Petrogenesis of eclogite and mafic
granulite xenoliths from South
Australian Jurassic kimberlitic
intrusions: Tectonic Implications
Thesis submitted in accordance with the requirements of the University of
Adelaide for an Honours Degree in Geology
Angus Tod November 2012
1
PETROGENESIS OF ECLOGITE AND MAFIC GRANULITE XENOLITHS FROM SOUTH AUSTRALIAN JURASSIC KIMBERLITIC INTRUSIONS: TECTONIC IMPLICATIONS
ANGASTON, EL ALAMEIN AND PITCAIRN ECLOGITES AND MAFIC GRANULITES
ABSTRACT
Jurassic kimberlites in South Australia have entrained sub lithospheric mafic granulites and eclogites from the eastern margin of the Australian Craton. This thesis looks at these rocks as a unique window into the sub-lithospheric mantle beneath the south eastern margin of Gondwana. Samples collected from Angaston, El Alamein and Pitcairn included eclogites, amphibole eclogites, amphibole granulites and feldspar rich granulites. These samples were prepared for analytical work at the University of Adelaide. Whole rock geochemistry was collected from x-ray fluorescence in the Mawson Laboratories. Mineral identification and geochemistry was determined by the Cameca SX 51 microprobe at Adelaide Microscopy. Geothermobarometry showed pressures between 6-30kbar, which represent 15-90km of depth and temperatures between 620-1200oC. These rocks experience very high pressure and temperatures and show petrological evidence of isobaric cooling path from the adiabat to the stable geotherm. Magma crystallisation models using MELTS program helped to determine the protoliths that appear to represent mafic underplates. The cumulate and melts that make up these xenoliths have been shown in this thesis to most likely have been derived from a MORB source that crystallised at high pressures (up to 30kbar). Pseudosections produced with the Theriak-Domino program were used to produce a metamorphic path and show that rock type is closely linked to emplacement depth and bulk composition. Radiogenic dating using Neodymium and Samarium system created isochron’s using IsoPlot and gave ages supporting protolith emplacement during the Neoproterozoic (≈670Ma) around the breakup Rodinia.
KEYWORDS: ECLOGITES, GRANULITES, EMAC, PETEROGENESIS AND METAMORPHISM
2
TABLE OF CONTENTS Angaston, El Alamein and Pitcairn Eclogites and Mafic Granulites ............................... 1 Abstract ............................................................................................................................. 1 Keywords: Eclogites, granulites, EMAC, Peterogenesis and metamorphism .................. 1 List of Figures and Tables ................................................................................................ 3 Introduction ...................................................................................................................... 8 Background ..................................................................................................................... 10
Setting ........................................................................................ Error! Bookmark not defined.
Petrology ................................................................................................................................. 14
Methods .......................................................................................................................... 22 Observations and Results ................................................................................................ 23 Discussion ....................................................................................................................... 47 Conclusions .................................................................................................................... 58 References ...................................................................................................................... 60 Appendix A: Methods .................................................................................................... 64 Appendix B Average garnet and clinopyroxene data (see extended appendix for all data) ........................................................................................................................................ 66 appendix C Hand Sample descriptions ........................................................................... 68 appendix DThinsection Descriptions .............................................................................. 69
3
LIST OF FIGURES AND TABLES
Figures Figure 1 Location map for the xenoliths which also shows locations of relevant locations for this thesis such as the Australian Craton, Tasman line and Kayrunnera xenoliths. Map adapted from Tappert et al (2011). .......................................................... 9 Figure 2 Modal percentages of minerals garnet (Gt), pyroxene (Px) and Feldspar (Fd) for the South Australian xenoliths. Xenoliths are distinguished by rock type is Figure 2.A and location of kimberlitic pipe withe two areas being El Alamein and Pitcairn in Figure 2.b. The rock type definitions are found in table 1 ............................................. 15 Figure 3 Photomicrographs of the three main types of textural xenoliths from Angaston, El Alamein and Pitcairn. Image A shows the common eclogitic texture such as triple points in plain polarised light (PPL) (Ai) and cross Polarised light (CPL) (Aii) in sample PA 6x2, Image B shows the common granulite texture and the main metamorphic reaction that defines theses suites of mafic rocks in PPL and CPL (Bii) in smaple Pit-M25, Image C shows gabbroic texture of xenoliths within a matrix of fine grained plagioclase and pyroxene in PPL (Ci) and CPL (Cii) in sample PA 7x9. ...................... 17 Figure 4 Photomicrographs of important textures found within the studied xenoliths. Image (A) shows the metamorphic reaction plagioclase + pyroxene ↔ garnet + quartz as seen at El Alamein. Exsolution features such as vermicular exsolution of clinopyroxene of garnet and orthopyroxene exsolution of clinopyroxene in plane polarised light (PPL) (Ai) and cross polarised light (CPL) in sample PA 7x1. Image (B) shows garnet mineral relationships within the granulites in the Pitcairn xenoliths. Garnets form as blebs around pyroxene in PPL (Bi) and CPL (Bii) in sample JS Kim. Image (C) shows the metamorphic reaction Plagioclase + pyroxene ↔ garnet + quartz as seen at El Alamein. Exsolution features such as garnet exsolution of clinopyroxene and orthopyroxene exsolution of clinopyroxene in PPL and CPL (Cii) in sample PA 7x1. ................................................................................................................................. 20 Figure 5 Photomicrograph of orthopyroxene crystal core of a complex carona structure in plane polarised light (PPL) (i) and cross polarised light (CPL) (ii) in sample PA 7x1. ........................................................................................................................................ 21 Figure 6 Modal percentages of end members for garnet of the South Australian xenoliths. The iron end member is almandine (Al), magnesium end member is pyrope (Py) and calcium end member is grossular (Go), xenoliths are distinguished by rock type in diagram 6a and xenolith location 6b ................................................................... 23 Figure 7 Modal percentages of end members for pyroxene of the South Australian xenoliths. The iron end member is Forsterite (Fs), iron calcium end member is Hedenbergite (Hd), magnesium end member is Enstatite (En) and the magnesium calcium end member is diopside (Di). Xenoliths are distinguished by rock type in diagram 6a and xenolith in location 6b. ......................................................................... 24 Figure 8 Graph of Jadeite cation % vs pressure. Pressure was calcutalted using Nimmis and Taylor (2000) clinopyroxene barometer. Data taken from table 2 and table 6. ....... 25 Figure 9 Modal percentages of end members for feldspars of the South Australian xenoliths. The Sodium end member is albite (Ab), calcium end member is anorthite (An) and potassium end member orthoclase (Or). Xenoliths are distinguished by rock type in diagram 6a and xenolith location 6b. .................................................................. 25 Figure 10 Plotted amphiboles from Pitcairn xenoliths using A-site occupancy by alkalis (Na + K) vs SiO2 to discriminate between the end members (HAWTHORNE et al.
