Middleton G.V. (Ed.) Encyclopedia of Sediments and Sedimentary Rocks

Подождите немного. Документ загружается.

ISOTOPIC

METHODS

SEDIMENTOLOCY

385

Van

Houton,

F.B., and

Bhattacharyya,

D.P., 1982.

Phanerozoic

ooltitic

ironstones—geologic record

and

facies

model.

Annual

Reviews

Earth

and

Planetary

Sciences, 10: 441-457.

Van

Houton,

F.B., and

Purucker,

M.E., 1984.

Glauconitic

peloids

and

chamositic

ooids—favorable

factors, constraints, and problems.

Earth-Science

Reviews, 20: 211-243.

Veizer,

J.,

Clayton,

R.H.,

Hinton,

R.W., von

Brunn,

V.,

Mason,

T.R.,

Buck,

S.G., and Hoefs, J., 1990.

Geochemistry

Precambrian

carbonates:

3—shelf

seas

and

non-marine

environments

of the

Archean.

Geochimica

Cosmochimica

Acta,

54: 2717-2729.

Veizer,

J.,

Clayton,

R.H., and

Hinton,

R.W., 1992.

Geochemistry

Precambrian

carbonates:

tV—liarly

Paleoproterozoic

(2.25

0.25 Ga)

seawater.

Geochimica

Cosmochimica

Acta,

56: 875-885.

Young, G.M., 1988.

Proterozoic

plate

tectonics,

glaciation

and

iron-

formations.

Sedimentary

Geology,

58: 127-144.

Young, T., 1993.

Sedimentary

iron

ores, tn

Patrick,

R.A.D., and

Polya,

D.A. (eds.).

Mineralization

in the

British

Isles.

London:

Chapman

and Hall, pp. 446-489.

Young, T.P., 1989.

Phanerozoic

ironstones: an

introduction

and

review.

In Young, T.P., and

Taylor,

W.E.G.

(eds.),

Phanerozoic

Ironstones.

Geological Society

London,

Special

Publication,

46, pp. ix—XXV.

Cross-references

Berthierine

Chlorite

Sediments

Glaucony

and

Verdine

Oolite

and

Coated

Grains

Seawater:

Temporal

Changes

in the

Major Solutes

Siliceous

Sediments

Stromatolites

Syneresis

ISOTOPIC METHODS

SEDIMENTOLOGY

Introduction

Isotopes are atoms whose nuclei contain the same number

of protons but a different number of neutrons. The nuclei of

unstable (i.e., radioactive) isotopes decay spontaneously

toward the stable condition through the emission of particles

and energy. Radioactive decay of a particular unstable nuclide

is generally accomplished by one or more of five decay

schemes, broadly characterized by beta (/J~), positron (^*)

and alpha (a) particle emission, electron capture (e.c.) and

tission. The decay schemes are accompanied by other nuclear

emissions, including neutrinos, anti-neutrinos, gamma radia-

tion, and X-radiation. The element undergoing radioactive

decay (the parent) is transformed into another element (the

daughter). The end-point of a radioactive decay scheme is the

production of a stable nuclide. The daughter(s) of radioactive

decay are referred to as radiogenic nuclides.

The nuclei of stable isotopes do not undergo radioactive

decay to form other isotopes. Nevertheless, the relative

abundances of an element's stable isotopes vary within and

among phases bearing that element. These differences,

commonly described as "fractionation" between two phases,

are controlled by a variety of equilibrium and kinetic

processes, including temperature and reaction rate.

Isotopic methods are utilized in sedimentology in two main

ways.

First, they provide a means to date sedimentary mate-

rials (commonly referred to as geochronology or radiometric

dating). This is done almost exclusively using radioactive

isotopes. Geoehronometers include the Rb-Sr, K-Ar, Sm-

Nd, and U-Th-Pb systems (including U-series disequilibrium

methods), and various cosmogenic and related radionuclides,

including '''C. The choice of geochronometer depends on the

composition and nature of the material to be studied, and its

estimated age. Radiometric methods can be used to provide

many types of sedimentological information, including deposi-

tional ages, ages of detrital grains, crustal residence times for

sediments, residence times of oceans, deposition rates, a

temporal chronology for Pleistocene and Holocene paleocli-

mate proxies (minerals, organic matter, etc.), ages of erosion

surfaces and landscapes, dates for diagenetic mineral forma-

tion, and temporal constraints on porewater evolution and its

movement in sedimentary basins.

Second, isotopes can be used to trace the origin of inorganic

and organic solids, liquids and gases within the sedimentary

system, and their interactions within and among the hydro-

sphere, biosphere, atmosphere and deeper lithosphere. The

stable isotopes of hydrogen, oxygen, carbon, nitrogen and

sulfur are of special interest because they can be fractionated

on a large scale by low-temperature processes. Commonly

utilized radiogenic isotope tracers include strontium-87

(expressed as * Sr/^'^Sr ratios) and neodymium-143 (expressed

as '''^Nd/'^'^Nd ratios). These isotopic tracers can be used in

identification of sediment provenance, characterization of

physicochemical conditions during deposition and weath-

ering/soil formation, determination of the temperature and

fluids involved in diagenesis, characterization of organic-

inorganic interactions within pore systems, description of tluid

evolution and migration in sedimentary basins, and re-

construction of paleoclimate.

Dating methods and radiogenic isotopic ratios

The principles of radiometric dating are described by Faure

(1986) and Bowen (1988). You are directed to these books for

detailed discussion and references concerning much of what

follows.

Basic equations

The decay rate of a radioactive parent nuclide (A') with time (/)

is described by:

-dN/dt = IN (Eq. 1)

where X is the decay constant. Long-lived radionuclides (e.g.,

'""^Sm, ^^Rb, •^^•^Th, ^•'^U, ^^^U, and ''°K have very small values

of X (e.g., 6.54 X 10"'^ yr"' for '''^Sm) and hence long

half-

lives (the time it takes for 50 percent of the parent to decay).

They can be used to date systems that are millions to billions of

years old. Other radionuclides have much larger decay

constants (e.g., '"C, /=1.21 x 10"" yr"') and hence much

shorter half-lives. This restricts their application to much

younger systems (e.g., ~ 10^ years for

'"*€).

Equation 1 can be integrated to produce the basic equation

of radioactive decay:

NJNQ

= e"'-' (Eq. 2)

where N,, is the number of parent atoms at r = 0. Equation 2

can be expressed in terms of the number of daughter atoms

(£))as

£) = £)o + A'(e'-'-1), (Eq. 3)

386

ISOTOPIC METHODS IN SEDIMENTOLOCY

where Do is the number of daughter atoms initially present in

the system (t =

0).

For radiometric dating to work, the system

(rock, mineral, etc.) cannot have lost or gained parent or

daughter atoms except through radioactive decay (a "closed"

system), and the material analyzed must be representative of

the system. If daughter atoms were present in the system prior

to closure, their abundance must be determined accurately.

K-Ar methods

For some geoehronometers (e.g., decay of

""K

to ''°Ar), there

are few daughter atoms in the system at the time of closure

{Do « 0), thus allowing the dating equation to take the form:

(AAr/^oiai)('"'K)(e'-' - 1) (Eq. 4)

40/

In the K-Ar system, a factor expressing a ratio of decay

constants (^ArZ-'-totai) must be introduced, since the parent C°K)

undergoes decay not only to ''"Ar but also to ''"Ca. Additional

complications arise because the daughter is a gas, and is easily

lost i"rom the system upon heating or over time. Correction for

diffusive argon loss is possible in some minerals (e.g.,

K-feldspar) if stepwise heating and the ""Ar/^^Ar method are

employed (^^Ar is produced by neutron bombardment of ^^K

in a reactor).

The '^KJ'^°Ar and ''°Ar/^'Ar methods can be used to obtain

depositional ages for some sediments. Sanidine and unaltered

biotite are preferred for such measurements in bentonites,

but clay minerals derived from the ash can also be used,

provided that they formed very early in diagenesis and that

their crystal growth is fully understood (Clauer and Chaud-

huri, 1995; Clauer et al., 1997). Syndepositional, K-rich

glauconite (glaucony) formed close to the sediment-water

interface in marine environments is also useful for obtaining

depositional ages.

