Boggs S. Principles of Sedimentology and Stratigraphy

Подождите немного. Документ загружается.

13.4 Magnetostratigraphy

467

scientists working independently and competitivelye group in northe Cal-

ifoia and one in Australia. The initial development of a magnetic polarity se-

quence by these groups of scientists is summarized by Cox (1973), Glen (1982),

McDouga (1977), and Watkins (1972). These scientists quickly realized the geo-

logic potential of magnetic reversals. If the absolute age of reversals could be es-

tablished, a quantitative time scale for reversals could be set up that would be

exemely useful for stratigraphic correlation and other purposes.

The first such polarity scales were achieved by measuring the ages and mag

netic polarities of young volcanic rocks on land by using potassium-argon (K-Ar)

techniques to estimate the ages of the rocks younger than about 5 million years

(Cox, Doell, and Dalrymple, 1963). During the 1960s, the geomagnetic time scale

evolved through numerous versions as new data points were added and age esti

mates were refined (McDougall, 1979). Figure 13.24 shows the version of the time

scale at had emerged by 1979. Note from Figure 13.24 that this polarity time

scale is subdivided into polarity "epochs," each named for a distinguished scien

tist who contributed to development of the field of geomagnetism; shorter

"events" were named for localities where definitive paleomagnetic characteristics

of specific groups of rocks has been studied, commonly localities where the rocks

were first sampled. We now understand that there is no fundamental distinction

between polarity epochs and polarity events and that polarity intervals

of a wide

spectrum of durations are possible (e.g., Butler, 1992, p. 210).

In addition to study of the polarity of volcanic rocks on land, a second very

important source of information about maetic reversal sequences is provided

by the linear anomaly patterns discovered in volcanic rocks of the ocean oor, par

ticularly along mid-oce ridges, and first interpreted by Vine and Matthews

(1963). (Magnetic anomalies are significant deviations from Earth's maetic

background on either a local or regional scale.) These linear "stripes" of normal

and reversed-polarity magnetic rocks (e.g., Fig. 13.25) are roughly parallel to ridge

crests and are typically 5-50 km wide and hundreds of kilometers long. These

anomalies were produced owg to reversals  Earth's maetic field as successive

e of oldest sediment

in contact with igneous

sement gives age of

seaoor & corresponding

anomaly

Sediment �

Igneous

base ment

Gauss

normal polarity

chron

Matuyama

reverse

polarity

chron

Mid

Uprising

magma

Brunhes

normal

polarity

chron

Fi 13.25

Schematic illustration of the

principles by which the

biostratigraphic age of sedi

ments (determined by

deep-sea drilling) overlying

igneous basement can be

used

to date a particular

marine magnetic anomaly.

[From Hailwood, E. A.,

1989, Magnetostratigraphy:

Geological Soc. Spec. Re

port 19, Blackwell Scientific

Publications. Fig. 17, p. 29,

reproduced by permission.]

468

Chapter 13 I Seismic, Sequence, and Magnetic Stratigraphy

Figure 13.26

Magnetic polarity time

scale for the Cenozoic-late

Mesozoic. Intervals of nor

mal polarity are shown in

black, reversed polarity in

white. Intervals lacking

magnetostratigraphic stud

ies or having uncertain va

lidity are cross-hachured.

[Compiled by Ogg, ). G.,

1995, Magnetic polarity

time scale of the Phanero

zoic, in Ahrens, T. j. (ed.),

Global Earth physics-A

handbook of physical con

stants: AGU Reference Shelf

1, American Geophysical

Union, Wa shington, D.C.,

Fig. 2, p. 252, reproduced

by permission.]

flows of lava erupted along ridge crests and cooled below the Curie point. Previ

ously magnetized volcanic rock was pushed or pulled aside from the ridges as new

volcanic rock formed and became magnetized. Vine and Matthews (1963) hypothe

sized that the linear magnetic anomaly pattes on the ocean floor correlate with

normal and reversed polarity intervals in the geomagnetic scale established on

land, allowing the ages of the anomalies to  estimated. The fact that the maetic

anomalies are roughly symmetrical about spreading dges was a critically impor

tant piece of evidence used in developing the concept of seafloor spreading.

The principal problem with the oceanic record is that it is very difficult to

date directly. Paleontologic ages on the oldest sediments overlying marine mag

netic anomalies are available where basement has been reached by Deep Sea

Drilling Program (DSDP) or Ocean Drilling Program (ODP) drill holes, but large

uncertainties are often associated with these age determinations. Figure 13.25 il

lustrates the method of dating seafloor anomalies by use of paleontologic data. By

making use of data from the ocean floor magnetic record, a detailed paleomagnetic

time scale dating back to the Jurassic has now been established (Fig. 13.26).

