Indermühle et al., 1999; Monnin et al., 2004). Both records are

shown in Figure C8. The CO

record clearly shows that the

carbon cycle was not in a steady state during the Holocene. The

concentration decreased from 268 ppmv at 10,500 yrBP to

260 ppmv at about 8,200 yr

BP and increased steadily in the

following 7,000 years to about 280 ppmv at the beginning of

Figure C7 Evolution of the CO

and CH

concentration increases during the transition from the last glacial epoch to the Holocene, compared

with records characterizing the temperature increases in Antarctica and Greenland (Monnin et al., 2001). Roman numbers mark the four time

intervals mentioned in the text.

Figure C8 CO

and CH

records representing the pre-industrial

Holocene (Blunier et al., 1995; Indermu

hle et al., 1998; Flu

ckiger et al.,

2002; Monnin et al., 2004). CH

concentrations in Greenland are on

average about 40 ppbv higher than in Antarctica.

CARBON DIOXIDE AND METHANE, QUATERNARY VARIATIONS 121

the last millennium. These variations are caused by the ocean

and the terrestrial biosphere. An uptake of CO

by the terres-

trial biosphere could be responsible for the decrease until

8,200 yr

BP, with a release of CO

from the slightly relapsing

biosphere and the ocean (due to an increase of the surface tem-

perature) accounting for the increase afterwards. Because car-

bon in the terrestrial biosphere is significantly depleted in the

carbon isotope

C, it is quite clear that d

C measured on

extracted from ice cores can help in estimating the relative

contributions by the ocean and terrestrial biosphere. Based on

presently available d

C results, the CO

release by the ocean

was responsible for about one third of the increase after 8,200

BP, and the release by the terrestrial biosphere responsible

for about two thirds.

The atmospheric CH

concentration decreased in Antarctica

from 720 ppbv at 10,500 yr

BP to about 575 ppbv at about

5,000 yr

BP. The results from the Greenland GRIP ice core

show a larger scatter and are about 45 ppbv higher on average.

The higher average in Greenland reflects the heterogeneous

latitudinal distribution of CH

sources and the fact that the

atmospheric residence time of CH

is only an order of magni-

tude longer than the inter-hemispheric exchange time. These

differences between measured CH

concentrations in Green-

land and Antarctica can be used to estimate a rough latitudinal

distribution of CH

sources as will be outlined in a later sec-

tion. The result of the estimate is that the CH

sources in the

tropics were increasing from the Last Glacial Maximum until

about 8,000 yr

BP, then decreasing until about 5,000 yrBP and

subsequently increasing again. Sources north of 30



N incre-

ased dramatically between the glacial maximum until about

10,000 yr

BP, but surprisingly started to decrease again until

8,000 yr

BP and increased again afterwards.

Remarkable is a significant CH

minimum that occurred for

a short duration. It is synchronous with a temperature drop of

about 7



C in Greenland and had a duration of 100–150 years.

The CH

minimum is more pronounced in the Greenland

record due to the higher accumulation rate. At Dome C in

Antarctica, where the accumulation rate is more than six times

lower, a signal of only about 100 year duration is attenuated by

about 50% (Spahni et al., 2003).

The last millennium

The CO

and CH

concentrations were rather constant during

the first eight centuries but increased dramatically during the

last two centuries to values never experienced during the last

650,000 years, as shown in Figure C9. The CO

concentrations

of different Antarctic ice cores are in good agreement, and also

agree with the short overlapping interval and continuous direct

atmospheric measurements, which started in 1958 (Blunier

et al., 1995; Etheridge et al., 1996; MacFarling Meure et al.,

2006). For the first eight centuries, they show a mean concen-

tration of about 280 ppmv. Measurements from Greenland ice

cores show significantly higher values, caused most probably

by artifacts due to higher temperature and higher impurity con-

centrations. Greenland results are, therefore, not shown in

Figure C9. The most detailed Antarctic record for the last mil-

lennium is available from the Law Dome ice core. The repro-

ducibility of the measurements is 1.2 ppmv and individual

results represent short time intervals due to a very high accu-

mulation rate at the coastal Law Dome. The Law Dome record

suggests an increase from about 279 ppmv at the beginning

of the millennium to 282 ppmv at

AD 1100, then a relatively

constant concentration until

AD 1600, followed by a decrease

to 277 ppmv, lasting until about

AD 1750.

