Odekon M. Encyclopedia of paleoclimatology and ancient environments

Подождите немного. Документ загружается.

Retallack, G.J., 2001. Soils of the Past: An Introduction to Paleopedology,

2nd edn. London, UK: Blackwell Science Ltd, 404pp.

Sheldon, N.D., Retallack, G.J., and Tanaka, S., 2002. Geochemical

climofunctions from North American soils and application to paleosols

across the Eocene-Oligocene boundary in Oregon. J. Geol., 110,

687–696.

Sheldon, N.D., 2005. Do red beds indicate paleoclimatic conditions?: A

Permian case study. Paleogeography, Palaeoclimatology, Palaeogeogra-

phy, 228, 305–319.

Walker, T.R., 1967. Formation of red beds in modern and ancient deserts.

Bull. Geol. Soc. Am., 78, 353–368.

Yapp, C.J., and Poths, H., 1992. Ancient atmospheric CO

pressures

inferred from natural goethites. Nature, 355, 342–344.

Cross-references

Arid Climates and Indicators

Atmospheric Evolution, Earth

Banded Iron Formations and the Early Atmosphere

Continental Sediments

Evaporites

Laterite

Mineral Indicators of Past Climates

Paleoclimate Proxies, an Introduction

Paleosols, Pre-Quaternary

Paleosols, Quaternary

Paleo-Precipitation Indicators

Sedimentary Indicators of Climate Change

ROCHES MOUTONNE

Roches moutonnées are classic landforms of glacial erosion

(Bennett and Glasser, 1996). They are giant stoss and lee forms

carved from bedrock, with smoothly abraded up-ice or stoss

faces and roughly quarried/plucked down-ice lee faces. They

can occur as isolated forms, can be superimposed on larger

asymmetric landforms, or occur as part of erosional landscapes

subject to regional scour. Individual forms vary in size from

less than 1 m to several hundred meters or even kilometers in

diameter (Rastas and Seppälä, 1981). In Sweden, for example,

large asymmetric hills (Flyggbergs) exceed 1 kilometer in

length; however, most roches moutonnées are much smaller.

Roches moutonnées are the product of both glacial abrasion

and glacial quarrying and it is this combination of processes

that results in the asymmetric profile. Glacial abrasion occurs

on the up-ice or stoss face of the landform, while glacial

quarrying occurs on the lee. Glacial quarrying involves two

processes, firstly the exploitation and propagation of rock frac-

tures such as joints and discontinuities, and secondly, the

entrainment and removal of bedrock blocks by ice. The charac-

teristics of the rock mass, therefore, are critical to quarrying:

the more fractured the rock mass, the more easily it can be

plucked by ice. In turn, this means that the morphology of

roches moutonnées is closely linked to the depth, orientation

and spacing of joints or discontinuities within the rock mass

(Gordon, 1981; Rastas and Seppälä, 1981). The exploitation

and propagation of joints is assisted by fluctuations in subgla-

cial water pressure (Röthlisberger and Iken, 1981; Iverson,

1991). In addition, the formation of a lee-side cavity between

the ice sole and the bed increases the stress gradients imposed

on the bedrock hummock by the overriding ice, which again

favors fracture propagation. Block entrainment may occur by

a variety of processes, not least of which is the regular closure

of the lee-cavity, causing the debris within it to be swept clear.

Therefore, roches moutonnées are indicative of subglacial con-

ditions in which there is high effective normal pressure on the

stoss side of a bedrock bump, but where this pressure is suffi-

ciently low on the down-ice side to allow a cavity to form.

Consequently, the up-ice side experiences glacial abrasion,

while the down-ice side experiences glacial quarrying. Roches

moutonnées tend, therefore, to form in areas of thin, fast-

flowing ice and in situations with abundant meltwater. They

may also be used as crude indicators of the direction of ice flow.

Matthew R. Bennett

Bibliography

Bennett, M.R., and Glasser, N.F., 1996. Glacial Geology: Ice sheets and

Landforms. Chichester, UK:Wiley, 364pp.

Gordon, J.E., 1981. Ice-scoured topography and its relationship to bedrock

structure and ice movement in parts of northern Scotland and West

Greenland. Geografiska Annaler, 63A,55–65.

Iverson, N.R., 1991. Potential effects of subglacial water pressure fluctua-

tions on quarrying. J. Glaciol., 37,27–36.

Rastas, J., and Seppäla, M., 1981. Rock jointing and abrasion forms on

roches moutonnées, SW Finland. Ann. Glaciol., 2, 159–163.

Rothlisbeger, H., and Iken, A., 1981. Plucking as an effect of water-

pressure variations at the glacier bed. Ann. Glaciology, 2,57–62.

