Odekon M. Encyclopedia of paleoclimatology and ancient environments

Подождите немного. Документ загружается.

Cross-references

Astronomical Theory of Climate Change

Continental Sediments

Cretaceous/ Tertiary (K-T) Boundary Impact, Climate Effects

Cretaceous Warm Climates

Cyclic Sedimentation (cyclothems)

Greenhouse (warm) Climates

Monsoons, Pre-Quaternary

Ocean Anoxic Events

Pre-Quaternary Milankovitch Cycles and Climate Variability

MESSINIAN SALINITY CRISIS

During the Messinian (uppermost Miocene), the Mediterranean

area was involved in a rapid and dramatic salinity crisis, as

indicated by the discovery of 1–2 km thick evaporite deposits

beneath its sea floor (Hsü et al., 1973). The problems in envi-

sioning viable depositional models for these evaporites gener-

ated a number of contrasting hypotheses regarding the timing

(synchronous versus diachronous), the origin (glacioeustatic

versus tectonic), the environmental conditions (deep versus

shallow water) and the impact (global versus regional) of the

salinity crisis. An international controversy arose mostly

because different sub-basins could have experienced different

evaporite deposition at the same time and direct data on the eva-

porite rocks in the deep Mediterranean depressions are limited

to only a few meters of boreholes. Other significant problems

were the absence of an accurate and reliable time frame for

the Messinian, the difficult application of biostratigraphy in

unfavorable hypersaline sediments, and lithologies that are not

well suited to magnetostratigraphy (diatomites and evaporites).

Originally, three main models were proposed to explain

the origin of the Mediterranean evaporite giant: (a) a shal-

low, rapidly subsiding basin able to create accommodation

space for shallow-water to supratidal evaporite accumulation

(Nesteroff, 1973); (b) a preexisting deep basin that became par-

tially cut off from world sea level, resulting in drawdown and

deposition of shallow-water to desiccated hypersaline facies

(Hsü et al., 1973); and (c) a preexisting deep basin in which

deposition occurred in deep hypersaline water (Busson,

1980). Other hypotheses integrate different environments of

deposition to explain the complex array of evaporite facies

throughout the Mediterranean Basin.

On a broad scale, the evaporite deposits beneath the deep

Mediterranean depressions indicate progressive restriction of

the Mediterranean-Atlantic connection. As suggested by seismic

data, the evaporites can be divided into two main sulfate-bearing

units, the Lower Evaporites (or First Cycle), containing thick

halite deposits, and the Upper Evaporites (Second Cycle), sepa-

rated by an interregional unconformity. The Second Cycle does

not have a marine signature like that of First Cycle and is not

evaporitic across all of the Mediterranean. In most land sections

it consists of detrital sediments with a brackish fauna, the

so-called “Lago Mare” (Lake Sea) facies, which represents a

dominantly freshwater body of Paratethyan origin, marking the

complete isolation of the Mediterranean from the ocean. Normal

marine conditions were suddenly re-established at the beginning

of the Pliocene as a consequence of flooding from the Atlantic.

Recently, astrochronology and integrated stratigraphy have

demonstrated that the onset of evaporite formation was a

synchronous event over the entire Mediterranean basin, dated

at 5.96 Ma (Krijgsman et al., 1999). Cyclic evaporite deposi-

tion is related to circum-Mediterranean climate changes driven

by changes in the Earth’s precession with an average periodi-

city of 21.7 kyr. Isolation from the Atlantic Ocean was estab-

lished between 5.59 and 5.33 Ma, causing a large fall in

Mediterranean water level followed by erosion (5.59–5.50 Ma)

and deposition of the non-marine “Lago Mare” sediments

(5.50–5.33 Ma). According to these data, the total duration of

the Messinian salinity crisis is approximately 640 kyr. This

view is not shared by all authors (Butler et al., 1995; Clauzon

et al., 1996; Riding et al., 1998), but the general consensus is

toward a marked drawdown of the Mediterranean water level.

Estimations of the water level drop range from a few hundred

meters up to 3,000 m (Tay et al., 2002), but some parts of

the Mediterranean appear to have remained under relatively

deep-water conditions throughout the entire salinity crisis

(Roveri et al., 2001).

Another common view is that the deeply incised subaerial

erosional surface that produced the incised canyons of the

Rhone, Ebro, Po and Nile Rivers represents the time interval

during which evaporites accumulated in the deep basins. More

controversial is the stratigraphic position of the marginal eva-

porites, which are placed by different authors either above or

below the main unconformity (Riding et al., 1998; Clauzon

et al., 1996), thus postdating or predating the desiccation phase

in the deep basins.

The compilation of a detailed chronology challenged the ori-

ginal view that the Mediterranean isolation could have been

triggered by glacioeustatic sea level lowering (Kastens, 1992).

The latest Miocene glaciation occurred from 6.26 to 5.5 Ma.

It was marked by 18 glacial-to-interglacial oscillations con-

trolled by the 41-kyr cycle of obliquity (Hodell et al., 2001).

Although the glaciation may have contributed to the isolation

of the Mediterranean, it preceded the base of the Lower Eva-

porites by 300 kyr, supporting tectonic or local climatic forcing

for the crisis inception (Krijgsman et al., 1999; Hodell et al.,

2001). This hypothesis is supported by the discovery of large

and rapid climate variations during the salinity crisis, as docu-

mented by biomarkers, stable hydrogen, and carbon isotope

compositions (Andersen et al., 2001). Moreover, the end of

the salinity crisis preceded the most prominent interglacial con-

ditions by 170 kyr, suggesting that sea level outside the Medi-

terranean began to rise prior to the base of the Pliocene. This

implies that tectonics were responsible for the end of the crisis

or there was a significant lag between sea level rise and the

resumption of normal marine conditions in the Mediterranean

(Hodell et al., 2001).

Stefano Lugli

Bibliography

Andersen, N., Paul, H.A., Bernasconi, S.M., McKenzie, J.A., Behrens, A.,

Schaeffer, P., and Albrecht, P., 2001. Large and rapid climate variability

during the Messinian salinity crisis: Evidence from deuterium concen-

trations of individual biomarkers. Geology, 29(9), 799–802.

Busson, G., 1980. Les grandes cuvettes evaporitiques en milieu detritique:

Comment elles se creusent, comment elles se remplissent. Bulletin

Centres de Recherches Exploration-Production Elf Aquitaine, 4,

557–588.

Butler, R.W., Lickorish, H., Grasso, M., Pedley, H.M., and Rarnberti, L.,

1995. Tectonics and sequence stratigraphy in Messinian basins, Sicily:

Constraints on the initiation and termination of the Mediterranean sali-

nity crisis. Geol. Soc. Am. Bull., 107, 425–439.

MESSINIAN SALINITY CRISIS 559

Clauzon, G., Sue, J.-P., Gautier, F., Berger, A., and Loutre, M.-F., 1996.

Alternative interpretation of the Messinian salinity crisis: Controversy

resolved? Geology, 24, 363–366.