The CH

record shows relatively constant values of about

700 ppbv until

AD 1750. Greenland values are slightly higher,

but in this case it is assumed that this is not an artifact but is

caused by the inter-hemispheric difference of the atmospheric

concentration. Records from Law Dome again provide results

that are more detailed. The concentration is about 640 ppbv at

Figure C9 CO

records from Antarctica and CH

records from

Greenland and Antarctica for the last millennium (Blunier et al. 1995;

Etheridge et al., 1996; MacFarling Meure et al., 2006).

Figure C10 Estimated CH

sources for different climatic epochs from

90–30



N, 30



N–30



S and 30–90



S based on concentration

differences between Greenland and Antarctica (Chappellaz et al., 1997;

llenbach et al., 2000).

122 CARBON DIOXIDE AND METHANE, QUATERNARY VARIATIONS

the beginning of the millennium and shows smaller variations but

also an increasing trend to a value of 700 ppbv at

AD 1750. The

increase could already be attributed to pre-industrial

anthropogenic influences, such as expansion of rice cultivation

and cattle.

Latitudinal distribution of pre-industrial CH

sources

Not only does the CH

concentration change with climatic varia-

tions but so does the inter-hemispheric concentration difference.

This combination of results can be used to make a rough estimate

of the latitudinal source distribution. In a simple model, sources in

the three boxes 90–30



N (northern box), 30



N–30



S (tropical

box) and 30–90



S (southern box) are taken into account

(Chappellaz et al., 1997). It is assumed that the source in the

southern box contributed 12 Tg yr

-1

during the cold periods of

the last glacial period and 15 Tg yr

-1

during the mild periods

of the last glacial epoch and the pre-industrial Holocene. The

model results are shown in Figure C10 (Dällenbach et al., 2000).

A remarkable, but still tentative, conclusion based on these

estimates is that the CH

variations during the last glacial epoch

were not mainly due to source changes in the tropics but in the

northern box.

Bernhard Stauffer

Bibliography

Blunier, T., and Brook, E.J., 2001. Timing of Millennial-Scale Climate

change in Antarctica and Greenland during the Last Glacial Period.

Science, 291(5501), 109–112.

Blunier, T., Chappellaz, J., Schwander, J., Stauffer, B., and Raynaud, D.,

1995. Variations in atmospheric methane concentration during the

Holocene epoch. Nature, 374,46–49.

Chappellaz, J., Blunier, T., Raynaud, D., Barnola, J.M., Schwander, J., and

Stauffer, B., 1993. Synchronous changes in atmospheric CH

and

Greenland climate between 40 and 8 kyr BP. Nature, 366, 443–445.

Chappellaz, J., Blunier, T., Kints, S., Dällenbach, A., Barnola, J.-M.,

Schwander, J., Raynaud, D., and Stauffer, B., 1997. Changes in the

atmospheric CH

gradient between Greenland and Antarctica during

the Holocene. J. Geophys. Res., 102(D13), 15987–15999.

Dällenbach, A., Blunier, B., Flückiger, J., Stauffer, B., Chappellaz, J., and

Raynaud, D., 2000. Changes in the atmospheric CH

gradient between

Greenland and Antarctica during the last glacial and the transition to the

Holocene. Geophys. Res. Lett., 27(7), 1005–1008.

Dansgaard, W., Johnsen, S.J., Clausen, H.B., Dahl-Jensen, D., Gundestrup, N.S.,

Hammer, C.U., Hvidberg, C.S., Steffensen, J.P., Sveinbjörnsdottir, A.E.,

Jouzel, J., and Bond, G., 1993. Evidence for general instability of past

climate from a 250 kyr ice-core record. Natur e, 364, 218–220.

EPICA members, 2004. Eight glacial cycles from an Antractic ice core,

Nature, 429, 623–628.

Etheridge, D.M., Steele, L.P., Langenfields, R.L., Francey, R.J., Barnola, J.-M.,

and Morgan, V.I., 1996. Natural and anthropogenic changes in atmospheric

over the last 1000years from air in Antarctic ice and firn. J. Geophys.

Res., 101,4115–4128.

Flückiger, J., Monnin, E., Stauffer, B., Schwander, J., Stocker, T., Chappellaz, J.,

Raynaud, D., and Barnola, J.-M., 2002. High resolution Holocene N

ice core record and its relationship with CH

and CO

. Global Biogeochem.

Cycles, 16(1), doi: 10.1029 /2001GB001417.

Indermühle, A., T.F. Stocker, F. Joos, H. Fischer, H.J. Smith, M. Wahlen,

B. Deck, D. Mastroianni, J. Tschuni, T. Blunier, R. Meyer, B. Stauffer,

1999. Holocene carbon-cycle dynamics based on CO

trapped in ice at

Taylor Dome, Antarctica, Nature, 398, 121–126.