Cross-reference

Glacial Geomorphology

ROCHES MOUTONNE

ES 873

SAPROPELS

The term sapropel is used in a generic sense to describe fine-

grained and unconsolidated sediments rich in organic matter

that have been deposited under water; in coal petrology,

sapropelic coals are those that formed from algal material

deposited in aquatic environments. The term sapropel is also

often used as a generic term for distinct dark-colored, organic

carbon-enriched sediment layers that are found in organic-

carbon-poor host sediments. In the Holocene sediment record

of several silled basins (e.g., the Baltic Sea, the Black Sea,

the Marmara Sea) an organic carbon-enriched sediment layer –

often termed sapropel – marks the change in depositional

environment and in biological production coincident with the

post-glacial transgression.

In a stricter sense, the term stands for a sedimentary feature

of the Mediterranean (and Red) Sea with specific connotations

regarding the origin: sediment cores and land exposures from

the Mediterranean Sea contain rhythmic series of sapropel

layers (cm to m thick) sandwiched in carbonate-rich and

organic carbon-poor hemipelagic sediments. These sapropels

are rich in organic carbon (up to 30% by weight), devoid of

benthic organisms and often laminated, and were deposited

during periods when the deep waters of the marine basins

were anoxic (Comas et al., 1996; Kidd et al., 1978; Emeis

et al., 2000). The duration of the numerous anoxic events in

the Mediterranean Sea was from a few hundred up to 7,000

years, and the most recent sapropel was deposited between

10,000 and 7,000 years ago. The observed temporal pattern

of sapropel deposition is clearly paced by the orbital rhy-

thms of precession, tilt and eccentricity and implies a causal

link with orbitally-induced climate change that resulted in

changes of the Mediterranean water balance (Cita et al., 1977;

Rossignol-Strick, 1983). The link between insolation and sapro-

pel deposition in the eastern Mediterranean Sea is very robust

for the last 5.33 million years and the astronomically-tuned

stratigraphy established here has set a standard for Pliocene

and Quaternary chronostratigraphy world-wide (Hilgen, 1991).

The close link between climate and sapropel formation

is firmly established, and a hydrological trigger for sapropel

formation is widely agreed. Sapropel periods were preceded

by and coincided with changes in surface and deep water circu-

lation that restricted the oxygen supply to the seafloor in the

deep basins of the Mediterranean Sea. At present, the deep

basins of the Mediterranean Sea are flushed by cold, oxygen-

rich waters that are generated during winters in the northern

sub-basins – the Gulf of Lyons, the Aegean, and the Adriatic

Seas. One prerequisite for anoxic conditions during periods of

sapropel formation was a decrease in surface water density

(relative to that of deepwater masses) that prevented sinking

of oxygenated surface water into the deep basins. Both freshen-

ing (indicated by decreasing ratios of stable oxygen isotopes in

the tests of planktonic foraminifers and changes in microfossil

assemblages) and increasing sea surface temperature (based

on evaluation of faunal assemblages and biomarkers) of surface

water preceded and accompanied all sapropel events. Because

oxygen demand for the mineralization of organic matter that

sank from the surface ocean to the deepwater masses contin-

ued, the dissolved oxygen in the denser and aging deepwater

eventually was depleted. The origin of fresh waters is less clear.

The original hypothesis (Olausson, 1961) was that meltwater

from the northern catchment (via the Black Sea) prevented

deepwater formation. Alternatively, the northern catchment of

the eastern Mediterranean Sea may have received increased

precipitation during maximum insolation, leading to decreased

salinity in the deepwater formation areas of the Mediterranean

Sea (Rohling and Gieskes, 1989). However, since approxi-

mately 600,000 years ago, sapropels were deposited under fully

interglacial, fully glacial, and intermediate climatic conditions.

Because conditions during glacials or cold stadials were

as unfavorable for runoff from the northern catchment as

for water-column stratification due to warming of the surface

layer, sapropel formation even under glacial conditions is

a strong argument for a source of freshwater in the monsoon

system that influences mainly the southern catchment. Accord-

ing to this hypothesis, sapropel formation would be closely

linked to tropical and even Southern Hemisphere climate.

A systematic correlation has been observed between the

distribution of sapropels and maxima of the so-called orbital

insolation monsoon index (Rossignol-Strick, 1983). Maxima

in the monsoon index point to an intensified African summer

monsoon, which may have led to enhanced continental humid-

ity in tropical Africa and ultimately enhanced discharge rates of

the Nile River. Alternatively, enhanced northward migration of

the Intertropical Convergence Zone (the meteorological equator)

into Northern Africa has been proposed for sapropel periods; this

would have channeled equatorial rainfall into the Mediterranean

Sea through what is now the Sahara Desert. A further possible

source of freshwater is the Mediterranean Sea itself: virtually

all sapropels coincide with significant warming of surface waters

at the transition from cold to warm climatic periods. This warm-

ing would on the one hand enhance the stratification of water

masses and would on the other hand result in increased evapora-

tion and rainfall in the Mediterranean catchment, effectively

pooling fresh and warm waters at the sea surface and impeding

deepwater formation. It is very likely that all of these sources

contributed in succession and in concert to prevent deep convec-

tion over several thousands of years.