Hodell, D.A., Curtis, J.H., Sierra, F.J., and Raymo, M.E., 2001. Correlation

of late Miocene to early Pliocene sequences between the Mediterranean

and North Atlantic. Paleoceanography, 16, 164–178.

Hsu, K.J., Ryan, W.B.F., and Cita, M.B., 1973. Late Miocene desiccation

of the Mediterranean. Nature, 242, 240–244.

Kastens, K.A., 1992. Did glacio-eustatic sea level drop trigger the Messi-

nian salinity crisis? New evidence from Ocean Drilling Program site

654 in the Tyrrhenian Sea. Paleoceanography, 7, 333–356.

Krijgsrnan, W., Hilgen, F.J., Raffi, I., Sierra, F.J., and Wilson, D.S., 1999.

Chronology, causes and progression of the Messinian salinity crisis.

Nature, 400, 652–655.

Nesteroff, W.O., 1973. Un modele pour les evaporites messinienes en Med-

iterranee: des bassin peu pro fonds avec depots d’evaporites iagunaires,

1973. In Drooger, C.W. (ed.), Messinian Events in the Mediterranean.

Amsterdam: Koninklijke Nederlandse Akademie Van Wetenschappen,

North-Holland Publishing Company, pp. 68–81.

Riding, R., Braga, J.C., Martin, J.M., and Sanchez-Almazo, I.M., 1998.

Mediterranean Messinian salinity crisis: Constraints from a coeval mar-

ginal basin, Sorbas, southeastern Spain. Mar. Geol., 146,1–20.

Roveri, M., Bassetti, M.A., and Ricci Lucchi, F., 2001. The Mediterranean

Messinian salinity crisis: An Apennine foredeep perspective. Sediment.

Geol., 140, 201–214.

Tay, P.L., Lonergan, L., Warner, M., Jones, K.A., and the IMERSE Work-

ing Group, 2002. Seismic investigation of thick evaporite deposits on

the central and inner unit of the Mediterranean Ridge accretionary com-

plex. Mar. Geol., 186, 167–194.

Cross-references

Astronomical Theory of Climate Change

Continental Sediments

Cyclic Sedimentation (cyclothems)

Evaporites

Neogene Climates

Plate Tectonics and Climate Change

Pre-Quaternary Milankovitch Cycles and Climate Variability

METHANE HYDRATES, CARBON CYCLING, AND

ENVIRONMENTAL CHANGE

Overview

Enormous amounts of methane reside as solid gas hydrate and

free gas bubbles in the pore space of marine sediment along conti-

nental margins. Most of this methane has formed through the

microbial breakdown of organic matter originally deposited on

the seafloor. At present, a small fraction escapes to ocean waters

through seafloor venting. Reaction with dissolved sulfate con-

sumes additional methane in pore waters of shallow sediment.

These potentially important carbon fluxes, which remain absent

from most global carbon cycle models, could change dramatically

though time because the stability and distribution of gas hydrate

and free gas depend on external conditions, particularly ocean

temperature. Chemical anomalies in sediment or sedimentary

rocks may document such variations in methane release and hint

at an under-appreciated mechanism for past climate change.

Chemistry and theoretical occurrence

At relatively low temperatures, high pressures a nd high gas

concentration, certain low-molecular-weight gases such as

methane, ethane, propane, carbon dioxide and xenon can

combine with water to precipitate crystalline substances

resembling ice (Figure M17). These solid compounds are

called clathrate hydrates of gas, or more conveniently “gas

hydrates,” because the gas molecules exist within cages of

water molecules. Three types of clathrate structures have

been identified: Structure I, Structure II and St ructure H.

Each of these structures contains small and large cavities that

can host guest gas molecules; their general formulae are

8C ·46H

O(S1),24C ·136H

O(S2),and4C ·34H

O(SH),

where C is the total number of available cavities. The type of

clathrate structure formed depends on the size and composition

of the guest gases. Moreover, the chemical composition may

differ from the ideal formulae because gas molecules need not

occupy all cavities (Sloan, 1998). Indeed, natural clathrate

hydrates of methane seem to be S1 with a composition of

approximately CH

 6H

O. For all three structures, the solid

gas hydrate can dissociate to constituent water and gas phases

with a decrease in pressure or increase in temperature, although

gas chemistry and crystal structure determine the precise condi-

tions (Sloan, 1998).

Gas hydrates naturally precipitate in the pore space of water-

bearing sediments when gas concentrations exceed saturation

conditions at appropriately high pressure, low temperature and

low salinity (e.g., Dickens and Quinby-Hunt, 1997). These criteria

are met in two general environments: Arctic regions beneath several

hundreds of meters of permafrost, and along continental margins

beneath several hundreds of meters of seawater (Kvenvolden,

1999). This article focuses on the latter because the marine

realm hosts the vast majority of natural gas hydrate (>99%)

(Kvenvolden and Lorenson, 2001).

Within deep-sea sediments, gas hydrates can potentially exist

between the seafloor and a relatively shallow, sub-bottom depth

where temperatures become excessively warm because of the

geothermal gradient (Figure M18). This depth interval is com-

monly called the gas hydrate stability zone (GHSZ) and its lower

bound generally increases with water depth because of greater

pressure and invariant salinity. Thus, across a continental margin,

the GHSZ forms a lens within sediment that starts at some shallow

water depth and thickens beneath deeper water (Figure M18). This

lens begins nominally at 250–500 m water depth at present,

depending on local bottom water temperature and gas com-

position (Dickens, 2001). Below the GHSZ, where temperatures

are too warm, free gas bubbles will exsolve from solution when

gas concentrations exceed saturation (Holbrook et al., 1996;

Figure M17 Example of natural gas hydrate recovered from Hydrate

Ridge off the coast of Oregon (Shipboard Scientific Party, 2002).

560 METHANE HYDRATES, CARBON CYCLING, AND ENVIRONMENTAL CHANGE

Dickens et al., 1997). Significant amounts of gas hydrate cannot

accumulate above the seafloor because, like ice, it has a density

less than liquid water and floats.

Natural distribution and composition

Although appropriate pressure and temperature conditions

exist throughout the deep ocean, insufficient concentrations

of requisite gases preclude gas hydrate formation in many

deep-sea sediment sequences. Along continental margins,

however, the decomposition of organic matter through m i-

crobial activity or thermal cracking generates enough low

molecular weight hydrocarbon gases to saturate pore waters

within portions of the GHSZ.

Gas hydrates can be recovered from at or near the seafloor

in a few places where hydrocarbon gases escape from sediment

along fluid conduits such as faults. Locations documented to

date include numerous sites in the Gulf of Mexico and on

Hydrate Ridge off the coast of Oregon (Figure M17 and

M19). Interestingly, these hydrates lie in contact with water

containing very low concentrations of hydrocarbon gases. Evi-

dently, a rapid supply of hydrocarbon gases can prevent the dis-

solution of gas hydrates on the seafloor.

