Odekon M. Encyclopedia of paleoclimatology and ancient environments

Подождите немного. Документ загружается.

LAKE-LEVEL FLUCTUATIONS

Introduction

Lake-level changes can occur as a consequence of climatic

change, tectonic activity, erosion at the outlet, or human activ-

ity. Water-level fluctuations associated with climate changes

are a response to variations in precipitation-evaporation (P-E)

over the watershed. A particular lake’ s sensitivity to P-E varia-

tions is primarily related to basin hydrology, or whether inflow

exits the lake basin via a surface outflow (overflowing, open)

or is confined to the lake basin (non-overflowing, closed).

The water level of a closed lake system, in the absence of tec-

tonic activity, erosion at the outlet, or human influence, repre-

sents an equilibrium state between: (a) input (catchment

runoff and groundwater inflow), and (b) output (evaporation

and sub-surface seepage). In a closed-basin system, changes

in effective moisture will cause the volume of the lake to either

increase or decrease until a new equilibrium is achieved. In

overflowing lakes, climatically driven lake-level changes are

limited to P-E decreases, because further increases are compen-

sated by greater outflow (Harrison and Digerfeldt, 1993). The

magnitude of lake-level response to climate change can be cal-

culated from known or estimated hydrological budgets if the

geomorphic characteristics of the lake and its watershed are

known, including bathymetry, catchment and shoreline topo-

graphy, soil type, and vegetation (Hostetler, 1995). Various

combinations of precipitation and evaporation can produce

the same effective moisture, but if other climate proxy data

can be used to constrain one of the variables, then the range

of the other can be determined (e.g., Barber and Finney, 2000).

A lake’s water-level history is the result of its response to

climate over a range of timescales (1–10

years). Large

water-level fluctuations at 10

to 10

-year timescales corre-

spond to global climate changes associated with glacial-

interglacial periods and global ice volume (e.g., Benson et al.,

1989; Finney et al., 1996). Lake-level data can provide impor-

tant Pleistocene and Holocene P-E information at <10

-year

timescales or better when chronological control for water-level

variations can be established (Abbott et al., 1997). Historical

gauge and climate observational records are available that

document lake-level variations and climate fluctuations at

-year timescales for the last 50–100 years (Street-Perrott

and Harrison, 1985).

Lakes are distributed globally, providing regional paleocli-

matic information for low latitudes (Kutzbach and Street-

Perrott, 1985; Street-Perrott and Perrott, 1990; Abbott et al.,

1997; Seltzer et al., 1998) and semi-arid regions at middle to

high-latitudes (Benson, 1981; Benson et al., 1989; Harrison

and Digerfeldt, 1993; Thompson et al., 1993; Yu et al., 1997;

Abbott et al., 2000; Anderson et al., 2005). In many tropical

and sub-tropical locations, lake-level variations are of particular

paleoclimatic significance. Region-wide changes in drought

patterns have been identified as coincident with significant cul-

tural change or collapse (e.g., Binford et al., 1997; Bookman

et al., 2004). Well-dated lake-level data have also been used

to identify feedbacks between tropical rainfall and global heat

transport by thermohaline circulation (e.g., Street-Perrott and

Perrott, 1990). In the middle latitudes, lake levels have been used

to determine quantitative estimates of paleoprecipitation (e.g.,

Benson, 1981; Hastenrath and Kutzbach, 1983, 1985), and to

verify paleoclimate model simulations (e.g., Webb et al., 1993).

Lake-level reconstructions

Past lake-level variations may be indicated by geomorphic evi-

dence within a lake basin. Raised shoreline features and wave-

cut platforms indicate higher lake levels (e.g., Gilbert, 1890;

Oviatt et al., 1994). Evidence for past lake-level variations is

also found within the stratigraphy of lacustrine deposits. It is

difficult to identify basin-wide sedimentary changes from

observations of a single core with a small cross section of

2–10 cm. Multiple cores collected on a shallow to deep water

transect provide more reliable lake-level information for

quantification of lake-level changes (e.g., Digerfeldt, 1986).

Sediment stratigraphy and facies changes may record water-

level shifts in several ways because lake-level fluctuations

affect numerous limnological and sedimentological processes.

Consequently, evidence from core studies may be subject to

more than one interpretation. However, consensus developed

from multiple lines of evidence can be used to build a compel-

ling lake-level reconstruction. Sedimentary changes can be

identified by visual stratigraphy, changes in color, pollen, grain

size distribution, and organic matter concentration. Multiple

lines of evidence commonly include physical, chemical,

paleolimnological, and paleoecological analyses and studies

of these are known as ‘‘multi-proxy’’ studies, with age control

determined by high precision geochronological techniques,

such as accelerator mass spectrometer (AMS) radiocarbon dat-

ing in combination with

210

Pb and

137

Cs assays for the youngest

sediments. Correlation of multiple cores recovered from differ-

ent water depths can be used to detect unconformities that form

hiatuses in the sediment record and determine the timing of

these events.

Stratigraphically-based multi-proxy lake-level reconstruc-

tions increase the ability to identify links between lake-level

reconstructions and climate fluctuations (Digerfeldt, 1986;

Dearing, 1997). Although the most useful combination of

multi-proxy studies varies from lake to lake, the three major

categories of sedimentary evidence are: (a) the distribution

and pattern of littoral vegetation, (b) sediment compositional

variations with varying depths, and (c) the shallowest depth

where fine-grained, organic-rich sediment accumulates (i.e.,

the sediment limit). Lundqvist (1925, 1927) pioneered the use

of core transects from shallow to deep water to investigate

how macrofossil content, pollen concentration, and sediment

composition varied with changing depth. Digerfeldt (1986)

and Dearing (1997) further modified these methods. The use

of core transects to investigate lake-level change is most com-

monly applied to lakes with small catchment to lake surface-

area ratios that have relatively shallow underwater slopes.

The distribution of shoreline (littoral zone) macrophyte

vegetation is largely determined by lake-water depth for

many closed-basin lakes. Characteristic zonation of emergent,

floating-leaved, and submerged vegetation extends outwards

from the shoreline. A decrease in water depth (lower lake level)

will commonly result in macrophyte vegetation migrating

toward the lake. However, such a vegetation response may

also occur over time because of progressive sedimentary infill-

ing. In contrast, an increase in water depth (higher lake level)

will result in landward macrophyte vegetation migration away

from the lake. It is more difficult to find alternative explana-

tions for landward expansion of shoreline vegetation than

higher lake level.