4
1997) pargasite (Pa), edenite (Ed), tschermakite (Ts), hornblende (Hb) and tremolite (Tr). ................................................................................................................................. 26 Figure 11Alkalis (Na2O + K2O) vs SiO2 whole rock geochemistry for the xenoliths plotted using IgPet (Carr 2002). Rock type boundaries described by Cox et al (1979). The South Australian xenoliths plot within the Basaltic region. This shows Angaston (red circles and blue squares) (Segui 2010), El Alamein (yellow crosses) and Pitcairn (green triangles) .............................................................................................................. 34 Figure 12 (A) Calculated CIPW Norm for the South Australian xenoliths, the proposed classification by Thompson (1984) for basalt based on their normative proportions of nepheline (Ne), olivene (Ov), albite (Ab), hypersthenes (Hy) and quartz (Qt). Red circles represent South Australian xenoliths (Segui 2010) which plot within the silica saturated and silica undersaturated portions. Green circles represent MORB (Jenner & O'Neill 2012) which plot in the silica oversaturated and silica saturated parts of the diagram. (B) Mole percent diagram (petrogenetic grid) relevant to variable precent melting (5% to the point where clinopyroxene disappears from the residue) of lherzolite over a pressure range of 0.5 to 3GPa (i.e., about 15-90km depth; pressure shown in bold). Each dashed line at a given pressure represents loci of melt compositions (molar normative) generated by progressive partial melting of lherzolite assemblage (ol + opx + cpx + melt) at that pressure (melt % increasing from left to right on each dashed curve). Each continuous line represents a fixed %melting curve. Aldo shown is the cpx out line. A lherzolitic source rock will lose cpx to the melt beyond this line. Sources of data: Takahashi and Kushiro (1983), Hirose and Kushiro (1993), and Baker and Stopler (1995). Note that it is mainly schematic and does not take into account the changing source composition that must happen as the melt in removed from the source. ............ 35 Figure 13 Plate of whole rock geochemical graphs of South Australian xenoliths (Segui 2010) with MORB (Jenner & O'Neill 2012) for comparison. Graphs A, B, C and D are MgO vs SiO2, CaO, TiO2 and Al2O3 respectively. Diamonds on Graph A and B show mineral compositions plagioclase (PLAG), clinopyroxene (CPX), orthopyroxene (OPX) and the mid ocean ridge basalt (MORB) melt composition, with two distinct trends; 1) a trend towards orthopyroxene showing orthopyroxene crystallisation driving the melt and 2) a trend clinopyroxene + plagioclase showing clinopyroxene + plagioclase driving crystalisation. Black arrow shows igneous variation trends, “M” is the direction towards melt differentiation and “C” is towards the cumulates or crystal extracts that must drive the magmatic trend. ...................................................................................... 36 Figure 14 Graph of barium (Ba) vs wt% MgO for the South Australian xenoliths (Segui 2010) (red circles) with MORB (Jenner & O'Neill 2012) (blue circle). South Australian xenoliths show a several order magnitude higher amounts. ........................................... 37 Figure 15 Isochrons calculated using IsoPlot (Ludwig 2003) graphs show 143Nd/144Nd vs 147Sm/144Nd (A) represents whole rock isotope data for the South Australian xenoliths from Angaston (Segui 2010), El Alamein and Pitcairn and gives an age 739±680Ma. (B) South Australian xenoliths (Segui 2010) (green triangles) and Neoproterozoic Cambrian and South Australian Adelaidean basalts (John Foden, per comms) (Blue diamond’s) and gives an age of 656 ± 92Ma. ......................................... 39 Figure 16 Pseudosection calculated for Pit M22 (see table 2) using THERIAK-DOMINO program (De Capitani & Petrakakis 2010), for the geologically realistic chemical system SiO2-Al2O3-FeO-Fe2O3-MgO-CaO-Na2O-K2O-H2O-TiO2 (NCKFMASHTO). The dataset used compiles the following a-x models which incorporate Fe3+ end-member minerals: garnet, biotite and melt (White et al. 2007),
5
orthopyroxene and magnetite (White et al. 2002), amphibole (Diener et al. 2007), clinopyroxene (Green et al. 2007), K-feldspar and plagioclase (Holland & Powell 2003) and ilmenite (White et al. 2000). Mn is not considered for the reasons given by White et al (White et al. 2007). Blue lines represent major introduction of a mineral to the assemblage (amphibole, garnet and plagioclase), arrow represent direction on pseudosection the mineral labled is introduced. The introduction of garnet to the assemblage turns to Gabbroic rock to granulite and the loss of plagioclase turns granulite to eclogite. Blue shaded polygon represents the mineral assemblage seen for Pit M22 and the blue star represents the pressure and temperature estimations for the sample (see table 6). ....................................................................................................... 43 Figure 17 Pseudosection calculated for Pit M25 (see table 2) using THERIAK-DOMINO program (De Capitani & Petrakakis 2010), for the geologically realistic chemical system SiO2-Al2O3-FeO-Fe2O3-MgO-CaO-Na2O-K2O-H2O-TiO2 (NCKFMASHTO). The Dataused compiles the following a-x models which incorporate Fe3+ end-member minerals: garnet, biotite and melt (White et al. 2007), orthopyroxene and magnetite (White et al. 2002), amphibole (Diener et al. 2007), clinopyroxene (Green et al. 2007), K-feldspar and plagioclase (Holland & Powell 2003) and ilmenite (White et al. 2000). Mn is not considered for the reasons given by white et al (2007). Red lines represents major introductions of mineral to an assemblage (amphibole, garnet and plagioclase), arrows represent direction on pseudosection the mineral in labled is introduced. The Introduction of garnet to the assemblage turns Gabbroic rock to granulite and the loss of plagioclase turns granulite to eclogite. Red shaded polygon represents the mineral assemblage seen for Pit M25 and the red star represents the pressure and temperature estimations for the sample (see table 6). ............................... 45 Figure 18 Pressure and temperature plot of geothermobarometry estimations for the SEA (O'Relly & Griffin 1985), EMAC (Pearson & O'REILLY 1991), Monk Hill (Tappert et al. 2011), Angaston (Segui 2010) and Pitcairn and El Alamein. Pressure and temperature estimations using garnet-clinopyroxene Fe-Mg thermometer (Ellis & Green 1979, Krogh 1988) and clinopyroxene barometer (Nimis & Taylor 2000). Data for UHP metamorphic rocks Refrence) schematic subduction metamorphic path taken from Agard (2009) and subduction data points taken from numerous sources(Gao 1999, Dale 2003, Janak 2004). Arrows right of Monk Hill Geotherm (Tappert et al. 2011) show the metamorphic path for the South Australian xenoliths. ................................................... 47 Figure 19 Ni-Cr (ppm) variation of the South Australian mafic xenoliths (Segui 2010) (red circles) and MORB data (Jenner & O'Neill 2012) (green circles). Trends on this Figure show melt fractionation curves for high pyroxene/olivene (high pressure) (blue line) and lower pyroxene/olivene trends (black line with yellow triangles) and the complimentary cumulate trend for a high pressure (black line with orange circles). Trends created using MELTS (Ghiorso & Sack 1995). The specific chosen starting basalt used was an olivene tholeiite from the Adelaidean Smithon basin in N.W. Tasmania. This was chosen as it clearly had experienced no crustal contamination (John Foden, per comms) ......................................................................................................... 51 Figure 20 % of melt remaining vs Temperature showing melt evolution path (red crossed) and the complimentary solid cumulate path (black dashes) created from the results of MELTS (Ghiorso & Sack 1995) modelling on the specific chosen starting basalt, an olivene tholeiite from the Adelaidean smithton basin in N.W. Tasmania. This was chosen as it clearly had experienced no crustal contamination (John Foden, per comms). The most favourable run made at pressure 8.5 kbar, low water content and
6
oxygen fugacity of QFM + 1as seen in the MELTS list. Minerals crystallised orthopyroxene (OPX), clinopyroxene (CPX), spinel (SP) and plagioclase (PLAG) ..... 53 Figure 21 MELTS (Ghiorso & Sack 1995) modelling of the south Australian xenoliths on Wt% MgO vs Wt% CaO (A) and SiO2 (B). Melt evolution path (blue crosses) and the complimentary solid cumulate path (black dashes) are shown on the diagram. Starting compostion is shown to be the orthopyroxenite (sample EA08 6). .................. 54
7
Tables Table 1 Distinguishing characteristics for xenoliths different rock types. ..................... 14 Table 2Whole rock major element geochemistry collected using XRF ......................... 29 Table 3Whole rock trace element geochemistry collected using XRF .......................... 31 Table 4 Radiogenic Isotope data .................................................................................... 38 Table 5 Equations used for Geothermobarometry .......................................................... 41 Table 6 Presure and temperature estimates where TEG79 (Ellis & Green 1979), TK88(Krogh 1988) and PNT95(Nimis & Taylor 2000) ............................................................................... 44
8
INTRODUCTION
Jurassic kimberlites occur at a number of locations in the mid north of South Australia.
They have transported a diverse range of lower crustal and upper mantle xenoliths.
Studies of these will help to explain the thermal and compositional evolution of the
lithosphere of South Eastern Gondwana. These xenoliths include mafic eclogite and
granulite, garnet peridotite (Tappert et al. 2011) and xenocrysts including diamonds
(Tappert et al. 2009), Cr diopside, garnet, picroilmenite and spinel. The mineral
assemblages of the eclogites range from garnet + clinopyroxene eclogite to amphibole +
garnet + clinopyroxene eclogite. The mineral assemblages of the granulites range from
Feldspar rich (Fd rich) to amphibole rich granulites but both granulite and eclogite
generally have a garnet and clinopyroxene dominated mineralogy. This project aims to
identify the protolith of the kimberlite transported eclogites and the pressure,
temperature and timing of their metamorphic evolution. This will lead to a better
understanding of the paleo-geotherm and structure along the south eastern margin of
Australian Craton. These xenoliths are unique samples of the lithospheric mantle
beneath the important transition zone between Precambrian craton and the Paleozoic
fold belt of Eastern Australia (Figure 1). We will test the hypothesis that these eclogites
and granulites are mafic under plates recording mantle derived magmatism supplied to
the evolving crust. It is possible they originated during Late Proterozoic rifting and
early Proterozoic subduction.