Good evidence for gas hydrates has been documented in mar-

ine sediment well beneath the seafloor at over 60 locations in all

oceans (Kvenvolden, 1999). Most of this evidence is indirect,

coming from geophysical methods and an understanding of how

sound waves propagate through sediment (Kvenvolden, 1999;

Kvenvolden and Lorenson, 2001). Relative to sediment with pore

space filled by water, the presence of solid gas hydrate and free gas

bubbles increases and decreases the velocity of sound, respec-

tively. A drop in sound velocity therefore occurs where gas

hydrate directly overlies free gas bubbles at the base of the GHSZ.

Seismic profiles generated on ships can image this acoustic inter-

face (Figure M19), commonly named a bottom-simulating reflec-

tor (BSR) because it represents a solid/gas phase boundary,

parallels the seafloor and cross-cuts sedimentary strata (Holbrook

et al., 1996; Kvenvolden, 1999). However, significant amounts of

gas hydrate can exist at locations without a BSR, when insufficient

gas concentrations to saturate pore waters occur at the bottom of

the GHSZ (Dickens, 2001). Remote sensing of gas hydrate using

current methods offers only a partial view of their distribution.

Considered unsafe for scientific drilling until the last decade,

only a few areas of the seafloor with gas hydrate have been drilled

so far in order to evaluate its vertical distribution and char-

acteristics. These places include the Blake Ridge offshore of

South Carolina, the Peruvian Margin, and Hydrate Ridge. In drill

holes at each of these locations, some gas hydrate samples were

recovered within the GHSZ. However, most gas hydrate eluded

recovery and dissociated to constituent methane and water as pres-

sure decreased and temperature increased during sediment core

retrieval (Paull and Matsumoto, 2000; Shipboard Scientific Party,

2002). This inference comes from the results of various proxy

techniques for measuring gas hydrate abundance. For example,

a special tool was used at each drilled location to collect sedi-

ment cores at in situ pressure. The amounts of hydrocarbon gases

released from these pressure cores (Figure M19) exceed gas hydrate

saturation at many depths (Dickens et al., 1997; Milkov et al.,

2003). In addition, the dissociation of gas hydrate is an endothermic

reaction that decreases the temperature of sediment and releases

fresh water to surrounding, relatively saline pore water. Anoma-

lously low temperatures and fresh pore waters measured after core

recovery can thus record the presence of gas hydrate in sediment

at in situ conditions (Paull and Matsumoto, 2000; Shipboard Scien-

tific Party, 2002). A major find resulting from this recent drilling is

that gas hydrate in these locations is widely dispersed in sediment at

low concentrations (<5% of pore space) except along local fluid

conduits where, in some cases, the amount of gas hydrate exceeds

the mass of sediment over m-scale intervals.

In most regions with gas hydrate, microbial production

of gas dominates, and the gas within recovered samples of

Figure M18 Theoretical location of crystalline gas hydrate and free gas bubbles below the seafloor before and after a 5



C rise in bottom

water temperature (after Dickens, 2001). The phase boundaries pertain to the pure methane system in pure water and seawater. Note that the

diagram pertains to an initial bottom water temperature of 10



C, as inferred for the Paleocene prior to the PETM.

METHANE HYDRATES, CARBON CYCLING, AND ENVIRONMENTAL CHANGE 561

hydrate comprises mostly methane, often in excess of 99%

(Kvenvolden and Lorenson, 2001). The microbial origin also

leads to an exceptional depletion in the isotope

C (Kvenvolden

and Lorenson, 2001). These microbial gas hydrates are Structure

I clathrates. In some regions such as the Gulf of Mexico, how-

ever, thermogenic gas dominates, and gas hydrate samples may

contain significant amounts of ethane and propane (Kvenvolden

and Lorenson, 2001). These mixed hydrocarbon gas hydrates are

often Structure II clathrates.

Global amount and significance

Estimates of the worldwide amount of gas hydrate in marine sedi-

ment remain poorly constrained and vary greatly. This reflects

the intrinsic difficulty in quantifying the abundance of a phase that

defies straightforward detection by remote methods and simple

recovery by conventional drilling methods.

In theory, the global amount of gas hydrate in the oceanic realm

can be calculated from two parameters: the volume of pore space

within the GHSZ along all continental margins (Figure M18),

and the average amount of pore space occupied by gas hydrate

within this volume (Dickens, 2001). Estimates for the first para-

meter range from 1 to 6  10

at present. On average, pore

space within this volume probably contains between 1% and

10% gas hydrate. This implies that present-day oceanic gas

hydrates hold somewhere between 1,000 and 22,000 Gt of carbon

(Gt = 10

g) as methane (Dickens, 2001), with most but not all

current literature typically suggesting about 10,000 Gt of carbon

(cf. Kvenvolden, 1999; Milkov et al., 2003). Not included in these

estimates are amounts of carbon stored in dissolved gas or in free

gas bubbles beneath gas hydrate, which have not been assessed

rigorously. Whatever the total, these amounts are enormous

considering that the atmosphere contains about 700 Gt C, and

the entire standing-stock of the terrestrial biosphere only accounts

for about 500–1,000 Gt C.

A dynamic view and a large capacitor

Oceanic gas hydrates and associated free gas constitute a major

pool of

C-depleted methane intimately connected to the glo-

bal carbon cycle (Figure M20). This view can be understood

by considering individual deposits of gas hydrate and free gas

in marine sediment as dynamic systems where carbon cycles

between the ocean, sediments, dissolved gas, gas hydrate and

free gas over time (Figure M19).

Figure M19 Seismic reflection profile across southern Hydrate Ridge on the Oregon Margin showing the bottom simulating reflector (BSR) and

the location of three sites drilled during Leg 204 (Shipboard Scientific Party, 2002). Also shown are methane concentrations determined

using pressure cores (Milkov et al., 2003), and the inferred dynamic cycling of methane in this region (modified from Dickens, 2003). The BSR

is an acoustic interface separating gas hydrate from underlying free gas bubbles.

Figure M20 A basic, steady-state view of the exogenic global carbon

cycle (ocean, atmosphere and biomass) with postulated

connections to gas hydrates and free gas (after Dickens, 2003).

562 METHANE HYDRATES, CARBON CYCLING, AND ENVIRONMENTAL CHANGE

Photosynthetic organisms in surface waters produce com-

plex organic molecules, which eventually sink through the

water column to the seafloor. Within the sediment, bacteria uti-

lize dissolved oxygen, nitrate, and sulfate to convert this orga-

nic matter to a series of products including carbon dioxide and

acetate. Other microbes, probably archaea, use these simple

carbon-based compounds to generate methane (e.g., Marchesi

et al., 2001). This dissolved gas migrates both laterally and

vertically via diffusion and fluid flow. Eventually, at suffi-

ciently high gas concentrations and appropriate pressure and

temperature conditions, gas hydrate can precipitate in pore

space. Sediment burial over time slowly brings these solid

methane-bearing molecules to higher temperatures. At some

relatively shallow depth in the sediment column, gas hydrate

is no longer stable and dissociates to gas-saturated water and

free CH

gas bubbles.