Dating of clay minerals presents a challenge. First, their

large surface area and general reactivity encourages sorption

of Ar and K on one hand, and their potential loss during

post-crystallization processes (both natural and during analy-

sis) on the other hand. Minimum depositional ages are

commonly obtained from poorly crystallized, low-K glauco-

nite for these and other reasons (Clauer and Chaudhuri, 1995).

Second, the same clay mineral in sediment can be of multiple

origins; for example, detrital illite from several sources may be

mixed with illite formed during several stages of diagenesis.

During burial diagenesis of shales, K-Ar dates for illitic clays

tend to decrease with increasing temperature and decreasing

particle size, in tandem with the smectite to illite reaction.

Diagenetic illite is concentrated in the tiner clay fractions.

However, even when the finest clay size-fractions (e.g.,

<0.1 um) are isolated, contamination by detrital illite can still

occur. This can induce a substantial error in the calculated age.

Likewise, diagenetic clays formed by transformation rather

than direct precipitation may inherit radiogenic isotopes

(including ''"Ar) from their precursor. Such problems of

contamination and inheritance are much less severe for

authigenic clays that have precipitated directly from solution

in the pores of sandstones.

A novel ''°Ar/K20 "pseudoisochron" method to identify

sources of detrital clay assemblages has been described by

Jeans et al. (2001). Pevear (1992) described a method of

40j^_40^j.

analysis that allows both the detrital and diagenetic

end-member ages to be obtained over a range of illitic clay

particle sizes. Such information contributes to our under-

standing of both the depositional and thermal histories of

many potential petroleum source rocks. Authigenic illitic clays

and K-feldspar, which are common in sandstone pores, can

also be dated. Ziegler and Longstaffe (2000a,b), for example,

used K-Ar dates for early K-feldspar and late illite to

recognize multiple stages of fluid flow in sandstones of the

Appalachian Basin, and related this behavior to regional and

local tectonic activity. Girard and Onstott (1991) used laser-

probe sampling and '"'Ar/'^Ar step-heating methods to obtain

separate dates for detrital K-feldspars and their secondary

overgrowths at a precision and on a microscopic scale not

possible using whole crystals. Likewise, '"'Ar/^'Ar dating of

authigenic K-feldspar in carbonate rocks has been used to

discern episodes of fluid tlow and to relate their timing to

regional tectonic events (Spotl etal., 1998).

Rb-Sr and Sm-Nd isochron methods

Radiometric dating using many geoehronometers, including

^^ ^' ""^ '"^ i

the decay of

and decay of

to '"^Nd, is

complicated by an initial quantity of stable daughter in the

system at t = 0. Estimation of Do is normally accomplished

using the isochron method. In this approach, isotopic analyzes

are obtained for several cogenetic samples containing varying

abundances of the parent and daughter elements (i.e., a range

of Rb/Sr or Sm/Nd ratios). The isotopic abundances are

expressed as a ratio using a common stable isotope of the

daughter as the denominator. For the Rb-Sr and Sm-Nd

geoehronometers, these expressions are:

and

(Eq. 5)

(Eq. 6)

On an isochron diagram for the Rb-Sr system, for example,

'*''Sr/*''Sr is plotted versus **'Rb/'*''Sr for each sample. The slope

of the straight line (e''-l) formed by the data from a closed

system provides the date, and the intercept of this line with the

y-axis at *^Rb/^^Sr = 0 provides the initial ratio for strontium

(*^Sr/*^Sriniiiai). In sedimentary rocks, straight lines on

isochron diagrams can also arise from mixing between source

materials of different ages ("mixing" lines). If the system has

been disturbed following closure (e.g., by weathering, interac-

tion with formation waters, hydrothermal alteration), a

straight-line fit to the isotopic data is not possible on the

isochron diagram, and a date cannot be obtained.

Feldspars, micas and illitic clays are typical targets of

the Rb-Sr isochron dating method in sediments. Ages of

deposition that are consistent with K-Ar dates and other

stratigraphic information are commonly preserved in well-

crystallized glauconite (K2O > 6.5 percent) (Clauer and

Ciiaudhuri, 1995). Rb-Sr dates for secondary K-feldspar

(adularia) in porous and permeable units can be used to define

episodic brine migration in sedimentary basins (Harper etal.,

1995).

Rb-Sr isochron dates have also been used to study the

smectite to illite reaction during burial diagenesis of shales,

and in particular to suggest that this reaction can occur

ISOTOPIC METHODS IN SEDIMENTOLOCY

387

episodically rather than continuously (Morton, 1985; Ohr

etal., 1991).

Use of the Rb-Sr method to study clay mineral formation

during weathering is more complex, because of the open nature

of these systems, and the potential for gain and loss of parent

and daughter isotopes. Progress is being made in this area by

using the "internal" isochron method described by Clauer and

Chaudhuri (1995). In this approach, each of acid-leached clay,

untreated clay and leacJiate solutions are analyzed to

determine whether the radiogenic ^'Sr in these f"ractions

evolved together or was derived from different sources.

Sm and Nd are generally not fractionated during sedimen-

tary processes. Hence, Sm-Nd isochron dating is rarely

attempted in sediments because the spread in Sm/Nd ratios

of cogenetic phases is normally insufficient. However, such

dating of diagenesis in argiliaceous sediments has been

performed using authigenic apatite, Fe-oxides/hydroxides

and fine clays (<0.2nm) (Awwiller and Mack, 1991; Ohr

et al., 1991). These Sm-Nd dates can survive isotopic

rehomogenization during deeper burial, hydrothermal activity

or metamorphism, unlike the Rb-Sr isotopic system, which is

typically reset.

As for the K-Ar system, detailed mineralogical character-

ization (microscopy, crystal structure, chemical composition)

that permits detrital and diagenetic minerals to be discerned

and separated is required when applying Rb-Sr and Sm-Nd

geoehronometers to sedimentary rocks. "Best-practices" have

been elaborated by Clauer and Chaudhuri (1995).

Model ages and epsilon (e) values

In many sedimentary systems, it is not possible to obtain the

initial ratio using the isochron method. In these cases, model

ages can be calculated by assuming an initial ratio, although

such results must be used with care. One approach is to derive

the initial ratio from some generalized reservoir by assuming

its linear evolution over Earth history. A common choice for

this reservoir in the Nd-isotope system is CHUR (Chondritic

Uniform Reservoir; DePaolo and Wasserberg, 1976). Differ-

ences between the '"^Nd/'^^Nd ratio (or ^^Sr/««Sr) of a sample

and the values of CHUR (or some other baseline) at a given

time are commonly expressed as the epsilon (e) parameter. The

'''•'Nd/'''^Nd value of £° (i.e., at present time) is described by:

(Eq. 7)

where CHUR* is the ratio at the present time. This

approach allows small but significant differences between

these ratios to be more easily expressed (Clauer and

Chaudhuri, 1995).

The sediment load of modern rivers, as well as fine-grained

clastic sedimentary rocks in general, acquire Sm/Nd ratios

and Sm-Nd isotopic compositions similar to their igneous and

metamorphic source rocks. These have '''^Nd/'^^Nd ratios and

hence E^a values characteristic of their age (e.g., <-20 for

Archean crust versus > 2 for recent island arc volcanic rocks;

Patchett et al,, 1999). The ENd values of clastic sedimentary

rocks can therefore provide useful information about their

provenance, plus insight into the crustal residence time of the

detritus (i.e., time elapsed since Nd was separated from the

mantle into the continental crust). Patchett etal. (1999) used

ENd values for sedimentary rocks from cratonic and syn-

orogenic sequences to infer temporal patterns in the dominant

sources of sediment in North America (600-450 Ma, con-

tinental shield; 450-150 Ma, Appalachian orogen; < 150 Ma,

Cordilleran orogen). Unterschutz et al. (2002) use a similar

approach to demonstrate a mixed volcanic arc and cratonic

provenance for metasedimentary rocks comprising the Triassic

Quesnel terrane of the southern Canadian Cordillera. These

data imply a pericontinental rather than exotic origin for the

Quesnel terrane during the assembly of this portion of the

Canadian Cordillera.

r ratios in seawater and sediments

The *^Sr/^*Sr ratios of stratigraphically well-identified marine

carbonates (brachiopod shells, abiotic marine carbonates),

phosphates (e.g., conodonts) and sulfates are used for regional

and global chemostratigraphic correlation. They also provide a

means to understand the interactions among global tectonic

and climatic processes (mountain building, seafloor spreading,

glaciation, erosion and weathering rates, sea level change, etc.)

over geological time (Veizer, 1989; Edmond, 1992). Because

Rb is not incorporated into most carbonates, phosphates and

sulfates, they acquire and retain the *^Sr/*^Sr ratio of the water

from which they precipitated, unless modified by interaction

with later fiuids. Such modification of strontium- (and oxygen-)

isotope compositions is a significant concern, given the relative

ease with which carbonate minerals recrystallize. Alteration

is normally detected using stable isotopic, major and trace

element, and cathodolumineseence methods, or avoided

through the use of more resistant phases (e.g., euconodonts)

(Derry etal., 1992; Asmerom etal., 1991; Ebneth etal., 2001).