Depending on the character of shoreline deposits, littoral

zone sediments may contain higher proportions of reworked

LAKE-LEVEL FLUCTUATIONS 489

mineral particles than offshore zones. This is caused by wave

action or wind-induced currents that winnow fine-grained sedi-

ments from shoreline deposits and transport them into deeper

water, toward the center of the lake. Generally, grain size

decreases gradually away from shore, forming a fining outward

facies. Lowstands lead to displacement of the shore towards the

center of the lake and may be recognized by an increase in

grain size and mineral quantity. Lake highstands, in turn, are

recognized by a decrease in grain size. Similarly, the level

below which fine organic or fine-grained mineral sediments

accumulate is usually determined by location with respect to

the shoreline. This level represents the limit between domi-

nantly erosional and depositional sedimentary processes. Lake

highstands will often be recorded as an increase in the sediment

limit and lowstands as a lowering of the sediment limit. However,

lower lake levels are accompanied by erosion and redeposition

of older sediments away from shore, these would previously

have accumulated in near-shore sedimentary environments.

The most appropriate location and group of lake-level

proxies to investigate must be determined from the limnologi-

cal and sedimentological properties of each individual water-

shed and lake basin. Additionally, geomorphic processes and

basin bathymetry are important factors to consider when inves-

tigating lake-level changes using core transect studies. Varia-

tions of shore character and lake basin bathymetry will dictate

whether sediment accumulation is continuous or interrupted

by periods of erosion or non-deposition with considerable

variability in sediment thickness and composition in shallow-

water environments. For example, vegetation changes can be

easily recognized in eutrophic lakes with high macrophyte pro-

duction and diversity, but in oligotrophic lakes, coarse minero-

genic material is often a more dominant signal. Lake-level

variations will most likely be recorded by changes in the

fine-grained sediment limit in lakes with sandy shorelines. In

stratified marl lakes, changes in the location of marl sediment

accumulation are often the dominant water-level indicator. Yu

et al. ( 1997 ) used sediment hiatuses, detritus layers, and terres-

trial moss layers from five cores with calcium-carbonate oxygen

isotopes to identify lake-level changes. Anderson et al. (2005)

used palynology, calcium carbonate proportions, and bulk

organic carbon and nitrogen isotopes on four cores from Mar-

cella Lake, Yukon Territory. Numerous studies have also investi-

gated relationships between changes in lake level and variation in

diatoms, ostracods and chironomids (Hoffman, 1998; Alin and

Cohen, 2003; Yang et al., 2003).

Another approach is to use seismic surveys to identify lake-

level changes and determine where to collect sediment cores.

The seismic profiles allow for the identification of erosion sur-

faces and on-lap and off-lap sedimentary structures, and pro-

vide information regarding changes in sediment thickness

across the lake basin. Abbott et al. (2000 ) used seismic profil-

ing and multi-proxy sedimentary analyses including bulk sedi-

mentology and palynology on Birch Lake, Alaska to determine

the sediment limit. Keely et al. (2005) used seismic profiling on

Lake Bosumtwi, Ghana. These seismic investigations improved

the determination of coring locations that yielded optimal infor-

mation on water-level changes.

Resolving the timing and rate of lake-level variations

requires an accurate and precise chronology. Early lake-level

reconstructions, prior to the advent of AMS radiocarbon dating,

were commonly based on beta-counting analytical techniques

using bulk sediment samples. Bulk sedimentary material is

typically composed of organic matter derived from aquatic

organisms and can be older than the true age of the deposit

because of a reservoir or hardwater effect. This reservoir effect

is the result of either a long lake-water residence time or the

presence of limestone in the drainage basin, which is a source

of ancient carbon. AMS radiocarbon dating requires samples

containing around 1 milligram of organic material for accurate

and precise ages. Furthermore, the technique eliminates uncer-

tainty regarding the source of the organic material because ter-

restrial macrofossils (as little as 0.5 mg) can be identified and

dated. However, it is still possible that terrestrial macrofossils

can be reworked from older sediments during water-level

changes. High resolution lake-level studies based on AMS

radiocarbon measurements of identified macrofossils have been

conducted on Lake Titicaca, Bolivia (see case study below);

Birch Lake, Alaska (Abbott et al., 1997, 2000); Marcella Lake,

Yukon, Canada (Anderson et al., 2005); and the Dead Sea

(Bookman et al., 2004).

Lake level variations are important for studying the mechan-

isms of climate change if both the spatial and temporal resolu-

tion of the data is sufficiently extensive and detailed to describe

patterns on the scale of the climate process. For example,

Brenner et al., (2001) used a synthesis of finely resolved

lake-level reconstructions from the circum-Caribbean and

Andean altiplano to evaluate paleoclimatic variations during

the period when the Mayan and Tiwanaku cultures arose. Addi-

tionally, Kutzbach and Street-Perrott (1985) and Street-Perrott

and Harrison (1985) verified paleoclimatic model simulations

since 15,000 y

BP at 3,000 year intervals by comparing the

model-based closed-lake area estimates in the Northern Hemi-

sphere tropics with the geologically-based lake-level evidence.

Holocene lake-level variations in Europe and Central and North

America are explained in other studies by the direct effects of

sea-level, solar insolation, and atmospheric circulation changes

(Hodell et al., 1991; Harrison and Digerfeldt, 1993; Wright

et al., 1993; Yu et al., 1997; Abbott et al., 2000; Anderson et al.,

2005). The sensitivity of lake systems to climate is evident by

the distribution, size, and thermal structure of lakes throughout

the world (Hostetler, 1995). Lake-level variations, by providing

effective moisture balance information unavailable from tem-

perature-sensitive paleoclimatic sources, are indispensable

archives of paleoclimatic information with which to study

climate change.

Case study: Lake Titicaca, Bolivia/ Peru

A multi-proxy high-resolution lake-level study of Lake Titicaca

documented periods of severe drought during the last 4,000

years in the central Andean altiplano in order to investigate

relationships between climatic and cultural change during the

late Holocene (Abbott et al., 1997). Fifteen sediment cores

were collected on a transect from shallow (1.9 m) to deep

(20.2 m) water from Lago Wiñaymarka, an isolated section of

the larger Lake Titicaca basin, connected by a 21 m-deep chan-

nel through the Tiquina Strait (Binford et al., 1997). Lake

Titicaca is a large (8,300 km

), deep (280 m) water body

located on the border of Bolivia and Peru. During the late

Pleistocene and Holocene, the lake alternated between an open

and closed hydrological system. It is in the upper reaches of a

much larger endorheic system that includes Lago Poopo and

the salares of central and southern Bolivia. Today, water drains

out of Lake Titicaca over a bathymetric sill at 3,804 m a.s.l.

down the Río Desaguadero from the southwest corner of Lago

Wiñaymarka (Wirrmann, 1992). The lake has undergone

490 LAKE-LEVEL FLUCTUATIONS

measurable water-level changes during the historic period (AD

1914 to present), ranging from 3,806.2 m a.s.l. in 1943 to

3,812.6 m a.s.l. in 1986, with an average annual fluctuation

of 0.8 m (Roche et al., 1992). The Lake Titicaca basin is parti-

cularly sensitive to drought because even with the overflowing

conditions that prevail today, only 1 to 3% of the lake water is

lost by surficial overflow. The majority of the balance is lost to

evaporation in the high altitude semi-arid environment.