Although much of the methane released from gas hydrate

at depth migrates upward to recycle through the above pro-

cesses, some quantity escapes these systems through one of

two pathways. In many places, upward moving methane

encounters sulfate diffusing down from the seafloor. This leads

to anaerobic oxidation of methane (AOM), a biochemical reac-

tion whereby consortia of archaea and bacteria convert methane

to bicarbonate ion across a sulfate/methane transition (SMT)

(e.g., Boetius et al., 2000). The SMT often occurs in the first

40 m below the seafloor (D’Hondt et al., 2002). However, in a

few places with gas hydrate, notably where faults extend from

the seafloor to depths below the GHSZ, high rates of fluid flow

along conduits discharge methane into deep waters of the ocean.

Aerobic oxidation of methane by bacteria probably consumes

most of this methane in the water column (Valentine et al., 2001).

Pressure and temperature affect the location and fluxes of

methane cycling within these dynamic carbon pools. As such,

gas hydrate systems may serve as a large “capacitor” in the glo-

bal carbon cycle, with relatively steady carbon inputs but

highly variable carbon outputs (Dickens, 2003). The burial

and degradation of organic carbon slowly contributes methane

carbon to gas hydrate systems over time. At steady-state con-

ditions, AOM and seafloor methane venting also slowly return

methane carbon to the ocean. However, the rates of venting

could change, perhaps dramatically, when deep ocean tem-

peratures or sea level vary. Consider, for example, a modest

rise in seafloor temperatures along continental margins from



Cto5



C(Figure M18). Once propagated into sediment,

this perturbation would significantly shrink the global

GHSZ, dissociating large amounts of gas hydrate to free

gas bubbles. The build-up of free gas within sediment would

lead to local pressures that exceed those from overlying water

and, in some cases, overlying sediment. Such overpressure

might cause sediment failure and rapid release of methane

from the seafloor to the ocean or atmosphere through slump-

ing or venting (Figure M21).

The capacitor concept is crucial for understanding gas

hydrates throughout time in three regards (Dickens, 2003).

First, widely accepted models for the global carbon cycle do

not contain gas hydrates and methane fluxes to and from the

seafloor. Nonetheless, these models may reasonably portray

the fundamental operation of the carbon cycle when methane

carbon inputs and outputs are small and roughly balanced,

which is probably the case at present. Second, a capacitor allows

for the production, storage and release of methane. Since sedi-

mentary rocks show that organic carbon has accumulated in

relatively cold deep waters (<15



C) over significant intervals

of the geological record, presumably gas hydrates and free gas

bubbles have always been a common phenomenon. Third, a

capacitor allows for major injections of

C-depleted carbon to

the ocean or atmosphere when external conditions change.

Potential climatic effects of seafloor methane release

Massive release of methane from marine gas hydrate systems to

the atmosphere could significantly impact climate and the

environment (Paull et al., 1991; Kvenvolden, 1999). Methane

is a greenhouse gas with a much higher warming potential than

carbon dioxide. Although methane molecules have a fairly

short lifetime in the present-day atmosphere (about 10 years

on average), this residence time depends strongly on atmo-

spheric concentrations of methane and other species. Addition-

ally, methane removal in the atmosphere eventually produces

carbon dioxide. Consequently, a large transfer of methane from

gas hydrate systems to the atmosphere could have a profound

and lasting effect on earth surface temperatures (Schmidt and

Shindell, 2003).

Microbes in the water column rapidly oxidize most methane

venting above modern gas hydrate systems (Valentine et al.,

2001). Therefore, direct injection of methane from gas hydrates

to the atmosphere presents a conceptual problem (Kvenvolden,

1999). Conceivably, methane could escape oxidation if it was

released sufficiently fast or if bubbles were coated with hydrate

(Heeschen et al., 2003). Dislodged gas hydrate could also float

toward the surface (Paull et al., 2002).

Substantial release of methane to the ocean could also affect

the marine environment, primarily by changing the chemistry

of deep ocean water (Dickens, 2000). Microbial oxidation of

methane removes dissolved oxygen and produces carbon diox-

ide (Valentine et al., 2001), which decreases pH upon reaction

with water, to form carbonic acid. In turn, increased acidity dis-

solves calcium carbonate, a common constituent of marine

shells and seafloor sediment.

Irrespective of whether methane oxidation occurred in the

atmosphere or ocean, its ultimate fate would be production of

carbon dioxide, which could propagate throughout the ocean,

atmosphere and terrestrial biomass. Consequently, a massive

release of

C-depleted methane would decrease the

ratio of all carbon upon the Earth’s surface (Dickens, 2000).

Figure M21 Schematic diagram highlighting the gas hydrate

dissociation hypothesis for the Paleocene / Eocene boundary

(modified from Katz et al., 1999). Large quantities of gas hydrate

convert to free gas bubbles, which escape through sediment

slumping (shown here) or venting to produce

C-depleted carbon

dioxide in either the ocean or atmosphere.

METHANE HYDRATES, CARBON CYCLING, AND ENVIRONMENTAL CHANGE 563

Ancient methane release

Events of extreme environmental change punctuate Earth’s

history. Across some of these aberrations, the

C ratio of

calcium carbonate and organic matter decreased at multiple

locations, suggesting a major and rapid input of

C-depleted

carbon to the atmosphere or ocean. Over the last 500 million

years, such intervals include the Permian/Triassic boundary

ca. 252 million years ago (Ma), several episodes in the Jurassic

through middle Cretaceous ca. 183–120 Ma, and the Paleo-

cene/Eocene boundary ca. 55 Ma. Increasingly, Earth scientists

have argued that these abrupt geochemical anomalies signify

massive release of methane from marine gas hydrate systems

(e.g., Katz et al., 1999; Hesselbo et al., 2000; Beerling et al.,

2002; Berner, 2002).

Of these events, the Paleocene/Eocene boundary (see

Paleocene-Eocene thermal maximum, this volume) warrants

special attention because evidence for massive methane release

is compelling. First, unlike older intervals, stratigraphic relation-

ships unequivocally demonstrate that the carbon isotope pertur-

bation was global and rapid. At least 40 different stable isotope

records, constructed using deep and shallow marine carbonate,

and terrestrial carbonate and organic matter, now show a pro-

minent drop in the

C ratio across the Paleocene/Eocene

boundary (e.g., Katz et al., 1999; Bowen et al., 2002; Zachos

et al., 2001). In records from open-ocean locations (Figure M22),

the initial decrease occurs over 10–40 cm while the return

to near initial values occurs over 100–400 cm, depending on

sedimentation rate. The exact shape and timing of the carbon

isotope excursion remain open issues, even at the same location.

Nevertheless, on the basis of different dating methods (see

Dickens, 2003), its decrease and return spanned <20 kyr and

<220 kyr, respectively. Given the abruptness of this change,

only methane release from the seafloor provides a satisfactory

explanation for the very large drop in carbon isotope ratios

(Dickens, 2003). Arguments for a comet impact (Kent et al.,

2003), while interesting, have little merit (Dickens and Francis,

2004; Schmitz et al., 2004).