Water acquires its ''^Sr/*''Sr ratio through rock-water

interaction (hydrothermal alteration, weathering, exchange).

Interaction with mantle-related volcanic rocks produces water

with abundant Sr but generally low ^'Sr/^'^Sr ratios (-0.7035),

which characterizes deep-sea hydrothermal input to the

oceans. The submarine hot-spring contribution to the oceans

may have varied with changes in the rate of seafioor spreading,

particularly during Paleozoic time, but likely represents a more

or less constant flux of constant Sr-isotope composition during

the Phanerozoic (Edmond, 1992).

Water associated with erosion and weathering of continental

rocks and their soils normally acquires low Sr contents but

relatively high ^'Sr/'^^Sr ratios (-0.712), mostly as a con-

sequence of interaction with feldspars and micas. The *^Sr/*^Sr

ratio of the leachate tends not to be as high as the

rocks because of preferential dissolution of plagioclase relative

to more radiogenic K-feldspar and biotite (Clauer and

Chaudhuri, 1995). However, much higher contents of highly

radiogenic Sr are released during erosion of Himalayan-style

metamorphic core complexes (Edmond, 1992).

The flux of continental Sr is carried to the ocean by rivers

and groundwater (Edmond, 1992; Basu et al,, 2001). The

^•'Sr/^^Sr ratio and Sr content of the meteoric water refiect the

lithologies intersected, and can also vary with rainfall amounts

in a region (Galy et al., 1999). Continental hot springs

associated with collisional orogens also contribute highly

radiogenic Sr ("Sr/**Srw0.77) to the Sr-budget of some

globally important fluvial systems, such as Himalayan rivers

(Evans et al., 2001). Variations in continental glaciation,

atmospheric CO2 levels, climate, vegetation, orogeny and

388

ISOTOPIC METHODS IN SEDIMENTOLOCY

eustatic sea level all help to determine the rock types exposed

at the Earth's surface, and the rates at which weathering and

erosion occur.

Marine carbonates and evaporites are the sink for Sr from

seawater, removing it from solution with no isotopic fractio-

nation. These rocks typically release high Sr contents and

intermediate ^'Sr/**Sr ratios (~ 0.7075-0.7080) to river water.

These ratios correspond to the balance between crustal

weathering and mantle-related processes at the time of

chemical sedimentation. The carbonates and evaporites can

be exposed by tectonic processes and eustatic changes in sea

level. When this happens, they tend to dominate the fluvial tlux

of Sr to the ocean. This feedback attenuates the variations in

Sr-isotope composition of seawater that otherwise would

ensue if it was determined only by the balance between

mantle-related activity and the weathering of sialic rocks

(Edmond, 1992).

The oceans presently have a uniform *^Sr/^''Sr ratio of

0.70906 ± 0.00003, which retlects the long residence time of Sr

in seawater. This value corresponds well with its proxy in

modern marine carbonate (0.70910 + 0.00004) (Burke etal.,

1982).

Its composition has tiuctuated widely over geological

time.

Values were high in the Late Cambrian (> 0.7092), but

gradually declined through wide fluctuations to as low as

0.70676 in the Jurassic; such low values also occurred in the

Permian. Since the Jurassic, the Sr-isotopic composition of

seawater has increased irregularly, with a particularly rapid

climb toward the present value beginning in the mid-Tertiary

(Burke et al., 1982; Edmond, 1992; Denison et al., 1998;

Ebneth etal,, 2001).

The strongest variations in the strontium isotopic composi-

tion of seawater reflect changes in continental input. Edmond

(1992) proposed that only Himalayan-style continental colli-

sions can produce a sufficiently high flux of radiogenic Sr to

account for the positive extremes in the seawater ^^Sr/^^Sr

record. During such major orogenic events, the Rb-Sr systems

of ancient continental rocks are disturbed, leading to the

redistribution of highly radiogenic Sr (^''Sr/^*Sr~0.769 in

Himalayan core complexes) into minerals such as plagioclase,

which weather easily following their exposure. During the late

Proterozoic, ^^Sr/^^Sr ratios also rose from very low (~ 0.707)

to high (~0.709) values. This may suggest a fundamental

change from rifting and production of juvenile oceanic crust to

continent-continent collisions, production of recycled con-

tinental crust, and its erosion (Asmerom etal., 1991). Many

research efforts continue to be directed toward finer-scale

definition of the Sr-isotope seawater curve, and the specific

variations arising from second-order processes, such as

glaciation (Blum and Erel, 1995).

'^Nd ratios In seawater and sediments

The '''•'Nd/''''*Nd ratios of seawater (commonly recast in terms

EiMd)

are controlled principally by erosion of the land masses

adjacent to the oceans. Nd, unlike Sr, has a very short

residence time in seawater relative to the mixing time of the

Earth's oceans. The Atlantic Ocean, which is surrounded

mainly by older sialic rocks, has a lower

ENJ

(-15 to -10) than

the Pacific Ocean (—5 to 0), which is surrounded mostly by

younger volcanic rocks related to subduction of oceanic crust

(Piepgras etal., 1979). As a result, the spatial variation in Nd-

isotope composition of modern seawater can be used to trace

circulation within and among oceans.

Variations in seawater '''^Nd/'''''Nd ratios can be tracked

through geological time through analysis of phosphates (e.g.,

conodonts) and carbonates (e.g., brachiopods). They acquire

Nd-isotope compositions characteristic of ambient seawater

shortly after deposition at the sediment-water interface, and

are largely unaffected by later alteration, including interaction

with meteoric water (Banner et al,, 1988). This makes

'''^Nd/'^'^Nd ratios more robust proxies of oceanic conditions

than strontium- or oxygen-isotope ratios. The Nd-isotope

compositions can be used to reconstruct the variation in the

exposed crust of the Earth, and to deduce plate tectonic

activity (i.e., the opening and closing of ocean basins) that

determined the distribution of the Earth's oceans (Shaw and

Wasserburg, 1985).

U-Th-Pb

methods

The U-Th-Pb system provides a wide range of geoehrono-

meters useful in the study of sediments. It comprises three

separate and complex decay series, ^^^U to

'^"^Pb,

^^^U to ^"^Pb

and ^•'^Th to ^"^Pb. In each case, the decay scheme contains

numerous intermediate daughters before the stable daughter is

achieved. The intermediate daughters have much shorter

half-

lives than the parent, and under the condition described as

secular equilibrium, the rate of the decay of these daughters

is controlled by the rate of decay of the parent (Faure,

1986).

For each of the U-Th-Pb decay schemes, model ages

can be obtained by assuming reasonable initial Pb-isotope

ratios (given relative to ^"^Pb). Combination of any two of the

U-Th-Pb geoehronometers permits the concordia approach

to dating to be used. The latter is particularly useful in studies

of sediment provenance, which focus on detrital zircon and

other U- or Th-rich phases.

The U-Th-Pb geoehronometers that are perhaps most

valuable in sedimentology are provided by certain intermediate

radioactive nuclides (commonly with high values of /. and

hence short half-lives). These intermediate daughters are used

in a wide variety of ways, collectively described as U-series

disequilibrium dating. Some allow dating of intervals between

those accessible using the '''C (<10^ years) and ''°K-'"'Ar

(> 10'' years) methods; others (e.g., ^'"Pb) permit dating of

modern materials (i.e., <100 years old). U-series disequilibrium

dating is possible because many intermediate daughters in the

decay series have geochemical properties different from their

parent. This allows parent and daughter to be separated during

processes such as chemical weathering, (bio)chemical sedimen-

tation, and adsorption onto clay minerals. For example,

uranium (as the uranyl ion) tends to remains in solution in

seawater, whereas Th is preferentially removed by precipita-

tion of zeolites or by adsorption onto particulate surfaces

(e.g., clays, organic matter, etc.). Radiometric dating is thus

accomplished using an isolated intermediate daughter (e.g.,

^•'"Th, -^^'Pa, ^'"Pb) that, free from replenishment by decay of

the absent parent, disintegrates at a rate determined solely by

its own (shorter) half-life. Alternatively, a daughter may be

excluded from a geologic material at the time of its formation,

leaving only the radioactive parent. The new daughter

production from that parent can be used to obtain the time

since the system was last open, at least until reestablishment of

secular equilibrium (Faure, 1986).