The immediate reason for the enrichment of organic matter

in sapropels is very much open to discussion. Oxygen condi-

tions at the seafloor, biological productivity at the sea surface,

and dilution by other sedimentary components are the primary

factors controlling concentration and accumulation of organic

carbon in marine sediments. Because the rates of sedimenta-

tion did not change significantly, the scientific discussion

focused on the roles of bottom-water anoxia and biological

productivity changes in the formation of organic carbon-rich

sediments (“productivity versus preservation”). Much of the

data gathered support the hypothesis of anoxic conditions

paired with (or possibly accelerated and in part caused by) an

increase in biological productivity during sapropel formation.

In line with this, many indicators suggest enhanced pro-

ductivity (assemblages of planktonic and – before and after the

sapropel events – benthic foraminifers, the accumulation rates

of barium). However, other authors see evidence that the eastern

Mediterranean Sea remained nutrient-limited (oligotrophic) and

that productivity patterns either shifted to a different type,

or were as low as today. A severe obstacle for environmental

reconstructions here as elsewhere is diagenesis that erases many

indicators after deposition and re-oxygenation of deepwaters.

Detailed investigation of diatom assemblages in one unusually

thick sapropel in which biogenic silica was preserved (in the

carbonate oozes and in almost all other sapropels, this opal is

lost to dissolution) showed that the original sapropel sequence

was rich in diatoms that formed a pronounced sequence of

laminations, including possible annual bloom layers. The flora

consisted of fragile mats of diatoms indicative of stratified,

oligotrophic conditions found in association with hydrographic

frontal systems. Collecting near the interface between upper

and lower water mass in the frontal system, the disintegration

of slowly grown mats with large biomass may sporadically

have resulted in large flux rates of opal and organic matter to

the seafloor (Kemp et al., 1999). Other authors (Sachs and

Repeta, 1999) proposed that the productivity level remained

similar to the present ultra-oligotrophic nutrient desert during

sapropel formation, because cyanobacterial nitrogen fixation

then as today apparently was a significant source of the essential

plant nutrient nitrate in the water column. In this scenario, anoxia

alone was responsible for greatly improved preservation of the

sparse rain of organic matter to the seafloor. More recently,

the role of phosphate liberated from anoxic sediments as a fuel

for enhanced fixation of atmospheric nitrogen by cyanobacteria

has been stressed (Struck et al., 2001; Slomp et al., 2002). In this

view, the eastern Mediterranean Sea was fertilized by phosphate

diffusing out of suboxic and anoxic sediments. This additional

phosphate may have been transported to the surface ocean by

ongoing shallow seasonal convection. It may thus have created

conditions conducive for massive nitrogen fixation from the

atmosphere that added nitrate, and initiated a positive feedback

between anoxia at the seafloor and biological productivity in

the euphotic zone. The anoxic and fertile sapropel periods ended

with the waning of freshwater inputs and warming, and renewed

flushing of the deep basins with oxygenated waters. In many

cases, the sapropel preceding this flushing was erased by oxygen

penetrating into the sediment, burning down the organic matter,

and leaving so-called ghost sapropels.

Conclusion

Formation of the Mediterranean sapropels is the result of

a combination of factors that include paleoclimate, paleocea-

nography, and the reactions of marine biogeochemical cycles

to changed boundary conditions. Whereas the mechanism of

sedimentation and carbon burial can be linked to changes

within the depositional environment, these very changes are

likely to be forced by processes outside the basin. External for-

cing is evident from the synchroneity of sapropel deposition i n

the entire (eastern) Mediterranean. Moreo ver, the rhythmi c

recurrence of environmental conditions that produce sapropels

on schedules of precessional cycles points to a climatically dri-

ven chain of events culminating in sapropel deposition. A ten-

tative chain of events postulates that carbon burial may have

been promoted by enhanced rates of marine productivity,

fuelled by an as ye t contested source of new nutrients, a corre-

sponding increase in oxygen utilization in deep water, and a

weakening of deepwater oxygen recharge which all occur at

the same time and are driven by the same external motor.

Kay-Christian Emeis

Bibliography

Cita, M.B., et al., 1977. Paleoclimatic record of a long deep-sea core from

the eastern Mediterranean. Quaternary Res., 8, 205–235.