This Thesis focuses on xenoliths from kimberlitic intrusions at three localities; El
Alamein, Pitcairn and Angaston (Figure 1). In this study we used a number of
techniques to help constrain the P-T-t path and the possible protoliths of the mafic
9
Figure 1 Location map for the xenoliths which also shows locations of relevant locations for this thesis such as the Australian Craton, Tasman line and Kayrunnera xenoliths. Map adapted from Tappert et al (2011).
xenoliths. Bulk geochemistry analysis were used to constrain the pressure, temperature
and time paths through the production of a pseudosection using the program
THERIAK-DOMINO program (De Capitani & Petrakakis 2010). Bulk geochemistry
was also used to make comparisons with other data sets. Mineral comparison analyses
were done using the electron microprobe to provide mineral geochemistry and
geothermobarometry determination. Geothermometry determination were based on the
exchange of Mg-Fe2+ between garnet (gt) and clinopyroxene (cpx) (Ellis & Green 1979,
Krogh 1988) and geobarometry estimations were made using the clinopyroxene site
10
occupancy barometer formulated by Nimis and Taylor (2000). The most common
mineral assemblage was Gt-cpx-rutile±kyanite±plag and made up the bulk of the mafic
xenolith mineral assemblage’s prograde and retrograde metamorphic reactions and
metamorphic/igneous textures were used to show the infered metamorphic path through
the pseudosections (Pearson et al. 1991, Jacob 2004).
BACKGROUND
Kimberlites are volatile rich, highly potasic, ultra mafic rock that ascend from the deep
mantle/lithospheric mantle (Sparks et al. 2006). Rocks from the mantle lithosphere and
lower crust, such as peridotite, granulite and eclogite, are commonly sampled as
xenoliths by the ascending kimberlites. Kimberlites have been studied due to their
association with diamonds for their economic importance; but also to provide rare
sampling windows into the lower crust and upper mantle. Eclogitic and granulitic
xenoliths are believed to form in three different ways (Jacob 2004, Griffin & O'Reilly
2007).
1) As a high pressure cumulate from ascending mafic magma.
2) Whole-sale under plating of ascending mafic magma at the Moho; forming
Gabbro, which eventually cools first into garnet granulite field then into the
eclogite facies.
3) Oceanic crust subducted at a convergent margin and converted to eclogite during
subduction, which is then tectonically accreted to the base of the lithosphere.
Many eclogites show MORB like chemistry and P-T values like the upper subducted
slab. This has led to many papers linking eclogites to kimberlite subduction processes
(Jacob 2004). Kimberlitic eclogites are thought to experience exceptionally high
11
pressures and temperatures when compared to eclogites that occur in glaucophane schist
terrain and those within crustal migmatite gneissic terrains (Coleman et al. 1965) that
are commonly associated with subduction zones. Eclogites found entrained within
kimberlites can be referred to as high Temperature (HT) and ultra high pressure
eclogites due to the forming temperatures of T>900oC and pressures P> 36Kbar.
Geological setting
The Precambrian Australian Craton is made up of a mosaic of Archean and Proterozoic
crustal elements on the western side of Australia. The Eastern side of Australia is made
up of the Paleozoic to Mesozoic Tasman fold belt, which shows a thinner crustal
package. The Tasman Line defines the boundary between these two building blocks that
make up Australia (Veevers & Conaghan 1984) (Figure 1). Xenoliths from kimberlite
pipes and dykes have been found in a number of places in south and south-eastern
Australia (O'Relly & Griffin 1985, Pearson . 1991, Song 1994, Segui 2010, Tappert.
2011). South Australian Kimberlites occur in the Mid North of South Australia at Port
Augusta, Eurelia, Terowie, Orroroo and in the Adelaide Hills (Figure 1). The Eurelia
and Adelaide Hills kimberlites have diamonds as xenocrysts and have, therefore, been
well studied. South Australian kimberlites intrude the Burra and Umbertana sequences
of the Adelaidean sediments (Colchester 1972, McCulloch 1982). They occur as pipes
with N.W.-S.E. trending dykes that are deeply weathered and have been dated to the
Jurassic (~180Ma) by Stracke et al (1979) and Tappert et al (2011). Xenoliths found in
South Australia are a mixture of country rock and nodules of ultra mafic to felsic rocks
derived from upper and lower lithosphere.
12
Paleogeothermal gradients have been determined at various locations on both sides of
the Tasman line in Australia where data is available. Xenolith suites from East of the
Tasman Line have been extensively studied by a number of previous authors
(Sutherland & Hollis 1982, O'Relly & Griffin 1985, Pearson & O'REILLY 1991,
Pearson et al. 1991, Pearson et al. 1995).These studies have led to the definition of the
South-eastern Australian geotherm (SEA). They concluded that the protoliths for these
mafic xenoliths were a number of mafic underplates from magmatic plumes that rose
during the opening of the Tasman sea (O'Relly & Griffin 1985). Pearson et al (1991)
studied xenoliths from the Pine Creek and Calcutteroo kimberlites and from the New
South Wales Kayrunnera alkali basalt (Tappert et al. 2011). These lie on the eastern
margin of the Australian Craton (EMAC) just to the west of the Tasman Line. They
used this data to define the EMAC geotherm, which appeared slightly cooler than the
SEA geotherm (REFRENCE). They concluded that the EMAC defined geotherm started
out with a similar crust-mantle boundary as seen in the SEA geotherm but have now
cooled due to equilibrations under different pressure and temperature conditions closer
to the Australian Craton (Pearson & O'REILLY 1991).
Song (1994) collected a number of kimberlitic xenoliths from Culcutteroo, Pine Creek
and Port Augusta (Figure 1) for geochemical and pressure temperature analysis for his
study on the evolution of the lithospheric mantle below S.E. South Australia. His work
concluded that the South Australian xenoliths reflected several periods of underplating
by basaltic liquids derived from deep lithosphere or athenosphere, subsequent
metamorphism, and metasomatism of the original melt (Song 1994). He also concluded
that his calculated P-T estimates illustrate that the lithosphere beneath the Paleozoic
13
Adelaide geosyncline is cooler than that of the Phanerozoic eastern Australia; indicating
of different petrology or geochemical composition, which is also shown between the
EMAC and SEA geotherms. Tappert et al (2011) in addition conducted research into the
Monk Hill garnet peridotites xenoliths (Figure 1) and xenocrysts. This locality is on the
eastern margin of the Australian Craton. Monk Hill P-T estimations also defined a
significantly lower geothermal gradient (40mW/m2) than that under south-eastern
Australia. However the estimates of Song (1994) and Tappert et al (2011) show a much
lower geotherm than the EMAC model of Pearson et al (1991). This may indicate that
the EMAC and SEA estimations are both still re-equilibrating to the stable geotherm in
the Jurassic, shown by the Monk Hill geotherm.
Tappert et al (2011) have showed the differences in goetherms are due to the
Kayrunnera volcanic rocks used in the Pearson et al (1991) P-T estimations. The
volcanic hosts for these xenoliths have Permian emplacement age, compared to the
Jurassic kimberlites in South Australia (Gleadow & Edwards 1978, Stracke et al. 1979),
and their geochemistry indicates they are alkaline basalts rather than kimberlites
(Ferguson & Sheraton 1979). Therefore, the earlier transport of the Kayrunnera volcanic
rocks reveals a different, and in this case, a hotter geothermal gradient compared to the
South Australian kimberlitic xenoliths.
The El Alamein and Pitcairn kimberlites are both located along the EMAC and are
assumed to be Jurassic in age (Stracke et al. 1979, Tappert et al. 2011). El Alamein is
the most northern and western kimberlite locality studied within the central Adelaide
fold belt of South Australia and was chosen to investigate any possible regional trends
14
in locality in mantle P-T conditions and mafic magma composition. Only limited prior
studies have been made at Pitcairn. The geochemistry and petrography is hoped to help
better constrain the paleogeotherm along the EMAC and to define the age, source and
tectonic signatures of their mafic protoliths. The inferred depth to the Moho in the Mid
North of South Australia is 35km (Branson et al. 1968)
Petrology Nineteen samples were collected for petrological descriptions from the Pitcairn and El
Alamein kimberlites (Figure 1) and were combined with 18 petrographical descriptions
of Angaston xenoliths from Segui (2010). The xenoliths were divided into five
lithological groups based on their mineral assemblage, mineral chemistry and textures.
The ratio Al(6)/Al(4) is a ratio used to distinguish eclogites from granulites based on the
clinopyroxene chemical composition and represents the ratio of Tschermaks to Jadeite
molecule within the mineral (Jacob 2004).