Indermühle, A., Monnin, E., Stauffer, B., Stocker, T., and Wahlen, M.,

2000. Atmospheric CO

concentration from 60 to 20 kyr BP from the

Taylor Dome ice core, Antarctica. Geophys. Res. Lett., 27(5), 735–738.

Joos, F., 2006. Radiative forcing and the ice core greenhouse gas record.

PAGES News, 13(3), 11–13.

MacFarling Meure, C., Etheridge, D., Trudinger , C., Steele, P., Langenfelds, R.,

van Ommen, T ., Smith, A., and Elkins, J., 2006. Law Dome CO

,CH

and N

O ice core records extended to 2000 years BP. Geophys. Res. Lett.,

33, L14810, doi: 10.1029/2006GL026152.

Monnin, E., Indermühle, A., Dällenbach, A., Flückiger, J., Stauffer, B.,

Stocker, T., Raynaud, D., and Barnola, J.-M., 2001. Atmospheric CO

Con-

centrations over the Last Glacial Termination. Science, 291(5501),

112–114.

Monnin, E., Steig, E.J., Siegenthaler, U., Kawamura, K., Schwander, J.,

Stauffer, B., Stocker, T.F., Morse, D.L., Barnola, J.-M., Bellier, B.,

Raynaud, D., and Fischer, H., 2004. Evidence for substantial accu-

mulation rate variability in Antarctica during the Holocene, through

synchronization of CO

in the Taylor Dome, Dome C and DML ice cores.

Earth Planet. Sci. Lett., 224, doi: 10.1016/ j.epsl.2004.05.007, 45–54.

Petit, J.R., Jouzel, J., Raynaud, D., Barkov, N.I., Barnola, J.-M., Basile, I.,

Bender, M., Chappellaz, J., Davis, M., Delaygue, M., Delmotte, M.,

Kotlyakov, V.M., Legrand, M., Lipenkov, V.Y., Lorius, C., Pépin, L.,

Ritz, C., Saltzman, E., and Stievenard, M., 1999. Climate and atmospheric

history of the past 420,000 years from the Vostok ice core, Antarctica,

Nature, 399,429–436.

Röthlisberger, R., Bigler, M., Wolff, E.W., Joos, F., Monnin, E., and

Hutterli, M.A., 2004. Ice core evidence for the extent of past atmo-

spheric CO

change due to iron fertilisation. Geophys. Res. Lett., 31,

L16207, 1–4.

Siegenthaler, U., Monnin, E., Kawamura, K., Spahni, R., Schwander, J.,

Stauffer, B., Stocker, T., Barnola, J.-M., and Fischer, H., 2005. Support-

ing evidence from the EPICA Dronning Maud Land ice core for

atmospheric CO

changes during the past millennium. Tellus, 57B,

51–57.

Sowers, T., 2001. N

O record spanning the penultimate deglaciation from

the Vostok ice core. J. Geophys. Res., 106(D23), 31903–31914.

Spahni, R., Schwander, J., Flückiger, J., Stauffer, B., Chappellaz, J., and

Raynaud, D., 2003. The attenuation of fast atmospheric CH

variations

recorded in polar ice cores. Geophys. Res. Lett., 30(11), 1571, doi:

10.1029/ 2003GL017093.

Spahni, R., Chappellaz, J., Stocker, T.F., Loulergue, L., Hausammann, G.,

Kawamura, K., Flückiger, J., Schwander, J., Raynaud, D., Masson-

Delmotte, V., and Jouzel, J., 2005. Atmospheric Methane and Nitrous

Oxide of the Late Pleistocene from Antarctic Ice Cores. Science,

310(5752), 1317–1321.

Cross-references

Antarctic Cold Reversal

Carbon Isotopes, Stable

Climate Variability and Change, Last 1,000 Years

Dansgaard-Oeschger Cycles

Deuterium, Deuterium Excess

Holocene Climates

Ice cores, Antarctica and Greenland

Last Glacial Maximum

Oxygen Isotopes

Pleistocene Climates

Quaternary Climate Transitions and Cycles

CARBON DIOXIDE, DISSOLVED (OCEAN)

The ocean contains about 60 times more carbon in the form of

dissolved inorganic carbon than in the pre-anthropogenic atmo-

sphere (600 Pg C). On time scales <10

yr, the ocean is the

largest inorganic carbon reservoir (38,000 Pg C) in exchange

with atmospheric carbon dioxide (CO

) and as a result, the

ocean exerts a dominant control on atmospheric CO

levels.