Comas, M.C., Zahn, R., Klaus, A., et al., (eds), 1996. Proceedings ODP,

Initial Results, Vol. 161. College Station, TX: Ocean Drilling

Program, 1023 pp.

Emeis, K.-C., Sakamoto, T., Wehausen, R., and Brumsack, H.-J., 2000. The

sapropel record of the Eastern Mediterranean Sea – Results of Ocean

Drilling Program Leg 160. Palaeogeogr. Palaeoclimatol. Palaeoecol.,

158, 259–280.

Hilgen, F.J., 1991. Astronomical calibration of Gauss to Matuyama

sapropels in the Mediterranean and implication for the Geomagnetic

Polarity Time Scale. Earth Planet. Sci. Lett., 107, 226–244.

Kemp, A.E., Pearce, R.B., Koizumi, I., Pike, J., and Rance, J., 1999.

The role of mat-forming diatoms in the formation of Mediterranean

sapropels. Nature, 398,57–61.

Kidd, R.B., Cita, M.B., and Ryan, W.B.F., 1978. Stratigraphy of eastern

Mediterranean sapropel sequences recovered during Leg 42A and their

paleoenvironmental significance. In Hsü K., and Montadert, J.L. (eds.),

Initial Reports of DSDP. Washington, US: Govt. Printing Office,

pp. 421–443.

Olausson, E., 1961. Studies in deep-sea cores. Rep. Swed. Deep Sea

Exped., 8, 337–391.

Rohling, E.J., and Gieskes, W.W.C., 1989. Late Quaternary changes

in Mediterranean intermediate water density and formation rate.

Paleoceanography, 4(5), 531–545.

Rossignol-Strick, M., 1983. African monsoons, an immediate climate

response to orbital insolation. Nature, 304,46–49.

876 SAPROPELS

Sachs, J.P., and Repeta, D.J., 1999. Oligotrophy and nitrogen fixation dur-

ing Eastern Mediterranean sapropel events. Science, 286, 2485–2488.

Slomp, C.P., Thomson, J., and deLange, G.J., 2002. Enhanced regeneration

of phosphorus during formation of the most recent eastern Mediterra-

nean sapropel (S1). Geochim. Cosmochim. Acta, 66(7), 1171–1184.

Struck, U., Emeis, K.-C., Rau, G.H., Voss, M., and Krom, M., 2001.

Biological productivity during sapropel S5 formation in the eastern

Mediterranean Sea – Evidence from stable isotopes of nitrogen and

carbon. Geochim. Cosmochim. Acta, 65(19), 3241–3258.

Cross-references

Coal Beds, Origin and Climate

Diatoms

Eccentricity

Intertr opical Convergence Zone (ITCZ) (see Encyclopedia of World

Climatology)

Monsoons, Quaternary

Obliquity

Ocean Anoxic Events

Precession, Climatic

SCANDINAVIAN ICE SHEET

The Scandinavian and adjacent ice sheets

The Scandinavian Ice Sheet is the most commonly used name

for an ice sheet that developed over Scandinavia many times

during the Quaternary period. It is sometimes called the

Fennoscandian Ice Sheet (Fenno = Finland), but as it expanded

from the Scandinavian mountains and did not always reach

Finland, the Scandinavian Ice Sheet is the most proper name.

At its maximum extent, it covered northwestern Europe as

far south as the Netherlands (Figure S1). In Eastern Europe it

almost reached the Black Sea in southern Ukraine.

During some periods of maximum development, the

Scandinavian Ice Sheet coalesced with the Barents-Kara Ice

Sheet, which was centered over the Barents and Kara Seas,

and expanded southwards onto Northern Russia and Siberia

(Figure S1). At such times, there was a contiguous ice sheet

stretching from north of Spitsbergen to the Netherlands and

from the Norwegian Sea to the Taimyr Peninsula in Siberia.

The Scandinavian Ice Sheet also occasionally met the British

Ice Sheet in the North Sea, but these two ice sheets were

topographically and dynamically more clearly separated than

those discussed above.

The older Scandinavian Ice Sheets

The strong glacial erosion associated with the last ice sheets

has removed most of the deposits of the older glacials and

interglacials. Glacial erosion also transformed the landscape

so that deep U-shaped valleys and fjords now are common,

especially in western Norway where vertical erosion of 1,900

m is inferred for part of Sognefjorden (Andersen and Nesje,

1992). The main record of older glaciations is the ice-rafted

debris (IRD) found in cores collected from the floor of the deep

Norwegian Sea. IRD indicates that the first ice sheets formed

in Scandinavia as early as 11 million years ago, whereas the

largest ice sheets grew after about 2.7 million years ago (Jansen

and Sjøholm, 1991; Mangerud et al., 1996).

