Odekon M. Encyclopedia of paleoclimatology and ancient environments

Подождите немного. Документ загружается.

Less-pronounced IRD layers have also been found at the

transition from MIS 5e to 5d and during MIS 5d. The two ice

rafting events C26 (at 115 ky

BP) and C25 (at 110 kyBP)

can be traced into the central part of the North Atlantic. A pro-

minent cooling of 3



C in SST at the transition from MIS 5e to

5d, derived from alkenone records from the Bermuda Rise,

may correlate with the C26 event in the North Atlantic. The

cold-water event C26 in the North Atlantic most likely relates

to the opening of woodlands in NW Europe. If this is so, the

event C26 marks the end of the classical Eemian in NW Europe

(Figure E4e–g). Based on pollen analysis and varve counts at

lake sediments from the famous Bispingen site in northern Ger-

many, it has been determined that Eemian woodlands existed

for around 11,000 years (Müller, 1974). Thus, the classical

Eemian lasted from 126 to 115 kyr.

Recurring cold events, averaging one event every 1,500

year, have been recognized in North Atlantic sediments of

MIS 5e and Eemian lake sediments in Central Europe (Bond

et al., 2001; Müller et al., 2005). By analogy with findings

from the Holocene in the North Atlantic, the trigger for this last

interglacial cyclic climate variability may have been changes in

solar activity, possibly amplified by changes in North Atlantic

ocean currents and/or the North Atlantic Oscillation. This sub-

orbital cyclicity during the time of minimum global ice volume,

however, was strongly subdued in comparison with the high-

amplitude climate fluctuations of the last glacial.

Terrestrial environments

The term “Eem” was introduced by Harting (1874), based on

the identification of a fauna of Lusitanian and even Mediterra-

nean shells in marine clays buried by the last glacial deposits

near the river Eem in the Netherlands. Today, it is customary

to interpret the Eemian in a pollen record as an interval domi-

nated by forest elements and associated with the climatic ame-

lioration of the last interglacial. The terms “Eemian” and “last

interglacial” should not be used synonymously. The Eemian,

is strictly speaking, a biostratigraphic representative of the last

interglacial in Europe. The onset of the Eemian is defined at the

point in the pollen record when tree taxa exceed 50% of the

total terrestrial pollen sum. The Eemian ends at the point when

even the percentages of coniferous trees decline significantly.

Obviously, this concept implies that, for reasons of climate gra-

dients and plant migration, the onset and end of the Eemian did

not occur at the same time in different areas of Europe. There-

fore, the Eemian is time-transgressive and should not be used

in a chronostratigraphic sense. In fact, it now appears that

Eemian woodlands persisted considerably longer in southern

Europe (126–110 ky

BP) than in northwest Europe (126–115

BP)(Figure E4f,g). This implies the establishment of steep

vegetation and climate gradients at the inception of the last gla-

cial. The steepening of vegetation gradients might have been

influenced by the steepening of meridional SST gradients in

the North Atlantic, which in turn is suggested to be connected

to a southward displacement of the North Atlantic Current at

around 115 ky

BP (Müller and Kukla, 2004).

The identification of the Eemian in a pollen record is based

on a specific composition and succession of forest elements.

In contrast to older interglacials, Eemian pollen records in

Europe do not contain Carya, Celtis, Eucommia, Pterocarya,

Sequoia, Thuja, and Tsuga as these taxa became extinct in this

region earlier in the Quaternary. A peculiarity of the Eemian

north of the Alps and Pyrenees is the absence of Fagus as

today this tree dominates quasi-natural forests in most of these

areas. The typical composition and succession of the Eemian is

known from master pollen records that reflect the biostratigra-

phy of several glacial-interglacial cycles (e.g., Reille et al.,

2000). Absolute dating of interglacial sediments provides a tool

to verify the biostratigraphic evaluation at other sites.

In Central Europe, the typical pollen zones of the Eemian

have a remarkable consistency. This consistency is believed

to be linked to an open seaway from the English Channel

through the North and Baltic Seas to the White Sea, which sup-

ported the influence of mild oceanic climates over the entire

region. The typical composition and succession of Eemian for-

est elements in Central Europe is shown in Figure E5. After a

pioneer phase with Juniperus, Betula, and Pinus, the Eemian

is characterized by the early spread of the deciduous taxa

Ulmus, Quercus, and Fraxinus and the subsequent spread of

Corylus. Within the Corylus phase there is a Taxus peak (some-

times featured as a double peak). Subsequently, an Eemian pol-

len record displays the spread of Carpinus and Abies. The

maximum amount of Carpinus and Abies in the late-temperate

phase varies in relation to altitude and proximity to mountain

ranges (Müller, 2000). The major spread of Picea, occurs in

the post-temperate phase of the Eemiopts. The final phase of

the Eemian shows the spread and dominance of Pinus before

coniferous forests decline.

In northern Europe, Eemian pollen records vary from those

in Central Europe. In general, the composition of taxa stays

the same. However, the further north a site is located, the more

reduced is the percentage of the temperate taxa Ulmus, Quer-

cus, Fraxinus, Corylus, and Carpinus to the benefit of Betula,

Pinus, Alnus, and Picea. In northern Scandinavia, Eemian pol-

len diagrams do not show a temperate phase anymore, but a

constant dominance of Pinus and Betula

. Nevertheless, the

long-distance wind transport of pollen resulted in grains from

Central Europe found even far in the north of Scandinavia

(Robertsson, 2000). There, the grains from temperate taxa

sometimes reflect the typical Eemian succession of Central

Europe.

In southern Europe, the boreal taxa Betula and Picea are

absent south of the Alps and Pyrenees or do not play an impor-

tant role in the record. The onset of the Eemian starts directly

with the spread of deciduous Quercus, Ulmus, Corylus, and

subsequently of Carpinus and Quercus evergreen. Mediterra-

nean taxa such as Pistacia, Olea, and Phillyrea contributed pol-

len during the first half of the Eemian. Deciduous Quercus and

Carpinus are the dominant taxa during the late-temperate phase

of the Eemian. Abies was of minor importance. There was no

post-temperate boreal forest phase such as in Central Europe,

therefore late-deciduous woodlands persisted to the end of the

Eemian at around 111 ky

BP (Tzedakis et al., 2002). Fagus has

been found south of the Alps in Italy and Austria during the

Eemian.