Ancillary information from sediment cores across the

Paleocene/Eocene boundary further supports a massive release

of methane from gas hydrate systems. The

O ratio of

carbonate shells precipitated by benthic foraminifera living on

the seafloor also decreased significantly at this time (e.g., Katz

et al., 1999; Zachos et al., 2001). This change suggests a major

rise in deep ocean temperatures, perhaps upwards of 6



which would affect the distribution of gas hydrate dramatically

(Figure M18 and M21). Deep marine sediment sequences

furthermore display clear evidence for a substantial drop in

dissolved oxygen and pronounced dissolution of calcium car-

bonate (e.g., Katz et al., 1999). In addition, these observations

appear to coincide with unusual slumping along continental

margins (Katz et al., 1999).

The cause and effects of presumed methane release during

the Paleocene/Eocene boundary remain open issues, in large

part because of intrinsic difficulties distinguishing the relative

timing of rapid environmental changes in 55 million year old

strata. However, all of the above observations closely correlate

with a prominent benthic foraminifera extinction (e.g., Katz et al.,

1999), extreme global warming (e.g., Zachos et al., 2001), and an

extraordinary diversification of terrestrial mammals (e.g., Bowen

et al., 2002).

Methane venting in the Quaternary

Major fluctuations in polar ice volume caused sea level to fall

by greater than 100 m several times during the last few hundred

thousand years. Several papers have speculated that these sea

level drops may have released substantial quantities of methane

from oceanic gas hydrate systems because of a decrease in the

pressure within seafloor sediment (e.g., Kayen and Lee, 1991;

Paull et al., 1991). If this idea is correct, methane must have

come from sediment in shallow water depths or through other

geological processes (e.g., sediment slumping) because the

GHSZ would remain relatively unchanged at deeper water

depths (Figure M18 and M21). A general correlation seems

to exist between known locations of gas hydrate and sediment

slumps formed during sea level low stands (e.g., Kennett et al.,

2002). However, there is little if any ancillary data (e.g., carbon

isotope deviations) to suggest that changes in sea level induce

significant variations in global methane outgassing from deep

marine sediment.

Climate records for the late Quaternary also indicate numer-

ous brief atmospheric warming events that occurred during

glacial times and that probably reflect some combination of

changing ocean circulation patterns and elevated greenhouse

gas concentrations. Notably, these “interstadial” events corre-

sponded to prominent highs in atmospheric methane. Most lit-

erature has attributed the methane peaks to increased emissions

from terrestrial wetlands (e.g., Severinghaus and Brook, 1999),

although there is limited evidence in the geological record for

their contemporaneous and requisite existence (Kennett et al.,

2002). On the other hand, several lines of evidence suggest that

water at intermediate depths in the ocean rose by several

degrees centigrade prior to the atmospheric warming events.

Recent papers have thus argued that these increases in bottom

water temperature would propagate heat into sediment, dissoci-

ate gas hydrate to free gas, and release methane to the ocean

and atmosphere through sediment failure (e.g., Kennett et al.,

2002). Sediment records from the California margin provide

some support for this interpretation. Immediately after episodes

Figure M22 Carbon isotope records across the Paleocene/ Eocene

boundary in different phases at three deep-sea locations (see Dickens,

2003). Original records have been placed on a common depth scale

with the carbon isotope minimum at 0.0 m. Note that the sedimentation

rates vary between sites, giving different shapes to the excursion.

Carbon isotopes are expressed in conventional delta notation (d

C) so

that negative values signify a drop in the

C ratio.

564 METHANE HYDRATES, CARBON CYCLING, AND ENVIRONMENTAL CHANGE

of intermediate water warming in the Santa Barbara Basin, the

C ratio of benthic foraminifera decreased (Kennett et al.,

2000), and the abundance of organic compounds produced

by methane-consuming bacteria increased (Hinrichs et al.,

2003). To date, however, no compelling evidence has been

provided to suggest that methane released from the seafloor

during interstadial events escaped microbial consumption in

the water column, or that atmospheric methane in fact initiated

climate change.

Conclusions

A large although unconstrained mass of

C-depleted hydrocar-

bons, mostly methane, exists as crystalline gas hydrates and

free gas bubbles along modern continental margins. Although

specimens of massive gas hydrate can be recovered from the

seafloor, the bulk of methane in these phases lies dispersed

within the upper few hundreds of meter of sediment. Most

methane in gas hydrates and free gas escapes collection during

conventional sediment recovery, but can be detected through

various proxy techniques. All information indicates that the

overall distribution of methane depends on carbon fluxes to

and from the ocean, and external conditions such as pressure

and temperature. In all likelihood, gas hydrates have occurred

throughout most of geological time. An open issue is whether

gas hydrate systems can suddenly discharge large amounts of

methane. If this is the case, rapid methane release could impact

the environment by contributing to atmospheric warming or by

changing ocean chemistry. A burgeoning body of literature

suggests that methane released from gas hydrate systems

explains unusual observations during certain intervals of

extreme environmental change in the ancient geological record,

such as the Paleocene/Eocene boundary. The idea that methane

contributes significantly to changes in Quaternary climate is

more controversial but has some support.

Gerald R. Dickens and Clayton Forswall

Bibliography

Bowen, G.J., Clyde, W.C., Koch, P.L., Ting, S.Y., Alroy, J., Tsubamoto, T.,

Wang, Y.Q., and Wang, Y., 2002. Mammalian dispersal at the

Paleocene/ Eocene boundary. Science, 295, 2062–2065.

Beerling D.J., Lomas M.R., and Grocke D.R., 2002. On the nature

of methane gas-hydrate dissociation during the Toarcian and Aptian

oceanic anoxic events. Am. J. Sci., 302,28–49.

Berner, R.A., 2002. Examination of hypotheses for the Permo-Triassic

boundary extinction by carbon cycle modeling. Proc. Natl. Acad. Sci.

USA, 99, 4172–4177.

Boetius, A., Ravenschlag, K., Schubert, C.J., Rickert, D., Widdel, F.,

Gieseke, A., Amann, R., Jørgensen, B.B., Witte, U., and Pfannkuche, O.,

2000. A marine microbial consortium apparently mediating anaerobic

oxidation of methane. Nature, 407,623–626.

D’Hondt, S., Rutherford, S., and Spivack, A.J., 2002. Metabolic activity

of subsurface life in deep-sea sediments. Science, 295, 2067–2070.

Dickens, G.R., 2000. Methane oxidation during the late Palaeocene

thermal maximum. Bull. Soc. Géol. France, 171,37– 49.

Dickens, G.R., 2001. The potential volume of oceanic methane hydrates

with variable external conditions. Org. Geochem., 32, 1132–1193.

Dickens, G.R., 2003. Rethinking the global carbon cycle with a large,

dynamic and microbially mediated gas hydrate capacitor. Earth Planet.

Sci. Lett., 213, 169–182.

Dickens, G.R., and Francis, J.M., 2004. Comment: A case for a comet

impact trigger for the Paleocene/Eocene thermal maximum and carbon

isotope excursion. Earth Planet. Sci. Lett., 217, 197–200.

Dickens, G.R., and Quinby-Hunt, M.S., 1997. Methane hydrate stability in

pore water: A simple theoretical approach for geophysical applications.