It is not known how many times the ice sheet expanded

south of Scandinavia. In Germany, there are tills from only

three glaciations, the Elsterian, Saalian and Weichselian (Ehlers,

1996). Thus, the classical view is that only at these times did the

ice sheet reach this far south. However, Scandinavian erratics

found in the Netherlands and glacial deposits in the North Sea

indicate that such expansion also happened before the Elsterian

(Mangerud et al., 1996).

The last ice age – the Weichselian

The last ice age in Northern Europe is named the Weichselian

from the German name of the Vistula River in Poland,

where glacial deposits from the Scandinavian Ice Sheet were

Figure S1 Map showing the extent of the Scandinavian, Barents-Kara and British Ice Sheets during two periods. Full line shows the Saalian

glaciation and stippled line the Late Weichselian glaciation. (Modified from Svendsen et al., 2004.)

SCANDINAVIAN ICE SHEET 877

described by German geologists shortly after 1900 (Ehlers,

1996). We now know that these deposits were formed late dur-

ing the last ice age, about 20,000 years ago. However, the name

Weichselian is accepted for the entire cold period following

the last interglacial in Europe, the Eemian. The subdivision

and time scale for the Weichselian is shown in Figure S2.

The growth and decay of the Scandinavian and adjacent

ice sheets is obviously much better known for the Weichselian

than for older glaciations (Ehlers and Gibbard, 2004). How-

ever, in most areas, the ice sheets were larger during the

Late Weichselian (about 20,000 years ago) than during earlier

phases of the Weichselian. Therefore, our knowledge of the

first 90,000 years of the Weichselian is limited.

Figure S2 shows a glaciation curve for western Norway

for the entire Weichselian (Mangerud, 2004). During the

last interglacial, the Eemian, the climate was warmer than at

present, and no glacier existed. The first ice advance (Gulstein)

almost reached the coast as early as 110,000 years ago.

Subsequently the ice front retreated during the Brørup intersta-

dial when open forests were established over much of Scandi-

navia (Lundqvist, 1992). As seen from the curve, the ice

margin subsequently advanced and retreated several times dur-

ing the Weichselian. There were probably several more fluctua-

tions than so far identified, although the record of ice rafted

detritus in cores from the Norwegian Sea supports the main

picture (Figure S2).

It is worth noting that the coast and probably much of the

interior of Scandinavia was ice free during and possibly even

after the Ålesund interstadial, which ended about 34,000 years

ago (Olsen et al., 2002; Mangerud et al., 2003). This means

that the ice sheet grew rapidly after this interstadial, until it

reached its maximum size about 24,000 years ago.

The extent of the Late Weichselian Ice Sheet is shown in

Figure S1. In Denmark, Germany and Poland, the limit of this

ice sheet has been known for more than a century (Ehlers,

1996). However, it is still debated whether the British and

Figure S2 The right hand curve shows fluctuations of the western margin of the Scandinavian Ice Sheet. The lower part of the curve, when the

ice front ended at the margin of the continental shelf corresponds with the Saalian ice limit in Figure S1. During the Eemian, the ice sheet

disappeared. During the Weichselian, the extent of the ice sheet fluctuated strongly, but about 24,000 years ago it again reached the margin of the

continental shelf, as shown in Figure S1. To the left are curves showing the number of ice rafted sand grains and their accumulation rates in

cores from the Norwegian Sea (Mangerud, 2004).

878 SCANDINAVIAN ICE SHEET

Scandinavian Ice Sheets met in the North Sea, although that is

assumed on the map. Furthermore, the limit of the ice sheet in

Northern Russia has been much debated. A serious problem

there is that permafrost with buried glacial ice melted at the

end of the Weichselian and created fresh glacial landscapes,

even though it had been deglaciated for tens of thousands of

years (Svendsen et al., 2004).

The Barents-Kara Ice Sheet

The Barents Sea is located immediately north of Scandinavia,

but the Scandinavian and Barents-Kara Ice Sheets did not

always react in concert. The Barents-Kara Ice Sheet was much

larger during the Early (about 90,000 years ago) and Middle

Weichselian (60,000– 50,000 years ago) than during the Late

Weichselian, in contrast to the Scandinavian and the North

American Ice Sheets (Svendsen et al., 2004). During both these

periods, the Barents-Kara Ice Sheet moved onto land in

Northern Russia, whereas during the Late Weichselian it termi-

nated in the sea ( Figure S1). During the Early Weichselian,

the northbound rivers in Northern Russia were blocked by the

ice sheet and large ice-dammed lakes formed between the

ice-sheet and the continental watershed to the south.