In North America, pollen investigations in interglacial sedi-

ments have not hitherto had the same significance as in Europe.

This might be due to the scarcity of interglacial lake sediments

or a lack of exploration drilling.

Other representatives of the last interglacial in terrestrial

environments are buried soils and geomorphologic features

testifying to the absence of glaciers. This is how the classic

Sangamonian interglacial has been described in North America.

The term “Sangamon” was introduced by Leverett (1898),

based on identification of a soil buried beneath till and loess

of the last glacial that he observed in railroad cuts in Sangamon

EEMIAN (SANGAMONIAN) INTERGLACIAL 305

County, Illinois. The Sangamon Geosol is a strongly developed

paleosol in the parent material from the penultimate glacial

(named Illinoian in the USA), and buried by the sediments of

the last glacial (Wisconsinan). This paleosol has been described

from many locations in the USA. It can be traced from the

Great Lakes to Texas, and is frequently used as a marker hori-

zon. The type of soil development varies with changes in par-

ent material and regional climate conditions. In its type area

at Athens Quarry, Illinois, the Sangamon Geosol is developed

in loess colluvium and shows a remarkable thick Ab-horizon

on top of a Bt-horizon. Overall, soil proxies show that the

Sangamon Geosol is more strongly developed than the Holo-

cene soil. However, the differences in soil properties between

the Sangamon Geosol and Holocene soil are interpreted primar-

ily to reflect soil development duration, with climatic imprint

being secondary. It is assumed that the Sangamon Geosol

formed over a much longer time interval than the last intergla-

cial, as is understood in Europe (Grimley et al., 2003). The

weathering features of the Sangamon Geosol at its type region

indicate that the formation persisted longer than the complete

Figure E5 Characteristic taxa composition and succession of an Eemian pollen record in Central Europe (data from Jammertal in SW Germany,

adopted from Mu

ller et al., 2005).

306 EEMIAN (SANGAMONIAN) INTERGLACIAL

MIS 5. The time represented in the Sangamon Geosol depends

upon whether burial occurred and pedogenesis ceased during

the early or late part of the Wisconsin glacial. In contrast to

the Sangamon Geosol, the Sangamonian Stage has been

defined as a chronostratigraphic unit by the Illinois State Geo-

logical Survey. The delimitation of the Sangamonian Stage is

arbitrary. Some authors adopt the boundaries of MIS 5,

whereas others restrict the Sangamonian Stage to MIS 5e.

The last interglacial paleosol in Europe, frequently found in

loess sections, is in general preserved as a well-developed tex-

tural Bt-horizon. The last interglacial soil is known as Rocourt

soil in NW Europe, as B1b of PK III in the Czech Republic,

and as the Salyn soil of the Mezin Complex in Central Russia.

The Bt-horizon is usually truncated and partly covered by

humus soil sediments (Antoine et al., 2001). In high accumula-

tion loess records, three humus zones have been found on top

of the truncated last interglacial Bt-horizon (Bibus et al.,

2002). The chernozem-like humus zones are separated from

each other and from the Bt-horizon by loess colluvium that

contains a mollusk fauna indicative of colder climate condi-

tions. Therefore, it has been concluded that the humus zones

developed during MIS 5a to 5c whereas the Bt-horizon is asso-

ciated with MIS 5e.

Conclusions

 The last interglacial is the most recent interval prior to the

Holocene when the global climate was as warm as or even

warmer than today.

 Eustatic sea-level changes, a proxy of mean global climate,

indicate that the last interglacial as a chronostratigraphic unit

lasted from 128 to 116 ky

BP.

 The Eemian is a biostratigraphic representative of the last

interglacial in Europe.

 The Sangamon Geosol is a pedostratigraphic representative

of ice-free conditions between the last and penultimate

glaciation in North America.

 Both the Sangamon Geosol and Eemian have time-transgres-

sive boundaries.

 MIS 5e, a close representative of the last interglacial in

deep-sea sediments, records the time when total global

volume of land-based ice was reduced.

Ulrich C. Müller

Bibliography

Antoine, P., Rousseau, D.-D., Zöller, L., Lang, A., Munaut, A.-V., Hatte,

C., and Fontugne, M., 2001. High-resolution record of the last

interglacial-glacial cycle in the Nussloch loess-Palaeosol sequences,

Upper Rhine area, Germany. Quaternary Int., 76–77,211–229.

Berger, A., and Loutre, M.F., 1991. Insolation values for the climate of the

last 10 million years. Quaternary Sci. Rev., 10, 297–317.

Bibus, E., Rähle, W., and Wedel, J., 2002. Profilaufbau, Molluskenführung

und Parallelisierungsmöglichkeit des Altwürmabschnittes im Lössprofil

Mainz-Weisenau. Eiszeitalter und Gegenwart, 51,1–14.

Bond, G., Kromer, B., Beer, J., Muscheler, R., Evans, M.N., Showers, W.,

Hoffmann, S., Lotti-Bond, R., Hajdas, I., and Bonani, G., 2001. Persis-

tent solar influence on North Atlantic climate during the Holocene.

Science, 294, 2130–2136.

Grimley, D.A., Follmer, L.R., Hughes, R.E., and Solheid, P.A., 2003. Mod-

ern, Sangamon and Yarmouth soil development in loess of unglaciated

southwestern Illinois. Quaternary Sci. Rev., 22, 225–244.

Hall, R.D., and Anderson, A.K., 2000. Comparative soil development of

Quaternary paleosols of the central United States. Palaeogeogr. Palaeo-

climatol. Palaeoecol., 158, 109–145.

Harting, P., 1874. De bodem van het Eemdal. Verlagen en Mededelingen

van de Koninklijke Akademie van Wetenschappen, afdeling. Natuur-

kunde II, 8, 289–290.