13.4 Magnetostratigraphy

469

detailed magnetic-polarity time scale for rocks older than the Jurassic has

been more difficult to establish because the continuous oceanic geomagnetic time

scale

cannot be extrapolated beyond the age of the oldest oceanic crust (about

160-170 m.y; older oceanic crust has been destroyed in subduction zones); howev

er, magnec reversals are known in terrestrial secons in rocks at least as old as 1.5

billion years (Conde, 1982). Although study of on-land stratigraphic sections from

various parts of the world has now provided many data on the paleomagnetic

cha racteristics of early Mesozoic and older rocks on land, our knowledge of the

polarity time scale for these rocks is much less refined than at for younger rocks.

Figures 13.27 and 13.28 show generalized polarity reversal patterns for the late

Mesozoic and Paleozoic; not all of these reversal pattes have been definitely ver

ified (Ogg, 1995). See also Gradstein et al. (1995).

Te rminology in Magnetostratigraphy

Because magnetostratigraphic polarity units are of primary interest in stratigra

phy, these units are formally subdivided into polarity superzones, polarity zones,

and polarity sub zones depending upon their duration (Table 13.3). The equivalent

geochronologic time units are referred to as chrons or superchrons. This terminol

ogy is recommended by the lUGS International Subcommission on Stratigraphic

Nomenclature and the IUGS/IAGA Subcommission on a Magnetic Polarity Time

Scale (see Salvador, 1994). The polari zone is the fundamental polarity unit for

subdivision of stratigraphic sections. A polarity zone may consist of strata with a

sgle direction of polarization throughout, may be composed of an intricate alter

nation of normal and reversed units, or may be dominantly either normal or re

versed but with minor subdivisions of the opposite polarity. A polari superzone

consists of two or more polarity zones, and a polari subzone is a subdivision of

a polarity zone. The North American Stratigraphic Commission follows approxi

mately the same scheme of nomenclature as that proposed by the JUGS (see the

North American Stratigraphic code in Appendix C). e principal polarity zone

names now in

use are the well-established names such as Brunhes, Matuyama,

Gauss, and Gilbert, which are used for the most recent 5 million years of Earth's

histor Historically, these units have been called epochs (Fig. 13.24); however, it is

now recommended that these "epochs" be called the Brunhes, Matuyama, Gauss,

and Gilbert polarity zones, or the Brunhes, Matuyama, Gauss, and Gilbert polari

ty chrons, if referring to time. Similarly, the so-called "events," such as the Jaramil

lo, Olduvai, and Reunion, should now be referred to as the Jaramillo, Olduvai,

and Reunion polarity subzones {in rocks) or subchrons (for time).

Applications of Magnetostratigraphy and Paleomagnetism

Coelaon

The primary application of magnetostratigraphy lies in its use as a tool for global

correlation of marine strata. Magnetostratigraphic correlation is particularly im

portant where paleontologic or lithologic correlation is difficult. It has special

significance

for international correlation because geomaetic reversals are

con

temporaneous, synchronous, worldwide phenomena. They have worldwide

scope owing to the fact that reversals of Earth's magnetic field affect the magnetic

eld everywhere on Earth at e same time. Because the polarity time scale can be

calibrated radiometrically or paleontologically, polarity events thus provide a pre

cise tool for chronostratigraphic (time) correlation. The first significant application

of magnetostratigraphic techniques to correlation and age determinaons of rocks

was correlation of linear ocean-floor magnetic anomalies to on-land sections of

volcanic strata whose ages had been determined by radiometric methods. These

correlation teciques were subsequently extended to cores of oceanic sediments.

470

Chapter 13 I Seismic, Sequence, and Magnetic Stratigraphy

Figure 13.27

Magnetic polarity time scale for

jurassic to Permian. Intervals of

normal polarity are shown in black,

reversed polarity in white. Intervals

lacking magnetostratigraphic stud

ies or having uncertain validity are

cross-hachured. [Compiled by

Ogg, ]. G., 1995, Magnetic polarity

time scale of the Phanerozoic, in

Ahrens, T. J. (ed.), Global Earth

physics

-A handbook of physical

constants: AGU Reference Shelf 1,

American Geophysical Union,

Washington, D.C., Fig. 3, p. 260,

reproduced by permission.]