U-series disequilibrium dating methods have proven most

useful in the study of Pleistocene sediments, commonly

through exammation of

ISOTOPIC

METHODS

SEDIMENTOLOCY

389

and

"°Th/^^'Pa ratios.

For

example, oceanic

sedimentation rates

can be

determined

obtaining these

ratios with increasing distance from

the

sediment-water

interface. Likewise, dates

can

commonly

obtained

for

young marine

and

nonmarine carbonates, given that their

crystal structure incorporates uranium upon precipitation,

but

almost always excludes thorium. The activity

^^"Th

such

carbonate derives from

the

decay

^^''U, which

turn

derived from •^^*U.

The

ability

date growth bands

speleothem

and

other terrestrial carbonate cements allows

temporal calibration ofthe paleoclimate information obtained

from oxygen, hydrogen, carbon

and

strontium isotopic

compositions

of the

same material (Schwarcz,

H.P., 1986;

Doraie etal., 1998; Kaufman etal., 1998). Corals can

used

similar fashion (Edwards

al., 1993),

and

provide

chronology

for

sea-level fluctuations.

Because

its short half-life (22.26 years), the abundance of

^'"Pb can

used

obtain sedimentation rates

lacustrine

and marine sediments that

are <100

years

old. It is a

radioactive daughter

gaseous ^^^Rn within

the

^•'^U decay

series.

Upon production, ^^^Rn

released

the atmosphere,

where

decays rapidly

^'°Pb. The ^'"Pb

swept quickly

from

the

atmosphere

precipitation,

and

accumulates

snow, ice, lakes, vegetation and sediments. The declining level

of '"Pb activity can then be used

calculate the time elapsed

since separation from

its

parent (Faure, 1986).

Cosmogenie and anthropogenic radionuclides

Another useful dating tool

Pleistocene

and

Recent sedi-

mentary environments

provided

cosmogenie nuclides.

These isotopes

are

produced

in the

Earth's atmosphere

interactions between cosmic radiation

and

oxygen, nitrogen

and argon. The most familiar

these reactions is the neutron-

induced formation

'''C from

'''N,

and its subsequent return

to '''N via

ji"

decay with

half-life

5730 years. Carbon-14

incorporated into

CO2,

which mixes rapidly within

the

atmosphere

and

hydrosphere,

and is

fixed

photosynthesis

into plants and the animals that eat them. Once

organism

dies,

replenishment

'''CO2 ends, and the time elapsed since

death

can be

determined from

the

remaining activity

'''C.

This model

complicated

many factors, including

variations

in the

radiocarbon content

of the

atmosphere

(e.g., Edwards etal., 1993; Beck etal., 2001). These have

resulted from changes

in the

flux

cosmic radiation

to the

Earth, release

"dead" carbon by combustion

fossil fuels,

addition

of '''C via

nuclear devices,

and

mass-dependent

fractionation among '^C, "C, and '''C.

In addition

organic matter from sediments

and

soils,

considerable effort

has

been devoted

'''C-dating

chemically

biologically precipitated calcium carbonate.

For accurate dating

fresh-water shells, their carbon must

have been derived wholly from

the

atmosphere,

and the '''C

content must have

not

been modified

postmortem

alteration. Complications also arise

marine environments

in environments where older, deep ocean waters

may

have

mixed with modern waters, both

which can supply carbon

to the shells

organisms (Faure, 1986). Carbon-14 measure-

ments have also been used

date groundwater and describe

fluid mixing

the subsurface (Clark and Fritz, 1997).

Tandem accelerator mass speetrometry permits much

smaller quantities

of '''C to be

analyzed than possible using

standard radiochemical methods.

It has

also made feasible

measurement

other cosmogenie (e.g., ^H, ^He,

'Be, '"Be,

'''C,

^^Al, •'"Si, ^^Cl, ^^Ar, *"Kr)

and

anthropogenie radio-

nuclides (e.g.,

'H,

^"Sr, '"Cs) (Faure, 1986; Clark and Fritz,

1997).

Cosmogenie radionuclides with long residence times

the atmosphere (e.g., ^^Ar, *'Kr) provide the potential

date

air bubbles trapped

in ice

cores (Faure, 1986). Others

are

removed quickly from

the

atmosphere

precipitation (e.g.,

'"Be,

^*A1,

^H,

^*C1,

'^^1).

They

can be

used

determine

sediment transport

and

deposition rates

marine

and

terrestrial environments,

measure accumulation rates

ice sheets,

and to

date groundwater

and

brines (Lao

et al.,

1993;

Wang etal., 1996; Yechielli etal., 1996).

The abundance

eosmogenie nuclides has also been used

to determine the age

erosion surfaces (Phillips etal., 1996).

The '"Be

and

•^^Al contents

terrestrial surfaces, espeeially

quartz, provide estimates

surfaee exposure time,

and

from

this,

erosion rates and the ages

exposed terrestrial surfaces

can

deduced (Nishiizumi etal., 1993; Bierman and Turner,

1995).

Cosmogenie

^^Cl

production

limestones

has

been

used

for

similar purposes (Stone etal., 1998). Anthropogenic

radionuclides (e.g., 'H,

'^'CS)

have also been used

date and

track mixing

modern groundwater,

and to

trace modern

geomorphic processes

sediments and soils (Clark and Fritz,

1997;

Vanden Bygaart and Protz, 2001).

Stable isotopes

The stable isotopes

hydrogen (^H

(or

D),

'H),

carbon ('^C,

'-C),

nitrogen ('^N, '"N), oxygen ('^O, '"^O, '^O)

and

sulfur

('^S,

•'''S,

^•'S, •'^S) are of

most interest

in the

study

sediments. However, boron ("B,

'"B;

Vengosh etai., 1991;

^^i;

Ding

e/a/.,

1996;Williams etai., 2001), silicon (^"Si,

Rocha

ai., 1998), chlorine

("Cl,

^^C\; Kaufmann

etai., 1993; Eggenkamp etal., 1995) and iron ('^Fe, "Fe, '*Fe,

^"Fe;

Zhu et al.,

2000; Bullen

et ai., 2001) are

attracting

increasing interest. Reviews

stable isotope systematics

pertinent

sedimentology inelude Fritz

and

Fontes (1980,

1986,

1989), Arthur etai. (1983), Faure (1986), Kyser (1987),

Longstaffe (1987, 1989, 1994, 2000), Savin

and Lee

(1988),

Bowen (1988, 1991), Swart etal. (1993), Clauer and Chaudhuri

(1995),

Clark and Fritz (1997), Hoefs (1997) and Kendall and

McDonnell (1998).

Stable isotope results

are

reported using

the

delta

(5)

notation

parts per thousand (%o

permill). For phase

-^A

[(RA

Rstandard)/Rstandard]

1000, (Eq.

where R is normally D/H, '^C/'-Q

''N/"'N,

'*O/"^O

^''S/''^S.

The (5-value

for

the standard

defined as 0%o exactly.

For hydrogen, this standard

VSMOW (Vienna Standard

Mean Oeean Water). Two standards

are in use for

oxygen,

VSMOW

and

VPDB (Vienna Peedee Belemnite). While

the

VSMOW standard

preferred amongst isotopists, the VPDB

standard

is in

common

use

among sedimentologists; VPDB

has

(5'*0 value

+30.91%o relative

VSMOW. Stable

carbon-isotope results

are

reported relative

VPDB, sulfur

isotopes

to CDT

(Canyon Diablo Troilite),

and

nitrogen

isotopes

atmospheric nitrogen (air). Most 5-values can

measured with

precision

of at

least +0.2%o, except

for

hydrogen (±2%o).