Kukla, G.J., Bender, M.L., Beaulieu, J.-L., de, Bond, G., Broecker, W.S.,

Cleveringa, P., Gavin, J.E., Herbert, T.D., Imbrie, J., Jouzel, J.,

Keigwin, L.D., Knudsen, K.-L., McManus, J.F., Merkt, J., Muhs, D.R.,

Mueller, H., Poore, R.Z., Porter, S.C., Seret, G., Shackleton, N.J.,

Turner, C., Tzedakis, P.C., and Winograd, I.J., 2002. Last Interglacial

Climates. Quaternary Res., 58,2–13.

Leverett, F., 1898. The weathered zone (Sangamon) between the Iowan

loess and Illinoian till sheet. J. Geol., 6, 171–181.

Litt, T., 1994. Paläoökologie, Paläobotanik und Stratigraphie des Jungquar-

tärs im nordmitteleuropäischen Tiefland unter besonderer Berücksichti-

gung des Elbe-Saale-Gebietes. Dissertationes Botanicae, 227,1–185.

McManus, J.F., Oppo, D.W., Keigwin, L.D., Cullen, J.L., and Bond, G.C.,

2002. Thermohaline circulation and prolonged interglacial warmth in

the North Atlantic. Quaternary Res., 58,17–21.

Muhs, D.R., 2002. Evidence for the timing and duration of the last intergla-

cial period from high-precision Uranium-Series Ages of corals on tecto-

nically stable coastlines. Quaternary Res., 58,36–40.

Müller, H., 1974. Pollenanalytische Untersuchungen und Jahresschichten-

zählungen an der eemzeitlichen Kieselgur von Bispingen/Luhe. Geolo-

gisches Jahrbuch, A21, 149–169.

Müller, U.C., 2000. A late Pleistocene pollen sequence from the Jammertal,

SW Germany with particular reference to location and altitude as fac-

tors determining Eemian forest composition. Vegetation History and

Archaeobotany, 9, 125–131.

Müller, U.C., and Kukla, G.J., 2004. North Atlantic Current and European

environments during the declining stage of the last interglacial. Geol-

ogy, 32, 1009–1012.

Müller, U.C., Klotz, S., Geyh, M.A., Pross, J., and Bond, G.C., 2005. Cyc-

lic climate changes during the Eemian interglacial in Central Europe.

Geology, 33, 449–452.

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

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

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

Ritz, C., Saltzmann, E., and Stievenard, M., 1999. Climate and atmo-

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

Antarctica. Nature, 399, 429–436.

Reille, M., Beaulieu, J.-L., Svobodova, H., Andrieu-Ponel, V., and Goeury,

C., 2000. Pollen analytical biostratigraphy of the last five climatic

cycles from a long continental sequence from the Velay region (Massif

Central, France). J. Quaternary Sci., 15, 665–685.

Robertsson, A.-M., 2000. The Eemian interglacial in Sweden, and compar-

ison with Finland. Geologie en Mijnbouw, 79, 325–333.

Shackleton, N.J., Sanchez-Goni, M.F., Pailler, D., and Lancelot, Y., 2003.

Marine Isotope Substage 5e and the Eemian Interglacial. Glob. Planet.

Change, 36, 151–155.

Turner, C., and West, R.G., 1968. The subdivision and zonation of intergla-

cial periods. Eiszeitalter und Gegenwart, 19,93–101.

Tzedakis, P.C., Frogley, M.R., and Heaton, T.H.E., 2002. Duration of Last

Interglacial conditions in Northwestern Greece. Quaternary Res., 58,

53–55.

Wijmstra, T.A., 1969. Palynology of the first 30 meters of a 120 m deep

section in northern Greece. Acta Botanica Neerlandica

, 18,511–527.

Cross-references

Astronomical Theory of Climate Change

Dating, Biostratigraphic Methods

Eccentricity

Glacial Eustasy

Heinrich Events

Ice-Rafted Debris (IRD)

Laurentide Ice Sheet

Obliquity

Oxygen Isotopes

Paleosols, Quaternary

Precession, Climatic

Pollen Analysis

Scandinavian Ice Sheet

Sea Level Change, Quaternary

Uranium-Series Dating

EEMIAN (SANGAMONIAN) INTERGLACIAL 307

ELECTRICAL CONDUCTIVITY

Electrical measurements are frequently used to determine the

chemical characteristics of ice cores in paleoclimate studies.

Electrical measurements on ice cores can be made quickly,

non-destructively, and with high spatial resolution, but they do

not provide information on specific ions. Electrical measure-

ments are used to identify the seasonal variations in chemistry

that define the annual layers in an ice core that are used to deter-

mine the age of ice, and to determine if the layers in an ice core

have been distorted by abnormal ice flow. For these tasks the

high spatial resolution and efficiency of the electrical measure-

ments is more important than information on specific ions.

Electrical measurements can be classified into two general cate-

gories: direct current and alternating current methods.

Direct current methods are conventionally referred to as the

Electrical Conductivity Method (i.e., ECM) (Hammer, 1983)

and measure the current flow between two electrodes with a

potential difference of about 1,000 V. The electrodes typically

have a surface area of about 16 mm

and are separated by about

1 cm. Direct comparison of measurements made with different

instruments is complicated because of differences in the elec-

trode geometry and the potential difference. Measurements are

typically made every 1 mm along a prepared flat surface of

the core that is parallel to the axis of the core. The ice lattice

prevents lateral movement of all ions except free protons, and

hence only protons can carry the direct current. The ECM

method is therefore sensitive only to the concentration of H

ions associated with strong acids. The ECM record is typically

presented as a continuous high spatial resolution record indica-

tive of the acidity of the ice core. It is particularly well suited

for indicating the presence of acids from volcanic activity

and identifying annual layers.

Alternating current methods are referred to as Dielectric

Property (DEP) measurements. They utilize a capacitor consist-

ing of ice as the dielectric material between two conductive

plates (Moore et al., 1989). A thin insulating layer prevents

electrical connection between the conductive plates and

the ice. The impedance of the ice-filled capacitor is typically

measured over a range of frequencies every 1 cm along

the core. The ice lattice prevents the lateral movement of all

ions except free protons, but the other ions can conduct

the electrical current by rotating around a stationary location.

The DEP method is therefore sensitive to the concentration of

a wide range of ions. The DEP record is typically presented

as a continuous record indicative of the total ion content of

the ice core. A variation on the DEP method is complex con-

ductivity that uses an alternating current with an electrode geo-

metry similar to the ECM method (Taylor et al., 2004).