The water-level history of Lake Titicaca was reconstructed

from the lithostratigraphy and 60 radiocarbon measurements.

First, the dated cores were used to identify stratigraphic levels

where erosion surfaces occurred, indicating that the lake level

was below that point (Figure L2). Second, for the periods with

erosional surfaces, the transect of cores was used to identify the

highest stratigraphic level where lacustrine sediments were pre-

served, indicating that water level was above that point. Using

this twofold stratigraphic approach, five periods of significant

lake-level depression forming five erosion surfaces were

defined by the following criteria: (a) scour marks, (b) mud

cracks, (c) abrupt transitions (<1 cm) characterized by

coarser-grained (fine-sand) sediments with high bulk density

(>1gcm

3

) overlying fine-grained organic-rich mud (> 20%

organic matter), (d) an abrupt increase in both iron and potas-

sium concentration associated with the reducing conditions

in water-saturated soils, and (e) highly fragmented shell mate-

rial in the overlying mud. The presence of one or more of

these characteristics combined with an abrupt change in the

radiocarbon age of adjacent strata indicates erosion or non-

deposition. Detailed core descriptions, smear-slide mineralogy,

and radiocarbon stratigraphy were used to delimit water-

saturated soils and erosion surfaces formed during low water

stands and sub-aerial exposure. Shallow-water subfacies

(<2 m water depth) were identified by: (a) the presence of

high concentrations of achenes (seeds) of the littoral sedge

Schoenoplectus tatora in a coarse-grained matrix (silt to sand),

(b) large amounts of aquatic plant macrofossils (Myriophyllum,

Chara, and Potamogeton), and (c) sediments containing >90%

CaCO

composed of calcified macrophyte coatings and frag-

mented mollusk shells.

This study demonstrated the presence of middle to late

Holocene hydrological fluctuations on the South American alti-

plano as shown by five periods of low water stands in Lake

Titicaca. The water level of Lake Titicaca fluctuated >22 m

during the past 3,500 years. Four of the lowstands were pro-

found (indicated water lowering depth) and occurred abruptly

over a period of 100–200 years. The earliest occurred during

a prolonged middle Holocene dry phase that ended after

3,500

BP. All of the cores collected in Lago Wiñaymarka

Figure L2 Lake-level reconstruction from Lake Titicaca based on lithostratigraphy from four cores and 60 AMS

C measurements. The Gray

Triangle symbol signifies the lowest stratigraphic level in cores A-D where an erosion surface occurs, indicating that lake level was below this point

at the time indicated. Likewise, the Black Triangle symbol indicates the highest stratigraphic level where lacustrine sediments are preserved,

implying water level was above this level.

LAKE-LEVEL FLUCTUATIONS 491

for this study penetrate into previously sub-aerially exposed

sediments, indicating that Lake Titicaca was >15 m below

overflow level (BOL) prior to 3,500

BP. A second lowstand of

5–8 m BOL occurred between 3,200 and 2,800

BP. Shallow-

water sub-facies between 3,500 and 2,800

BP suggest the lake

remained below the overflow stage during this period. The

third lowstand of 10–12 m BOL ended after 2,200

BP. The

duration of this dry phase is estimated at 200 years. The fourth

lowstand of 10–12 m BOL ended after 1,450

BP. The duration

of this phase remains unknown, but is likely to have been near

250 years based on sediment accumulation rates. The final low

lake level of 7–12 m BOL began prior to 900

BP and ended

after 600

BP. Shallow-water subfacies suggests that water level

probably remained low until 500

BP. The final lowstand is coin-

cident with the decline of raised-field agriculture and collapse

of Tiwanaku culture.

Mark B. Abbott and Lesleigh Anderson

Bibliography

Abbott, M.B., Binford, M.W., Brenner, M., and Kelts, K., 1997. A 3500

C yr high-resolution record of water level changes in Lake Titicaca,

Bolivia/ Peru. Quaternary Res., 47, 169–180.

Abbott, M.B., Finney, B.P., Edwards, M.E., and Kelts, K.R., 2000. Lake-

Level reconstructions and paleohydrology of Birch Lake, Central

Alaska, based on seismic reflection profiles and core transects. Qua-

ternary Res., 53, 154–166.

Alin, S.R., and Cohen, A.S., 2003. Lake-level history of Lake Tanganyika,

East Africa, for the past 2500 years based on ostracod-inferred water-

depth reconstruction. Palaeogeogr. Palaeoclimatol. Palaeoecol., 199,

31–49.

Anderson, L., Abbott, M.B., Finney, B.P., and Edwards, M.E., 2005. Paleo-

hydrology of the southwest Yukon Territory, Canada, based on

multi-proxy analyses of lake sediment cores from a depth transect.

The Holocene, 15(8), 1172–1183.

Barber, V.A., and Finney, B.P., 2000. Late Quaternary paleoclimatic recon-

structions for interior Alaska based on paleolake-level data and hydro-

logic models. J. Paleolimnol., 24,29–41.

Benson, L.V., 1981. Paleoclimatic significance of lake level fluctuations in

the Lahontan Basin. Quaternary Res., 16, 390–403.

Benson, L.V., Curry, D.R., Dorn, R.I., Lajoie, K.R., Oviatt, C.G.,

Robinson, S.W., Smith, G.I., Stine, S., and Thompson, R.S., 1989.

Variations in sizes of four Great Basin lake systems during the past

35,000 years. Palaeogeogr. Palaeoclimatol. Palaeoecol., 78, 241–268.

Binford, M.W., Kolata, A.L., Brenner, M., Janusek, J., Abbott, M.B., and

Curtis, J., 1997. Climate variation and the rise and fall of an Andean

civilization. Quaternary Res., 47, 171–186.

Bookman, R., Yehouda, E., Agnon, A., Stein, M., 2004. Late Holocene

lake levels of the Dead Sea. Geol. Soc. Am. Bull., 116, 555–571.

Brenner, M., Hodell, D.A., Curtis, J.H., Rosenmeier, M.F., Binford, M.W.,

and Abbott, M.B., 2001. Abrupt climate change and Pre-Columbian

cultural collapse. In Markgraf, V. (ed.), Interhemispheric Climate

Linkages. San Diego, CA: Academic Press, pp. 87–103.

Dearing, J.A., 1997. Sedimentary indicators of lake-level changes in the

humid temperate zone: A critical review. J. Paleolimnol., 18,1–14.

Digerfeldt, G., 1986. Studies on past lake-level fluctuations. In Berglund,

B.E. (ed.), Handbook of Holocene Palaeoecology and Palaeohydrol-

ogy. Chichester: Wiley, pp. 127–

143.