The average concentration of inorganic carbon in the ocean is

2.3 mmol kg

1

and its residence time is 200 ka.

Dissolved carbon dioxide in the ocean occurs mainly in

three inorganic forms: free aqueous carbon dioxide (CO

(aq)),

bicarbonate (HCO



), and carbonate ion (CO

2

). A minor form

CARBON DIOXIDE, DISSOLVED (OCEAN) 123

is true carbonic acid (H

) whose concentration is less than

0.3% of [CO

(aq)]. The sum of [CO

(aq)] and [H

]is

denoted as [CO

]. The majority of dissolved inorganic carbon

in the modern ocean is in the form of HCO



(>85%), as

described below.

Carbonate chemistry

In thermodynamic equilibrium, gaseous carbon dioxide

(CO

(g)), and [CO

] are related by Henry’s law:

ðgÞ

½ ð1Þ

where K

is the temperature and salinity dependent solubility

coefficient of CO

in seawater (Weiss, 1974). The concentra-

tion of dissolved CO

and the fugacity of gaseous CO

f CO

, then obey the equation [CO

] ¼K

f CO

, where the

fugacity is virtually equal to the partial pressure, pCO

(within

1%). The dissolved carbonate species react with water,

hydrogen and hydroxyl ions and are related by the equilibria:

þ H

O ¼



HCO



þ H



2

þ 2H

ð2Þ

The pK

values (¼log(K

)) of the stoichiometric dissociation

constants of carbonic acid in seawater are pK

¼5.94 and

¼9.13 at temperature T ¼15



C, salinity S ¼35, and sur-

face pressure P ¼1 atm (Prieto and Millero, 2001). At typical

surface seawater pH of 8.2, the speciation between [CO

[HCO



], and [CO

2

] is 0.5%, 89%, and 10.5%, respectively,

showing that most of the dissolved CO

is in the form of

HCO



and not in the form of CO

(Figure C11 ). The sum of

the dissolved carbonate species is denoted as total dissolved

inorganic carbon (SCO

SCO

¼ CO

½þHCO





þ CO

2



ð3Þ

This quantity is also represented as DIC, TIC, TCO

, and C

Another essential parameter to describe the carbonate system

is the total alkalinity (TA), a measure of the charge balance in

seawater.

TA ¼ HCO





þ 2CO

2



þ BOHðÞ





þ OH



½H

½þminor compounds

ð4Þ

SCO

and TA are conservative quantities, i.e., their concentra-

tions measured in gravimetric units (mol kg

1

) are unaffected

by changes in pressure or temperature, for instance, and they

obey the linear mixing law. Therefore, they are the preferred

tracer variables in numerical models of the ocean’s carbon

cycle. Of all the carbon species and carbonate system para-

meters described above, only pCO

, pH, SCO

, and TA can

be determined analytically. However, if any two parameters

and total dissolved boron are known, all parameters (pCO

[CO

], [HCO



], [CO

2

], pH, SCO

, and TA) can be calculated

for a given T, S, and P (cf. Zeebe and Wolf-Gladrow, 2001).

Buffering

The dissolved carbonate species in seawater provide an effi-

cient chemical buffer to various processes that change the pro-

perties of seawater. For instance, the addition of a strong acid

such as hydrochloric acid (naturally added to the ocean by

volcanism), is strongly buffered by the seawater carbonate

system. In distilled water, the addition of HCl leads to an

increase of [H

] and [Cl



] in solution in a ratio 1:1. This is

not the case in seawater. For example, the addition of 1 mmol

1

HCl to distilled water at pH 7 reduces the pH to very close

to 6. The same addition to seawater at pH 7 and SCO

2,000 mmol kg

1

at T ¼15



C and S ¼35 only reduces the

pH to 6.997. The seawater pH buffer is mainly a result of the

capacity of CO

2

and HCO



ions to accept protons.

One specific buffer factor, the so-called Revelle factor, is

important in the context of the oceanic uptake of anthropogenic

. The Revelle factor, R, is given by the ratio of the relative

Figure C11 Concentrations of the dissolved forms of the carbonate system in seawater (so-called Bjerrum plot). SCO

= 2,000 mmol kg

1

temperature T =15



C, salinity S = 35, and pressure P = 1 atm (after Zeebe and Wolf-Gladrow, 2001).

124 CARBON DIOXIDE, DISSOLVED (OCEAN)