The complex conductivity method is also sensitive to the

total ion content of the ice but has more spatial resolution

than the DEP method.

The DEP and ECM methods are typically presented as a sin-

gle profile along the length of an ice core. Multitrack measure-

ments utilize multiple adjacent profiles that are measured on a

flat surface parallel to the axis of the core (Taylor and Alley,

2004). This is presented as a two dimensional image of the con-

ductivity in a vertical plane through the ice core. This enables

the inclination and continuity of annual layers to be observed.

Kendrick Taylor

Bibliography

Hammer, C.U., 1983. Initial direct current in the buildup of space

charges and the acidity of ice cores. J. Phys. Chem., 87(21),

4099–4103.

Moore, J.C., Mulvaney, R.L., and Paren, J.G., 1989. Dielectric strati-

graphy of ice; a new technique for determining total ionic concentra-

tions in polar ice cores. Geophys. Res. Lett., 16(10), 1177–1180.

Taylor, K.C., and Alley, R.B., 2004. Two dimensional electrical stratigra-

phy of the Siple Dome, Antarctica ice core. J. Glaciol., 50, 231–235.

Taylor, K.C., Alley, R.B., Meese, D.A., Spencer, M.K., Brook, E.J.,

Dunbar, N.W., Finkel, R., Gow, A.J., Kurbatov, A.V., Lamorey, G.W.,

Mayewski, P.A., Meyerson, E., Nishiizumi, K., and Zielinski, G.A.,

2004. Dating the Siple Dome, Antarctica ice core by manual and com-

puter interpretation of annual layering. J. Glaciol., 50, 453–461.

Cross-reference

Ice cores, Antarctica and Greenland

EOLIAN DUST, MARINE SEDIMENTS

Introduction

Charles Darwin was the first scientist to comment formally on

the red dusts blowing from Africa out across the Atlantic, an

event he observed while on the Beagle (Darwin, 1846). Marine

geologists have known since the late 1950s that dust storms

originating in the arid and semi-arid regions of continents carry

dust long distances, and that sea-floor sediments in regions far

from continents, out of the reach of river-borne sediment, are

dominated by eolian dust.

Atmospheric scientists find that most of the dust is trans-

ported by a few spring storms. These storms raise the dust high

up into the troposphere where it is transported for long dis-

tances and eventually removed by both dryfall and rainout

(Prospero et al., 1983; Pye, 1987). Satellites now routinely

track dust storms that move from the semiarid regions of cen-

tral China all the way across the North Pacific to California,

and from the Sahara across the Atlantic Ocean to the Caribbean

and Florida. Since most of the world’s continental surface area

is in the Northern Hemisphere, the Northern Hemisphere is far

dustier than the Southern.

Sources of atmospheric dust

Three main source regions provide dust to the atmosphere and

ocean. The two largest are the deserts and basins of central

China, north of the Tibetan Plateau, and the Sahara and Sahel

regions of North Africa. The third important source of dust is

the deserts of Arabia and the Horn of Africa, which provide

dust to the northwestern Indian Ocean. The amount of material

available for dust transport depends on the climate of the

source region, with arid and semi-arid climates providing the

most material, and more humid – vegetated – regions provid-

ing less. There are, however, hyperarid regions where rain falls

very rarely. In these regions, there is not enough water to break

down the feldspar minerals that are common in continental

rocks into clays. Therefore, this weathering process does not

create enough fine-grained material for long-range transport.

What little material there was has been stripped away by the

wind long ago, a condition known as deflation. The deserts

308 ELECTRICAL CONDUCTIVITY

of Australia fit this description; they are hyperarid, deflated,

and supply surprisingly little dust to the oceans. As a result

of this physical and climatic geography, there is about an order

of magnitude less dust deposited in the Southern Hemisphere.

The total amount of dust delivered to the ocean is estimated

to be about 0.1 to 0.2  10

tyr

1

, which is roughly 1% of

the total river load of 15 to 20  10

tyr

1

(Rea, 1994). None-

theless, pelagic clays composed of eolian dust characterize the

surface sediments of about one third of the ocean floor, an area

similar to that of all land surfaces.

Size of dust grains

The large spring dust storms raise a broad mixture of grains

from the desert floors, but after a transport distance of 1,000

or 2,000 km, the sand and coarse silt sized grains have fallen

out and the remaining suspended minerals are fine silt and

clay-sized. The size of these distal dust grains, now preserved

in abyssal sediments, reflects the dynamic equilibrium between

the supporting energy of the transporting winds and gravita-

tional settling. Equilibrium grain size changes very little with

increasing distance from the source region, and maps of the

median grain size of dust in North Pacific sediments show

little variation across this broad expanse of ocean (Rea and

Hovan, 1995). Thus, at any one location, the size of these equi-

librium dust grains preserved in deep sea sediments is a record

of the energy of the transporting winds, with stronger winds

resulting in coarser grains. Numerous grain size studies of dust

show that it has a median diameter of 2–4 microns. The

hypothesis that dust grain size provides a record of wind

strength was tested by examining dust captured in sediment

traps during an experiment conducted in the Arabian Sea.

There, the dust grain size was coarsest when winds were stron-

gest, but not necessarily when the most material was being

transported (Clemens, 1998).

This observation indicates one more aspect of dust deposi-

tion that is important: that dust grain size is independent of

the amount (and source) of dust deposited. Cross-plots of grain

size and dust supply data from any one core show a random

scatter. This independence means that scientists may discover

the source area climate from the dust supply data and the wind

intensity from the grain size data, both from the same set of

samples.

Composition, mineralogy and geochemistry

The first modern reports on the subject of Saharan dust in

the central Atlantic recognized the red-stained quartz grains

accumulating in Atlantic sediments as being from the Sahara.

Marine geologists studying the surface sediments of the North

Pacific have understood for more than 40 years that the band

of quartz-rich sediment extending east from the deserts of

central and eastern Asia represented dust coming from China.

Studies comparing the detailed mineralogy of the dust with that

of the surface sediment show them to be indistinguishable.

