Microsoft Word - Electronic Appendix B_1-27-11.doc - Electronic_Appendix_B

Electronic appendix

Article: Pilet et al. (2011) Monte Carlo simulations of metasomatic enrichment in the lithosphere and implications for the source of alkaline basalts

Electronic A

ppendix

: Evolution

of isotopic compositions as a function of

time for metasomatized hydrous lithosphere

Electronic Appendix

, we describe the method

used to calculate the isotopic evolution of

metasomatized

hydrous lithosphere

assum

ing

that the

liquids which generate this metasomatic enrichment are

produced by

partial melting of a MORB source

(Fig. 1; main article)

As indicated in the article, two possible scenarios have

been proposed for the production of alkaline magmas by melting

of metasomatized lithosphere:

(

1) Shortly after

or coincident with metasomatism, the lithosphere experiences a thermal perturbation or decompression and

thereby melts

in situ

; or (2) the metasomatized lithosphere is recycled into the convecting mantle by

subduction

or delamination and melts during later upwelling (e.g., in a plume or at a ridge).

Therefore, the time interval

between the metasomatism of this lithosphere and its subsequent melting (which produces the alkaline lavas)

could range from 0 to 1

–

Ga (see the main article for further explanation).

Here we provide two examples of calculations for isolation times of 0.15 Ga and 1.5 Ga.

The ranges in Sr, Nd,

and Pb isotopic compositions for metasomatized lithosphere were calculated in two steps: (1)

We calculated the

isotopic composition of depleted MORB mantle (DMM to E

DMM) 0.15 Ga

or 1.5 Ga

ago assuming that the

isotopic compositional range of these two MORB mantle end members was similar to the range of isotopic

compositions observed in MORB toda

and that they were formed by an instantaneous depletion event at 2.5

Ga. (2) The isotopic evolution of the metasomatized lithosphere from 0.15 Ga

or from 1.5 Ga,

to the present was

calculated using the Rb/Sr, Sm/Nd, and U/Pb ratios calculated for each

of the 25,000 Monte Carlo simulations

Present day i

sotopic composition of depleted mantle

We estimated the present day isotopic composition for DMM and E

DMM in order to reproduce the

compositional range observed in MORBs (data from

the

PetDB databa

se)

. Since

mixing between two fixed

composition

s will not

reproduce the complete composition

range observed in MORB

, we introduce

some

isotope variability to our estimated DMM and E

DMM compositions;

neodymium, strontium, and lead isotopic

compositions

for D

MM and E

DMM are shown in Fig.

Fig.

1. a

)

Sr/

Sr versus

143

Nd/

144

Nd and b

)

206

Pb/

204

Pb versus

207

Pb/

204

Pb isotopic diagrams for the estimated composition

of DMM and

DMM sources compared with MORB data (from

the

PetDB databa

se). The

light

d points (forming a solid light

red field) correspond to the present day calculated isotopic compositions of

the model

metasomatized hydrous lithosphere

The Rb/Sr, Sm/Nd

and U/Pb ratios of DMM and E

DMM were estimated assuming that these two re

servoirs

were produce simultaneously at 2.5 Ga from a primitive source.

Figure

2 shows the isotopic compositions of

these two res

ervoirs as a function of time.

For comparison, we have also plotted the Sr, Nd

and Pb isotopic

evolution of

the

MORB source

estimated by several other authors (see caption to Fig.

2).

Electronic appendix

Article: Pilet et al. (2011) Monte Carlo simulations of metasomatic enrichment in the lithosphere and implications for the source of alkaline basalts

Fig.

2. a)

Sr/

Sr, b)

143

Nd/

144

Nd, c)

206

Pb/

204

Pb, and d)

207

Pb/

204

Pb isotopic compositions versus time for estim

ated DMM and

DMM sources. P

anel

a, b, c

, and d also show

the evolutio

n of isotopic compositions for

the

MORB source (DMM) estimated

Rehkamper and Hofmann (1997)

Stracke

et al.

(2003)

, and

Workman and Hart (2005).