For oxygen and sulfur, which comprise three and four stable

isotopes respectively,

it is

normally sufficient

measure only

I8Q^I6Q

^j^j

3''§/32§ ratios. However, non-mass dependent

fractionation

in the

Earth's stratosphere makes

390 ISOTOPIC METHODS IN SEDIMENTOLOCY

•""S/'^S and "S/"S ratios of interest in sedimentary systems

where feedback from the atmosphere might be measurable

(e.g., the Earth's early sulfur cycle, Farquhar et ai., 2000;

studies of oceanic productivity, Luz and Barkan, 2000).

described earlier, the ion-probe has the potential to replace

tedious and error-prone physical and chemical methods

previously used to isolate such samples for stable isotopic

analysis (Ayalon and Longstaffe, 1992).

Recent advances in analytical methods

Most stable isotope measurements are made using the gas-

source, triple-collecting, dual-inlet, stable isotope ratio mass-

spectrometer. In the traditional approach, which dominated

until the late 1990s, samples of solids or liquids were converted

quantitatively into the appropriate gas (hydrogen, carbon

dioxide, nitrogen, sulfur dioxide, etc.) using complex "off-line"

preparation methods. These gases were then brought sepa-

rately to the mass spectrometer for analysis. However, recent

advances in stable isotope analysis of hydrogen, carbon,

oxygen, nitrogen and sulfur now make collection of these data

much more accessible to sedimentologists. In the new

approach, known as continuous-fiow, stable isotope ratio

mass-spectrometry, "on-line" preparation devices (periph-

erals) are used to produee and eoncentrate the appropriate

pure gas from a wide variety of solids, liquids and gases. This

pure gas is swept by a stream of helium from the peripheral

directly into the attached mass spectrometer for analysis.

This conceptually simple, but fundamental change in

analytical approach has enormous implications for stable

isotope research in sedimentology. Very small quantities of gas

(routinely to ~200nmols) can now be analyzed. Sample size

largely ceases to be a limitation on the ability to measure stable

isotope ratios. Lasers, for example, allow precision sampling

and appropriate gas production from small (~30-100nm)

regions of many solids (e.g., carbonate, sulfides, some

silieates). Other peripherals (gas ehromatographs, elemental

analyzers, thermochemieal eonvertors, thermogravimetrie bal-

ances,

etc.) allow similarly small quantities of water, organic

matter and other materials to be analyzed in continuous-fiow

mode. Continuous-flow coupling of a gas chromatograph to a

stable isotope ratio mass-spectrometer, for example, allows

eompound-specific stable isotopic fingerprints to be obtained

for complex organic materials. Likewise, the coupling of a

thermogravimetrie balance and a stable isotope ratio mass-

spectrometer makes possible separation and isotopie analysis

of waters of hydration and hydroxyl groups from elays and

other hydrous minerals. Such data can be used to determine

temperatures and fiuid evolution in many sedimentary

environments.

The mass spectrometers and their on-line peripherals are

automated, which allows large numbers of samples to be

analyzed in short periods of time. This approach largely

replaces labor-intensive, arcane and complex conventional

procedures with a methodology that is accessible to most

researchers. Aeeordingly, research problems requiring a large

number of analyzes (typical of sedimentology) can be under-

taken routinely.

Development of the ion-probe is also of interest to

sedimentologists. Direct stable isotopic measurement of solids

in thin section is now possible for some materials, with a

precision that begins to approach that of conventional dual-

inlet or continuous-flow methods. The ion-probe, for example,

has been used to obtain the oxygen-isotope compositions of

quartz overgrowths in sandstones, and from that, to infer

temperature and fiuid compositions during secondary quartz

cementation (Hervig et ai., 1995). Like the laser methods

Stable isotopic variations in water

Seawater and fresh (meteoric) water have characteristic

hydrogen- and oxygen-isotope compositions, as do formation

waters whose isotopic compositions evolved during rock-water

interaetion and mixing in sedimentary basins. Direct isotopic

measurements of oceans, lakes, rivers, shallow porewaters, and

formation waters from deep in sedimentary basins reveal much

about the origin and evolution of these fluids. The stable

carbon and hydrogen isotopic compositions of hydrocarbons

in sedimentary systems provide analogous information about

their sources, as well as hydrocarbon generation and degrada-

tion (Hoefs, 1997).

Seawater. tjnmodified ocean water has i5'*O and <5D values

of about 0%o, which generally vary only slightly because of

dilution by fresh water near major rivers. High latitude,

shallow epicontinental seas that receive a large infiux of

surface runoff also can acquire isotopic compositions sub-

stantially less than 0%o, especially in surface layers (e.g., the

Cretaceous Western Interior Seaway). Evaporation ean also

cause variation in the stable isotopic composition of seawater

(and fresh water). Because "'O and 'H are preferentially

partitioned into the vapor phase during evaporation, surface

waters within a restricted basin are progressively enriched in

'^O and D, at least until extreme levels of solute concentration

are reached. The isotopic composition of evaporated bodies of

water is described by:

5D = md^^O. (Eq. 9)

The value of m decreases as relative humidity declines. It is

characteristic of a given region, and the resulting distribution

of 5D and 5'*O water values along this slope defines a "local

evaporation line".

Continental glaciation can affeet the oxygen and hydrogen

isotopic composition of seawater. Continental ice caps are

major reservoirs of '^O- and 'H-rich water; melting of the

present Greenland and Antarctic ice caps would lower the

d'^0 value of the ocean by ~0.7%o. Studies of Pleistocene ice-

cover fiuctuations suggest that the variation then was no larger

than ~ l%o (Savin and Yeh, 1981).

Whether the oxygen isotopic composition of the ocean has

changed with time on a larger scale remains unresolved. "Least

altered" shell materials (mostly brachiopods) and marine

carbonate cements suggest a trend toward '*O depletion of

seawater with increasing age (to — 7%o for the Cambrian)

(Wadleigh and Veizer, 1992). Data from ophiolites and mud-

rocks support the concept of a 0%o ocean throughout much of

geologic time (Muehlenbachs and Clayton, 1976; Land and

Lynch, 1996). This behavior may refiect greater susceptibility

of carbonates versus silicates to post-formational isotopic

alteration. It may also arise from different water compositions

in the deep, open ocean versus shallow, epicontinental seas.

Seawater sulfate. Ocean water is the Earth's major reservoir

of dissolved sulfate. Modern seawater sulfate has an average

^"'S value of -f21.0%o, and (5'*O value of ~-|-9.6%o. The ^^^S

values of sulfate minerals are very similar to the seawater from

which they evaporated, whereas their 5'*O value are enriched

ISOTOPIC METHODS IN SEDIMENTOLOGY

391

by ~3.5%o (Hoefs, 1997). The stable isotope composition of

seawater sulfate has varied throughout geologieal time,

especially its d^'^S values, which have ranged from —hlO to

-f30%o (Claypool etal., 1980). The variations in 5'*0 values

are smaller and do not necessarily track those of sulfur, for

reasons yet poorly understood (Hoefs, 1997).

Stable sulfur-isotope behavior in sedimentary environments

is complex. The isotopic composition of seawater sulfate at

any time is the baseline from which the (5-values of diagenetic

marine sulfide and residual sulfate are derived. However, their

(5-values also depend on whether open communication with

seawater was maintained in the sediment during sulfide

formation (i.e., an open versus closed system), and the nature,

rate and temperature of the processes that drive reduetion of

sulfate to sulfide. The isotopic fractionations can be consider-

able;

on average, diagenetic pyrite in sediments is depleted of

•""S relative to the dissolved sulfate by ~50%o (Longstaffe,

1989;

Hoefs, 1997).

The changes in sulfur and oxygen isotopic compositions of

dissolved sulfate in seawater arise from variations in the input

and output of sulfate to and from the oceans. The volumes

and isotopie composition of these fiuxes are controlled by

microbial reduction of sulfate to sulfide and evaporite

formation in seawater, contributions of sulfate and sulfide to

the oceans from continental weathering, the eycling of

seawater through mid-oeeanic ridges, and the volume of

seawater

itself.

Variations in seawater-sulfate isotopic compo-

sition through geological time can be linked to the same kinds

of global tectonic and eustatic processes that drive variations

in the ^'Sr/**Sr ratio of seawater. Variations in the S- and

Sr-isotope seawater curves show many similarities through

geological time. By comparison, the (5"C values of marine

carbonates, which capture the fluctuations in inorganic versus

organie eontributions to marine bicarbonate, tend to be

antipathetic to marine sulfate

5^'*S

values (Hoefs, 1997).

