The Layered Series of the Skaergaard Intrusion and its Liquid Line

The Layered Series of the Skaergaard Intrusion and its Liquid Line of Descent*
Peter Thy,1 Christian Tegner,2 Jakob K. Jakobsen,3 and Charles E. Lesher1
Deparment of Geology. University of California, Davis, USA, 3Nordic Volcanological Centre, Reykjavik, Iceland, and 2Department of Earth Sciences, Universitry of Aarhus, Denmark
1
Cryptic Mineral Variations of the Skaergaard Layered Series
The Skaergaard Intrusion and its Liquid Line of Descent
Thy, P., Lesher, C.E., Nielsen, T.F.D., and Brooks, C.K., 2006. Experimental constraints on the Skaergaard liquid line of descent.
Lithos 92, 154-180.
Toplis, M.J. and Carroll, M.R., 1996. Differentiation of ferro-basaltic magmas under conditions open and closed to oxygen:
implications for the Skaergaard intrusion and other natural systems. J. Petrol. 37, 837-858.
Wager, L.R. and Deer, W.A., 1939. Geological investigations in East Greenland. Part III. The petrology of the Skaergaard intrusion,
Kangerrdlugssuaq, East Greenland. Meddelelser om Grønland 105 (4) (reissued 1962).
pla
gio
cla
se
+
MZ
LZc
LZb
LZa
+
HZ
olivin
e
Stratigraphic Height (Meter)
Total FeO wt.%
ol
+ F ivine
e-T
i ox
ides
+a
ugit
e
UZc
Tegner
25
Wager
+ox
20
UZc
15
UZc
Toplis
Hunter
10
EG4507
5
Jakobsen
Tinden Sill
0
40 45 50 55 60 65 70 75
Schematic cross section through
the intrusion showing the main
series and mineral zone divisions.
1700
1500
UZa
1100
MZ
700
700
Legend
Average/SD
All Analyses
Corona Structure
Melanogranophyre
500
LZb
1140
1120
ox
1100
B
1080
ox
LZb
A
LZa
ol+pl
C
Contoured for constant ilmenite and
magnetite increase in steps of 20 %.
48
50
52
54
56
58
60
-1
SiO2 wt.%
-0.5
1040
B
C
14
46
1060
1.5 mt/il
aug
16
aug
MZ
1.0 mt/il
LZb
==
100
LZa
LZa
20
40
80 0
60
0.02
0.04
0.06
0.08
0
Mn per formula unit
0.2
0.4
0.8 0
0.6
0.01
Plagioclase An mol. %
0.02
0.03
0.04
0.05 0
0.01
K per formula unit
0.02
1020
SH/UBS
UZc
1900
1700
UZb
1500
UZa
1300
1100
MZ
900
LZc
700
Legend
Average/STD
All Analyses
500
=
LZb
=
=
=
Ferrobustamite
Melanogranophyre
300
=
=
=
=
100
1000
0
0.04
Fe* per formula unit
Ti
Ti
0.03
2100
Stratigraphic Height (Meter)
18
b
il
mt/
0.5
LZc
300
100
Temperature (oC)
20
Legend
Average/SD
All Analyses
Melanogranophyre
500
=
1160
ol+pl
FMQ
Total FeO wt. %
22
LZc
Augite
26
MZ
=
=
LZc
SiO 2 wt.%
A
==
==
900
900
300
A. Using ilmenite and magnetite modes obtained as part of this study.
B. Using experimental results extrapolated to low temperatures.
C. Using magnetite modes calculated by Toplis & Carroll (1996).
UZ
UZa
1300
1100
Forward Modeling of the Skaergaard Liquid Line of Descent
24
UZb
1700
1300
Olivine Fo mol. %
a
UZb
1500
0
LZa
0.5
0.0
Delta fO2
0.2
0.4
0.6
Mg/(Mg+Fe) ratio
0.8
0.02
0.04
Ti per formula unit
0.06
0.04
0.08
0.12
Al per formula unit
0.25
0.50
0.75
1.00
1.25
1.50
0.25
0.50
0.75
1.00
1.25
1.50
Ti/Al atomic ratio
Fe per formula unit
Il/(Il+Mt)
Ap
0.01
0.02
0.03
0.04
0.05
Mn per formula unit
0.01
0.02
0.03
Na per formula unit
Modal Variations of the Skaergaard Layered Series
Pl
Stratigraphic Height (Meters)
Future Studies. The new petrographic understanding of the Skaergaard
layered series allows accurate modeling of the liquid line of descent, crystallization, and solidification processes that have not been possible based on
the existing information.