Finney, B.P., Scholz, C.A., Johnson, R.C., Trumbore, S., 1996. Late

Quaternary lake-level changes of Lake Malawi. In Johnson, T.C.,

Odada E.O. (eds.), Limnology, Climatology and Paleoclimatology of

the East African Lakes. Australia: Gordon and Breach Publishers,

495–508.

Gilbert, G.K., 1890. Lake Bonneville. United States Geological Survey

Monograph, volume 1, pp. 213.

Harrison, S.P., and Digerfeldt, G., 1993. European lakes as palaeo-

hydrological and palaeoclimatic indicators. Quaternary Sci. Rev., 12,

223–248.

Hastenrath, S., and Kutzbach, J.E., 1983. Paleoclimatic estimates from

water and energy budgets of East African lakes. Quaternary Res., 19 ,

141–153.

Hastenrath, S., and Kutzbach, J.E., 1985. Late Pleistocene climate and

water budget of the South American Altiplano. Quaternary Res., 24,

249–256.

Hodell, D.A., Curtis, J.H., Jones, G.A., Higuera-Gundy, A., Brenner, M.,

Binford, M.W., and Dorsey, K.T., 1991. Reconstruction of Caribbean

climate change over the past 10,500 years. Nature, 352, 790–793.

Hoffman, W., 1998. Cladocerans and chironomids as indicators of

lake level changes in north temperature lakes. J. Paleolimnol., 19,

55–62.

Hostetler, S.W., 1995. Hydrologic and thermal response of lakes to climate:

Description and modeling. In Lerman, A., Imboden, D.M., Gat, J.R.

(eds.), Physics and Chemistry of Lakes (2nd edn.). New York /Amsterdam:

Springer, pp. 63–82.

Keely, B., Scholz, C.A., King, J.W., Peck, J., Overpeck, J.T., Russell, J.M.,

and Amoako, P.V.O., 2005. Late Quaternary low stands of Lake

Bosumtwi, Ghana: Evidence from high resolution seismic reflection

and sediment core data. Palaeogeogr. Palaeoclimatol. Palaeoecol.,

216, 235–249.

Kutzbach, J.E., and Street-Perrott, F.A., 1985. Milankovitch forcing of fluc-

tuations in the level of tropical lakes from 18 to 0 kyr

BP. Nature, 317,

130–134.

Lundqvist, G., 1925. Utvecklingshistoriska insjostudier I Sydsverige.

Sveriges Geologiska Undersokning, C 330.1–129.

Lundqvist, G., 1927. Bodenablagerungen und Entwicklungstypen der Seen.

Die Binnengewasser, 2,1–122.

Oviatt, C.G., Sack, D., and Curry, D.R., 1994. The Bonneville basin, Qua-

ternary, western U.S. In Gierlowski-Kordesch, E., and Kelts, K. (eds.),

Global Geologic Record of Lake Basins (vol. 1). Cambridge, UK:

Cambridge University Press, pp. 371–375.

Roche, M.A., Bourges, J.C., and Mattos, R., 1992. Climatology and

hydrology of the Lake Titicaca basin. In Dejoux, C., and Iltis, A.

(eds.), Lake Titicaca: A Synthesis of Limnological Knowledge. Boston:

Kluwer, pp. 63–83.

Seltzer, G.O., Baker, P.B., Cross, S., Fritz, S.C., and Dunbar, R., 1998.

High-resolution seismic reflection profiles from Lake Titicaca, Peru/

Bolivia. Geology, 26, 167–170.

Street-Perrott, F.A., and Harrison, S.P., 1985. Lake levels and climate

reconstruction. In Hecht, A.D. (ed.), Lake Levels and Climate Recon-

struction. New York, NY: Wiley, pp. 291–340.

Street-Perrott, F.A., and Perrott, R.A., 1990. Abrupt climate fluctuations in

the tropics: The influence of Atlantic ocean circulation. Nature, 343,

607–612.

Webb, T. III., Ruddiman, W.F., Street-Perrott, F.A., Markgraf, V.,

Kutzbach, J.E., Bartlein, P.J., Wright, H.E., and Prell, W.L., 1993.

In Vegetation, lake levels, and climate in eastern North America for

the past 18,000 years. Wright, H.E., Kutzbach, J.E., Webb, T. III.,

Ruddiman, W.F., Street-Perrott, F.A., and Bartlein, P.J. (eds.), Global

climate since the Last Glacial Maximum. Global climate since the Last

Glacial Maximum, Minneapolis, MN: University of Minnesota Press,

pp. 514–535.

Wirrmann, J.P., 1992. Morphology and bathymetry. In Dejoux, C., and

Iltis, A. (eds.), Lake Titicaca: A Synthesis of Limnological Knowledge.

Boston: Kluwer, pp. 16–23.

Wright, H.E., Kutzbach, J.E., Webb, T. III., Ruddiman, W.R., Street-

Perrott, F.A., and Bartlein, P.J., 1993 (eds.). Global Climate Since the

Last Glacial Maximum. Minneapolis, MN: University of Minnesota

Press, pp. 569.

Yang, X., Kamenick, C., Schmidt, R., and Sumin, W., 2003. Diatom based

conductivity and water level inference models from eastern Tibetan

(Qinghai-Xizang) Plateau lakes. J. Paleolimnol., 30,1–19.

Yu, Z., McAndrews, J.H., and Eicher, U., 1997. Middle Holocene dry

climate caused by change in atmospheric circulation patterns: Evidence

from lake levels and stable isotopes. Geology, 25, 251–254.

Cross-references

Lacustrine Sediments

Paleoclimate Proxies, an Introduction

Paleolimnology

Paleo-Precipitation Indicators

Radiocarbon Dating

492 LAKE-LEVEL FLUCTUATIONS

LAST GLACIAL MAXIMUM

Introduction

The Last Glacial Maximum (LGM) occurred when ice sheets

and glaciers throughout the world were at their maximum

extent during the last ice age. Once dated as 18,000 radiocar-

bon (21,500 calibrated/sidereal/“calendar”) years ago, the

LGM is now believed to have occurred between 23,000 and

19,000 (cal) years ago (Clark and Mix, 2002). However, debate

continues regarding its extent, timing and the nature of its ice

sheets (below). The LGM is believed to have resulted from cli-

mate change caused by changes in the orbital configuration of

the Earth. However, the LGM was synchronous in both hemi-

spheres, suggesting that poorly understood feedback factors

probably operated. Somehow, they provided sources of precipi-

tation for ice growth and maximum expansion at the LGM.

The LGM is an important “time-slice” for paleogeographic

reconstruction and computer simulation because the configura-

tion of the Earth differed so greatly from that of the present.