Porewater in sitallow marine sediments. The (5'^0 and dD

values of porewater in the first ~800 m of Cenozoic oceanic

sediments commonly decrease to values less than 0%o with

increasing depth from the sediment-water interface (Savin and

Yeh, 1981). Numerous explanations have been offered for this

behavior, including primary variations in porewater composi-

tion, and production of secondary clays, carbonates and

zeolites. Low-temperature erystallization of '*O-rich second-

ary minerals has the potential to deplete the porewater

reservoir of '*O in a quasi-closed system with a high rock/

water ratio (Longstaffe, 2000). However, neoformation of

secondary minerals cannot easily account for the lowering of

(5D values in oceanic porewater, which can exceed 10%o. Low-

D water (<200%o) formed during oxidation of organic matter,

methane and hydrogen may eontribute to the composition of

porewater in oceanic sediments, as may exchange with

methane or hydrogen from deeper sources (Lawrence and

Taviani, 1988).

Meteoric (fresiij water. Meteorie water has a large but

systematic variation in isotopic composition (5'*Ow-55 to

0%o; (5D«-400 to + 10%o) described by the Global Meteoric

Water Line (GMWL) (Craig, 1961):

(5D = 8(5'^O + 10(%o). (Eq. 10)

This regularity is largely the produet of Rayleigh distillation

within the hydrological cycle. In the simplest situation, water

evaporated from the ocean is progressively depleted of '*O and

D through successive evaporation and condensation while

travelling aeross the continents. This distillation produces a

geographically controlled distribution of oxygen- and hydro-

gen-isotope eompositions, with fresh water becoming progres-

sively depleted of the heavier isotope at higher altitudes and

latitudes. The depletion is further accentuated by lower average

air-mass temperatures and larger values of anquid-vapor toward

higher latitudes and altitudes, and by the re-evaporation of

meteoric waters from the continental surface (Yurtsever and

Gat, 1981).

Formation water. The 5'*0 and 5D values of formation

water from sedimentary basins commonly (but not always)

define a trend ("formation-water line") with a lower slope than

the GMWL. Along this trend, hotter and more saline

formation waters are generally more '^O-rich than cooler, less

saline samples. The formation-water line for a specific

sedimentary basin commonly intersects the GMWL at an

isotopic composition typical of local surface and shallow

groundwater. This suggests that meteoric water comprises an

important fraction of many formation waters (Clayton etai.,

1966).

However, the deviation of the formation-water line

from the GMWL (equation 10) indicates that other processes

are also at play. These include evaporation (see equation 9),

isotopic exchange between pore-lining minerals and porewater,

release of water by the smectite to illite reaction, mixing

between seawater and fresh water or their evolved equivalents,

and perhaps membrane filtration (see Longstaffe, 2000 and

references therein).

The extent of formation-water '^0-enriehment via exchange

depends on the isotopic composition of the exchanging solid

phases, temperature, mineral/water ratio, and the exehange-

ability of the minerals in the pore system. Extensive exchange

may be plausible in carbonate-rich rocks, but is unlikely in

silieielastic units, which exchange oxygen isotopes much less

easily than carbonates. Isotopic exchange involving clays or

organic matter may also contribute to the (5D values of

formation waters. Fluid movement from shales to sandstones

in settings where the smectite to illite reaction controls shale

porewater chemistry can explain the '^O-rich nature of some

formation waters, but this requires very high (originally

smectite-rich) shale to sand ratios within the basin. Mixing

between evaporitie brines derived from seawater and (later)

meteoric water can also explain the stable isotopic composition

of many formation waters.

It seems unlikely that there is a single explanation for the (5D

and (5'^0 values of formation waters in sedimentary basins.

Porewater evolution and movement follow different pathways,

depending on the geological circumstances of each sedimen-

tary basin. By understanding the range of stable isotopic

compositions that can be expected for seawater, meteoric

water and formation water, one ean track the evolution of

fiuids in a sedimentary basin, using the proxies provided by

secondary minerals formed in the pore system.

Stable isotopic variations in sediments and

sedimentary rocks

Ciiemical sediments. Chemical sediments and sedimentary

rocks generally have very high (5'^O values (up to +45%o),

because they acquire their isotopic compositions at low

temperatures in oxygen isotopic equilibrium with water,

during which enriehment in '^O of tlie solid phase is large.

Modern marine limestones, for example, commonly have *

392

ISOTOPIC

METHODS

SEDIMENTOLOCY

values near

30%o

(VSMOW); fresh-water varieties have variably

lower values that reflect the composition of the meteoric water

involved. Dolomitization and other diagenetic modifications

have a substantial impact on the (5'*'0 values of carbonates; for

example, post-crystallization interaction with meteoric water

invariably leads to a lowering of their (5'^0 values (Arthur el

al., 1983). Care is needed in the study of ancient marine

carbonates to obtain samples that have preserved their original

oxygen isotopie compositions (Carpenter etal., 1991).

Stable carbon-isotope variations. Marine carbonate rocks

normally have <5''^C values of ~0%o, whieh reflect the balance

between inorganic ("C-rich) and organic ('^C-rich) carbon in

the total dissolved marine carbonate reservoir at the time of

formation (Land, 1980; Hoefs, 1997). Variations in the stable

carbon-isotope composition of the oceanic reservoir through

geologic time are normally related to changes in the amount of

organic ("C-poor) versus inorganic ('^C-rich) carbon added to

or removed from the dissolved carbon reservoir.

The shifts in stable carbon-isotope composition of marine

carbonates have normally been no more than +2%o through-

out most of geological time, but relatively sudden shifts of

larger magnitude and global distribution are known. Such

decreases at the Precambrian/Cambrian, Permian/Triassic, and

Cretaceous/Tertiary boundaries are commonly correlated to

upwelling of '^C-poor, anoxic, oceanic bottom-waters and

substantially reduced organic productivity (extinction events?)

in shallow marine environments (Veizer etal., 1980; Hoefs,

1997;

Kimura and Watanabe, 2001). Very high (5'^C values for

marine carbonates (>-|-10%o) at ~2.2Ga to 2.0 Ga, and at

the end of the Proterozoic, have been linked to periods of

vastly increased burial of organic carbon. This beinavior has

been attributed by some to the complete freezing of the

Earth's oceans (tlie "Snowball Earth" hypothesis) (Knoll

el al., 1986; Baker and Fallick, 1989; Derry et al., 1992;

Hoffman etal., 1998).

Most carbonate-bearing sediments existed in zones of

microbial activity during sedimentation and early diagenesis.

In sediments with open access to the marine bicarbonate

reservoir (presently 5'-'CsiO%o), the volume of carbon dioxide

added to the system from microbial activity is too small to

detect. However, when organic carbon comprises a significant

fraction of the bicarbonate supply to the porewater, distinctive

carbonate (5'^C values result (see Longstaffe, 1989 and

references therein). For example, carbonate rocks invaded by

meteoric water typically acquire lower (5'^C values because

of the flux of low-'^C, soil-derived CO2 ((5'^C«-25%o). Large

variations in carbonate 5^^C values (5'^C« -30 to + 30%o) are

also characteristic of carbonate rocks intercalated with coals

and other organic matter, such as organic-rich shales, oil-

shales and oil-sands. Calcite with (5'^C values as low as —70%o

is associated with methane oxidation at seeps.

Growth of carbonate concretions in organic-rich mudstones

and shales represents a special case where the CO2 contribu-

tion from marine bicarbonate trapped with the porewater is

small compared to that produced during diagenetic reactions.

In the simplest case, each carbonate growth band acquires a

^'"^C value characteristic of the processes dominant at that

depth from the sediment-water interface (Irwin etal., 1977).

Consumption of organic matter in the shallowest zones is

dominated by microbial oxidation. A zone of microbial sulfate

reduction then follows in marine and brackish water environ-

ments. Both zones are characterized by CO2 with d^^C values

of —25%o. Still deeper zones are characterized by microbial

fermentation, which leads to production of '^C-rich CH4; in

some extreme case, CO2 reduction follows. Residual CO2

available for fixation in carbonate minerals is therefore

variably enriched in '-^C (avg. i5'^Cw -I-I5%o). At still greater

depths, residual organic matter is gradually destroyed by a

variety of abiotic, thermally driven reactions, which release

CO2 with 5'-^C values of —25%o to -10%o (see Longstaffe,

1989 and references therein).