Age Geologic

Polari

{Ma) Age

Chron

Age Geologic

(Ma)

Age

Until very recently, correlation of sediment cores by use of magnetic polaty

events had its greatest application in the study of marine sediments younger an

about 6 to 7 million years. Correlation was previously restricted to very young

rocks because the magnetic time scale had not been developed beyond about 7

Ma, and because most gravity and piston cores of ocean-oor sediment did not

penetrate deep enough to sample older sediments. As mentioned, the detailed

geomagnetic time scale has subsequently been extended to about 160-170 Ma.

13.4 Magnetostratigraphy

471

Furthermore, deeper coring by use of hydraulic piston cores now makes it possi

ble to obtain undisturbed cores of sediments as old as about middle Miocene.

Longer cores obtained during the Deep Sea Drilling Program and Ocean Drilling

ogram by rotary coring methods have recovered rocks as old as about middle

Jurassic; however, these rotary cores are commonly too badly disturbed to provide

unambiguous paleomagnetic data. Because paleomagnetic methods have now

been extended to coelation of on-land sections, this development opens up the

possibility of even more extensive future use of paleomagnetic methods for corre

lating on-land stratigraphic sections. Magnetostratigraphy thus becomes an im

portant tool for inteational correlation of older, on-land strata as the magnetic

polarity time scale is extended farther back into geologic time.

Figure 13.29 provides an easily visualized example of paleomagnetic correla

tion in cores of young oceanic sediment. Beginning with the Brunhes Normal

Epoch (polarity chron) at the top of the cores, the correlation can be carried down

ward on the basis of the patterns of reversed and normal polarity. With longer

Figure 13.28

Magnetic polarity time scale

for the Paleozoic. Intervals

of normal polarity are

shown in black, reversed

polarity in white. Intervals

lacking magnetostrati

graphic studies or having

uncertain validity are cross

hachured. Unverified Paleo

zoic polarity intervals from

the Soviet scale (Khramov

and Rodionov, 1981) are

shown as shades of gray.

[Compiled by Ogg, J.G.,

1995, Magnetic polarity

time scale of the Phanero

zoic, in Ahrens, T. J. (ed.),

Global Earth physics-A

handbook of physical con

stants: AGU Reference Shelf

1, American Geophysical

Union, Washington, D.C.,

Fig. 4, p. 265, reproduced

by permission.]







.··Ncla�



a�et



ap�c.�l •:

Magnetosatiaphic

polarity units

Polarity supzone

(106-107 years duraon)

Polarity zone

(105 -106 years duraon)

l'oJantv subzone

years

Description

Magnetostratigraphic unit composed

of two or more polarity zones

Magnetostratigraphic unit distin

guished by a single direction of

magnec polarization or by a

distinctive alteation of normal and

revised polarities

Subdivision of a polarity zone

Note: e Chapter 15 for an explanation of geoch ronologic and chr onostratigraphic nnits.

Geochronologic

(time

)

equivalent

Chron

(or superchron)

Chron

(or subchron)

Chronostratigraphic

equivalent

Chronozone

(or superchronozone)

Chronozone

( subchronozone)

Example

New name

Brunhes Normal Polarity

Zone or Brunhes Normal

Polarity Chron

Jaramillo

J'OJanry

subzone  Jaramillo

Polarity su bchron

Old name

Brunhes Normal

Epoch

Jaramillo Event

13.4 Magnetostratigraphy

473

ctic

Depth Ocn

(meters) T3-67 -6

North Central South North South Indian

Pacific 0. Pacific 0. Pacific 0. Atlantic 0. Atlantic 0. Ocean

V21-173 V24-58 C9·115 V22·230 V12-18 C12-327



Even

Age

(.y.)

cores and older sediment, correlation becomes more difficult because the magne·

tostratigraphic record consists of many sets of reversals (Figs. 13.24 and 13.26) that

may look very much alike. Correlation of these reversal patterns may require in

dependent radiometric or paleontologic age evidence to first establish stratigraph

ic position. Paleomagnetic reversal patterns are particularly useful for correlating

long distances across biogeographic boundaries where correlation bv fossils, even

planktonic fossils, may be difficult owing to the fact that different

iogeographic

provinces are marked by different fossil assemblages. Figure 13.29 shows that pale

omagnec coelation can be carried across the Arctic, Pacific, Indian, and Atlantic

ocean basins, each of which is characterized by sediments composed of different

lithologies with different fossil assemblages. In a similar manner, Figure 13.30 illus

trates how on-land stratigraphic sections can be correlated on the basis of magnet

ic polarity reversal stratigraphy. Additional examples of correlation of on-land

sections by means of magnetostratigraphy are given by Butler (1992), Aissaoui,

McNeill, and Hurley (1993), and Belkaaloul et a!. (1997).