Extensive mid-latitude ice sheets existed in North America,

northwest Europe and, controversially, in northern Russia and

Siberia. The shape of ocean basins and coastlines was different

because water abstracted from the ocean and locked in the ice

sheets lowered sea level by between 120 and 130 m. This

altered the chemistry of the ocean, which became more saline

and isotopically (d

O) heavier. Atmospheric circulation was

modified by the large ice sheets: for example, the LGM

Laurentide Ice Sheet (LIS) split the jet stream into two arms.

Thick ice sheets caused the Earth’s crust to subside glacio-

isostatically, the extent of which is partly revealed by raised

shorelines fashioned as the crust recovered and sea level rose

during and after deglaciation. Modeling these effects has pro-

vided constraints on the rheology of the Earth’s mantle and

crust. The concentration of greenhouse gases in the atmosphere

(CO

and CH

) was lower than at present, at a time of reduced

ocean and continental temperatures throughout the world.

Vegetational zones were compressed by latitude and altitude.

General aridity prevailed as deserts expanded and extensive

dust storms deposited sand and silt (loess) deposits. However,

in some regions that are arid today, extensive lakes occurred,

such as Lake Bonneville and Lake Lahontan in America

(Benson, 2004).

History

The term “Last Glacial Maximum” is relatively recent and

refers to the time of maximum global ice volume inferred from

the oxygen isotope stratigraphy (d

O) of ocean sediments

(CLIMAP, 1976, 1981). While the d

O signal is a global

one, it conceals a wide variety of temporal responses by differ-

ent ice masses caused by regional climate variation as well as

the glaciological and environmental characteristics of indivi-

dual ice sheets and glaciers. This was recognized when the

LGM was re-defined as having occurred between 23,000 and

19,000 (cal) years ago (Clark and Mix, 2002).

The LGM concept has older antecedents that identified the

maximum extent of continental ice sheets during the “latest”

or “last” glaciation (Flint, 1971; Sibrava et al., 1986). This

was based on field mapping of “fresh” and relatively unmodi-

fied glacial and glaciofluvial landforms, such as moraines,

drumlins and eskers. First developed in the midwestern USA,

where the ice maximum became known as the “Classical Wis-

consin” (Flint, 1971), this method was also applied to Great

Britain and Ireland where a “newer drift” (LGM) was distin-

guished from an older one. However, this criterion is not

always reliable: the “fresh” moraines of the Warthe glaciation

in Germany were once though to be LGM, but are older than

the Eemian (“last interglacial”). Only the late Weichselian gla-

ciation is of LGM age.

Techniques

Glacial geology in the field provides evidence regarding ice

source regions, ice movement and the extent of LGM ice. Esti-

mating the precise age of the glaciation is essential for determin-

ing its timing and regional variability. Since 1950, radiocarbon

dating has been used to accomplish this essential task, with

accelerator mass spectrometry (AMS) providing greater preci-

sion and accuracy. Where

C dating of organic samples is impos-

sible, cosmogenic rock exposure ages can now date glacial

boulders and glaciated rock surfaces (Gosse and Phillips,

2001). Uranium series dating of corals provides information on

the LGM sea level – a means of estimating continental ice

volume. Within this dating framework, a whole range of environ-

mental disciplines use data from continents and oceans to com-

plete the picture beyond the LGM ice sheets. These data form

the basis of computer simulations of the LGM climate and,

reciprocally, provide validation or question such simulations

(Ruddiman, 2000). LGM ice sheets have also been simulated

as three-dimensional thermodynamic ice-sheet models.

Extent of glac iation

Two versions of the possible extent of ice in the Northern

Hemisphere at the LGM were assembled by Denton and

Hughes ( 1981 ) and CLIMAP (1981): a minimum and a maxi-

mum reconstruction (Figure L3). The main differences involve

the extent of glaciation in northern Russia and Siberia and

Arctic Canada (below). In the Southern Hemisphere, the

Antarctic ice sheet extended across the continental shelf to its

edge and large ice caps expanded in the Himalayas, South

America and New Zealand. High altitude small ice caps and

glaciers at lower latitudes expanded in East Africa, the Andes,

New Guinea and Hawaii. However, the greatest expansion

occurred in the mid- to higher-latitudes of North America and

Europe, for which there is no present-day analogue.

The extent of ice at the LGM is well established on the

southern margins of Northern Hemisphere mid-latitude ice

sheets. The LIS, which was coextensive with the Cordilleran

Ice Sheet of the Rocky Mountains, advanced to Long Island,

Ohio, Illinois, Iowa and South Dakota. In Europe, the Fenno-

Scandian Ice Sheet advanced to Berlin and Warsaw. However,

the northern margins of LGM ice continue to be the subject

of debate (Andrews, 1987). Ongoing controversy continues

regarding northwestern Russia and eastern Siberia, where a

major expansion of the Barents Sea and Kara Sea ice sheets

may predate the LGM. Much of the evidence is submarine

and insufficient reliable age estimates are available on land.

Ice sheet reconstructions

The LGM reconstructions of CLIMAP were based to a large

extent on a “marine-based” ice sheet model from Antarctica

(Denton and Hughes, 1981). For example, in North America,

the LIS reconstruction was a high-elevation, over 4 km thick,

single ice dome centered over Hudson’s Bay. Similarly, the

LAST GLACIAL MAXIMUM 493

Fenno-Scandian ice sheet was centered over the Gulf of

Bothnia, where it exceeded 3 km in thickness. These became

standard ice-sheet boundary conditions for computer simula-

tions of the LGM climate. More recent ice sheet modeling sug-

gests an ice thickness of over 3.8 km for the LIS, with mean

volumetric estimates of 10

(Marshall et al., 2002).

Glacial geology in the field, however, especially indicators

of ice movement from glacial erratics, is not consistent with

the CLIMAP reconstruction. Instead the field evidence points

to a multi-domed and thinner LIS. The LIS was less extensive

in the Canadian Arctic and, as such, supports the minimum

view of LGM glaciation (Andrews, 1987; Dyke et al., 2002).

A thinner, multi-domed ice sheet is believed to have responded

more rapidly to changes in climate: for example, Heinrich

events, massive ice discharges into the ocean once thought to

have lasted about 2,000 years, are now believed to have lasted

only 250 years. Such rapid changes are consistent with climate

events inferred from ice core records in Greenland.

Timing of the LGM

Table L1 shows estimates of ice volume and the amount of sea

lowering at the LGM. Recent estimates based on hundreds of

AMS

C and uranium series ages from corals in Barbados

suggest that the lowest sea level occurred 26,000 years ago

(Fairbanks and Peltier, 2004).

Radiocarbon dating indicates that the LIS may have advanced

rapidly to its maximum extent as early as 27–29 cal ka

BP,well

before the LGM, and remained near that limit until 17 ka (cal).

