composition of the earth and its reservoirs: geochemical...

44
Composition of the Earth and its reservoirs: Geochemical observables Cin-Ty A. Lee Rice University MYRES-I 2004

Upload: dodat

Post on 13-Apr-2018

220 views

Category:

Documents


4 download

TRANSCRIPT

Composition of the Earth and its reservoirs:

Geochemical observables

Cin-Ty A. Lee

Rice University

MYRES-I 2004

The Earth is dynamic and heterogeneous

Not to scale

?CoreMantle

ContinentalCrust

Base of lithosphereOceanicCrust

Atmosphere

Ocean

Plum

e

Mid-oceanRidge

5

10

0

P (G

Pa)

Temp (oC)500 1000 1500 2000

100

200

300

Depth (km

)

H2 O

Saturated

B

5

10

0

P (G

Pa)

Temp (oC)500 1000 1500 2000

100

200

300

Depth (km

)

H2 O

Saturated

5

10

0

P (G

Pa)

Temp (oC)500 1000 1500 2000500 1000 1500 2000

100

200

300

Depth (km

)

H2 O

Saturated

B

The Earth is differentiating:But are we mixing or unmixing reservoirs?

Not to scale

?CoreMantle

ContinentalCrust

Base of lithosphereOceanicCrust

Atmosphere

Ocean

Plum

e

Mid-oceanRidge

Gastronomy

Tackley, 2000

Definition of heterogeneity

Heterogeneity: a qualitative term describing how “well-mixed” a system is

Reservoirs: chemically distinct regions in a system that have physically defined boundaries

At what point do we consider a heterogeneity a reservoir?

SamplingLengthscale

L1 ~ lo

L2 > lo

L3 >> lo

Tell me what lengthscale you’re interested in…

Tackley, 2000

3000 km

SP

CPX

OL

OPX

1 cm

Forms of compositional heterogeneities

Major - wt. %

Minor - ~0.1 wt. %

Trace - <100 ppm

Isotopic

What types of compositional heterogeneities lead to variations in

physical parameters?

Three Steps to Bliss

1. Bulk Earth Composition

2. Composition of reservoirs

3. Quantifying the size and distribution of reservoirs

Bulk Earth

Bulk SilicateEarth

“PrimitiveMantle”

Core

DepletedMantle

PrimitiveMantle

Continental& Oceanic

Crusts

Core

+

+

+

In theBeginning TodayFirst 30 Ma

Bulk Earth

Bulk SilicateEarth

“PrimitiveMantle”

CoreCore

DepletedMantle

DepletedMantle

PrimitiveMantle

PrimitiveMantle

Continental& Oceanic

Crusts

CoreCore

+

+

+

In theBeginning TodayFirst 30 Ma

STEP 1Towards a Bulk Earth

Composition

Bulk Silicate Earth(Primitive Mantle)

+

Core

What types of samples can we work with?

Mantle rocks (xenoliths, massifs, ophiolites)

Lavas/Magmas

Sediments

Meteorites?

0

1

2

3

4

5

6

7

35 40 45 50

0

1

2

3

4

5

6

7

35 40 45 50

CaO

wt.

%A

l 2O3

wt.

%

MgO (wt. %)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

35 40 45 50

Na 2

O w

t. %

Melting trend

MANTLE PERIDOTITES

The search for the holy grail

Do samples of “Primitive Mantle” exist?

Can we use “primitive” meteorites, e.g. undifferentiated meteorites as a proxy for the

undifferentiated Earth?

Yes and No

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

0.5 1.0 1.5 2.0

.

Mg/Si

Al/Si

EHEL LL

L HCR CICOCM

CVCK

BSE

Terrestrialarray Cosmochemical

Fractionationtrend

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

0.5 1.0 1.5 2.0

.

Mg/Si

Al/Si

EHEL LL

L HCR CICOCM

CVCK

BSE

Terrestrialarray Cosmochemical

Fractionationtrend

Drake and Righter, 2002; after:

Clayton, R. N., Mayeda, T. K., Goswami, J. N. & Olsen, E. J. Oxygen isotope studies in ordinary chondrites. Geochim. Cosmochim. Acta 55, 2317-2337 (1991).

42.

Clayton, R. N. Oxygen isotopes in meteorites. Annu. Rev. Earth Planet. Sci. 21, 115-149 (1993).