Geochronolo



Although magnetostratigraphic sequence in itself does not normally provide un

equivocal ages for geologic events preserved in strata, correlation of magnetic po

larity zones or anomalies from areas where the ages of magnetic events have been

established by radiometric methods or paleontologic data to areas where the ages

of e strata are unknown, or poorly known, provides a mea of estimating the

ag of events in the new areas. Magnetostratigraphic geochronology may be par

ticularly useful in determining ages within stratigraphic successions that are so

nonfossiliferous that little biostratigraphic age control exists. For example, Heller

d Tu ngsheng (1984) used magnetostratigraphic methods to work out the chronol

 of nonfossiliferous Chinese loess (eolian) deposits.

Magnetostratigraphic chronometry can also provide absolute ages for sedi

mts that have been zoned by fossils and whose ages have been estimated from

fos

sil data, as described by Hailwood (1989, p. 51). For example, McFadden and

Hunt (1998) correlated magnetopolarity zones within the Oligocene-Miocene

Arikaree Group of northwestern Nebraska to the maetic polarity time scale to

tablish absolute ages of important land-mammal zones (Fig. 13.31). The ability

to correlate a local magnetic polarity zonation to the magnetic polarity time scale

quires either a unique pattern of magnetic reversals with one or more distinctive

polarity intervals, excellent biochronological data (Chapter 14), and/ or high

rolution radioisotope age determinations from interbedded datable rocks such

Figure 13.29

Paleomagnetic correlations

of cores from the Arctic, Pa

cific, Indian, and Atlantic

oceans. Cores have differ

ent lithologies and fossil as

semblages. [From Opdyke,

N., 1972, Paleomagnetism

of deep-sea cores: Reviews

Geophysics and Space

Physics, v. 10, Fig. 20, p.

244, American Geophysical

Union, Wa shington D.C.,

reproduced by permission.]

474 Chapter 13 I Seismic, Sequence, and Magnetic Stratigraphy

Figure 13.30

Correlation of Late Creta

ceous through Cenozoic

stratigraphic sections in the

Umbrian Apennines on the

basis of magnetostratigra

phy. The sections are also

correlated with the seafloor

magnetic anomaly se

quence (right column). Age

from biostratigraphic

(foraminiferal) zonation

(Miocene-Santonian) is

given at the left of the col

umn showing dominant

lithology. Magnetic anom

aly numbers and paleonto

logic calibration points

(shown by arrows) are

noted at the left side of the

seafloor polarity column.

[Redrawn and modified

from Lowrie, W., and W. Al

varez, 1981 , One hundred

million years of geomag

netic history: Geology, v. 9,

Fig. 1, p. 393.]

Miocene

400

Oligocene

300

Eocene

200 .

Paleocene

Maastrichtian

>00



Santonian

-- Magnetic

SECTIONS

 Contessa quarry

® Contessa road

@ Contessa highway

0 Bottaccione

®Moria

@ Furlo upper road

-0: Gubbio

Polarity

.normal

Oreversed

as volcanic rocks (Chapter 15). In the case of the Arikee Group rocks, which are 

tween about 18 and 30 million years old, correlation  solid because Earth's geomag

nec history during the past 30 mlion years was relavely "bus" with numerous

polarity transions. Thus, e correlation is comparatively easy and unambiguous.

Another example is the use of maetostratigraphy of sediment cores or on

land secons as a tool for esmang the ages of volcanic eruptio that took place ei

ther

on land or in the ocean. This  done by determining e ages of erupted ash that

fell or were washed into on-land depositional sites or the ocean and preserved as ash

beds  sediments (e.g., Shane et al., 1996). By establishing the magnetic chronology

from the reversal pattes in cores or outcrop sections, geologists can deteine e

ages of ash layers in the sediments by reference to the paleomaetic time scale, a

technique called tephrochronolo.