For the North Pacific region, the average composition of the

<2 mm mineral component is dominated by illite, which is

30–40% of the total, with subequal amounts of quartz, chlorite,

kaolinite, and plagioclase (10 –15% each). Chlorite increases in

abundance to the north, and smectite becomes a noticeable

minor component nearer to North America (Leinen, 1989).

In the past decade, various technical advances with regard to

geochemical analyses have made it possible to perform other

studies that provide important information about dust transport

and deposition. Trace element geochemistry and radiogenic

isotopic analysis of the dust have allowed scientists to resolve

questions of source and how sources have changed over time.

The first broadly-based study was of the geochemistry of the

surficial sediment of the North Pacific. It showed that dust from

China dominates the Pacific surficial sediment all the way

south to the Intertropical Convergence Zone (ITCZ) and almost

all the way east to North America. North America, on the other

hand, supplies very little dust to the North Pacific Ocean. The

neodymium, lead, and strontium isotope geochemistry of the

dust isolated from sea-floor sediment samples is identical to

that of Chinese loess, with a similar bulk geochemistry to that

of Paleozoic shales (Jones et al., 1994, 2000). Over the years,

scientists have used both the mineralogy, especially the clay

mineralogy, and the geochemistry of the dust to determine its

provenance, with excellent results. As a result of this kind

of work, it is possible to understand the changing sources of

dust to the oceans, and that the dust in the Greenland ice cores

and Antarctic ice cores originates from Asia and southern

Argentina, respectively (see Biscaye et al., 1997).

History of dust deposition

Long-term changes in dust deposition

Several studies of the long-timescale record of dust deposition

have been carried out in both the Atlantic Ocean, downwind

from the Sahara, and in the North Pacific, downwind from

the deserts of central and eastern Asia. Most of this work has

aimed to determine the mass accumulation rate of the dust,

which is interpreted in terms of the aridity of the continental

source regions. Some of these studies have also analyzed the

grain size of the dust as an indication of wind strength, and

aimed to determine the geochemistry of the dust in order to

understand changes in the source region. Furthermore, all of

this long-term work has been carried out in the Northern Hemi-

sphere; the history of the Southern Hemisphere is assumed to

be similar but that has yet to be shown conclusively. Deposits

of nearly pure eolian sediment are known as pelagic clays or

red clays after their brick-red to brown color. There are very

few fossils in these cores, and because of their slow sedimenta-

tion rate, often 30–50 cm per 10

yr, magnetic reversal strati-

graphy generally does not work. Thus, age control is poor;

often done on the basis of fossil fish teeth, or ichthyoliths.

Improving the age control in red clays is an ongoing research

topic (Gleason et al., 2002).

Two North Pacific cores have been studied in detail: a giant

piston core LL44-GPC3 (this coding refers to the ship that

raised the core, the AT&T cable-laying ship Long Lines, and

that this core was the third Giant Piston Core recovered), and

the cores from Ocean Drilling Program (ODP) Site 885/886.

Core GPC3 came from the central North Pacific at about

30.3



N, 157.8



W, and ODP Site 885/886 is at 44.7



168.3



W. Both sites now underlie the zonal winds of the

prevailing westerlies. The record at GPC3 spans the entire

Cenozoic, covering the last 70 million years. The iridium layer

corresponding to the meteorite impact at the Cretaceous-

Tertiary boundary that may have caused the ultimate extinction

of the dinosaurs lies in the lower portion of the core. The mate-

rials recovered farther north at Site 885/886 provide a very

good record of the last 12 million years. Together, these two

records define the eolian depositional record for the latest

Cretaceous and Cenozoic time.

Dust accumulation recorded in core GPC3 (Figure E6;

Janecek and Rea, 1983; Rea et al., 1985) shows a very low

EOLIAN DUST, MARINE SEDIMENTS 309

input of dust through most of the Cenozoic. There is doubling

of flux values at about 25 Ma, moving upcore from the Paleo-

gene to the Neogene, and a further large increase in dust input

at 3 or 4 Ma. This history can be interpreted in terms of chan-

ging climates, in this case increasing aridity, in the eolian

source region. The size of the eolian grains is a measure of

the wind intensity. In the younger part of the record, the size

of the dust grains increases towards the present from low

values in the Eocene, likely a result of increasing wind inten-

sity accompanying the mid to late Cenozoic cooling of the

higher latitudes. The big surprise in the GPC3 data was

the very large change in dust grain size in the lower part of

the core. We have found this same shift in grain size in other

cores from the North Pacific. It represents, apparently, a sudden

decline in wind intensity associated with several global changes

that occurred in conjunction with the Paleocene/Eocene

boundary (Rea et al., 1990) and the onset of unusually warm

Eocene climates.

Sediments recovered from ODP Site 885/886 are character-

ized by a reliable magnetic reversal stratigraphy and so are

much better constrained as to age and rate than those from

GPC3. Nevertheless, the major late Cenozoic increase in dust

flux seen in GPC3 is also the most significant change in this

record (Figure E7). This order of magnitude increase in flux

begins at 3.6 Ma and represents a significant drying in the

eolian source regions in China. Other data indicate a rapid

uplift of the northeastern portion of the Tibetan Plateau then,

the deserts of central China being cut off from the moisture

source to the south (Rea et al., 1998). The interpretation of the

dust flux peak at about 8 Ma is less clear. The size of the eolian

grains increases in younger materials. The increase is not regular,

and the last shift towards coarser dust grains, hence stronger

winds, begins about 4.5 Ma. Many studies of late Cenozoic cli-

mate change have shown that major Northern Hemisphere gla-

ciation, the ice age we are in currently, began at 2.6 Ma. Other

kinds of paleoclimate records suggest that this time of increasing

continental ice cover was preceded by an increase in wind inten-

sity two million years earlier, and an increase in the atmospheric

dust loading one million years earlier. It can be concluded that

different aspects of the world’s climate system change sequen-

tially, one after the other, rather than all at the same time.