The differences in the Rb/Sr and

Sm/Nd ratios of these different works illustrate the difficulty in constrai

ning the composition

of the DMM mantle at 2.5 Ga

panel

and b,

the isotopic

evolution

of bulk silicate Earth (

BSE

)

was calculated using the Rb/Sr

and Sm/Nd ratios

and present

day is

otopic compositions reported by W

orkman and Hart (2005)

; dashed line

in panel a and horizontal line at

ε

Nd = 0 in panel

2) Isotopic composition of metasomatic

veins

isolated for 0.15 Ga

To illustrate how we calculate

the isotopic

evolution of a specific composition of

metasomatized hydrous

lithosphere

isolated for 0.15

Ga, we use

the example described in

Electronic

Appendix

a) Composition of the source

In the example described in

Electronic

Appendix

, the source is composed of 51.9 % of the E

DMM

component (

the remainder is DMM). We used these proportions and

est

imates of the isotopi

c compositions of

DMM and E

DMM to calculate

the isotopic composition and the Rb/Sr, Sm/Nd, and U/Pb ratios

of the source.

Parameters of the mantle source (assuming that this source is composed of 51.9% E

DMM and 48.1% DMM):

Isotopic composition of the

mantle

source 0.15 Ga ago

Using the trace

element ratios an

estimate

of the isotopic composition of the source, we can calculate

its

isotopic composition

at 0.15 Ga.

Electronic appendix

Article: Pilet et al. (2011) Monte Carlo simulations of metasomatic enrichment in the lithosphere and implications for the source of alkaline basalts

Mantle source isotopic composition at 0.15 Ga:

Thes

e ratios correspond to the isotopic composition of the metasomatic vein formed 0.15 Ga ago.

c) Isotopic evolution of

metasomati

zed

hydrous lithosphere

from 0.15 Ga to the present

The

calculated Rb/Sr, Sm/Nd, and U/Pb ratios for

the

metasomatized hydrous

lithosphere

(hydrous cumulates +

trapped liquid)

shown in Fig. 5d (in the main article) were used to evaluate the isotopic evolution of the

metasomatized lithosphere. By using only these ratios, w

have neglected the role of the much larger volume of

unme

tasomatized lithosphere on the isotopic evolution of a parcel of heterogeneous lithosphere that consists of a

small volume of a metasomatized hydrous component (hydrous cumulates + trapped liquid) and a much larger

volume of umetasomatized material. Two ar

guments support this assumption: (1)

the

metasomatized hydrous

lithosphere

is characterized by high trace

element

contents relative to DMM and E

DMM, and thus the Rb/Sr,

Sm/Nd, and U/Pb ratios of a heterogeneous

lithosphere

will essentially be those of the

metasomati

zed hydrous

lithosphere; (2)

the relatively low temperatures in the lithosphere and the relatively short time scale of this

calculation

(0.15 Ga) limits the diffusive

interaction between metasomatized

and unmetasomatized lithosphere.

Isotopic

compositions of

metasomatized hydrous lithosphere

isolated for 0.15 Ga:

* The Rb/Sr, Sm/Nd, and U/Pb ratios are calculated using the composition of

metasomatized hydrous

lithosphere

from our Monte Carlo si

mulation (reported in

Appendix A

of the supple

mentary information and in Fig.

of the main article).

ese four

isotopic

ratios

represent one of

the light

gray

point

s shown

in Fig. 9a

, panels I

(in the main

article)

; light gray because the example in Appendix A has 18% trapped liquid.

A similar

calculation was used

to evaluate the isotopic evolution of

metasomatized hydrous lithosphere

over a

period of

0.5 Ga

; results are

shown in Fig. 9b

, panels I

3) Isotopic composition of metasomatic lithosphere isolated for 1.5 Ga

While melting of met

asomatized lithosphere “

in situ

”, i.e., prior to being

cycled into the convecting mantle, is

capable of matching the isotopic compositions of many alkaline lavas observed in continental or oceanic

setting

, some alkaline magmas exhibit more extreme isoto

pic compositions and it has been suggested that such

compositions can be produced by

cycling the metasomatized lithosphere through the convecting mantle and

thus greatly increasing the time for in

growth of extreme isotopic ratios (Halliday

et al.

, 1995

;

Niu & O'Hara,

2003

;

Pilet

et al.

, 2005). Here we calculate the isotope evolution of recycled

metasomatized hydrous lithosphere

that is isolated for 1.5

Ga. In contrast to the calculation above (involving a 0.15 Ga isolation time), here we

cannot neglect t

he composition of the unmetasomatized lithosphere, since for isolation times of 1 to 2 Ga, some

sizable fraction of the metasomatic veins can be expected to re

equilibrate with the surrounding peridotite

(Kogiso

et al.

, 2004). Thus we use the weighted aver

age composition of a parcel of

heterogeneous lithosphere

that consists of

metasomatized hydrous lithosphere

and

depleted peridotite

(assumed

to have a composition equal

DMM).

This heterogeneous lithosphere we define as metasomatized peridotite.