However, the Cordilleran Ice Sheet in the Puget Sound region

did not reach its maximum extent until 4,000 years later. Simi-

larly, the Antarctic Ice Sheet reached its maximum extent at

different times in different sectors. The Scandinavian and British

ice sheets were not joined at the LGM (compare with

Figure L3), but may have been joined somewhat earlier. They

reached their maximum extent about >26 (cal) ka along western

margins, but about 23–22 (cal) ka on their southern margins.

The age of the LGM on the northern margin of the LIS

is unknown and may be much younger than on its southern

margin. All that is known is that deglaciation commenced about

15,000 years ago, an age probably close to maximum expansion

(Andrews, 1987;Dykeetal.,2002). Uncertainty in establishing

the age of the LGM in some regions arises because suitable sam-

ples are unavailable for radiocarbon dating, a deficiency now

being addressed by cosmogenic rock exposure ages.

David Q. Bowen

Bibliography

Andrews, J.T., 1987. The Late Wisconsin glaciation and deglaciation of the

Laurentide Ice Sheet. In Ruddiman, W.F., and Wright, H.E. Jr. (eds.),

North America and Adjacent Oceans during the Last Deglaciation.

Boulder, CO: Geological Society of America. Geology of North

America; K-3, pp. 13–38.

Benson, L., 2004. Western Lakes. In Gillespie, A.R., Porter, S.C., and

Atwater, B.F. (eds.), The Quaternary Period in the United States.

Amsterdam/ New York: Elsevier, pp. 185–204.

Clark, P.U., and Mix, A.C., 2002. Ice Sheets and sea level of the Last

Glacial Maximum. Quaternary Sci. Rev., 21,1– 7.

CLIMAP (Climate Long Range investigation, Mapping and Prediction).,

1976. The surface of the ice age Earth. Science, 191, 1131–1144.

CLIMAP., 1981. Seasonal Reconstructions of the Earth’s Surface at the

Last Glacial maximum. Geological Society of America Map Series,

MC-36. Boulder, CO: Geological Society of America.

Denton, G.H., and Hughes, T.J., 1981. The Last Great Ice Sheets. New

York: Wiley, pp. 1–484.

Dyke, A.S., Andrews, J.T., Clark, P.U., England, J.H., Miller, G.H., Shaw, J.,

and Veillette, J.J., 2002. The Laurentide and Innuitian ice sheets during

the Last Glacial Maximum. Quaternary Sci. Rev., 21,9–31.

Fairbanks, R.G., and Peltier, W.R., 2004. Reanalysis and Extension of the

Barbados Sea Level Record. Eos Transactions American Geophysical

Union, 85 (47).

Table L1 Estimates of ice equivalent sea level (m) for LGM ice sheets

(adapted from Clark and Mix, 2002)

Ice sheet CLIMAP min. CLIMAP max. Model min. Model max.

Antarctica 24.5 24.5 14 21

NorthAmerica 77 92 82.4 82.4

Greenland 1 6.5 2 3

Eurasia 20 34 13.8 18

All Others 5 6 6 6

TOTAL 127.5 163 118.2 130.4

Figure L3 (a) The CLIMAP minimum reconstruction of Northern Hemisphere LGM ice sheets; (b) The CLIMAP maximum reconstruction of Northern

Hemisphere ice sheets. L: Laurentide Ice Sheet; C: Cordilleran Ice Sheet; I: Innuitian Ice Sheet; G: Greenland Ice Sheet; B: Great Britain and

Ireland Ice Sheet; S: Scandinavian (Fenno-Scandian) Ice Sheet; Ba: Barents Sea Ice Sheet; K: Kara Sea Ice Sheet. This does not show smaller ice caps:

for example, in the western USA or the Pyrenees. See text for the location of the smaller Southern Hemisphere ice sheets and ice caps.

494 LAST GLACIAL MAXIMUM

Flint, R.F., 1971. Glacial and Quaternary Geology. New York: Wiley.

Gosse, J.C., and Phillips, F.M., 2001. Terrestrial in situ cosmogenic

nuclides: Theory and application. Quaternary Sci. Rev., 20, 1475–1560.

Marshall, S.J., James, T.S., and Clarke, G.K.C., 2002. North American Ice

Sheet reconstruction at the Last Glacial Maximum. Quaternary Sci.

Rev., 21, 175–192.

Ruddiman, W.F., 2000. Earth’s Climate. New York: W.H. Freeman.

Sibrava, V., Bowen, D.Q., and Richmond, G.M., 1986. (eds.). Glaciations

in the Northern Hemisphere. (Project 24 International Geological Corre-

lation Programme). Quaternary Sci. Rev., 5,1–511.

Cross-references

Atmospheric Circulation during the Last Glacial Maximum

CLIMAP

Cordilleran Ice Sheet

Cosmogenic Radionuclides

Cryosphere

Eemian (Sangamonian) Interglacial

Glacial Geomorphology

Glacial Megalakes

Glacial Sediments

Heinrich Events

Ice cores, Antarctica and Greenland

Ice-Rafted Debris (IRD)

Laurentide Ice Sheet

Oxygen Isotopes

Paleoclimate Modeling, Quaternary

Paleotemperatures and Proxy Reconstructions

Pollen Analysis

Radiocarbon Dating

Scandinavian Ice Sheet

Sea Level Change, Quaternary

Thermohaline Circulation

Uranium-Series Dating

Wisconsinan (Weichselian, Würm) Glaciation

LAST GLACIAL TERMINATION

Late Quaternary climates are characterized by successive gla-

cial and interglacial periods. While glacial inceptions are

usually progressive, the transitions from full glacial conditions

to full interglacial ones occur within about 10 kyr or less. Dur-

ing these remarkable time periods, named terminations, the cli-

mate system experiences considerable reorganizations. The

Last Glacial Termination, between approximately 20 and 8

BP, is of particular interest since it can be studied in detail

and can be dated with good accuracy using radiocarbon meth-

ods. If, by definition, the major event is the disappearance of

the large Northern Hemisphere ice sheets (Laurentide and Fen-

noscandian) with a resulting sea-level rise of about 120 m, this

is associated with many other important changes. First, the

warming recorded in Greenland ice cores or in marine cores

at high northern latitudes does not smoothly follow the decreas-

ing size of continental ice areas. On the contrary, abrupt warm-

ings and coolings, associated with changes in ocean circulation

punctuate the transition. Second, many well-dated climatic

proxies suggest that some generalized warming that was

recorded in tropical or southern marine cores occurred before

any major reduction of northern ice sheets. Third, the rise in

atmospheric carbon dioxide concentration from glacial levels

around 180–200 ppm towards interglacial levels around

260–280 ppm also occurred several thousands of years before

the continental ice volume decrease. All these facts are not

easily explained in the context of the Milankovitch theory,

which claims that astronomical changes affect the mass balance

of ice sheets, and consequently lead to global climatic changes.