41.

Clayton, R. N. & Mayeda, T. K. Oxygen isotope studies in carbonaceous chondrites. Geochim. Cosmochim. Acta 63, 2089-2104 (1999).

40.

It is safe to say thatEarth is derived from Earth-like

materials

. . . but all is not lost

CI chondrites

10-2 1 102 104 106 108

108

106

104

102

1

10-2

He H

N

C O

FeMg

SAl

NiCo

TiCu

GeSr

BRb

PbLiBe

ThSola

r Ele

men

tal a

bund

ance

(per

106

Si a

tom

s)

CI carbonaceous chondrite(per 106 Si atoms)

10-2 1 102 104 106 108

108

106

104

102

1

10-2

He H

N

C O

FeMg

SAl

NiCo

TiCu

GeSr

BRb

PbLiBe

ThSola

r Ele

men

tal a

bund

ance

(per

106

Si a

tom

s)

CI carbonaceous chondrite(per 106 Si atoms)

The Early Evolution of the Inner Solar System: A Meteoritic Perspective

C. M. O'D. Alexander, A. P. Boss, R. W. Carlson

SCIENCE, Volume 293, Number 5527, pp. 64-68

H He

Li 1225

Be B 964 C N O F 736

Ne

Na 970

Mg 1340

Al 1680

Si 1311

P 1267

S 648 Cl Ar

K 1000

Ca 1520

Sc 1644

Ti 1590

V Cr 1300

Mn 1190

Fe 1336

Co 1351

Ni 1354

Cu 1037

Zn 660

Ga 997

Ge As 1157

Se 684

Br Kr

Rb 1080

Sr Y Zr 1750

Nb Mo 1600

Tc Ru 1600

Rh Pd 1334

Ag 952

Cd In 470 Sn 720

Sb 912

Te 680

I Xe

Cs Ba La-Lu Hf Ta W 1800

Re 1800

Os 1800

Ir 1600

Pt 1411

Au 1225

Hg Tl 428 Pb496

Bi 451

Po At Rn

Fr Ra Ac-Lr

La 1500

Ce Pr Nd Pm Sm Eu 1290

Gd Tb Dy Ho Er Tm Yb 1420

Lu 1590

Ac Th1590

Pa U 1540

Np Pu Am Cm Bk Cf Es Fm Md No Lr

Refractory, >1400 K Moderately volatile ~800-1250 K

Transitional ~1250-1350 K Highly volatile <800 K

50% condensation temperatures (pressure 1E-4 bars)

Lithophile – silicate loving

Siderophile – Fe loving

Atmophile - atmosphere

Planetary differentiation

Refractory

Moderately volatile

Volatile

Nebular differentiation

PM

Establish concentrations in “Primitive Mantle” using refractory lithophile element ratios for chondrites

Fig. 6 fromMcDonough & Sun (1995)

STEP 2. RESERVOIR DOGS

Not to scale

?CoreMantle

ContinentalCrust

Base of lithosphereOceanicCrust

Atmosphere

Ocean

Plum

e

Mid-oceanRidge

xx

WMDs

Partial melting as a major differentiation process

A

P (G

Pa)

Temp (oC)

Depth (km

)

B500

1000

1500

2000

2500

0

1

2

3

4

5

6

7

0

50

100

150

200

Dry solidus

Liquidus

Dry meltingregime

Mid-ocean Ridge

Major Element Effects

BulkSilicateEarth

0

5

10

15

20

25

30

35

40

45

0 10 20 30 40 50 60 70 80

basalt

Continental CrustAndesite

Dacite Rhyolite

MgO

(wt.

%)

SiO2 (wt. %)

Lower Crust

Partition coefficient

Solid residue

melt

D = CsolidCmelt

D > 1 compatible in solid

D < 1 incompatible in solid

0.01

0.1

1

10

100

1000

Cs Rb Ba Th U Nb La Ce Pr Sr Nd Zr Hf SmGd Tb Dy Ho Y Er Yb Lu

MORB

Plume

Cont.Crust

Melts and crustnormalized toprimitive mantle

Increasing DMore compatibleincompatible

PrimitiveMantle

Using magmas as “windows” to the mantle

MORB source is depleted in highly incompatible elements (DMM = depleted MORB mantle)