Table 1 Distinguishing characteristics for xenoliths different rock types.
Lithology Mineralogy Chemistry Texture Eclogitic rock
Gt+ CPX±Qt±OPX±Am
Al(6)/Al(4) ≥2 (CPX)
Triple point equi- to sub granular euhedral texture
Amphibole Eclogite
CPX+Gt+Am±Qt±OPX
Al(6)/Al(4) <2 (CPX) Modal Am≥10%
Triple point equi- to sub granular euhedral texture
Orthopyroxenite
OPX+CPX±Am±Gt
Modal OPX>70%
_
Feldspar rich granulite
Pl+CPX+Kfd±Opx±Gt±Am±Qt
Modal Pl+K+Fd 40-50%
Diverse stages of exsolution textures, rutile inclusions
within garnet and or clinopyroxene
Amphibole granulite
Pl+Am+CPX+Kfd±Opx±Gt±Qt
Modal Pl+K+Fd 40-50%
and Am >10%
Diverse stages of exsolution textures, rutile inclusions
within garnet and or clinopyroxene
Gabbroic Rock
Pl+CPX+OPX+Am±Il±Ru
Doleritic microstructure, lath shaped plagioclase,
clinopyroxenes showing an intergrowth with amphibole
and minor ilmenite
15
The El Alamein (Figure 1) xenoliths were collected from weathered kimberlitic sills
near Port Augusta in the South Australian Mid North. Ten samples were examined and
used for petrological description and mineral geochemistry. The Pitcairn xenoliths
(Figure 1) were collected from a location 160km East S.E. of Port Augusta. Nine
samples were examined and used for peterological descriptions and mineral
geochemistry and whole rock geochemistry.
Figure 2 Modal percentages of minerals garnet (Gt), pyroxene (Px) and Feldspar (Fd) for the South Australian xenoliths. Xenoliths are distinguished by rock type is Figure 2.A and location of kimberlitic pipe withe two areas being El Alamein and Pitcairn in Figure 2.b. The rock type
definitions are found in table 1
Eclogitic Rock
The Eclogitic rocks are dominantly composed of garnet (30-60%), clinopyroxene (20-
40%, with essential rutile (1-10%) ± ilmenite ± spinel (Figure 2). They have
equigranular textures with well-equilibrated triple point grain boundary textures
indicative of equilibrated rocks (Figure 3A). These rocks show unstrained textures with
clinopyroxene and garnet making up 90% of the rocks composition. Rutile is in
abundance within this rock type and occurs at the junction between clinopyroxene and
garnet. The little amount of plagioclase that is present (0-3%) exists as small crystals
16
17
Figure 3 Photomicrographs of the three main types of textural xenoliths from Angaston, El Alamein and Pitcairn. Image A shows the common eclogitic texture such as triple points in plain polarised light (PPL) (Ai) and cross Polarised light (CPL) (Aii) in sample PA 6x2, Image B shows the common granulite texture and the main metamorphic reaction that defines theses suites of mafic rocks in PPL and CPL (Bii) in smaple Pit-M25, Image C shows gabbroic texture of xenoliths within a matrix of fine grained plagioclase and pyroxene in PPL (Ci) and CPL (Cii) in sample PA 7x9.
with their shape dictated by the other surrounding minerals. Alteration within the
eclogite rock type is readily seen along fractures and mineral boundaries. The Alteration
minerals are carbonates and micas.
Amphibole Eclogite
The amphibole eclogite rocks are dominantly composed of garnet (30-50%),
clinopyroxene (20-40%,), Amphibole (>10%) with essential ruitle (1-10%) ± ilmenite ±
spinel (Figure 2). These rocks show triple point textures indicative of equilibrated rocks
(Figure 3A). They also show layering textures defined by garnet-rich layers and then
clinopyroxene rich layers. Amphibole demonstrates a strong relationship to rutile
forming as inclusions within amphibole. The low amounts of plagioclase (0-3%) within
this rock type is rimmed by garnet.
Feldspar (Fd) rich Granulite and Amphibole Granulite
These Fd-rich granulite and amphibole granullite rocks are mafic rocks. The Fd-rich
granulites are composed of clinopyroxene (20-45%), garnet (15-40%), plagioclase (20-
55%) ± ilmenite ± rutile (Figure 2). The amphibole granulites are composed of
clinopyroxene (20-40%), garnet (15-35%), plagioclase (20-55%), Amphibole (>10%) ±
ilmenite ± rutile (Figure 2). They are characterised by fine to medium grained
interlocking sub- to equiangular textures (Figure 3B). The diagnostic mineral for these
rock types is large elongate plagioclase which display simple and multiple twinning.
18
Reddish-pyrope garnet occurs as sub to euhedral grains that form rings around the
clinopyroxenes, llmenite and plagioclase. Clinopyroxene are subhedral and dark green,
their boundaries with plagioclase are straight and curved with garnet. Ilmenite,
pyroxene, amphiboles and garnets all show exsolution textures implying the breakdown
of this mineral assemblage. Within this rock type we see three types of exsolution: 1)
garnet and amphibole exsolution from clinopyroxenes (Figure 4C and 5); 2)
clinopyroxene as vermicular exsolution in garnet (Figure 3A), and; 3) clinopyroxene
exsolution from orthopyroxene (Figure 5). These textures are believed to show the
transition from primary igneous (gabbroic rock) textures towards granulite metamorphic
textures under static stress.
Orthopyroxenite1
There was one orthopyroxenite sample found from El Alamein (EA 08 6) with 70%
orthopyroxne (OPX) and 10% clinopyroxene (CPX) and minor Garnet. Orthopyroxenite
are formed as early high pressure cumulates from fractionated magmas
Gabbroic rock
These xenoliths show plagioclase laths and prismatic clinopyroxenes that are
intergrown with amphibole and minor illmenite and rutile (Figure 3C). Garnet rings
encompass the clinopyroxenes in some of these rock type.
1 No thin section was created for this sample so the petrography was concluded from the hand sample only.
19
20
Figure 4 Photomicrographs of important textures found within the studied xenoliths. Image (A) shows the metamorphic reaction plagioclase + pyroxene ↔ garnet + quartz as seen at El Alamein. Exsolution features such as vermicular exsolution of clinopyroxene of garnet and orthopyroxene exsolution of clinopyroxene in plane polarised light (PPL) (Ai) and cross polarised light (CPL) in
sample PA 7x1. Image (B) shows garnet mineral relationships within the granulites in the Pitcairn xenoliths. Garnets form as blebs around pyroxene in PPL (Bi) and CPL (Bii) in sample JS Kim.
Image (C) shows the metamorphic reaction Plagioclase + pyroxene ↔ garnet + quartz as seen at El Alamein. Exsolution features such as garnet exsolution of clinopyroxene and orthopyroxene
exsolution of clinopyroxene in PPL and CPL (Cii) in sample PA 7x1.
Mineral Relationships
Several of the xenoliths studied from El Alamein contain microstructures that show
evidence of arrested local equilibrium and re-equilibrium in different locations
throughout the slide. Evidence for this includes: 1) Double and single corona of
garnet/garnet + CPX around OPX on the boarder of plagioclase that is readily seen in
the garnet granulites from El Alamein (Figure 3B, 4A and 4B) and within the Pitcairn
xenoliths. 2) Garnet and pyroxene exsolution of relict pyroxenes’ (Figure 4A and 4B).
3) The replacement of plagioclase by garnet and quartz assemblage (Figure 3A, 4A and
4B).
Pearson and O’Reilly (1991) also noted these microstructures and deduced that they
were the result of the breakdown of the primary assemblages, mainly olivine +
plagioclase and pyroxene + plagioclase as a response to cooling from igneous
temperatures. Garnet coronas can be explained by Equation (1):
𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝 + 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝 = 𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔 + 𝑞𝑞𝑞𝑞𝑞𝑞𝑞𝑞𝑞𝑞𝑞𝑞 (1)
The garnets form as blebs or rims around pyroxenes or as lamellae or blebs in
pyroxenes (Figure 4B). Garnet-clinopyroxene symplectites also form at the interface of
plagioclase OPX relict boundaries and can be expressed in Equation 2 that depicts the
transition from granulite to eclogite.
21
𝑜𝑜𝑜𝑜𝑜𝑜ℎ𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜 + 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝 = 𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔 + 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 + 𝑞𝑞𝑞𝑞𝑞𝑞𝑞𝑞𝑞𝑞𝑞𝑞 (2)
This reaction also shows the compositional change from a more jadeite-rich pyroxene to
a more omphacite-rich pyroxene due to the breakdown of albite.