Understanding the dynamics of the Last Glacial Termination is

therefore a critical step both for the theory of glacial-intergla-

cial cycles as well as for understanding millenial scale climatic

variability.

Astronomical forcing and ice volume changes

Terminations attracted the attention of paleoclimatologists

when they examined the first continuous marine isotopic

records. The characteristic sawtooth character of climatic

changes was first outlined by Emiliani (1966). The notion of

terminations was introduced by Broecker and van Donk

(1970) and it was already clear at this time that the astronomi-

cal theory would not easily explain this structure (Broecker and

van Donk, 1970): “... the cause of the primary sawtoothed

cycle is still an open question.”

Though this early definition of terminations was descriptive,

it is interesting to note that terminations do systematically cor-

respond to deviations that would be predicted by a linear astro-

nomical theory of Quaternary climate, as illustrated in

Figure L4. If the origin of these sawtoothed 100 kyr glacial-

interglacial cycles is linked to astronomical parameters, it

remains to be explained what amplifying feedbacks are

involved during terminations. When examining the last five

or ten terminations, there is no direct correspondence between

the timing or intensity of terminations and the insolation for-

cing. On the contrary, termination V, between isotopic stages

12 and 11, was one of the largest transitions, although it

occurred about 425 ky

BP when insolation changes were quite

small. Conversely, termination III, between stages 8 and 9,

was not a large transition although seasonal insolation changes

were considerable about 243 ky

BP.

Probably the simplest way to explain terminations is to envi-

sion the sawtooth-like 100 kyr cycles as relaxation oscillations.

In this concept, the ultimate cause of terminations would be found

in the preceding glacial maximum, and the insolation forcing only

acts as a favorable external influencing factor. This can quite suc-

cessfully explain the timing of events (Paillard, 1998)aswellas

the amplitude and phase of terminations (Parrenin and Paillard,

2003). However, a mechanistic understanding of terminations

needs a closer inspection of the evolution of different climatic

parameters like temperatures at different Earth locations, deep

oceanographic properties, sea level, or greenhouse gas concentra-

tions. A detailed sequential chronology of these changes is avail-

able for the last termination.

The sequence of events during the Last Glacial

Termination

The summer insolation forcing at high northern latitudes (taken

at the summer solstice) steadily increased from the Last Glacial

Maximum (LGM) about 21 ky

BP up to 11 kyBP, in the middle

of the termination, at the peak insolation time. However, the

mid-transition actually occurred around 12 ky

BP, while the

maximum insolation forcing happened later, around 11 ky

for June insolation, around 8 kyBP for August insolation, and

around 9 ky

BP for the averaged summer forcing. This highlights

the fact that the phase relationship between astronomical for-

cing and global ice sheet volume is poorly understood on a mil-

lennial time scale. However, as shown in Figure L5, the

temperature variations have a much more complex history.

LAST GLACIAL TERMINATION 495

At high northern latitudes, an abrupt warming was observed in

Greenland around 14 ky

BP, as well as in many other locations

in Europe and in the North Atlantic. The following warm inter-

val (Bølling-Allerød) only lasted about two thousand years and

ended with an abrupt cooling event into the Younger Dryas

time interval. This cold period was shorter than a thousand

years and also ended abruptly with a warm period that progres-

sively led into the Holocene and the present-day climate. The

same pattern is also visible in the atmospheric methane (CH

)

concentration recorded in Greenland ice cores. However,

records from Antarctica or from the Southern Ocean have

almost the opposite pattern, with a cooling occurring in the

middle of the transition, the Antarctic Cold Reversal, during

the northern warm Bølling-Allerød interval. The explanation

for such rapid climatic changes, as shown for example by

231

Pa/

230

Th records, involves major changes in the Atlantic

deep circulation, which transports heat from one hemisphere

to the other. This opposition has been named the “bipolar-

seesaw” (Stocker, 1998). These abrupt changes recorded during

the last termination are a clear indication that the internal varia-

bility of our climate is of primary importance in accounting for

the dynamics of glacial-interglacial transitions.

The atmospheric CO

concentration closely follows the evo-

lution of Antarctic temperatures, although some secondary fea-

tures are probably linked to the northern climate (Monnin et al.,

2001). Interestingly, this southern temperature rise started

around 18 ky

BP, when the northern ice sheets were still close

to their maximum volume. Explaining this southern warming

with an astronomical forcing appears quite problematic,

although the annual mean insolation and the obliquity forcing

may have some importance (Loutre et al., 2004). More than

half of the glacial-interglacial pCO

increase, from 180 to

240 ppm, occurred during this early phase, between 18 ky

and 14 kyBP, while sea level rose by about only 20 m above

glacial levels. Furthermore, an early warming has also been

found in many other locations, like the tropics (Lea et al.,

2000). It is therefore reasonable to assume that this early atmo-

spheric CO

rise and the associated warming had a critical role

in the triggering of the deglaciation. In this context, the key to

understanding the terminations probably lies in the carbon

cycle (Paillard, 2001), and the fundamental question is not in

understanding why pCO

was low during LGM, but why it

rose rapidly just afterwards. If maximum global ice volume

was a trigger for some Southern Ocean warming and pCO

Figure L4 The top curve is a classical reconstruction of glacial-interglacial cycles, based on the d

O of foraminifera, and interpreted here as

normalized ice volume V (Bassinot et al., 1994). On the diagram below, the derivative (dV /dt) is plotted as a function of the insolation forcing.

A clear linear relationship is obtained, in large part due to the tuning methodology used to build a timescale for the ice volume data. However,

marked points are significantly below the regression line (heavy line). It is then possible to define terminations as time periods during which ice

volume decreases much faster than a linear astronomical theory could predict (below the dotted line). They occur about every 100 kyr, as can be

seen on the top panel where the same marked points are represented.

496 LAST GLACIAL TERMINATION

Figure L5 From top to bottom, daily insolation forcing at the summer solstice at 65



N (Berger, 1978); sea level reconstructed from corals

(Fairbanks, 1989; Bard et al., 1996); Greenland temperature based on oxygen isotopes (Dansgaard et al., 1993); atmospheric CH

concentration

recorded at Dome C (Monnin et al., 2001); intensity of the deep oceanic circulation based on

231

Pa /

230

Th ratio (McManus et al., 2004); atmospheric

concentration (Monnin et al., 2001); Antarctic temperature based on deuterium measured at Dome C (Stenni et al., 2001). Cold events are

highlighted: Heinrich event 1 (H1), Youger Dryas (YD), Antarctic Cold Reversal (ACR).