J. Geophys. Res., 102, 773–783.

Dickens, G.R., Paull, C.K., Wallace, P., and ODP Leg 164 Scientific Party,

1997. Direct measurement of in situ gas volumes in a large gas hydrate

reservoir. Nature, 385, 426–428.

Heeschen, K.U., Trehu, A.M., Collier, R.W., Suess, E., and Rehder, G.,

2003. Distribution and height of methane bubble plumes on the

Cascadia Margin characterized by acoustic imaging. Geophys. Res.

Lett., 30, Art. No. 1643.

Hesselbo, S.P., Gröcke, D.R., Jenkyns, H.C., Bjerrum, C.J., Farrimond, P.,

Bell, H.S.M., and Green, O.R., 2000. Massive dissociation of gas

hydrate during a Jurassic oceanic anoxic event. Nature, 406, 392–395.

Hinrichs K.-U., Hmelo L.R., and Sylva, S.P., 2003. Molecular fossil record

of elevated methane levels in late Pleistocene coastal waters. Science,

299, 1214–1217.

Holbrook, W.S., Hoskins, H., Wood, W.T., Stephen, R.A., Lizarralde, D.,

and Leg 164 Science Party, 1996. Methane hydrate and free gas on the

Blake Ridge from vertical seismic profiling. Science, 273,1840–1843.

Katz, M.E., Pak, D.K., Dickens, G.R., and Miller, K.G., 1999. The source

and fate of massive carbon input during the Latest Paleocene Thermal

Maximum. Science, 286, 1531–1533.

Kayen, R.E., and Lee, H., 1991. Pleistocene slope instability of gas hydrate-

laden sediment on the Beaufort Sea Margin. Mar. Geotech., 10, 125–141.

Kennett, J.P., Cannariato, K.G., Hendy, I.L., and Behl, R.J., 2000. Carbon

isotopic evidence for methane hydrate instability during Quaternary

interstadials. Science, 288, 128–133.

Kennett, J.P., Cannariato, K.G., Hendy, I.L., and Behl, R.J., 2002. Methane

Hydrates in Quaternary Climate Change. Washington, DC: American

Geophysical Union, 216pp.

Kent, D.V., Cramer, B.S., Lanci, L., Wang, D., Wright, J.D., and Van der

Voo, R., 2003. A case for a comet impact trigger for the Paleocene/

Eocene thermal maximum and carbon isotope excursion. Earth Planet.

Sci. Lett., 211,13–26.

Kvenvolden, K.A., 1999. Potential effects of gas hydrate on human

welfare. Proc. Natl. Acad. Sci. USA, 96, 3420–3426.

Kvenvolden, K.A., and Lorenson, T.D., 2001. The global occurrence of

natural gas hydrate. In Paull, C.K., and Dillon, W.P. (eds.), Natural

Gas Hydrates: Occurrence, Distribution and Detection. Washington,

DC: American Geophysical Union. Geophysical Monograph Series

124, pp. 3–18.

Marchesi, J.R., Weightman, A.J., Cragg, B.A., Parkes, R.J., and Fry, J.C.,

2001. Methanogen and bacterial diversity and distribution in deep gas

hydrate sediments from the Cascadia Margin as revealed by 16S rRNA

molecular analysis. FEMS Microbiol. Ecol., 34, 221–228.

Milkov, A.V., Claypool, G.E., Lee, Y.-J., Xu, W., Dickens, G.R., Borowski,

W.S., and ODP Leg 204 Scientific Party, 2003. In situ methane concen-

trations at Hydrate Ridge offshore Oregon: New constraints on the glo-

bal gas hydrate inventory from an active margin. Geology, 31, 833–836.

Paull, C.K., and Matsumoto, R., 2000. Leg 164 overview, in Proceedings

of the Ocean Drilling Program, Scientific Results, vol. 164, pp. 3–10.

College Station, TX: Ocean Drilling Program.

Paull, C.K., Ussler W., and III, Dillon, W.P., 1991. Is the extent of glacia-

tion limited by marine gas-hydrates? Geophys. Res. Lett., 18, 432–434.

Paull, C., Brewer, P., Ussler, W., Peltzer, E., Rehder, G., and Clague, D.,

2002. An experiment demonstrating that marine slumping is a mechan-

ism to transfer methane from seafloor gas-hydrate deposits into the

upper ocean and atmosphere. Geo-Mar. Lett., 22, 198–203.

Schmidt, G.A., and Shindell, D.T., 2003. Atmospheric composition,

radiative forcing, and climate change as a consequence of a massive

methane release from gas hydrates. Paleoceanography, 18, article

10.1029/ 2002PA000757.

Schmitz, B., Peucker-Ehrenbrink, B., Heilmann-Clausen, C., Aberg, G.,

Asaro, F., and Lee, C.T.A., 2004. Basaltic explosive volcanism, but

no comet impact, at the Paleocene-Eocene boundary: High-resolution

chemical and isotopic records from Egypt, Spain and Denmark. Earth

Planet. Sci. Lett., 225,1–17.

Severinghaus, J.P., and Brook, E.J., 1999. Abrupt climate change at the

end of the last glacial period inferred from air trapped in polar ice.

Science, 286, 930–933.

Shipboard Scientific Party, (2002). Leg 204 Preliminary Report. ODP Prelim.

Rpt., 104 [Online]. Available from World Wide Web: http://www-odp.

tamu.edu/publications/prelim/204_prel/204PREL.PDF.

Sloan, E.D. Jr., 1998. Clathrate Hydrates of Natural Gases. New York:

Marcel Dekker, 705pp.

Valentine, D.L., Blanton, D.C., Reeburgh, W.S., and Kastner, M., 2001.

Water column methane oxidation adjacent to an area of active hydrate

METHANE HYDRATES, CARBON CYCLING, AND ENVIRONMENTAL CHANGE 565

dissociation, Eel River Basin. Geochim. Cosmochim. Acta, 65,

2633–2640.

Zachos, J., Pagani, M., Sloan, L., Thomas, E., and Billups K., 2001.

Trends, rhythms, and aberrations in global climate 65 Ma to present.

Science, 292, 686–693.

Cross-references

Carbon Cycle

Carbon Dioxide, Dissolved (Ocean)

Carbon Isotopes, Stable

Paleocene-Eocene Thermal Maximum

Paleogene Climates

Pleistocene Climates

MID-PLIOCENE WARMING

The middle Pliocene is the most recent period in Earth historywith

temperatures as warm as those we expect the Earth to experience

by the latter part of the 21

Century. Estimates of middle Pliocene

warming are from 2.0



Cto3.2



C(Figure M23), placing them in

the range of the Intergovernmental Panel on Climate Change

(IPCC) projections of global temperature rise by the year 2100.

Yet, in many respects, the mid-Pliocene is very much like the

modern world. Terrestrial geography and ocean basin configura-

tion are largely unchanged, many floral and faunal species are still

extant, and altered internal and external forcings (e.g., mid-ocean

ridge spreading rates, solar luminosity) are not factors as they

would have been for much more ancient periods. While most

scientists are confident that we have identified the primary for-

cings triggering the present warming trend, no one is certain of

the ultimate climate impact of those forcings, or their associated

feedback processes. The middle Pliocene presents us with some-

what the reverse situation: extensive global data reveal the mature

state of a warmer world, essentially the resulting climate impact of

a prior global warming. The forcings that led to Pliocene warming

may be only partially identified, but in both the mid-Pliocene

and the present day, the myriad of feedbacks that exist between

the forcing triggers and their eventual impacts remain the topic

of extensive research.