Figure 5. Photomicrograph of orthopyroxene crystal core of a complex carona structure in plane polarised light (PPL) (i) and cross polarised light (CPL) (ii) in sample PA 7x1.
Slide PA78X shows an OPX crystal at the core of a complex corona structure and may
illustrate either re-equilibration of primary Al-rich orthopyroxene with garnet, or an
intermediate stage in the transformation of olivene + plagioclase to garnet +
clinopyroxene (Pearson et al. 1991) (Figure 5).
Segui (2010) discussed a similar group of rocks located within the Angaston kimberlite
dykes and diatremes which also lies on the eastern margin of the Australian Craton
(Figure 1). The Angaston mafic xenoliths however had an important addition to their
suite, the inclusion of kyanite. This inclusion of kyanite to the assemblage was in the
form of fine grained crystals that make up a matrix and symplectites with clinopyroxene
22
that aid in the breakdown of plagioclase with garnet (Segui 2010). Re-equilibration
textures mainly that of garnet corona and garnet and pyroxene exsolution were also
discussed.
METHODS
Samples were collected in the field from locations labelled (Angaston, El Alamein and
Pitcairn) and transported in clear bags to Mawson Laboratories. Sample descriptions
were made at location sites.Sample preparation was conducted at Adelaide University
Mawson Laboratories. A diamond saw, stainless steel jaw crusher, tungsten mill and
Franz separator were used to preparethe samples for major element geochemistry, trace
element geochemistry and radiogenic isotope geochemistry.
Major mineral geochemistry was conducted on in-situ mineral grains using the Cameca
SX 51 microprobe at Adelaide Microscopy. Whole rock major element analysis was
conducted using fused glass discs and x-ray fluorescence (XRF). Total FeO was
determined by the digestion in HCl in the presence of CO2 with the solution then being
titrated using potassium di-chromate with BADS as an indicator. Whole rock trace
element geochemistry was obtained by using pressed pellets and XRF.
Phase diagram calculations are based on whole-rock compositional data. Water content
is based of weight loss on ignition. Analysis via titration methods provides the
concentration of ferric iron. Due to the inevitable hydration and oxidation of the sample
during retrogression and/or weathering at the surface, the H2O and Fe2O3 values of the
composition used for phase diagram calculations should be considered a maximum
estimate.
23
Radiogenic Isotope work was carried out on seven whole rock samples which were
prepared at the Mawson Laboratories. Neodymium and Samarium analysises were
carried out on a Finigan MAT 262 TIMS at Adelaide University. The Nd and Sm values
were calculated using depleted mantle values (from Goldstein et al 1984).
OBSERVATIONS AND RESULTS
Mineral Geochemistry
Mineral geochemistry was collected using the SX 51 microprobe at Adelaide
Microscopy and used to help with geothermobarometry, mineral identification and end
members.
Figure 6. Modal percentages of end members for garnet of the South Australian xenoliths. The iron end member is almandine (Al), magnesium end member is pyrope (Py) and calcium end member is
grossular (Go), xenoliths are distinguished by rock type in diagram 6a and xenolith location 6b
Garnet
The garnets analysed from both localities were homogenous throughout with no
zonation along the rims. Figure 6 shows garnets plotted within a compositional end
member triangle with almandine (Fe), pyrope (Mg) and grossular (Ca) end members.
24
The Almandine composition of all garnets is roughly <50% with most differences
occurring within the pyrope and gossular end member for garnet. Pitcairn displays a
more pyrope-rich end member for garnet and El Alamein towards the Almandine end
member. The rock types ternary diagram also depicts subtle differences in garnet end
member compositions of garnet. The Fd-rich granulites display a higher amount of
pyrope and gossular than the other rock types. The rock types that have >10%
amphibole also display a trend towards the almandine end member, away from the
samples with <10% amphibole.
Figure 7. Modal percentages of end members for pyroxene of the South Australian xenoliths. The iron end member is Forsterite (Fs), iron calcium end member is Hedenbergite (Hd), magnesium end member is Enstatite (En) and the magnesium calcium end member is diopside (Di). Xenoliths are distinguished by rock type in diagram 6a and xenolith in location 6b.
Pyroxene
The pyroxenes found in all samples were homogenous, showing no zoning. The
dominant clinopyroxene throughout all samples was a type of clinopyroxene with a high
diopside (Ca) (Figure 7). No systematic difference can be seen between the
compositions of clinopyroxene through location of xenolith or type of xenoliths.
Orthopyroxene is only present in two samples from El Alamein and show enstatite as
the dominating end member. Figure 8 shows a plot of pressure (Nimis & Taylor 2000)
vs Jadeite (table 6) which illustrates a positive correlation, where omphacite is
25
consumed by jadeite during increased metamorphism (Fleet & Zussman 2003). The
percentage of Jadeite was calculated as the sodium cation amount.
Figure 8 Graph of Jadeite cation % vs pressure. Pressure was calcutalted using Nimmis and Taylor (2000) clinopyroxene barometer. Data taken from table 2 and table 6.
Figure 9 Modal percentages of end members for feldspars of the South Australian xenoliths. The Sodium end member is albite (Ab), calcium end member is anorthite (An) and potassium end member orthoclase (Or). Xenoliths are distinguished by rock type in diagram 6a and xenolith location 6b.
Feldspar
The composition of feldspars in the Pitcairn and El Alamein xenoliths lay within the
plagioclase feldspar series. Feldspars were not identified within the eclogitic rock type
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0 5 10 15 20 25 30 35
Jade
ite %
Pressure (kbar)
26
group. The majority of plagioclase plot towards the albite (Na) end member (Figure 9).
Regional differences can be seen between El Alamein and Pitcairn feldspars. Pitcairn
feldspars display a broader range of plagioclase compositions (30-90%albite), with
some plotting towards the orthoclase (K) end member. El Alamein Feldspars all plot
along the plagioclase series and have a much tighter range of composition (55-75%).
Figure 10 Plotted amphiboles from Pitcairn xenoliths using A-site occupancy by alkalis (Na + K) vs SiO2 to discriminate between the end members (HAWTHORNE et al. 1997) pargasite (Pa), edenite (Ed), tschermakite (Ts), hornblende (Hb) and tremolite (Tr).
Amphibole
There is a large difference in abundance of amphibole between the Pitcairn and El
Alamein xenoliths, as El Alamein displays significantly lower amounts of amphibole if
any at all. The chemistry of Amphibole is very important due to its component of water,
which may indicate that Pitcarin mantle may have been more volatile enriched. The
composition of the amphiboles found at Pitcairn display a trend towards pargasite (Pr),
which is the silica-poor Alkali (Na+KA-site) high end member (Figure 10). Pargasite is
known to form in high temperature metamorphic regions such as in contact aureoles
27
with igneous intrusions (HAWTHORNE et al. 1997). The samples also seem to show
trends towards the pargasite end member as you move from the core of the mineral to
the rim of the mineral.
Whole Rock Geochemistry Whole-rock geochemical analyses were conducted on selected xenoliths from the
Pitcairn, El Alamein and Angaston localities. These added to the existing data set of
whole rock analyses of Angaston xenoliths produced by Segui (2010). Candidate
samples for these analyses were restricted by the availability of sufficiently large and
homogeneous samples (usually those larger than ~ 300 gms). Such candidates are far
more abundant at Angaston than at other localities. These data are reported in Table 2
Although most of these xenoliths are now dominated by metamorphic
mineralogies and textures (though as discussed earlier there are good examples where
some xenoliths and some textures are in transition from igneous to metamorphic), a key
issue is the nature and age of their protoliths. It seems likely they have mafic igneous
precursors and if this is the case then key questions include: what is the age and tectonic
significance of the igneous event? Do they have equivalent volcanic suites recognizable
in the geological record of S.E. Gondwana?
Amongst the samples analysed were a few examples of the host kimberlite (Angaston
Kim, EA08 2 and EA08 4). These are ultramafic rocks characterized by having very low
SiO2 and Al2O3 and very high MgO and moderate to high TiO2. They have very high
ignition loss, dominated by CO2. Their trace elements are characterised by high Ni and
Cr and high light rare earth element (LREE) and high LREE/ heavy rare earth elements
28
(HREE) and very high Ba and Sr. One sample analysed is a lower crustal garnet gneiss
(CAX-M10), with a psamo-pelitic composition.