Jan Mangerud

Bibliography

Andersen, B.G., and Nesje, A., 1992. Quantification of Late Cenozoic gla-

cial erosion in a fjord landscape. Sveriges Geologiska Undersökning Sr.

81,15–20.

Ehlers, J., 1996. “Quaternary and glacial geology.” Chichester, UK:

Wiley.

Ehlers, J., and Gibbard, P. (eds.), 2004. “Quaternary Glaciations – Extent

and Chronology. Part 1: Europe.” Amsterdam: Elsevier.

Jansen, E., and Sjøholm, J., 1991. Reconstruction of glaciation over

the past 6 million years from ice-borne deposits in the Norwegian

Sea. Nature, 349, 600–604.

Lundqvist, J., 1992. Glacial Stratigraphy in Sweden. Geological Survey of

Finland Special paper 15, pp. 43–59.

Mangerud, J., 2004. Ice sheet limits on Norway and the Norwegian continen-

talshelf.InEhlers,J.,andGibbard,P.(eds.),“Quaternary Glaciations –

Extent and Chronology. Part 1: Europe.” Amsterdam: Elsevier .

Mangerud, J., Jansen, E., and Landvik, J., 1996. Late Cenozoic history of

the Scandinavian and Barents Sea ice sheets. Glob. Planet. Change,

12,11–26.

Mangerud, J., Løvlie, R., Gulliksen, S., Hufthammer, A.-K., Larsen, E.,

and Valen, V., 2003. Paleomagnetic correlations between Scandinavian

Ice-Sheet fluctuations and Greenland Dansgaard-Oeschger Events,

45,000–25,000 y

BP. Quaternary Res., 59,211–220.

Olsen, L., Sveian, H., van der Borg, K., Bergstrøm, B., and Broekmans, M.,

2002. Rapid and rhythmic ice sheet fluctuations in western Scandinavia

15–40 Kya – areview.Polar Res., 21,235–242.

Svendsen, J., Alexanderson, H., Astakhov, V., Demidov, I., Dowdeswell, J.,

Funder, S., Gataullin, V., Henriksen, M., Hjort, C., Houmark-Nielsen, M.,

Hubberten, H., Ingólfson, O., Jakobsson, M., Kjær, K., Larsen, E.,

Lokrantz, H., Lunkka, J., Lyså, A., Mangerud, J., Matiouchkov, A.,

Murray, A., Möller, P., Niessen, F., Nikolskaya, O., Polyak, P.,

Saarnisto, M., Siegert, C., Siegert, M., Spielhagen, R., and Stein, R.,

2004. Late Quaternary ice sheet history of Northern Eurasia. Quaternary

Sci. Rev., 23,1229–1271.

Cross-references

Eemian (Sangamonian) Interglacial

Glaciations, Quaternary

Ice-Rafted Debris (IRD)

Interstadials

Last Glacial Maximum

Wisconsinan (Weichselian, Würm) Glaciation

SEA LEVEL CHANGE, LAST 250 MILLION YEARS

Introduction

One of the most fundamental geological observations has been

that the seas once inundated large areas of the Earth’s surface

that are now exposed on land. Early observers attributed these

marine incursions to one or multiple “Noachian” floods, but

empirical science prevailed as uniformitarian concepts were

applied to the stratigraphic record. Charles Lyell (1830) recog-

nized that changes in sea level due to volcanic activity were

responsible for the submergence and subsequent emergence

of the temple at Serapis near Naples, Italy (see discussion by

Gould, 1999). Observing cycles of shallow-water sedimenta-

tion across Europe, Lyell reasoned that apparent rises in sea

level could explain the flooding of the continents (geologists

call these “transgressions”) and the subsequent retreat of the

seas (called “regressions”). For over a century after Lyell’s

work, geologists mapped these advances and retreats of the

sea, noting that during intervals such as the middle Cretaceous

(ca. 80 My

BP), much of the continents were inundated, whereas

at other times, such as today, sea level was much lower. Geol-

ogists have equated these transgressions and regressions with

global sea level (or “eustatic”) changes.

Transgressions and regressions may be related to eustatic

sea level changes, but they also are caused by processes of sub-

sidence or uplift, and changes in sediment supply. These pro-

cesses are revealed by 20th century tide gauge data that

indicate a global sea-level rise of 1.5–2mmyr

1

(Church

et al., 2001), due to the effects of ocean warming and melting

of glaciers. However, tide gauge data for the mid-Atlantic Uni-

ted States uniformly show a rise of greater than 3 mm yr

1

,in

part due to regional subsidence (Psuty and Collins, 1986).