Data collection, process studies, and numerical simulations,

all of which provide both validation and predictive capacity, are

targeted at narrowing estimates of climate change. The middle

Pliocene world provides an excellent paleo-laboratory for test-

ing the sensitivity of the physical models that we rely upon

for estimating future warming impacts. It challenges our under-

standing of the sensitivity of key components of the climate

system and how we simulate that system: polar vs. tropical sen-

sitivity, the role of ocean circulation in a warming climate, the

hydrological impact of altered storm tracks, and the regional

Figure M23 Simulations of mid-Pliocene surface air temperatures show that the Earth was more than 2



C warmer than during the mid

century. Warming was greatest in high latitudes, particularly in regions where land and sea ice extent were lower. Slight cooling in some

regions results from increased elevation, or increased standing water in the case of the Amazon River basin.

566 MID-PLIOCENE WARMING

climate impacts of modified atmospheric and oceanic energy

transport systems.

Issues in middle Pliocene warming

The interpretation of a mid-Pliocene global warming, where tem-

peratures were significantly higher than modern, is supported

by both terrestrial and marine paleoclimate proxies (e.g., Dowsett

et al., 1996; Dowsett et al., 1999;Haywoodetal.,2005;

Thompson, 1991;Thompson,1992; Thompson and Fleming,

1996; Wara et al., 2005). Marine microfaunal evidence suggests

that warming was greatest at high latitudes (Dowsett et al.,

1996), resulting in a reduced equator-to-pole temperature gradient

and tropical oceans that were, in the end, far less sensitive to

change. Estimates of paleo-CO

levels including analyses of the

stomatal density of fossil leaves (Kürschner et al., 1996;vander-

Burgh et al., 1993), analyses of d

Cratiosofmarineorganiccar-

bon (Raymo and Rau, 1992), and measurements of differences

between the carbon isotope composition of surface and deep

waters (Shackleton et al., 1992) all suggest an elevation of atmo-

spheric CO

of roughly 60–100 ppmv above pre-industrial con-

centrations. Present-day atmospheric CO

levels are close to 384

ppmv (in 2008) and are, therefore, within the envelope predicted

to have existed during the middle Pliocene (360–400 ppmv).

Increases in additional greenhouse gases, such as methane, may

have accompanied the CO

increase during the Pliocene. How-

ever, an open question is whether the middle Pliocene climate

represents the lag effect of modest greenhouse warming, revealing

how the oceans and ice-sheets integrate, redistribute, and even

amplify climate impacts through longer-term feedback mechan-

isms (i.e., ocean circulation, sea level rise, permafrost melting,

etc.).

Paleoclimate modeling studies have suggested that the mid-

dle Pliocene pattern of warmer sea surface temperatures and

surface air temperatures is more consistent with enhanced

ocean heat transports than with a simple increase in the atmo-

spheric CO

level (Figure M24) because low latitude regions

show such limited warming compared to high latitudes. How-

ever, there is some disagreement between proxy records of tro-

pical temperatures, particularly in the Pacific Ocean. Several

recent studies using Mg/Ca ratios and alkenones find that the

eastern equatorial Pacific may have been warmer than pre-

viously thought. These findings are somewhat more consistent

with moderately increased greenhouse gas levels, and have

been used to suggest that the middle Pliocene warming may

have been characterized by a tropical sea-surface temperature

pattern that is akin to a permanent El Niño. Although no study

has yet determined a specific set of forcings and feedbacks that

were responsible for triggering changes in ocean circulation dur-

ing the Pliocene, data suggest that a combination of increased

greenhouse gases and altered ocean heat transports, and perhaps

additional factors, acted concurrently (linked by feedback

relationships) to generate the middle Pliocene warming (e.g.,

Dowsett et al., 1996; Raymo et al., 1996).

Enhanced ocean heat transports may have been the result of

strengthened thermohaline circulation (THC) and/or increased

flow of surface ocean currents. The latter process has not been

widely accepted, since generating enhanced wind stress under

a regime of reduced latitudinal temperature gradients and a

decreased dynamic circulation of the atmosphere is difficult

(Crowley, 1996). Recently, however, Haywood et al. (2000)

showed that regionally enhanced wind stress fields over the North

Atlantic, North Pacific and Southern Ocean occur in middle Plio-

cene experiments using the Hadley Centre’s General Circulation

Model (GCM). Such regional increases may be sufficient to

strengthen the gyre circulations in these ocean basins and cause

an increase in meridional heat transport. Moreover, greater wind

stress curl and consequent gyre strengthening in the North Pacific

is supported by paleoceanographic evidence along the California

margin (Ravelo et al., 1997).

Other factors must also be considered. Surface conditions in

the middle Pliocene were somewhat different from present-day:

vegetation distributions were altered (Thompson and Fleming,

1996) and sea level was probably around 15–25 meters higher

than modern (e.g., Dowsett and Cronin, 1990). In addition, both

the Greenland and Antarctic Ice Sheets were probably smaller

(though the size of the Antarctic Ice Sheet remains a key Pliocene

controversy), and topographic relief may not have reached modern

proportions in the Western Cordillera, the East African Rift region

and on the Tibetan Plateau (Raymo et al., 1988; Ruddiman and

Kutzbach, 1989; Raymo and Ruddiman, 1992).

Middle Pliocene warming and estimates of future

climate change

Despite these differences, the middle Pliocene displays more

similarities with our anticipated future climate than almost

any other period in Earth history. Foremost among these is

the magnitude of warming in the middle Pliocene, which is

similar to future global climate change simulations at high lati-

tudes and over mid-latitude continental interiors. If forcings

and associated feedbacks are proportional to the ultimate cli-

mate impacts, then the middle Pliocene is probably closer to

what we can expect from future global warming than are the

warm climates of the Holocene or Pleistocene interglacials.

The potential to use the middle Pliocene as an analogue

for future climate change is also reinforced by the fact that

many key boundary conditions, such as the position of the con-

tinents, vegetation types and the intact state of the isthmus of

Panama, were the same as or similar to those of the present

day (Crowley, 1991; Crowley, 1996; Maier-Reimer et al.,

Figure M24 Modern and middle Pliocene sea surface temperatures

show no significant difference in the tropics, in contrast to mid and

high latitude regions, suggesting a role for altered ocean heat

transports in surface warming.

MID-PLIOCENE WARMING 567

1990). The opportunity to conduct detailed data-model compar-

isons is also excellent as there is:

1. Abundant paleoclimatic information from both the marine

and terrestrial realm (relative to older time periods);

2. A large number of extant species, making micropaleontolo-

gically-based climate interpretations feasible;

3. A reduced chance that diagenesis will hamper geochemical

analyses and more precise dating of proxy data, and thus

more precise validation of equilibrium GCM simulations

(Chandler et al., 1994).

