Transcript an analogue study from - The University of Texas at Dallas

Crust formation in the early Earth
prior to the onset of subduction
-an analogue study from
the Oman ophiolite
Hugh Rollinson
Sultan Qaboos University, Oman
Given that
(a) The Earth’s earliest crust was mafic, and
(b) The Earths early continental crust was
TTG in composition
Given that
(a) The Earth’s earliest crust was mafic, and
(b) The Earths early continental crust was
TTG in composition
It is proposed that the mechanisms whereby
felsic rocks (trondhjemites) form from mafic
crust in a geologically recent oceanic
environment is a powerful analogue for the
formation of felsic crust in the Hadean and
early Archaean
Given that
(a) The Earth’s earliest crust was mafic, and
(b) The Earths early continental crust was
TTG in composition
It is proposed that the mechanisms whereby
felsic rocks (trondhjemites) form from mafic
crust in a geologically recent oceanic
environment is a powerful analogue for the
formation of felsic crust in the Hadean and
early Archaean
A consequence of this proposal is that such
crust was not created in a subduction setting
Historically this idea had been neglected, because
Ophiolitic trondhjemites
thought to be the product
of crystal fractionation
and so of minor importance
Historically this idea had been neglected, because
Ophiolitic trondhjemites
thought to be the product
of crystal fractionation
and so of minor importance
1000
LILE and LREE depleted
Av TTG
100
Av CC
10
Oman trondhjemites
1
0.1
Rb Ba Th U
K Ta La Ce Pb Sr Nd Hf Zr Sm Ti Y Yb
Ophiolitic and Archaean
trondhjemites have
different compositions
So what has changed …….?
1. New models
of oceanic
magma chambers
large long-lived
magma chambers
are out
So what has changed …….?
2. New experimental
evidence for water
saturated partial
melting of oceanic
gabbro at low
pressures
(data from Koepke et
al. 2004 @ 0.2 Gpa)
An
experimental
melt compositions
Albite
enrichment
To
Tdh
Ab
Gd
Gr
Or
So what
has
changed …….?
3. Field evidence for
partial melting
- the occurrence
of hornblende gabbros
- abundant agmatites
So what
has changed?
4. Geochemical
evidence for
high temperature
sea water
interaction
with mafic
and felsic melts
Oman amphiboles have
elevated chlorine
(Coogan 2003)
trondhjemite (WR)
Elevated
Sr-isotope
ratios
Elevated
O-isotope
ratios
trondhjemite (WR)
Plagioclase-amphibole
temperatures 830-970 C
(Bosch et al 2004)
Ophiolite data from Bosch et al. (J. Pet.,2004) with Trondhjemite data from
Gregory and Taylor (JGR, 1981)
Geochemical similarities…………
1000
NMZ
LIMPOPO
BELT
LEWISIAN
100
10
OMAN
1
Oman vs Archaean TTG
0.1
0.1
1
10
Y (ppm)
Low Sr; high Y
Oman trondhjemites overlap
Limpopo belt TTGs, and to
a lesser extent the classical
TTG trend (eg Lewisian)
100
Geochemical similarities…………
Low Sr; low MgO
Similar to early Archaean
but not late Archaean TTGs
1000
NMZ
LIMPOPO
BELT
LEWISIAN
100
2000
1500
10
OMAN
1000
t >3.0 Ga
t < 3.5 Ga
500
t >3.5 Ga
1
Oman
Oman vs Archaean TTG
0
0.1
0
0.1
1
10
Y (ppm)
Low Sr; high Y
Oman trondhjemites overlap
Limpopo belt TTGs, and to
a lesser extent the classical
TTG trend (eg Lewisian)
t < 3.0 Ga
5
10
15
Na2O + CaO %
100
6
4
t < 3.0 Ga
t >3.0 Ga
t < 3.5 Ga
2
Oman
t >3.5 Ga
0
50
60
70
SiO2 (wt %)
80
Can this work for early Archaean felsic rocks at Isua ?
(Kamber et al. 2005)
Unit B1 – includes
felsic volcanic rocks
1000
100
10
Oman trondhjemites
Abundant
mafic
rocks
1
0.1
Rb Ba Th U
K Ta La Ce Pb Sr Nd Hf Zr Sm Ti Y Yb
Enriched in LILE, K, LREE & Pb
Low Sr, Ta, Ti
NE limb of Isua Greenstone belt, W Greenland
So how does/did it work ?
Oman
• Partial melting of low K
gabbro section of
oceanic crust
• Access of water
(seawater)
• Heat source – active
magmatism of ridge
• Melting in presence of
plagioclase and amphibole
absence of garnet
So how does/did it work ?
Oman
• Partial melting of low K
gabbro section of
oceanic crust
• Access of water
(seawater)
• Heat source – active
magmatism of ridge
• Melting in presence of
plagioclase and amphibole
absence of garnet
Early Earth
• Partial melting of ?high K
mafic crust
• Access of water
• Heat source –
? Radioactive heating
• Melting in presence of
plagioclase +/- garnet
That’s it !