29
Table 2. Whole rock major element geochemistry collected using XRF
Sample Mg# SiO2 %
Al2O3 %
TiO2
% Fe2O3T
% FeO %
Fe2O3
% MnO %
MgO %
CaO %
Na2O %
K2O %
P2O5 %
SO3 %
LOI %
Total %
Metapelite CAX-‐M10 0.53 63.47 19.48 0.83 6.14 2.27 3.62 0.08 2.55 0.59 0.81 5.30 0.03 0.01 0.70 99.99 DS012-‐3 0.51 45.62 12.61 1.48 14.70 8.70 5.03 0.25 9.20 11.21 2.63 0.89 0.08 0.03 1.21 99.90 Orthopyroxenite EA08-‐6 0.83 52.52 3.09 0.11 9.47 6.36 2.41 0.17 30.06 1.35 0.16 0.04 0.00 0.03 1.32 98.31 Eclogitic rock EA08-‐1 0.59 42.78 13.56 1.46 13.06 9.24 2.79 0.21 13.45 10.83 1.30 0.21 0.18 0.13 2.93 100.1 DS012-‐5 0.43 42.05 12.42 2.79 18.01 11.32 5.43 0.18 8.64 11.82 1.36 0.17 0.07 0.29 1.15 98.95 Mafic Gneiss RT-‐CAL 2X1 0.74 49.75 10.85 0.14 10.12 6.57 2.82 0.18 18.74 8.05 0.72 0.14 0.04 0.06 1.01 99.80 RT-‐CAL 2XL10 0.54 53.42 17.35 0.45 8.62 3.90 4.29 0.05 4.63 1.81 3.70 5.82 0.07 0.09 2.31 98.30 EA08-‐9 0.52 50.78 19.19 0.97 7.15 4.32 2.35 0.05 4.65 3.42 4.72 3.58 0.08 0.20 3.57 98.36 Amphibole Eclogite PIT-‐M10 -‐ 51.27 21.49 0.31 5.30 -‐ -‐ 0.08 5.00 9.65 3.78 1.45 0.06 0.10 0.44 98.92 PIT-‐M22 0.49 49.53 11.50 1.24 12.93 11.18 0.51 0.20 10.85 11.45 1.37 0.47 0.14 0.02 0.47 100.1 Amphibole granulite JS KIM PITR 0.44 48.14 16.50 1.52 11.76 8.22 2.63 0.13 6.43 10.14 3.37 0.63 0.07 0.24 0.83 99.76 PIT-‐M26 0.46 48.98 13.97 1.65 13.45 9.41 3.00 0.19 8.04 10.96 1.99 0.72 0.18 0.02 0.60 100.7 DS012-‐4 0.39 37.21 11.91 5.42 17.49 12.75 3.32 0.19 8.31 12.02 2.09 0.53 3.11 0.11 0.85 99.24
30
Table 2 continued
Sample Mg#
SiO2 %
Al2O3 %
TiO2
% Fe2O3T
% FeO%
Fe2O3
% MnO %
MgO %
CaO %
Na2O %
K2O %
P2O5 %
SO3 %
LOI %
Total %
Fd rich Granulite DS012-‐1 0.81 49.39 11.89 1.21 12.63 2.43 9.93 0.20 10.64 11.69 1.54 0.50 0.14 0.03 0.90 100.7 DS012-‐2 -‐ 45.47 16.68 0.79 10.41 -‐ -‐ 0.15 9.41 12.05 2.38 0.20 0.10 0.47 1.20 99.32 DS012-‐6 0.61 46.77 13.81 1.41 10.27 6.63 2.91 0.17 10.49 12.38 2.61 0.50 0.24 0.03 1.37 100.0 DS012-‐7 0.48 44.99 13.88 1.63 14.66 9.51 4.09 0.22 8.93 13.09 2.04 0.17 0.08 0.12 0.58 100.3 EA08-‐3 0.63 42.99 13.35 1.40 12.76 7.94 3.93 0.21 13.31 10.60 1.56 0.23 0.16 0.12 3.01 99.71 EA08-‐5 0.60 42.70 13.75 1.43 13.06 8.61 3.49 0.21 13.02 10.83 1.46 0.19 0.14 0.14 2.70 99.64 EA08-‐8 0.58 42.81 13.74 1.44 13.17 9.53 2.58 0.21 13.30 10.61 1.42 0.19 0.14 0.15 2.75 99.94 PIT-‐M9 0.69 48.73 17.79 0.84 8.31 3.27 4.68 0.12 7.13 10.36 3.33 0.71 0.10 0.32 1.13 98.86 PIT-‐M20 0.60 49.31 14.61 0.48 9.06 6.74 1.57 0.16 10.20 13.46 1.75 0.46 0.10 0.05 0.71 100.3 PIT-‐M21 0.53 50.10 14.39 0.17 10.75 8.33 1.49 0.19 9.50 11.90 1.65 0.34 0.06 0.26 0.29 99.59 PIT-‐M23 0.51 43.71 23.53 0.86 7.29 5.98 0.64 0.07 6.32 10.61 2.87 0.82 0.11 0.19 1.20 97.58 PIT-‐M24 0.61 49.97 16.34 0.33 8.01 5.56 1.83 0.12 8.65 9.43 3.28 1.29 0.06 0.25 0.88 98.61 PIT-‐M25 0.57 51.00 21.19 0.31 5.21 3.80 0.99 0.08 5.01 9.99 3.79 1.35 0.06 0.09 1.01 99.10 Kimberlite ANGAS KIM 0.94 17.30 3.09 1.51 5.08 0.66 4.34 0.08 10.60 30.59 0.37 0.95 0.63 0.09 17.3 87.61 EA08-‐2 0.78 20.43 4.38 3.20 11.90 6.05 5.17 0.20 21.53 13.66 0.19 0.31 0.20 0.21 19.9 96.14 EA08-‐4 0.79 26.75 6.53 1.54 12.70 6.10 5.92 0.20 22.89 9.60 0.18 0.23 0.19 0.03 17.3 98.18 EA08-‐7 0.71 28.86 8.41 1.71 13.98 8.67 4.35 0.23 20.88 9.36 0.15 0.23 0.21 0.03 15.3 99.35
31
Table 3. Whole rock trace element geochemistry collected using XRF
Xenolith Type OPXenite Psamo
pelite Psamo-‐pelite
Eclogitic rock
Eclogitic rock
Amph Eclogite
Amph Granulite
Amph Granulite
Amph Granulite
Amph Eclogite Kimberlite Kimberlite
Location El Alamein Pitcairn Angaston Angaston El Alamein Pitcairn Pitcairn Angaston Pitcairn Pitcairn El
Alamein El Alamein
Sample EA08-‐6 CAX-‐M10 DS012-‐3 DS012-‐5 EA08-‐1 PIT-‐M22 PIT-‐M26 DS012-‐4 JS KIM PIT
PIT-‐M10 EA08-‐4 EA08-‐7
Zr 3.7 192.8 52.4 30.3 84.6 81.9 100.3 43.7 33.5 71.1 86.9 92.0 Nb 0.5 16.4 45.9 2 12.1 4.7 7.3 4.4 22.1 5.4 31.4 23.7 Y 0.8 29.3 28.3 18.9 24.2 24.4 27.4 53.6 17.7 21.9 23.1 26.1 Sr 20.2 175.8 197.6 381.8 140.8 103.2 188.8 527.0 508.5 144.9 376.7 284.8 Rb 3.3 172.4 16.3 12.7 15.1 9.3 14.8 16.2 9.8 11.1 12.8 11.8 U 0.1 1.0 1.6 -‐0.7 -‐1.3 0.5 -‐0.5 0.2 2.3 -‐0.6 1.4 -‐0.2 Th -‐0.2 6.8 0.8 -‐3.1 1.0 -‐0.7 2.3 -‐1.1 -‐0.8 2.5 4.8 2.9 Pb 2.1 75.2 -‐0.9 -‐8.3 -‐3.7 2.7 3.2 0.4 -‐0.6 1.5 2.6 3.2 Ga 4.7 26.1 17.5 12.3 16.4 21.4 21.1 15.6 18.3 17.7 4.9 7.9 Zn 49 62 81 92 100 84 92 89 45 74 61 72 Ni 802 68 347 430 392 216 133 161 89 152 362 419 Cu 87 57 35 220 169 62 96 19 55 36 220 205 Ba 533 1171 502 9492 2469 128 1131 2281 3344 227 389 497 Sc 33.5 9.9 60.9 59.7 40.1 44.9 41.5 39.9 33.1 42.8 53.3 49.8 Co 81 95 105 108 57 152 69 91 64 142 54 61 V 138 84 507 894 376 361 359 577 342 339 513 490 Ce -‐10 48 43 22 39 21 24 80 17 23 58 63 Nd 8 16 27 14 23 17 14 70 8 18 24 28 La -‐1 24 26 10 17 8 5 31 3 8 31 36 Cr 5602 73 17 77 325 434 244 -‐18 117 405 252 217
32
Table 3 continued Xenolith
Type Kimberlite Kimberlite Fd Rich Granulite
Fd Rich Granulite
Fd Rich Granulite
Fd Rich Granulite
Fd Rich Granulite
Fd Rich Granulite
Fd Rich Granulite
Fd Rich Granulite
Fd Rich Granulite
Fd Rich Granulite
Location Angaston El Alamein
El Alamein
El Alamein
El Alamein
El Alamein Pitcairn Pitcairn Pitcairn Angaston Angaston Angaston
Sample Angas Kim EA08-‐2 EA08-‐5 EA08-‐8 EA08-‐9 EA08-‐3 PIT-‐M9 PIT-‐M20 PIT-‐M21 DS012-‐6 DS012-‐7 DS012-‐1 Zr 161.6 47.2 79.4 79.5 224.0 82.0 21.1 25.2 2.8 54.7 25.0 33.2 Nb 109.7 18 7 7.6 16.9 15.5 0.1 1.9 1 8.4 2.8 6.1 Y 40.1 17.9 22.6 23.1 20.0 24.2 5.4 12.5 5.6 22.6 27.4 17.5 Sr 1016.7 476.3 159.7 328.2 674.6 146.5 624.7 130.3 181.7 188.5 269.6 758.0 Rb 73.7 15.9 11.9 12.2 45.0 15.6 18.2 5.1 4.3 12.5 8.3 10.2 U 4.0 2.2 0.6 0.9 0.8 2.2 -‐1.0 0.3 0.8 0.3 0.8 1.3 Th 16.5 -‐1.1 -‐0.5 -‐0.1 -‐0.3 -‐1.3 -‐3.2 1.6 -‐5.2 -‐0.2 -‐2.6 -‐2.8 Pb 4.7 -‐0.4 0.1 2.4 5.9 -‐0.9 -‐4.0 2.6 -‐4.2 -‐0.5 -‐4.7 -‐4.4 Ga 3.8 5.5 17.0 13.7 22.7 13.2 17.3 12.8 13.1 15.9 16.7 15.8 Zn 42 44 94 94 30 89 29 40 43 70 86 65 Ni 288 394 327 473 58 369 99 150 164 298 122 159 Cu 34 22 166 160 84 187 7 35 2 34 85 10 Ba 1901 3466 2155 1876 3350 2229 4947 1220 5573 630 3518 4693 Sc 9.1 38.8 42.0 44.7 22.9 41.9 17.5 45.5 49.9 40.4 57.0 33.8 Co 35 40 100 98 53 81 59 88 105 120 96 91 V 112 345 391 390 179 390 90 229 252 335 529 244 Ce 130 44 27 25 105 44 10 11 3 29 10 21 Nd 59 17 17 18 29 24 2 6 2 19 9 11 La 107 34 10 8 70 24 -‐1 1 0 11 1 7 Cr 317 38 248 521 140 239 164 437 247 531 239 324