Continental Crust is enriched in highly incompatible elements

Plume is enriched or more primitive in character

Trace-elements may suffer from fractionation during magmatic processes

Isotopes in general are NOT fractionated

Solid residue

meltSm/Nd < 1

Sm/Nd > 1

Rb/Sr > 1

Rb/Sr < 1

Cont. Crust

Depleted Mantle

87Rb – 87Sr

time

BulkEarth

time

BulkEarth

4.559 Gy 0

87S

r/86S

r

Depleted Mantle

Cont. Crust

147Sm – 143Nd

time

BulkEarth

time

BulkEarth

4.559 Gy 0

143 N

d/14

4 Nd

Depleted Mantle radiogenic Ndunradiogenic Sr

Continental Crustunradiogenic Ndradiogenic Sr

Hofmann, 1997Nature 385: 219-229

At least the upper part of the Earth’s mantle is depleted in highly incompatible elements

This depleted portion appears to be complementary to the continental crust

How much of the mantle is depleted?

Mass Balance Magic

Cont. Crust MORB PM DMM DMM/PM Relative Size

of DMMCs 2.6 0.007 0.021 0.0007 0.0 0.7Rb 58 0.56 0.6 0.056 0.1 0.6Ba 390 6.3 6.6 0.63 0.1 0.3U 1.42 0.047 0.0203 0.0047 0.2 0.4K 15606 600 240 60 0.3 0.4Sr 325 90 19.9 9 0.5 0.1

Concentration (ppm)

Possibly 30 to 70% of mantle has been

processed to make Cont. Crust

Mass Balance Says NOTHING about

geometry

1.0

Relative Size of DMM0.9

0.80.7

0.6

0.5

0.40.3

0.2

0.10.0

Cs Rb Ba U K Sr

What do hotspots tell us?(Ocean Island Basalts – OIB)

Hofmann, 1997Nature 385: 219-229

Hotspot magmas appear to be variably enriched in incompatible elements

Source feature

Or

Sampling problem?

87Sr/86Sr 143Nd/144Nd

Sampling problem is a possibility

BUT average isotopic composition of some OIBs differ from that of MORBs

OIB source must differ slightly in composition from that of MORB source

IF OIBs are associated with plumes +IF plumes derive from the lower mantle

The lower mantle must be slightly enriched.

Do plumes exist?

Meibom & Anderson 2003

Other Reservoirs in the Silicate Earth

Subducted oceanic crust

SCLM – subcontinental lithospheric mantle

Subducted sediment

Primordial mantle???

Tackley, 2000

Where do we go from here?

More of the same data?

New data?

NEW QUESTIONS?

YOU can be an armchair geochemist

Use compiled datasetsGEOROC

RidgePetDBGERM

But don’t abuse the data sets

New directions New data

DIRECTION 1

Is a homogeneous Bulk Silicate Earth a valid assumption?

Was the primordial mantle stratified? If so, how does this affect geochemical mass balance conditions?

Is the lower part of the mantle Fe-rich?

Can we ignore mass fluxes across the core-mantle boundary?

What data may be able to answer these questions?

Short-lived radionuclides182Hf-182W, 146Sm-142Nd, 107Pd-107Ag

More precise data on first series transition elementsFe, Mn, Ni, Co, Cr, V, Sc

Direction 2

Constrain the size and distribution of reservoirs

Geochemistry, Geophysics, Geodynamics, Petrology

G3PDo geochemical heterogeneities reflect variations in

physical properties that are detectable by geophysical methods?

Clearly, major-element variations affect physical properties of the mantle.

The physical effects of trace-element and isotopic heterogeneities are unclear. However, trace-element and isotopic heterogeneities may be correlated with variations in oxygen fugacity and water content, both of which may have profound effects on elasticity, conductivity, and viscosity.

2FeO + O2 = Fe2O3

Future?

• Effects of fO2 and H2O on physical properties of peridotite

• Characterizing the variation of fO2 and H2O in the mantle and what processes control these parameters

Geochemical research onpartitioning behavior of redox sensitive elementsvalence state of redox-sensitive elementsquantification of H2O content in the mantle

Conclusions

Geochemistry has provided a considerable knowledge base for our understanding of how the Earth works.

Further progress will require interdisciplinary collaboration and the generation of new questions and areas of focus.

More geochemical data will obviously be helpful, but existing databases should be mined to reveal areas that need more refinement.