LAST GLACIAL TERMINATION 497

increase, as required by the relaxation oscillation paradigm, it is

natural to look at the evolution of the Antarctic Ice Sheet at this

time. Since it is largely driven by sea-level changes, the Ant-

arctic continental ice reached its maximum by entirely covering

the continental shelves several thousands of years after the

LGM (Denton and Hughes, 2002). A possible scenario is to

assume that this could affect sea ice formation and bottom

water formation around Antarctica. If we furthermore hypothe-

size that low CO

levels during glacial times are linked to

Southern Ocean bottom water characteristics, then expansion

of the Antarctic Ice Sheet could lead to the atmospheric

pCO

increase. This would account both for the triggering of

terminations and for the observed sequence of events (Paillard

and Parrenin, 2004).

Millenial scale variability and terminations

The role of the millenial-scale climatic variability on the dynamics

of the last termination remains an open question. As shown in

Figure L5, the southern warming and pCO

increase are associated

with a major decrease in the deep Atlantic Ocean circulation. This

oceanic slowdown is traditionally explained by the massive ice-

berg discharge from the Laurentide Ice Sheet, known as Heinrich

event 1, which occurred around 17 ky

BP. A reasonable hypothesis

would be to put these millennial scale changes at the center of the

dynamics of terminations, but many questions still need to be

resolved: What triggers Heinrich events? Are deep circulation

slowdowns caused solely by iceberg discharges? What is the

role and dynamics of the Younger Dryas and, more generally, of

Dansgaard-Oeschger events? How can a Southern Ocean warm-

ing explain a rapid pCO

increase by 40–60ppminonly3or

4 thousand years? All these questions are currently being closely

examined by paleoclimatologists, and their answers should lead

to a new understanding of the Last Glacial Termination. They also

highlight how far we have come from the traditional Milankovitch

theory in which Quaternary ice volume changes were tentatively

explained only by insolation and ice sheet mass balance. If

Dansgaard-Oeschger events, the Younger Dryas, Heinrich events,

associated Antarctic warmings, and finally glacial-interglacial

cycles appear today to be rather different topics, the detailed his-

tory of the Last Glacial Termination suggests that understanding

the links between these different aspects of climatic variability

could provide the clue to a revised and a more integrated theory

of Quaternary climates.

Didier Paillard

Bibliography

Bard, E., Hamelin, B., Arnold, M., Montaggioni, L., Cabioch, G., Faure, G.,

and Rougerie, F., 1996. Deglacial sea-level record from Tahiti corals and

the timing of global meltwater discharge. Nature, 382,241–244.

Bassinot, F.C., Labeyrie, L.D., Vincent, E., Quidelleur, X., Shackleton,

N.J., and Lancelot, Y., 1994. The astronomical theory of climate and

the age of the Brunhes-Matuyama magnetic reversal. Earth Planet.

Sci. Lett., 126,91–108.

Berger, A.L., 1978. Long-term variations of daily insolation and

Quaternary climatic change. J. Atmos. Sci., 35, 2362–2367.

Broecker, W.S., and van Donk, J., 1970. Insolation changes, ice volumes

and the O

record in deep-sea cores. Rev. Geophys. Space Phys., 8,

169–197.

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. Nature, 364, 218–220.

Denton, G.H., and Hughes, T.J., 2002. Reconstructing the Antarctic ice

sheet at the Last Glacial Maximum. Quaternary Sci. Rev., 21, 193–202.

Emiliani, C., 1966. Paleotemperature analysis of Caribbean cores P6304–8

and P6304–9 and a generalized temperature curve for the past 425,000

years. J. Geol., 74, 109.

Fairbanks, R.G., 1989. A 17,000 year glacio-eustatic sea level record:

Influence of glacial melting rates on the Younger Dryas event and deep

ocean circulation. Nature, 342, 637–642.

Lea, D.W., Pak, D.K., and Spero, H.J., 2000. Climate Impact of late

Quaternary Equatorial Pacific sea surface temperature variations.

Science, 289, 1719–1724.

Loutre, M.F., Paillard, D., Vimeux, F., and Cortijo, E., 2004. Does mean

annual insolation have the potential to change the climate ? Earth

Planet. Sci. Lett., 221,1–14.

McManus, J.F., Francois, R., Gherardi, J.-M., Keigwin, L.D., and Brown-

Leger, S., 2004. Collapse and rapid resumption of Atlantic meridional

circulation linked to deglacial climate changes. Nature, 428, 834–837.

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

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

concentrations over the Last Glacial Termination. Science, 291,

112–114.

Paillard, D., 1998. The timing of Pleistocene glaciations from a simple

multiple-state climate model. Nature, 391, 378–381.

Paillard, D., 2001. Glacial cycles: Toward a new paradigm. Rev. Geophys.,

39, 325–346.

Paillard, D., and Parrenin, F., 2004. The Antarctic ice-sheet and the trigger-

ing of deglaciations. Earth Planet. Sci. Lett., 227, 263–271.

Parrenin, F., and Paillard, D., 2003. Amplitude and phase of glacial cycles

from a conceptual model. Earth Planet. Sci. Lett., 214, 243–250.

Stenni, B., Masson-Delmotte, V., Johnsen, S., Jouzel, J., Longinelli, A.,

Monnin, E., Röthlisberger, R., and Selmo, E., 2001. An oceanic cold

reversal during the last deglaciation. Science, 293, 2074–2077.

Stocker, T.F., 1998. The seesaw effect. Science, 282,61–62.

Cross-references

Antarctic Cold Reversal

Astronomical Theory of Climate Change

Bølling-Allerød Interstadial

Dansgaard-Oeschger Cycles

Heinrich Events

Last Glacial Maximum

Millennial Climate Variability

North Atlantic Deep Water and Climate Change

Quaternary Climate Transitions and Cycles

Sea Level Change, Quaternary

Thermohaline Circulation

Younger Dryas

LATE PALEOZOIC PALEOCLIMATES

Introduction

The climate of the late Paleozoic, particularly the Pennsylva-

nian and Early Permian, was very similar to the present-day cli-

mate. During both intervals, the South Pole experienced

significant glaciation and the North Pole experienced intermit-

tent glaciation (Hambrey and Harland, 1981; Zachos et al.,

2001). Because of this, the term “ice house” has been used

to describe the climate of both intervals. Paradoxically, al-

though the poles were frigid, warm, wet climates prevailed at

the equator during both intervals. In both intervals, glaciation

occurred in the context of a major continental collision:

between Laurussia and Gondwana to form Pangaea in the late

Paleozoic; and between India and Asia to form Eurasia in the

Cenozoic. Finally, in both intervals, an extended period of

“hot-house” climate, with warm poles and shallow pole-to-

equator temperature gradients, preceded glaciation (Ziegler

et al., 1981; Zachos et al., 2001; Raymond and Metz, 2004).

498 LATE PALEOZOIC PALEOCLIMATES