Clasticsedimentaryrocks. The bulk oxygen-isotope composi-

tion of clastic sedimentary rocks is highly variable

(5'*0 = +8%o to +25%o) because it reflects varying contribu-

tions of (i) detrital minerals and rock fragments that normally

retain oxygen-isotope compositions characteristic of their

original high-temperature source, (ii) '*O-rich weathering

products (~-|-10%o to -l-30%o; clays, altered feldspars and rock

fragments, etc., and (iii) '^O-rich cements. To summarize,

clastic sedimentary rocks are typically richer in '*0 than

igneous rocks

(-|-5%o

to -|-10%o) because they contain some

minerals that acquired their isotopic compositions at low

temperatures in oxyen isotopic equilibrium with water (Savin

and Epstein, 1970a,b; Longstaffe and Sehwarcz, 1977). Shales

and mudstones generally are enriched in '^O relative to

sandstones because of the former's much higher content of

clay minerals and secondary silica.

Diagenetic minerals such as carbonate, silica and clays are

of particular interest because they are the product of a complex

mineralogical, fluid and thermal history. They form by

precipitation from solution, and by transformation or altera-

tion of pre-existing phases. Their isotopic composition is

controlled by several parameters, including composition of the

fluid, the degree of isotopic equilibrium during crystallization

(including microbial effects), the mass balance for the element

of interest between the mineral and the pore fluid, and the

extent of isotopic exchange following crystallization.

Clay minerals are the most important mineral reservoirs

of hydrogen in sedimentary rocks, with (5D values that nor-

mally range from -100%o to -20%o, typical of phases formed

during weathering (Lawrence and Taylor, 1971, 1972; Savin

and Hsieh, 1998; Longstaffe, 2000). However, lower values

(~—150%o) are known for clay minerals formed or exchanged

at high latitudes or altitudes or during diagenesis in the pres-

ence of low-D porewater (Longstaffe, 1989; Longstaffe and

Ayalon, 1990). Soil clays typically display a systematic varia-

tion in 5D and 5'^0 values that describe a trend parallel to, but

offset from the GMWL. The position of these "elay weath-

ering lines" relative to the GMWL is determined by the

hydrogen- and oxygen-isotope mineral-water fractionations for

the temperature of clay formation (Sheppard et al., 1969;

Lawrence and Taylor, 1971,1972; Mizota and Longstaffe, 1996).

Equilibrium stable isotope fractionation and

geothermometry

The difference between two phases in the ratios (R) of an

element's "heavy" to "light" isotopes (e.g., oxygen in

CaCO3(soiid) arid H2O(iiqui(j)) is known as fractionation. The

stable isotope fraetionation faetor between two phases A and

B is defined as:

A-B =

RA/RB

(Eq. II)

Values of a are normally very close to unity. Eor example,

C02(gas) in equilibrium with H20(nqujd) is enriched in O

ISOTOPIC

METHODS

SEDIMENTOLOCY

393

relative to the water, hence the value of a is positive. At 25°C,

a'^Oco,-H,o=

1.0412

(Friedman and O'Neil, 1977). As for all

temperature-dependent fractionations, this value increases at

lower temperatures and decreases at higher temperatures.

Equilibrium fractionations derive from small differences in the

thermodynamic properties between compounds containing an

isotope with a higher (e.g., '*O) versus a lower mass number

(e.g., '*O). This relative mass difference exerts important

control on the size of equilibrium fractionations. As a result,

the largest fractionations in nature are typically observed for

stable hydrogen isotopes.

The ability to calculate values of a from

(5A

and 5^

-B = (<5A +

1000)/(5B

1000),

(Eq. 12)

is of special significance in equilibrium systems, because a is

temperature (T) dependent. Experimentally determined a-

values for mineral-mineral or mineral-water pairs (aA-e)

normally fall on smooth curves when 10^1naA_B (known as

permill fractionation) is plotted versus 10^/7^ or some

variation of that function (7" in Kelvin). The smectite and

water system provided a typical example of an equilibrium

oxygen-isotope geothermometer (Savin and Lee, 1988):

lO^lna^ecite-water = 2.58(

10«)

^'^ - 4.19. (Eq. 13)

Such relationships are commonly used to deduce the crystal-

lization temperature of an authigenic mineral, provided that the

composition of the mineralizing fluid is known. Stable isotope

geothermometers of use in sedimentary systems have been

summarized by Kyser (1987), Savin and Lee (1988), Longstaffe

(1989) and Sheppard and Gilg (1996). New or revised

equations appear regularly (e.g.. Bird elal., 1994; Vitali etal.,

2001).

Stable isotope geothermometers are useful in low

temperature systems provided that (i) equilibrium was attained

between phases A and B, and (ii) subsequent isotopic exchange

was negligible. Knowledge of the rock/water ratio for the

element of interest is also important when considering the

stable isotopic behavior of mineral-water systems (Longstaffe,

1989,

1994).

Isotopic Exchange. Temperature, grain-size, surface area,

crystal chemistry, mineral/water ratio and, in some cases,

microbial activity are important factors controlling isotopic

exchange between minerals and water (see Longstaffe, 2000

and references therein). Isotopic exchange between minerals

and water requires dissolution and recrystallization of the solid

phase. Consequently, post-crystallization isotopic exchange

between many minerals and water is insignificant at surface

temperatures. For example, post-crystallization oxygen-

isotope exchange between quartz and water is not detected at

<200°C, even for the finest particles. By comparison, such

exchange can affect calcite and, to a lesser extent, dolomite,

because these minerals are much more easily recrystallized at

low temperatures in the presence of porewater. The ^'"'C value

of carbonates is generally less affected by such processes, as the

mineral/water ratio for carbon is very high in most cases.

Stable isotopic exchange in clay-water systems is complex,

and manifested differently in clay-dominated shales and

mudstones (where porewater is most affected) versus fluid-

dominated systems such as porous and permeable sandstones

(where clay is most affected). Exchange of oxygen isotopes

between water and structural sites in clay minerals is extremely

slow in low-temperature environments, including modern

ocean sediments (Savin and Epstein, 1970a,b; Yeh and

Eslinger, 1986). Without neoformation, oxygen-isotope

exchange is insignificant in marine environments, even for

very fine (<0.1 |im), smectite-rieh clay assemblages. The rate of

oxygen-isotope exchange between clay minerals and water

increases as temperatures rise, with significant exchange occur-

ring by 300°C. Mineralogical reactions, such as the smectite to

illite conversion, facilitate this behavior (Yeh and Savin, 1977).

Hydrogen-isotope exchange between clay minerals and

water also occurs slowly at low temperatures, but the rate is

faster than for oxygen. Measurable hydrogen-isotope exchange

can take place at <100°C; by 200°C, extensive hydrogen-

isotope exchange can occur unaccompanied by oxygen isotopic

effects. Some studies suggest that hydrogen-isotope exchange

between seawater and detrital clay minerals from 2 million to

3 million year old ocean sediments was restricted to the

<0.1 nm size-fraction (Yeh and Epstein, 1978); others suggest

that exchange with expandable clay minerals occurs readily at

surface temperatures. Significant post-formational hydrogen-

isotope exchange between non-expandable clays and pore-

water can also occur over millions of years at 30°C to 40°C

without a corresponding effect on the oxygen-isotope compo-

sition of the clay (Longstaffe and Ayalon, 1990).

Applications of equilibrium stable

isotope fractionation

Stable isotope geothermometry underlies the vast body of

research in which oxygen-isotope compositions of ice cores,

foraminifera, brachiopods, diatoms and biogenic phosphate in

the oceans, and speleothem carbonates on land are used to

reconstruct the Earth's past climate (Shackleton, 1968;

Emiliani and Shackleton, 1974; Sehwarcz, 1986; Bowen,

1991;

Petit etal., 1999; McDermott etal., 2001; Norris etal.,

2002;

Stanton et al., 2002). These stable isotope proxies for

temperature are coupled with radiometric, biostratigraphic

and paleomagnetic information, which provide an absolute

chronology for the scale and rate of climate change.

Stable isotope geothermometers for oxygen and hydrogen

are also used to identify the type(s) of water (seawater, fresh

water, brine) involved in formation of chemical sediments,

provided that independent temperature data are available (e.g.,

see equation 13). Likewise, this approach has been combined

with stable carbon-isotope, ^'Sr/^'^Sr and "*'Nd/""'Nd ratios to

explain dolomitization, and to identify multiple fluid regimes

responsible for diagenesis of carbonate rocks (e.g.. Land, 1980,

1991;

Banner etal., 1988; Coniglio etal., 1988; Chaftez and

Rush, 1995; Benito etal., 2001).