Although sea level is rising globally, relative sea level (in this

region a term encompassing both the effects of subsidence/

uplift and eustatic change) is rising much faster. High sediment

supply can cause regression even during a relative sea level

rise. For example, relative sea level is rising rapidly in the Mis-

sissippi Delta region due to the effects of global sea-level rise

and rapid regional subsidence. Consequently, this region is gen-

erally experiencing a rapid transgression. However, near the

mouth of the Mississippi, the high supply of sediments results

in regression as the delta builds upward and into the Gulf. These

modern examples illustrate the differences between a eustatic

rise, a relative rise in sea level, and a marine transgression.

Global sea level has risen and fallen many times in response

to growth and decay of the Northern Hemisphere Ice Sheets

that have dominated the last 2.5 Myr (“the Ice Ages”). Drilling

of submerged reef terrace records off Barbados (Figure S3)

and Tahiti provided Rosetta Stone, which showed a 120 m

lower sea level than present during the Last Glacial Maximum

(Fairbanks, 1989; Bard et al., 1996). Uplifted reef terrace

records (corrected for long-term uplift) have extended global

sea-level estimates back to 130,000 years ago (Fairbanks and

Matthews, 1978; Chappell et al., 1996); however, it has proven

difficult to firmly extend reef terrace records further back in

time due to dating problems.

Pre-Quaternary sea level change

Oxygen isotope ratios (d

O) provide a potential means for

reconstructing sea-level change over the past 100 Myr. d

SEA LEVEL CHANGE, LAST 250 MILLION YEARS 879

values reflect the effects of temperature and ice-volume (hence

sea level) changes in the shells of single-celled organisms

called foraminifera. By calibrating foraminiferal d

O varia-

tions to the reef terrace record, geologists can make an estimate

of global sea-level changes ( Figure S3). However, it is difficult

to extend oxygen isotopic sea-level estimates beyond the past

million years because the temperature history of the planet

prior to this is less well known. In addition, diagenetic altera-

tion potentially overprints Late Cretaceous to early Tertiary

O values (ca. 100–25 My BP), whereas records older than

100 Myr are typically altered.

Our primary information on sea-level change prior to the

Quaternary ( >2My

BP) is derived from stratigraphic records

of continental inundation. Transgressions have flooded the con-

tinents on roughly 100-Myr timescales (e.g., Sloss, 1963),

reflecting changes in the rates of ocean crust production.

Finer-scale cycles (1–10 Myr) have been attributed either to

sea-level change (Suess, 1885) or to tectonic controls (Sloss,

1963; Stille, 1924; Grabau, 1936). These cycles are observed

in the sediment and rock record with deeper water environ-

ments r e p r e s en ti n g t r an s g r e s s i o n s a n d s h a l l o w e r w a te r

environm en ts representin g regressions.

Studies at Exxon Production Research Company (EPR)

revolutionized our view of the stratigraphic record of sea-level

change. Sea level falls and tectonic changes both produce ero-

sion surfaces termed “unconformities” ; packages of sediments

on continental margins bracketed by unconformities are called

“sequences. ” Researchers at EPR made a revolutionary break-

through in using seismic reflection profiles to identify

sequences and then using these sequences to estimate the ages

and magnitudes of past sea level changes ( Figure S4; Haq et al.,

1987; Vail et al., 1977). These estimates proved highly contro-

versial, in part because of the proprietary nature of the data

used to construct them and in part because of flaws in the

method used to estimate the amplitudes of sea-level change.

Nevertheless, more than a decade of studies, described below,

have been carried out, validating the general number and tim-

ing of eustatic events published by EPR – a landmark achieve-

ment. However, estimation of eustatic amplitudes remains one

of the thorniest problems in geology.

A large mid-Oligocene (ca. 30–32 My

BP) sea level lowering

illustrates problems with estimating amplitudes. Vail et al.

( 1977) initially concluded that this event comprised over

400 m of global sea level lowering, a remarkably large drop.

Geophysical studies showed that the methodology employed

in these estimates did not properly account for the effects of

basin subsidence and water depth and thus grossly overesti-

mated the amplitudes (see summary in Christie-Blick et al.,

1990). A further decade of study by EPR scientists provided

an estimate of 160 m of lowering for this mid-Oligocene

event (Figure S5), an amplitude that still strained credulity.

Stable isotopic studies have long suggested amplitudes closer

to 30–60 m for this event (e.g., Miller et al., 1985, 1998). Rig-

orous studies of borehole transects on the New Jersey margin

yield an estimate for this event of 50 m (Kominz and Pekar,

2001) using a method called backstripping (see below for defi-

nition), and this estimate is probably correct within  10 m.

Despite these recent studies, the amplitude of sea-level change

remains one of the poorest constrained boundary conditions of

the Earth.