33
Table 3 conitinued Xenolith type
Fd Rich Granulite
Fd Rich Granulite
Fd Rich Granulite
Fd Rich Granulite
Fd Rich Granulite
Fd Rich Granulite
Location Angaston Pitcairn Pitcairn Pitcairn Pitcairn Pitcairn
Sample DS012-‐2 PIT-‐M23 PIT-‐M24 PIT-‐M25 RT-‐CAL 2X1 RT-‐CAL 2XL10
Zr 34.7 28.4 11.5 19.6 4.6 303.8 Nb -‐0.2 4.1 1.5 1 0.5 6.7 Y 13.3 14.1 9.2 5.7 3.8 10.2 Sr 1122.2 572.2 364.6 616.7 50.2 656.1 Rb 16.9 10.7 19.8 14.9 3.5 121.0 U 0.8 0.8 -‐0.1 -‐1.0 -‐1.1 0.6 Th 0.8 -‐3.0 -‐2.3 -‐3.6 1.7 1.9 Pb -‐3.2 -‐5.0 -‐1.1 0.0 -‐2.5 4.7 Ga 17.5 14.5 12.6 16.7 9.9 22.8 Zn 39 16 53 24 72 29 Ni 101 163 145 89 358 31 Cu 78 6 93 14 156 21 Ba 3517 9691 6751 3867 558 4579 Sc 25.5 16.2 28.9 17.9 41.2 12.9 Co 87 53 55 46 119 21 V 177 172 136 87 146 166 Ce 26 18 13 10 -‐1 57 Nd 11 4 5 2 7 16 La 7 2 2 1 -‐2 32 Cr 175 316 400 189 1174 95
34
The bulk of the samples, the main subject of this study, are mafic rocks with SiO2 in the
range of 40 – 53%. They plot in the “basalt” and “picritic basalt” fields on the Total
Figure. 11Alkalis (Na2O + K2O) vs SiO2 whole rock geochemistry for the xenoliths plotted using IgPet (Carr 2002). Rock type boundaries described by Cox et al (1979). The South Australian xenoliths plot within the Basaltic region. This shows Angaston (red circles and blue squares) (Segui 2010), El Alamein (yellow crosses) and Pitcairn (green triangles)
Alkalis v SiO2 plot (Figure 11). The suites from the three localities are geochemically
similar. They have high Mg#, mostly >0.6. They are mostly critically undersaturated
and some are Ol-Hy normative (Figure 12). They show positive correlations of MgO
with CaO, FeO (T), Ni and Cr, and show negative correlations of MgO with SiO2 with
TiO2. They have high Ni, Cr and Sc, and relatively low Zr (mostly between 50 and 100
ppm). On a MORB-normalised incompatible trace element diagram (“spider plot”), they
have show flat middle to heavy REE patterns, with associated elements including the
HFSE (Zr, Nb), these elements are MORB-like in relative concentration. They have
somewhat elevated LREE, U, Th and alkalis. Their most distinctive feature is the
presence of very significant Ba enrichments and somewhat lesser Sr.
35
Figure. 12 (A) Calculated CIPW Norm for the South Australian xenoliths, the proposed classification by Thompson (1984) for basalt based on their normative proportions of nepheline (Ne), olivene (Ov), albite (Ab), hypersthenes (Hy) and quartz (Qt). Red circles represent South Australian xenoliths (Segui 2010) which plot within the silica saturated and silica undersaturated portions. Green circles represent MORB (Jenner & O'Neill 2012) which plot in the silica oversaturated and silica saturated parts of the diagram. (B) Mole percent diagram (petrogenetic grid) relevant to variable precent melting (5% to the point where clinopyroxene disappears from the residue) of lherzolite over a pressure range of 0.5 to 3GPa (i.e., about 15-90km depth; pressure shown in bold). Each dashed line at a given pressure represents loci of melt compositions (molar normative) generated by progressive partial melting of lherzolite assemblage (ol + opx + cpx + melt) at that pressure (melt % increasing from left to right on each dashed curve). Each continuous line represents a fixed %melting curve. Aldo shown is the cpx out line. A lherzolitic source rock will lose cpx to the melt beyond this line. Sources of data: Takahashi and Kushiro (1983), Hirose and Kushiro (1993), and Baker and Stopler (1995). Note that it is mainly schematic and does not take into account the changing source composition that must happen as the melt in removed from the source.
In order to demonstrate that the mafic xenoliths do indeed have igneous geochemical
trends, Figure 13 shows the variation of these samples compared with a large global
data set of MORB compositions (Jenner & O'Neill 2012). The Figure 13 plots show
MgO plotted against SiO2, CaO, TiO2 and Al2O3. In these Figures the igneous variation
trend of the MORB suite is indicated with an arrow. The direction towards ‘M’ is that of
melt differentiation, that towards ‘C’ is cumulates or crystal extracts that must drive this
magmatic trend. As can be seen, the xenolith suite shows significant overlap with the
MORB field, leading us to conclude that: 1) these are indeed metamorphosed
36
Figure 13 Plate of whole rock geochemical graphs of South Australian xenoliths (Segui 2010) with MORB (Jenner & O'Neill 2012) for comparison. Graphs A, B, C and D are MgO vs SiO2, CaO, TiO2 and Al2O3 respectively. Diamonds on Graph A and B show mineral compositions plagioclase (PLAG), clinopyroxene (CPX), orthopyroxene (OPX) and the mid ocean ridge basalt (MORB) melt composition, with two distinct trends; 1) a trend towards orthopyroxene showing orthopyroxene crystallisation driving the melt and 2) a trend clinopyroxene + plagioclase showing clinopyroxene + plagioclase driving crystalisation. Black arrow shows igneous variation trends, “M” is the direction towards melt differentiation and “C” is towards the cumulates or crystal extracts that must drive the magmatic trend.
mafic igneous rocks, and 2) that their parent magmas were MORB-like. It is also
obvious that many of the xenolith samples are scattered outside the MORB field,
generally towards higher MgO and lower SiO2, in the ‘C’ direction of the indicated
trend. This is most likely because the xenoliths have experienced some crystal – melt
sorting, many being melt-depleted cumulate-enriched samples (originally gabbros). The
polygons on Figures 13 A and B illustrate the compositional space created by the
mixture of melt and the probable primary igneous phases (orthopyroxene,
clinopyroxene, plagioclase ± spinel). Using the compositional space there appears to be
37
no apparent control on garnet; inferring it was not crystallised as a primary phase but
rather metamorphic.