Eolian records from the Atlantic and Indian Oceans indicate

a late Cenozoic drying of Saharan and Sahelian Africa. The

general sense of more dust coming into the oceans during

the past three or four million years holds for these studies,

but the timing of the aridity increases does not seem to be

exactly the same for northern Africa and eastern Asia. A drying

of the Northern Hemisphere continents in general concor-

dance with Northern Hemisphere cooling and the ultimate

beginning of the Plio-Pleistocene glacial ages has been clearly

demonstrated by all studies of dust depositional history.

Figure E6 The Cenozoic record of dust deposition from core

LL44-GPC3. Grain size values are median diameter in phi-units, where

phi is the negative logarithm to the base 2 of the grain diameter in

millimeters (f = log

). Thus 9f =2mm, 8f =4mm, etc.

Figure E7 The record of dust deposition and grain size from core ODP

Site 885 /886 for the last 12 million years.

310 EOLIAN DUST, MARINE SEDIMENTS

Orbital timescale changes in dust deposition

The classic terrestrial, Quaternary record of climate change is

found in the loess horizons, interspersed with soils, that sweep

across the Loess Plateau of China. Loess, a wind blown silt,

forms during the relatively cold and dry glaciations, and the

soils form during the warmer and wetter interglacial periods.

In marine geology, the classic record of Quaternary climate

change is found in the oxygen-18 (d

O) values of seawater,

which respond to the amount of glacial ice piled onto the con-

tinents. Linking these two paleoclimate records has been a

long-standing goal of paleoclimatologists. The chance to

achieve this objective was provided by a core raised from

the Shatsky Rise, an oceanic plateau in the northwest Pacific.

A study of dust deposition in this core captured the past

500,000 years (Figure E8) of the joint continental and marine

history of climate change (Hovan et al., 1991). The flux of

dust to the North Pacific varies by a factor of three to five

on these glacial-interglacial timescales, with fluxes reaching a

maxima during glaciations. As the dust flux maximal also

correspond with the timing of loess formation, it is possible

to determine the temporal relation between loess formation

and the oxygen isotopic record generated from the same core.

Spectral analysis of these and other eolian records of sui-

table resolution have shown the importance of the orbital varia-

bility that was first well quantified by Milankovitch in these

paleoclimatic records (Pisias and Rea, 1988; Clemens and

Prell, 1990; Hovan et al., 1991).

Provenance studies

The geochemistry of the dust accumulated on the sea floor pro-

vides clues as to the source of the dust and the nature of any

one source region. Knowing which continent, Asia or America,

has provided the dust to the core location is a critical first step

in the interpretation of source area climate. Here, again, the

best-studied core is GPC3. Kyte and others (1993) analyzed

samples from throughout GPC3 and were able to demonstrate

that the dust in the upper portion of that core came from Asia,

and dust in the lower part came from America. This observa-

tion regarding a change in the bulk geochemistry from a

younger material of Asian affinity to an older sediment of

American affinity was also noted by Pettke et al. (2002) who

measured the Nd and Pb isotopic ratios of the dust. In the pre-

sent climate regime, the change in dust geochemistry from

Asian to American happens at the Intertropical Convergence

Zone, and not, as might be expected, at the boundary between

the zonal westerlies and the tradewinds. If this observation

remains true during the Cenozoic, then the mid-Cenozoic geo-

chemical transition observed in GPC3 marks the past location

of the ITCZ, a location far north of the equator, perhaps



–25



N (Rea, 1994; Rea et al., 2000; Pettke et al., 2002).

Figure E8 The record of dust deposition in the northwestern Pacific Ocean for the past 500,000 years. The column on the left shows the

classic Chinese loess stratigraphy from the section at Xifeng. Data from core V21-146 raised from Shatsky Rise at 37.7



N, 163.0



E includes eolian

flux and the d

O values measured by N. Shackleton on benthic foraminifera. The right column is the standardized d

O record of the marine

geologists developed by the SPECMAP project. Even numbers denote glacial stages where isotopic values are more positive.

EOLIAN DUST, MARINE SEDIMENTS 311

It is also possible to learn more about the eolian source areas

than just where they are. For instance, Pettke et al. (2000)

examined samples from ODP Site 885/886 and determined

the

Ar/

Ar ages of six dust samples whose stratigraphic ages

ranged from 1.2 to 11.1 Ma. The mineral ages of those sam-

ples, dominated by illite, is about 200 million years. The 200

Myr age was interpreted to represent the formation age of

the illites, and demonstrates that the processes of weathering,

erosion, and late Cenozoic transport of those illites has not

affected their argon-argon ages. This implies that other geo-

chemical attributes of dust minerals may reflect the ultimate

origin of those minerals and not the continental climate regime

immediately prior to their transport to the sea.

Summary

Wind-blown dust forms a minor portion of deep sea sediment

on a volumetric basis, but much of the world’s ocean basins

are characterized by a terrigenous sediment component that is

dominated by eolian dust. These dust deposits provide paleocli-

matic information not available from any other source. Studies

of wind-blown dust fall into three categories: determination of

its mass accumulation rate, an indicator of source area aridity;

dust grain size, an indicator of transport energy; and dust com-

position, used for provenance studies. Dust records provide a

useful and direct method to examine the evolution, usually cli-

mate related, of continental surfaces, and the only method to

examine the geologic history of the Earth’s wind systems

(Rea, 1994). This last aspect is particularly important, as atmo-

spheric circulation is an equal partner with ocean circulation as

a global agent of climate change.

David K. Rea

Bibliography

Biscaye, P.E., Grousset, F.E., Revel, M., Van der Gaast, S., Zielinsky, G.A.,

Vaars, A., and Kukla, G., 1997. Asian provenance of glacial

dust (stage 2) in the Greenland Ice Sheet Project 2 ice core, Summit,

Greenland. J. Geophys. Res., 102, 26765–26781.

Clemens, S.C., 1998. Dust response to seasonal atmospheric forcing: Proxy

evaluation and calibration. Paleoceanography, 13, 471–490.

Clemens, S.C., and Prell, W.L., 1990. Late Pleistocene variability of Ara-

bian Sea summer monsoon winds and continental aridity: Eolian

records from the lithogenic component of deep-sea sediments. Paleo-

ceanography, 5, 109–145.

Darwin, C., 1846. An account of the fine dust which often falls on vessels

in the Atlantic Ocean. Q. J. Geol. Soc. Lond., 2,26–30.