Recognizing that it had the unique ability to provide a glo-

bal dataset, the international Ocean Drilling Program (ODP),

operator of the drillship JOIDES Resolution, entered into sea-

level research in the 1990s. ODP identified four major goals

in sea-level research: (a) test the synchrony of sea-level events

in widely separated locations; (b) estimate the amplitude of sea-

level changes; (c) evaluate various models that seek to explain

the stratigraphic response to sea-level oscillations; and (d)

determine the mechanisms that control sea level. To accomplish

these goals, a strategy of drilling transects of boreholes on pas-

sive margins was adopted.

The New Jersey sea level transect

The New Jersey margin was selected by ODP as an ideal

location to begin the investigation of the Late Cretaceous to

Cenozoic history of sea-level change because of its rapid

sedimentation, tectonic stability, good chronostratigraphic

control, and abundant seismic well log and borehole data

(Miller and Mountain, 1994). To evaluate sequences and sea-

level changes, the “New Jersey Sea-Level Transect” was

designed as a series of boreholes from the onshore New Jersey

Coastal Plain across the continental shelf to the slope and

rise (Figure S5). The transect initially focused on the past

34 Myr (Oligocene-Recent; Miller and Mountain, 1996), a

time when large ice sheets waxed and waned, potentially

changing sea level by up to 100 m. More recent studies

Figure S3 Comparison of the global estimate for sea-level derived from Barbados terraces (Fairbanks, 1989) with deep-sea benthic foraminiferal

O values from Pacific core V19-30 scaled to sea-level assuming 0.11% /10 m of sea-level change (Fairbanks and Matthews, 1978) (Wright, Miller,

Sheridan, unpublished).

880 SEA LEVEL CHANGE, LAST 250 MILLION YEARS

Figure S4 Cenozoic portion of the Haq et al. (1987) sea-level record.

SEA LEVEL CHANGE, LAST 250 MILLION YEARS 881

have extended back to 100 MyBP (Late Cretaceous; Miller

et al., 2003).

Seismic profiles provide a means of imaging strata like a

sonogram images structures within the human body. Seismic

profiles exhibit stratigraphic geometries resulting from erosion,

potentially due to sea-level lowerings (“sequence boundaries”).

The seismic surfaces reveal themselves in borehole sediments

as erosion surfaces. The New Jersey Transect dated these

sequence boundaries by drilling in relatively deep water

(400–1,400 m) on the continental slope and rise and using

various tools for age control (fossils, magnetostratigraphy,

Sr-isotopic stratigraphy). Drilling onshore in New Jersey also

dated these sequences in a more proximal nearshore to

paleo-shelf setting, where time gaps or hiatuses were longer.

Nevertheless, the onshore hiatuses (white gaps between blue

boxes; Figure S6) correlate remarkably well with offshore

sequence boundaries (m1, m2, etc. on Figure S6). This estab-

lishes a regional correlation between these hiatuses and

associated sequence boundaries, and a relation to at least

regional-scale sea-level lowerings.

Comparison of the New Jersey sequence stratigraphic

record with oxygen isotopes provides a link between regional

sea-level lowerings and global changes in ice volume (hence

sea level) for the past 34 Ma (Figure S6). Oxygen isotopic

O) increases in deep-sea Atlantic cores correlate with the

sea-level lowerings in New Jersey (Figure S6). This demon-

strates that a significant portion of oxygen isotopic changes

must be attributed to changes in ice volume. Still, we cannot

precisely determine the extent of the role of ice growth versus

cooling of bottom waters on any individual d

O increase. For

example, a large d

O increase occurred across the Oligocene/

Miocene boundary (ca. 23 Ma), correlating with the m6

continental slope and the O6/Kw0 onshore sequence boundaries

(Figure S6). If this were attributed entirely to ice volume (pre-

sumably from Antarctic ice, because Northern Hemisphere Ice

Sheets did not begin to grow until late Miocene to Pliocene), this

change would be equivalent to a 90 m eustatic sea level drop. It

is likely however, that bottom waters cooled at the same time as

the glaciation; during the Quaternary, about two-third of the d

signal is controlled by ice volume and one-third by temperature

(Fairbanks, 1989). Applying this rule of thumb to the older

record indicates about 60 m, not 90 m, of eustatic lowering;

however, it is still possible that the temperature effects were lar-

ger and sea level change would have been correspondingly less.

Figure S5 Map showing seismic profiles and boreholes, New Jersey sea-level transect.

882 SEA LEVEL CHANGE, LAST 250 MILLION YEARS

Figure S6 Comparison of offshore New Jersey, onshore New Jersey, global oxygen isotopes, ages of Bahamas reflections, and Haq et al. (1987) sea-level record (Miller et al., 1998).

SEA LEVEL CHANGE, LAST 250 MILLION YEARS 883