The ability to use numerical climate model simulations to study

this past global warming is also greatly enhanced by the develop-

ment of the PRISM (Pliocene Research, Interpretation and Synop-

tic Mapping) series of quantitative digital data sets produced by the

United States Geological Survey since 1992. The ocean surface

maps were updated using data from more than 170 sites world-

wide, and are available as the PRISM2 data set of middle Pliocene

conditions (Dowsett et al., 1999). Efforts are now underway to

extend the PRISM data to include 3-dimensional temperature data

for the mid-Pliocene oceans. These data are designed specifically

for use in paleoclimate modeling studies utilizing GCMs. As such,

the PRISM2 reconstruction represents the most detailed global

paleoclimate reconstruction for any time other than the Last Gla-

cial Maximum (21 ky

BP.), and the only major reconstruction of a

globally warmer time period.

Mark A. Chandler

Bibliography

Chandler, M.A., Rind, D., and Thompson, R.S., 1994. Joint investigations

of the middle Pliocene climate II: GISS GCM Northern Hemisphere

results. Palaeogeogr. Palaeoclimatol. Palaeoecol., 9, 197–219.

Crowley, T.J., 1991. Modeling Pliocene warmth. Quaternary Sci. Rev., 10,

275–282.

Crowley, T.J., 1996. Pliocene climates: The nature of the problem. Mar.

Micropaleontol., 27,3–12.

Dowsett, H., Barron, J., and Poore, R.Z., 1996. Middle Pliocene sea surface

temperatures: A global reconstruction. Mar. Micropaleontol., 27,

13–26.

Dowsett, H.J., and Cronin, T.M., 1990. High eustatic sea level during

the middle Pliocene: Evidence from southeastern U.S. Atlantic coastal

plain. Geology, 18, 435–438.

Dowsett, H.J., Barron, J.A., Poore, R.Z., Thompson, R.S., Cronin, T.M.,

Ishman, S.E., and Willard, D.A., 1999. Middle Pliocene paleoen-

vironmental reconstruction: PRISM2. US Geol. Surv. Open File Rep.,

99–535.

Haywood, A.M., Valdes, P.J., and Sellwood, B.W., 2000. Global scale

paleoclimate reconstruction of the middle Pliocene climate using the

UKMO GCM: Initial results. Global Planet. Change, 25, 239–256.

Haywood, A.M., Dekens, P., Ravelo, A.C., and Williams, M., 2005.

Warmer tropics during the mid-Pliocene? Evidence from alkenone

paleothermometry and a fully coupled ocean-atmosphere GCM. Geo-

chem. Geophys. Geosyst., 6, Q03010 (doi:10.1029 /2004GC000799).

Kürschner, W.M., van der Burgh, J., Visscher, H., and Dilcher, D.L., 1996.

Oak leaves as biosensors of late Neogene and early Pleistocene paleoat-

mospheric CO

concentrations. Mar. Micropaleontol., 27, 299–312.

Maier-Reimer, E., Mikolajewicz, U., and Crowley, T., 1990. Ocean general

circulation model sensitivity experiment with an open Central American

isthmus. Paleoceanography, 5, 349–366.

Ravelo, A.C., et al., and ODP Leg 167 Shipboard Scientific Party, 1997.

Pliocene carbonate accumulation along the Californian margin. Paleo-

ceanography, 12, 729–741.

Raymo, M.E., and Rau, G., 1992. Plio-Pleistocene atmospheric CO

levels

inferred from POM delta

C at DSDP Site 607. Eos Trans. AGU, 73

(43), 95–96.

Raymo, M.E., and Ruddiman, W.F., 1992. Tectonic forcing of Late

Cenozoic climate. Nature, 359,117–122.

Raymo, M., Grant, M., Horowitz, M., and Rau, G.H., 1996. Mid-Pliocene

warmth: Stronger greenhouse and stronger conveyor. Mar. Micropa-

leontol., 27, 313–326.

Raymo, M.E., Ruddiman, W.F., and Froelich, P.N., 1988. Influence of

Late Cenozoic mountain building on ocean geochemical cycles.

Geology, 16, 649–653.

Ruddiman, W.F., and Kutzbach, J.E., 1989. Forcing of Late Cenozoic

Northern Hemisphere climate by plateau uplift in southern Asia and

the American West. JGR Atmos., 94(D15), 18,409–18,427.

Shackleton, N.J., Le, J., Mix, A., and Hall, M.A., 1992. Carbon isotope

records from Pacific surface waters and atmospheric carbon dioxide.

Quaternary Sci. Rev., 11, 387–400.

Thompson, R.S., 1991. Pliocene environments and climates in the Western

United States. Quaternary Sci. Rev., 10,115–132.

Thompson, R.S., 1992. Palynological data from a 989-ft (301–m) core of

Pliocene and Pleistocene sediments from Bruneau, Idaho. US Geol.

Surv. Open File Rep., 92–713.

Thompson, R.S., and Fleming, R.F., 1996. Middle Pliocene vegetation:

Reconstructions, paleoclimate inferences, and boundary conditions for

climate modeling. Mar. Micropaleontol., 27,27–49.

vanderBurgh, J., Visscher, H., Dilcher, D.L., and Kürschner, W.M., 1993.

Paleoatmospheric signatures in Neogene fossil leaves. Science, 260,

1,788–1,790.

Wara, M.W., Ravelo, A.C., and Delaney, M.L., 2005. Permanent El

Niño-like conditions during the Pliocene warm period. Science, 309:

758–761 (doi:10.1126/ science.1112596).

Cross-references

Greenhouse (Warm) Climates

Paleoclimate Modeling, Pre-Quaternary

Pleistocene Climates

MILLENNIAL CLIMATE VARIABILITY

Introduction

One of the most puzzling features of Earth’s past climate is a

series of strikingly abrupt and robust variations that occur roughly

every 1,000–7,000 years. First recognized in temperature records

from Greenland ice cores and North Atlantic deep-sea sediments,

these millennial climate cycles are now known to have a

global or nearly global footprint, and they have been found in

records spanning at least the last half million years. Paleoclimate

records reveal at least four modes of millennial climate change:

Heinrich events, Bond cycles, and Dansgaard/Oeschger cycles

that occur exclusively within glacials, and unnamed persistent

cycles that punctuate both glacials and interglacials. (Note that

the term cycle, as used here, refers to a repeating process that is

not necessarily periodic.)

Records of millennial climate variations have drawn wide-

spread attention because they produced the first convincing evi-

dence that Earth’s climate system is much less stable than

once thought and can shift suddenly on human time scales.

The millennial cycles typically exhibit threshold behavior;

i.e., abrupt transitions from one mode of operation to another

during which temperature changes of several degrees Celsius

occur within decades if not years. Although these striking

temperature jumps appear to have taken place predominantly

during glaciations, results of some modeling experiments

imply that they might be induced in the near future by contin-

ued global warming. Even though the likelihood of that

happening might be low, the impact on society is potentially

so severe that understanding why the climate system is prone

568 MILLENNIAL CLIMATE VARIABILITY