A related application is in determination of rates of sedimentation for deep

sea sediments. Paleomagnetic correlation of deep-sea cores with rocks on land

whose ages have been determined radiometrically allows absolute ages to be as

signed to the boundaries between different geomagnetic events in the cores. The

thickness of sediments between horizons within the cores whose ages are thus

Upper

Geomagnetic Polarity

Time Scale

Compo� Magnetic

Polan Stratigraphy

Arikaree Gup,

Sioux County NE

Hiatus







- w

13.4 Magnetostratigraphy

475

Figure 13.31

Correlation of the Arikaree Group and

lower Runningwater Formation,

northwest Nebraska, to the magnetic

polarity time scale of Berggren et al.

(1995). [From MacFadden, B. j., and R.

M. Hunt, ]r., 1998, Magnetic polarity

stratigraphy and correlation of the

Arikaree Group, Arikareean (late

Oligocene-early Miocene) of north

western Nebraska, in Terry, D. 0., H. E.

LaGarry, and R. M. Hunt, Jr. (eds.), De

positional environments, lithostratigra

phy and biostratigraphy of the White

River and Arikaree Groups (late Eocene

to early Miocene, North America), GSA

Spec. Paper 325, Fig. 16, p. 162.]

determined can then be used to calculate the sedimentation rate. For example, if

we assume that 10 m of sediment were deposited in a given area of the ocean dur

ing the time represented in a core by the Matuyama Reversed Polarity chron ex

tending from 2.4 to 0.7 Ma, a me interval of 1.7 million years, the sedimentation

rate

for this area of the ocean can be calculated as 10 m/1.7 million yr 5.8 m/m.y.

Pa leoclimatolo

Ages of sediment cores determined by paleomagnec methods have also been

used to study paleoclimate oscillations during the Quaternary and late Pliocene.

For example, the magnetostratigraphy of deep-sea sediments provided a means of

estimating the ages of ice-rafted debris in piston cores. It also furnished a method

of studying the timing of siliceous ooze deposition, which reflects increased bio

logic productivity owing to increased oceanic upwelling and resulting increase in

nutrients during cooler periods. Quantitative determinations of variations in mi

crofossil assemblages, particularly planktonic foraminiferal assemblages, have

also been studied in relation to climatic cycles. Some microfossil species are much

476

Chapter 13 1 Seismic, Sequence, and Magnetic Stratigraphy

more abundant during cooling trends, whereas others are more abundant during

warming trends. Also, fluctuations in oxygen-isotope ratios in carbonate shells in

deep-sea sediments are related to climate changes (Chapter 15). Use of paleomag

netic methods to estimate the ages of these climate-related biologic oscillations,

fluctuations in oxygen isotope ratios, and variations in distribution of ice-rafted

material in the oceans has added significantly to our owledge of climatic fluctu

ations on land. It has also radically changed our ideas about the number of climat

ic cycles that occurred during the Quaternary. We now know, for example, that

many more cycles of cooling and warming took place than the four major glacial

advances and treats postulated from land-based studies.

Study of Displaced Te anes

The magnetic inclination of ancient magnetized rocks is now being used ext

sively as a tool to examine presumed movements of continental masses and smaller

blocks (see McElhinny and McFadden, 2000, for details and case histories). By

measurg the remanent magnetic inclination and declination  ancient rocks, ge

ologists can reconstruct the original geographic position of these rocks at the time

they formed. These studies have shown not only that major continents have shifted

their positions with time, but also that many smaller blocks of rock have moved

from their original locations. That is, these blocks are now located in different lati

tudes from those in which they formed. Quite commonly the blocks have diffe rent

lithologies and structural attitudes from those of adjoining areas. These exoc

blocks are often called suspect terranes, referring to the probability that they are

not now in the geographic positions in which they originally formed. There is

growing evidence that large portions of many continental margins are made up of

a coage of these suspect terranes, assembled by seaoor spreading and subduc

tion processes over long periods of time from different parts of Earth.

Other Applications of Paleomagnetism

The study of paleomagnetism can be applied to a number of other geologic prob

lems, not all of which, strictly speaking, are stratigraphic problems. ese appli

tions include dating of archaeological materials; tracing the source or provenance

of these materials; study of magnetic fabrics in sedimentary rocks (e.g., paleocur

rent analysis); study of apparent polar-wanderg paths of Earth through time;

and paleogeographic and tectonic plate reconstruction (Aissaoui, McNeill, and

Hurley1 1993; Butler, 1992; Khramov, 1987; McElhinny and McFadden, 2000; Piper1

1987; Ta rling, 1983; Ta uxe1 20021 Ch. 6; Van der Voo, Scotese, and Bonhommet,

1984; Van der Voo

1993