Gleason, J.D., Moore, T.C., Rea, D.K., Johnson, T.M., Owen, R.M., Blum,

J.D., Hovan, S.A., and Jones, C.E., 2002. Ichthyolith strontium isotope

stratigraphy of a Neogene red clay sequence: Calibrating eolian dust

accumulation rates in the central North Pacific. Earth Planet. Sci. Lett.,

202, 625–636.

Hovan, S.A., Rea, D.K., and Pisias, N.G., 1991. Late Pleistocene continen-

tal climate and oceanic variability recorded in northwest Pacific sedi-

ments. Paleoceanography, 6, 349–370.

Janecek, T.R., and Rea, D.K., 1983. Eolian deposition in the northeast Paci-

fic Ocean: Cenozoic history of atmospheric circulation. Geol. Soc. Am.

Bull., 94, 730–738.

Jones, C.E., Halliday, A.N., Rea, D.K., and Owen, R.M., 1994. Neody-

mium isotopic variations in North Pacific modern silicate sediment

and the insignificance of detrital REE contributions to seawater. Earth

Planet. Sci. Lett., 127,55–66.

Jones, C.E., Halliday, A.N., Rea, D.K., and Owen, R.M., 2000. Eolian

inputs of lead to the North Pacific. Geochimica et Cosmochimica Acta,

64, 1405–1416.

Kyte, F.T., Leinen, M., Heath, G.R., and Zhou, L., 1993. Cenozoic sedi-

mentation history of the central North Pacific: Inferences from the

elemental geochemistry of core LL44-GPC3. Geochimica et Cosmochi-

mica Acta, 57, 1719–1740.

Leinen, M., 1989. The pelagic clay province of the North Pacific Ocean.

In Winterer, E.L., Hussong, D.M., and Decker, R.W. (eds.), The

Eastern Pacific Ocean and Hawaii. Boulder, CO: Geological Society

of America. The geology of North America, Volume N, pp. 323–335.

Pettke, T., Halliday, A.N., Hall, C.M., and Rea, D.K., 2000. Dust produc-

tion and deposition in Asia and the North Pacific Ocean over the past

12 Myr. Earth Planet. Sci. Lett., 178

,1–17.

Pettke, T., Halliday, A.N., and Rea, D.K., 2002. Cenozoic evolution of

Asian climate and sources of Pacific seawater Pb and Nd derived from

eolian dust of sediment core LL44-GPC3. Paleoceanography, 17,

3.1–3.13.

Pisias, N.G., and Rea, D.K., 1988. Late Pleistocene paleoclimatology of the

central Equatorial Pacific: Sea-surface response to the southeast trade-

winds. Paleoceanography, 3,21–37.

Prospero, J.M., Charlson, R.J., Mohnen, V., Jaenicke, R., Delany, A.C.,

Moyers, J., Zoller, W., and Rahn, K., 1983. The atmospheric aerosol

system: An overview. Rev. Geophys. Space Phys., 21, 1607–1629.

Pye, K., 1987. Aeolian Dust and Dust Deposits. New York: Academic

Press, 334p.

Rea, D.K., 1994. The paleoclimatic record provided by eolian deposition in

the deep sea: The geologic history of wind. Rev. Geophys., 32,

159–195.

Rea, D.K., and Hovan, S.A., 1995. Grain size distribution and depositional

processes of the mineral component of abyssal sediments: Lessons from

the North Pacific. Paleoceanography, 10, 251–258.

Rea, D.K., Leinen, M., and Janecek, T.R., 1985. Geologic approach

to the long-term history of atmospheric circulation. Science, 227,

721–725.

Rea, D.K., Zachos, J.C., Owen, R.M., and Gingerich, P.D., 1990. Global

change at the Paleocene-Eocene boundary: Climatic and evolutionary

consequences of tectonic events. Palaeogeogr. Palaeoclimatol.

Palaeoecol., 79,117–128.

Rea, D.K., Snoeckx, H., and Joseph, L.H., 1998. Late Cenozoic eolian

deposition in the North Pacific: Asian drying, Tibetan uplift, and cool-

ing of the Northern Hemisphere. Paleoceanography, 13, 215–224.

Rea, D.K., Moore, T.C., Jr., and Lyle, M., 2000. Atmospheric and oceanic

circulation dynamics in the equatorial Pacific of the Paleogene world.

GFF, 122, 135–136.

Cross-references

Astronomical Theory of Climate Change

Cenozoic Climate Change

Dust Transport, Quaternary

Eolian Sediments and Processes

Ocean Drilling Program (ODP)

SPECMAP

EOLIAN SEDIMENTS AND PROCESSES

Introduction

The word “eolian” (or “aeolian”) comes from the Greek “Aeolus.”

Aeolus, in Greek mythology, was the god of the winds. “Eolian”

thus refers to any process of sediment entrainment, transport and

deposition by wind. Study of eolian processes and sediments is a

part of the disciplines of geomorphology, sedimentology and

paleoclimatology. Extensive reviews of the use of eolian deposits

in paleoclimatology can be found in Pye (1987), Pye and Tsoar

(1990), Lancaster (1995), Livingstone and Warren (1996)and

Kohfeld and Harrison (2001).

Eolian deposits of interest to paleoclimatologists include, in

order of decreasing particle size, windblown sand, loess, and

long-range-transported (“LRT” or “aerosolic”)dust.Syersetal.

312 EOLIAN SEDIMENTS AND PROCESSES

(1969) presented a useful fivefold classification of eolian materials

based on dominant particle size, transport pathway and range and

persistence in the atmosphere. In their scheme, there are continua

of (a) coarse-to-fine grained materials, (b) kilometer-to-10

ranges of transport, and (c) ground level-to-interplanetary space

levels of transport (Figure E9). Thus, eolian sand is in a boundary-

layer zone of transport (within 2 m of the ground surface)

and loess occupies a range of relatively short-to-intermediate-dis-

tance transport within the lower troposphere. It is important to

note, however, that some near-source loesses contain eolian sand.

Similarly, most loess deposits contain some particles that are in the

size range of aerosolic dust (10–1 mm) and these particles are cap-

able of much longer-distance transport in the middle and upper tro-

posphere. There is even a fraction of most loesses that contains

particles in the stratospheric dust particle size range (<1 mm).