The stable isotopic eompositions of carbonate and silicate

cements from siliciciastic rocks have been used in a similar

fashion. A common application is to deduce the mixing and

movement of seawater, meteoric water and their evolved brines

on basinwide scales over time, and to relate this behavior to the

generation, migration and degradation of hydrocarbons (e.g.,

Longstaffe and Ayalon, 1987, 1991; Ayalon and Longstaffe,

1988;

Tilley and Longstaffe, 1989; Longstaffe, 1993, 1994,

2000;

Ziegler and Longstaffe, 2000a,b). Neoformed minerals

(pedogenic clays, hydroxides, carbonates) from paleosols and

regoliths can also be used to deduce the 5'*0 and ^D values of

ancient meteoric water. These results provide critical insight

into continental paleoclimate because the oxygen- and hydro-

gen-isotope compositions of meteoric water are strongly

determined by temperature and the precipitation-evaporation

regime at the time of soil formation (Lawrence and Taylor,

1971,

1972; Bird etal., 1992, 1994; Mizota and Longstaffe,

394

ISOTOPIC METHODS IN SEDIMENTOLOCY

1996;

Stern etal., 1997; Savin and Hsieh, 1998; Yapp, 1998;

Girard etai,, 2000).

Kinetic stable isotope fractionation and

its applications

Kinetic isotope effects, which occur during incomplete and

unidirectional processes, are a second important source of

stable isotopic fractionation in sedimentary systems. These

occur during rapid evaporation of water, and are also

especially important in biologically mediated systems. For

example, '^C02 is selected preferentially over "CO2 during

photosynthesis (Hoefs, 1997). The magnitude of the fractiona-

tion depends mostly on reaction rates and pathways. For

example, significantly larger stable carbon-isotope fractiona-

tions characterize the C3 versus the C4 photosynthetic path-

ways.

Hence, stable carbon-isotope analysis of ancient organic

matter and pedogenic carbonates in sediments and soils can

provide significant paleoecological and paleoclimatic informa-

tion (Cerling, 1984; Cerling etal., 1989, 1997),

Kinetic isotope fractionation associated with microbial

activity in sedimentary systems is of special importance. The

stable carbon- and hydrogen-isotope compositions of methane

and carbon dioxide emanating from wetlands, for example,

can be used to identify the reaction pathways that produce

these greenhouse gases (Hornibrook et al,, 1997), As discussed

earlier, many early diagenetic processes are also driven by

microbially mediated reactions. These produce characteristic

stable carbon- and sulfur-isotope compositions in secondary

carbonates and pyrite (Irwin etal,, 1977; Curtis etal., 1986;

Longstaffe, 1989; Habieht and Canfield, 1997, 2001; Canfield,

2001),

A full understanding of these reactions allows stable

carbon and sulfur isotopic data for these phases, particularly

from concretions, to be used to reconstruct ancient deposi-

tional environments. Likewise, one can deduce the geochem-

ical conditions at and beneath the sediment-water interface

during early diagenesis. Depositional and early diagenetic

interplay between fresh and marine waters as sea level

tluctuated in shallow epeiric seas can also be elucidated, using

the stable carbon-, sulfur-, and oxygen-isotope results for

secondary minerals (Irwin et al,, 1977; Curtis et al,, 1986;

Longstaffe, 1987, 1989, 1994; Mozley and Burns, 1993;

McKay etal,, 1995; McKay and Longstaffe, 2002),

Equilibrium oxygen-isotope mineral-water geothermo-

meters are generally used to identify the water involved during

formation of early diagenetic minerals in sedimentary systems.

However, microbial activity is known to mediate the pre-

cipitation of some oxygen-bearing phases, such as siderite. The

oxygen-isotope fractionation associated with microbially

produced siderite at low temperatures is different than the

equilibrium (inorganic) fractionation for this system. Use of

the latter will result in calculated 5'^O values for porewater

that are lower than produced by the former; this can lead to

incorrect identification of paleoenvironments (Mortimer and

Coleman, 1997), The potential consequences of such microbial

interventions on mineral-water oxygen isotopic fractionation

in sedimentary and early diagenetic environments remain to be

explored fully.

Conclusion

The use of radioactive, radiogenic and stable isotopes is now

common in sedimentology. These methods are used to study

rates of deposition and erosion, characterize depositional

environments, identify the provenance and dispersal of

sediment, make regional and global chemostratigraphic

correlations, understand diagenetic processes and geochemical

conditions at and near the sediment-water interface, define the

timing and nature of regional fluid flow in sedimentary basins,

deduce global paleoclimate both in terrestrial and marine

environments, infer variations in global biogeochemical cycles

and their significance, and make global paleogeographic and

tectonic reconstructions. The most important advances over

the last decade are the ease with which isotopic measurements

can now be made, and the fact that a comprehensive approach

to the use of isotopic methods is now the norm. Multiple

isotopic chronometers and tracers are employed together to

test and constrain hypotheses much more rigorously than

possible from use of a single isotopic tracer or geochron-

ometer. The hallmark of effective utilization of isotopic

methods is their integration with sedimentological, strati-

graphic and other geological data.

Fred J. Longstaffe

Bibliography

Arthur, M.A,, Anderson, T,F,, Kaplan, LR., Veizer, J., and Land, L.S.,

1983,

Stable Isotopes in Sedimentary Geology. Tulsa: SEPM Short

Course, 10,

Asmerom, Y,, Jacobsen, S.B,, Knoll, A,H,, Butterfield, N,J,, and

Swett, K,, 1991, Strontium isotopic variations of Neoproterozoic

seawater: implications for crustal evolution,

Geochimica

et Cosmo-

chimica Acta, 55: 2883-2894,

Ayalon, A,, and Longstaffe, F,J,, 1988, Oxygen-isotope studies of

diagenesis and pore water evolution in the Western Canada

Sedimentary Basin: evidence from the Upper Cretaceous basal

Belly River sandstone. Alberta, Journal of Sedimentary Petrology,

58:

489-505,

Ayalon, A,, and Longstaffe, F.J., 1992, Isolation of diagenetic silicate

minerals in clastic sedimentary rocks for oxygen isotope analysis: a

summary of methods. Israel Journal of Earth Sciences, 39: 139-

148,

Awwiller, D.N,, and Mack, L.E., 1991, Diagenetic modification of

Sm-Nd model ages in Tertiary sandstones and shales, Texas Gulf

Coast. Geology, 19: 311-314.

Baker, A,J,, and Fallick, A.E., 1989, Heavy carbon in two-billion-year-

old marbles from Lofoten-Vesteralen, Norway: implications for

the Precambrian carbon cycle, Geochimica et Cosmochimica Acta,

53:

1111-1115,

Banner, J.L,, Hanson, G.N,, and Meyers, W,J,, 1988, Rare earth

element and Nd isotopic variations in regionally extensive

dolomites from the Burlington-Keokuk formation (Mississippian):

implications for REE mobility during carbonate diagenesis, Jour-

nalof Sedimentary Petrology, 58: 415-432,

Basu, A.R,, Jacobsen, S.B,, Poreda, R.J,, Dowling, C,B,, and

Aggarwal, P.K,, 200L Large groundwater strontium flux to the

oceans frorn the Bengal Basin and the marine strontium isotope

record. Science, 293: 1470-1473,

Beck, J,W, etal., 2001, Extremely large variations of atmospheric

'""C concentration during the last glacial period. Science, 292:

2453-2458,

Benito, M,I,, Lohmann, K.C., and Mas, R., 2001. Discrimination of

multiple episodes of meteoric diagenesis in a Kimmeridgian reefal

complex. North Iberian Range, Spain, Journal of Sedimentary

Research, 71 : 380-393,

Bierman, P,R,, and Turner, J,, 1995, '"Be and ^''Al evidence for

exceptionally low rates of Australian bedrock erosion and the likely

existence of pre-Pleistocene landscapes. Quaternary Research, 44:

378-382,

Bird, M,I,, Longstaffe, F,J,, Fyfe, W,S,, and Bildgen, P,, 1992,

Oxygen-isotope systematics in a multiphase weathering system in

Haiti, Geochimica et Cosmochimica Acta, 56: 2831-2838,