Based on

field observations of

metasomatic veins

in the French Pyrrenees

(in Lherz in particular), we assume that the

metasomatized hydrous

lithosphere

comprises 1 % of the total parcel of heterogeneous lithosphere.

a) I

sotopic composition of the

mantle

source 1

5 Ga ago

Electronic appendix

Article: Pilet et al. (2011) Monte Carlo simulations of metasomatic enrichment in the lithosphere and implications for the source of alkaline basalts

The mantle source composition is calculated using the same approach that we used in the example above (for an

isolation time of 0.15 Ga). The composition of this deple

ted mantle source at present and 1.5 Ga ago is

This composition corre

sponds to

the isotopic composition of

metasomatic veins formed 1.5 Ga ago.

b) Rb/Sr, Sm/Nd, U/Pb trace

element ratios of metasomatized

peridotite

We estimate the trace

element ratios of metasom

atized peridotite consisting of 1%

metasomatized hydrous

ithosphere

embedded in 99%

depleted peridotite similar to DMM. Based on the estimate

DMM composition

and

the composition of

metasomatized hydrous lithosphere

(equal to

metasomatic veins plus metasomatized

peridotite)

we calculated the Rb/Sr, Sm/Nd,

and U

/Pb trace

element ratios

of re

equilibrate metasomatized

peridotite.

*The Rb/Sr, Sm/Nd

, and U/Pb

of DMM correspond to the value

estimate

in Fig

The Rb, Sm

, and U

content

s of DMM were

calculated

using

our estimate

of Rb/Sr,

Sm/Nd, and U/Pb t

race

element ratios and the Sr,

Nd, and

Pb content

reported by

Workman and Hart (2005) for DMM.

Using

the trace

element ratios that control the evolution of each of the isotopic systems, we recalculate the

isotopic composition of metasomatized peridotite

at 1.5 Ga.

c) I

sotopic evolution of metasomatized

peridotite from 1.

5 to the present

After 1.5 Ga

the isotopic ratios of t

he metasomatized peridotite were

calculated using the initial isotopic ratios

and the corresponding trace

element ratios.

This isotopic composition

represents one of

the light point

s shown in Fig.

c, panels I

(in the main article)

For long periods of isolation, the fraction of metasomatized

hydrous lithosphere

in the bulk system

(

depleted

mantle +

metasomatized hydr

ous lithosphere

)

controls the isotopic composition of the

metasomatized peridotite

(i.e., the

bulk system

)

. The following table lists the isotopic composition

of metasomatized peridotite after 1.5

Ga that contains between 1 and 100%

metasomatized hydrous

lithosphere

Electronic appendix

Article: Pilet et al. (2011) Monte Carlo simulations of metasomatic enrichment in the lithosphere and implications for the source of alkaline basalts

The calculation shows, that as the fraction of

metasomatized hydrous lithosphere

increases, the isotopic

compositions become more extreme given a 1.5 Ga isolation time.

References

Halliday, A. N., Lee, D.

C., Tommasini, S., Davies, G

. R., Paslick, C. R., Fitton, J. G. & James, D

. E. (1995). Incompatible

trace

elements in OIB and MORB and source enrichment in the sub

oceanic mantle.

Earth and Planetary Science Letters

133

, 379

395.

Kogiso, T., Hirschmann, M. M. & Reiners, P. W. (2004).

Length scales of mantle heterogeneities and their relationship to

ocean island basalt geochemistry.

Geochimica et Cosmochimica Acta

, 345

360.

Niu, Y.

& O'Hara, M. J. (2003).

Origin of ocean island basalts: A new perspective from petrology, geochemistry

, and mineral

physics considerations.

ournal of Geophysical Research

108

, 220

i:10.1029/2002JB002048

Pilet, S., Hernandez, J., Sylvester, P. & Poujol, M. (2005). The metasomatic alternative for ocean island basalt chemical

heterogeneity.

Earth and P

lanetary Science Letters

236

, 148

166.

Rehk

mper, M. & Hofmann, A. W. (1997). Recycled ocean crust and sediment in Indian Ocean MORB.

Earth and Planetary

Science Letters

147

, 93

106.

Stracke, A., Bizimis, M. & Salters, V. J. M. (2003). Recycling oceanic cr

ust: Quantitative constraints.

Geochemistry,

Geophysics, Geosystems

8003,

doi:10.1029/2001GC000223.

Workman, R. K. & Hart, S. R. (2005). Major and trace element composition of the depleted MORB mantle (DMM).

Earth

and Planetary Science Letters

231

, 53