These particles are capable of transport at very high altitudes in

the stratosphere over thousands of kilometers of distance.

Eolian sand

Eolian sand consists of all deposits of sand-sized particles (2.00–

0.05 mm diameter) that have been transported by the wind.

Although most investigators think of dunes when considering

eolian sand, there are also low-relief landforms composed of

wind-blown sand called sand sheets. Most investigators, how-

ever, have directed attention to dunes in studies related to

paleoclimatology. Hack (1941), in studying the origin of dunes

in the Navajo Country of northeastern Arizona, proposed that

the degree of activity of sand dunes (as well as the specific type

of dune form) is a function of the balance between sand supply,

wind strength and duration, and amount of vegetation cover.

Sixty years of study since that time have confirmed this basic

model. Lancaster (1988) showed that the degree of activity of

linear dunes in the Namib and Kalahari Deserts of southwes-

tern Africa is a function of the amount of time that wind is

above the threshold velocity for entrainment of sand grains

and the balance between precipitation and potential evapo-

ration, a proxy for amount of vegetation cover. Muhs and

Holliday (1995) confirmed that the Lancaster “dune mobility

index,” as it has come to be called, explains the varying

degrees of dune activity in the Great Plains and desert south-

west of the United States.

The importance of the Hack-Lancaster models of dune

activity to paleoclimatology is that they provide a basis for

interpreting past periods of eolian sand deposition or eolian

sand stability. Assuming that there is an adequate supply of

sand, the entrainment and transport of sand by wind will occur

when a vegetation cover is lacking. Plant-covered dunes remain

stable because much of the shear stress from wind is imparted

to the vegetation rather than to the sediment grains. Lack of a

protective vegetation cover usually occurs when potential eva-

potranspiration exceeds precipitation, a condition that charac-

terizes arid, semiarid and subhumid regions. Thus, in a

stratigraphic sequence of eolian sand and intercalated buried

soils (paleosols), periods of relatively greater aridity are repre-

sented by eolian sand and periods of relatively greater humidity

are represented by paleosols (Figure E10).

Because they are transported by the wind, stabilized eolian

sands are also a direct reflection of past atmospheric circulation.

Figure E9 Diagram showing the continuum of eolian materials, with range of transport as a function of dominant particle size and generalized

zone of transport in the atmosphere (based on Syers et al., 1969).

EOLIAN SEDIMENTS AND PROCESSES 313

Based on observations of actively moving sand dunes, eolian

geomorphologists have developed an extensive classification

of dune types that can be related to dominant wind directions

(Pye and Tsoar, 1990; Lancaster, 1995; Livingstone and

Warren, 1996). Linear dunes have long axes that parallel the

dominant sand-moving winds and barchanoid ridges have long

axes that are orthogonal to the predominant wind direction.

Some dune types with superficially similar forms are actually

formed by opposing directions of dominant winds. For exam-

ple, barchan dunes have a crescentic shape with arms that point

downwind, whereas parabolic dunes have a crescentic shape

with arms that point upwind. Recognition of the differences

between these two dune types requires field study. Neverthe-

less, stabilized dunes that have preserved their original forms

provide valuable information about past wind regimes.

Field studies have shown that combined geomorphic, strati-

graphic and geochronologic studies can document regional-scale

changes in atmospheric circulation. For example, Lancaster et al.

(2002) showed that linear dunes in western Mauritania are of three

different ages, based on optically stimulated luminescence (OSL)

ages, 25–15 ka (last-glacial age), 13–10 ka (Younger Dryas),

and <5 ka (late Holocene). These investigators found that last-

glacial dunes are oriented northeast and Younger Dryas-age

dunes are oriented north-northeast (Figure E11). In contrast, late

Holocene dunes are oriented north-to-south, consistent with mod-

ern northerly winds that characterize the dominance of the Azores

high over the region in summer. Thus, the greater proportion

of easterly and northeasterly winds that built the last-glacial and

Younger-Dryas-aged dunes reflect a different circulation regime

at those times.

Other regions also show considerable change in the directions

of dune-building winds over short timescales. In the central Great

Plains of the United States, late Holocene dunes are parabolic

forms and have orientations that are very similar to modern

dune-forming winds. For example, late Holocene parabolic dunes

in the Great Bend area of central and western Kansas have arms

that point to the southwest, consistent with modern southwesterly

winds. However, prior to the late Holocene (perhaps during a

mid-Holocene warm-dry period), winds apparently came from

the northwest, based on geochemical trends (Arbogast and Muhs,

2000). Thus, dune-forming winds shifted at least 90



from

their mid-Holocene direction to their present direction in the

late Holocene. These data suggest a more zonal circulation pat-

tern in the central Great Plains during the mid-Holocene than

occurs today.

Loess

Loess, or windblown silt, is one of the most extensive surficial

deposits on the Earth and may occupy as much as 10% of the

land’s surface, worldwide. Loess consists dominantly of parti-

cles between 50 and 2 microns, although some sand (>50

microns) and clay (<2 microns) are typically present. Unlike

eolian sand, loess does not form distinctive landforms. Com-

monly, loess forms a blanket over a pre-existing landscape

without much geomorphic expression. Loess ranges in thick-

ness from a few decimeters to many tens of meters. It covers

extensive areas of central North America, Alaska, Europe,

Russia, Central Asia, China, Argentina and New Zealand and

is the parent material for some of the world ’s most fertile soils.

Extensive areas of loess occur near regions of active or past

glaciation or in semiarid regions where there are abundant

supplies of silt-rich sediments.

Loess has several distinctive characteristics that are impor-

tant in paleoclimatology. Because much loess is glaciogenic,

Figure E10 Stratigraphy and uncalibrated radiocarbon ages of late Holocene eolian sands from sections in the Brandon Dune Field, southern

Manitoba, Canada. Episodes of eolian sand deposition represent relatively arid periods in the region, whereas paleosols represent less-arid periods

when vegetation stabilized the dunes (redrawn from Wolfe et al., 2000).

314 EOLIAN SEDIMENTS AND PROCESSES