UZa
McBirney
90-22
Liquid Line of Descent. The lack of knowledge of oxide modes in low temperature experimental investigations of the Skaergaard liquid line of descent
has been a main hindrance for accurate predictions (Thy er al., 2006). The
actual modes of oxides in the Skaergaard gabbros can instead be used to
extrapolate to the low melt fractions of the UZ for which the phase relations
and the T-fO2 variation remain uncertain. The forward modeling results
reveal a fundamental conflict between the suggestion that the iron content
of Skaergaard liquids increases during Fe-Ti oxide fractionation and the previous observation that at the same time oxygen fugacity (fO2) drops by two
log-units below the FMQ buffer (Toplis and Carroll, 1996). Such a drop
would require an unacceptably high proportion of Fe-Ti oxides and high
magnetite content in the fractionating assemblage that is not supported by
the gabbros. It is suggested that the strong reduction in the UZc may be related to open system crystallization and exchange of oxygen between the
crystallizing magma and the host basalts.
+o
livin
e
UZc
30
UZb
il
mt/
1.5
Mineral Modal Variation. Mineral modes were calculated on a weight basis
by least-squares linear approximation of mineral compositions to bulk rock
major element compositions. There is a systematic but irregular decrease in
the proportion of plagioclase upsection from LZa to UZc. The boundary between LZa and LZb is marked by an abrupt increase in the clinopyroxene
mode into LZb. The olivine mode is low throughout LZ, while olivine is rare
or absent in MZ. In UZa, olivine re-appears and increases sharply into UZb
and UZc. The average gabbros of the LZa and LZb contain systematic low
amounts of Fe-Ti oxides. The transition to the LZc appears to be gradual
with highest averages of ~16 % ilmenite and ~8 % magnetite reached just
within LZc and thereafter decrease to 5 % ilmenite and <1 % magnetite in
the UZc. The magnetite/ilmenite ratio in the gabbros is constant throughout
the layered series (~ 0.54).
+a
p a ti t e
LS
1900
90-22
Mineral Cryptic Variation. Cryptic variations reveal a systematic near
linear decrease in plagioclase composition from An66 at the base of the LZa
and An30 at the top of the UZc. In contrast, the mafic minerals (olivine and
augite) show systematic upward increases in iron content that progressively
become more marked in UZ. The compositions of olivine are affected by the
appearance of Fe-Ti oxides in the LZc and shift towards more magnesian
compositions. The minor element Mn increases systematically in olivine and
augite until UZc and the appearance of bustamite. The Al content in augite
shows a systematic decrease while the Ti content are remarkably constant
until UZ where it systematically increases.
35
SH
UZc
MBS
1900
UZc
2100
UZc
Interpretations of the liquid line of descent.
UBS
SH/UBS
SH/UBS
2100
90-22
Petrography and Zone Division. The petrography largely confirms previous results of Wager and Deer (1939). Plagioclase, olivine, and clinopyroxene occur throughout the layered series. Augite occurs in the LZa in distinctly lower modal amounts and has a predominantly interstitial texture
compared to the zones above. Olivine occurs throughout the LS, except for
the MZ where it is only locally present. The amount of low-Ca pyroxene (or
inverted pigeonite) decreases upwards in the LZ and MZ, and eventually
disappears in UZ (includes subsolidus grains). Magnetite and ilmenite
appear simultaneously in the LZc and continue throughout the MZ and UZ,
with ilmenite consistently being the dominant oxide. The modal abundances
of ilmenite and magnetite systematically decrease upwards in the stratigraphy from the base of the LZc to the UZc. Apatite defines the UZb with only
sporadic occurrences below as late crystallized interstitial grains.
Manganese-rich, ferrobustamite appears together with ferrohedenbergite in
the UZc, together with minor quartz and orthoclase. Biotite is present in LZa
near the intrusive contact.
Plagioclase
Olivine
The Skaergaard Intrusion
-
A new reference profile through the Layered Series of the Skaergaard intrusion has been established using GPS positioned surface samples collected
during fieldwork in 2000 and mineral exploration cores drilled in 1990 essentially in the same profiles as used by Wager and Deer (1939). The profile
offers a detailed picture of the Skaergaard layered series. Cryptic variation
was determined by electron microprobe analyses of the main minerals. Bulk
cumulate compositions were analyzed by XRF and ICPMS. Modal variation
was determined by least-squares approximations of bulk gabbro compositions to the constituent minerals.
OI
Cpx
Pl/(Pl+Cpx)
Opx
Il
Mt
UZc
2000
UZb
1500
UZa
MZ
1000
LZc
500
* 33rd IGC MPI-06 Symposium “Layered intrusions and the evolution
of magma chambers: A
tribute to J. Richard
Wilson.”
LZb
LZa
0
0
0.2 0.4 0.6 0.8
0.1 0.2 0.3 0.4
0.2 0.4 0.6 0.8
1
2
3
4
0.1 0.2 0.3 0.4
0.1 0.2 0.3 0.4
Mineral Content in Wt. %
0.1 0.2 0.3 0.4
0.5
1
1.5
2
0.05
0.15
0.04