Figure 14 Graph of barium (Ba) vs wt% MgO for the South Australian xenoliths (Segui 2010) (red circles) with MORB (Jenner & O'Neill 2012) (blue circle). South Australian xenoliths show a several order magnitude higher amounts.
As we mentioned earlier, one very strikingly ubiquitous feature of these rocks’ trace
elements is their extraordinary Ba content. Figure 14 illustrates that these have Ba
several orders of magnitude higher than MORB. We attribute this feature to
contamination by the kimberlite that transported the xenoliths to the surface.
Interestingly, this is a very selective contamination because other trace elements (e.g.
Nb, HREE and Zr) are not affected.
Radiogenic Isotopes
The Finigan MAT 262 TIMS at the University of Adelaide was used to obtain
Samarium-Neodymium radiogenic isotopic data. These analyses were carried out to aid
in identifying the xenoliths’ protoliths and their age, and to compare with the
Neoproterozoic-Cambrian basalt data (Foden et al. 2002) (John Foden per Comms,
38
2012). Seven samples were selected to best represent the three xenolith suites. Three
samples were chosen from each El Alamein (EA08 #) and Pitcairn (Pit M#), and
represent the granulites and eclogites, and one was chosen from Angaston (DS012 #) to
add to Segui’s (2010) isotope data. All data is recorded in Table 2. These data were
plotted using isoplot (Ludwig 2003) with those of Segui (2010), and for comparison
data, Cambrian-Neoproterozoic basalts from S.E. Australia (Foden et al. 2002) (John
Foden per Comms, 2012). The xenoliths isotopic data shows initial εNd0 between -
22.51 (EA08 9) and +8.09 (34.3.2) and Neoproterozoic data εNd700 between -12.02
(EA08 9) and +7.32 (34.4.9).
Table 4 Radiogenic Isotope data
Sample Sm(ppm) Nd(ppm) 147Sm/144Nd 5% error 144Nd/143Nd 2 σ εNd0 εNd700
EA08 3 4.87 21.83 0.1349456 6.75E-‐04 0.512624 8.20E-‐06 0.282851 5.2595917
EA08 5 4.64 17.26 0.1626150 8.13E-‐04 0.512608 1.28E-‐05 0.593011 2.4536666
EA08 9 3.97 30.31 0.0792297 3.96E-‐04 0.511484 8.40E-‐06 22.51296 12.024218
Pit M25 0.66 2.49 0.1603349 8.02E-‐04 0.512565 1.08E-‐05 1.433760 1.7837394
Pit M24 1.14 4.57 0.1508940 7.54E-‐04 0.512249 2.87E-‐05 7.588201 3.4835258
Pit M22 2.22 14.2 0.0945688 4.73E-‐04 0.512481 1.33E-‐05 3.064541 6.0865036
DS012 #1 3.81 8.97 0.2569307 1.28E-‐03 0.512566 1.46E-‐05 1.408401 6.8313195
The Nd-Sm isotopic ratios have been compiled with Segui’s (2010) isotope values for
whole rock samples (Figure 15A). An isochron calculated using the whole rock data
from Pitcairn, Angaston and El Alamein gives an age of 739±680Ma (Figure 15A). This
age is believed to be approximated to the age of the protoliths of the metamorphosed
xenoliths. It is possible that the hint of a steeper trend in the 143/144Nd vs 147Sm/144Nd
isochron diagram is due to inclusion of sample EA08 9, which have some crustal
contamination of the original magmatic protolith by entrained crustal material. This
process has been documented and studied by Rudnick et al (1986)
39
Figure 15. Isochrons calculated using IsoPlot (Ludwig 2003) graphs show 143Nd/144Nd vs 147Sm/144Nd (A) represents whole rock isotope data for the South Australian xenoliths from Angaston (Segui 2010), El Alamein and Pitcairn and gives an age 739±680Ma. (B) South Australian xenoliths (Segui 2010) (green triangles) and Neoproterozoic Cambrian and South Australian Adelaidean basalts (John Foden, per comms) (Blue diamond’s) and gives an age of 656 ± 92Ma.
40
The Late Neoproterozoic age of the mantle xenoliths indicate they probably correlate
well with the suites of Neoproterozoic to Early Cambrian mafic magmas that were
intruded into the Australian passive margin during Rodinia breakup and Gondwana
assembly. These suites include Cambrian and South Australian Adelaidean basalts, such
as the Wooltana and Depot creek basalts and the Gairdner and Broken Hill dykes and
equivalent aged suites on King Island and in Western Tasmania (Foden et al. 2006).
The Nd isotopic composition of these suites show broad correlation with those of the
xenoliths studied (Figure 15B). The El Alamein, Pitcairn, Angaston, the Cambrian and
South Australian Adelaidean data shows positive εNd value at 700Ma but not the
contaminated (EA08 9). When the Cambrian and South Australian Adelaidean basalts
and the El Alamein, Pitcairn and Angaston data is collated the isochron gives the age
656±92Ma, which indicates they both originated in the Late Proterozoic.
Segui (2010) recorded Early Jurassic ages for the Angaston garnet and clinopyroxene
separates. His conclusion was that this age represents the closure temperature (Tc) of the
Sm-Nd system for garnet and clinopyroxene. The clolsure temperature is debated to be
between 600-800oC (Mezger et al. 1992) and the associated age is likely to rep[resent
the timing of the emplacement by the kimberlite, when the minerals (clinopyroxene and
garnet) cooled below their Tc.
Pseudosection and Geothermobarometry P–T pseudosections were calculated for sample Pit –M22 (Amphibole eclogite) (Fig
16,Table 2) and Pit-M25 (Fd rich- eclogite) (Fig 17, Table 2) using the THERIAK-
41
DOMINO software program (De Capitani & Petrakakis 2010), for the geologically
realistic chemical system SiO2–Al2O3–FeO–Fe2O3–MgO–CaO–Na2O–K2O–H2O–TiO2
(NCKFMASHTO). The dataset used compiles the follwing a–x models, which
incorporate Fe3+ end-member minerals: garnet, biotite and melt (White et al. 2007),
orthopyroxene and magnetite (White et al. 2002), amphibole (Diener et al. 2007),
cordierite (Holland & Powell 1998), Clinopyroxene (Green et al. 2007), K-feldspar and
plagioclase (Holland & Powell 2003) and ilmenite (White et al. 2000). Mn is not
considered for the reasons given by White et al. (2007). Mineral abbreviations are as
follows: opx – orthopyroxene; g – garnet; sp – spinel; bi – biotite; ksp – K-feldspar;
ANAB – plagioclase; ilm– ilmenite; mt – magnetite; q – quartz; liq – silicate liquid ⁄
melt. The THERIAK–DOMINO software calculates equilibrium mineral assemblages
for specific bulk-rock compositions that minimises Gibbs-free energy at a given point in
P-T space. Sample Pit M22 (Table 2, Figure 16) petrography defines a peak assemblage
of garnet + clinopyroxene + amphibole. Sample Pit M25 (Table 2, Figure 17)
petrography defines a peak assemblage of garnet + clinopyroxene + plagioclase ±
amphibole
Table 5 Equations used for Geothermobarometry
Thermometer Equation
(Ellis & Green 1979,
Krogh 1988)
→⅓Mg3Al2Si2O12 (pyrope) + CaFeSi2O6 (hedenbergite) ↔
⅓Fe3Al2Si3O12 (almandine) + CaMgSi2O6 (diopsode)
Barometer Equation
(Nimis & Taylor 2000)
CaMgSi2O6 (diopside) + CaCrAlSiO6 (Ca Cr tschermak’s) ↔
½(Ca2Mg)Cr2Si3O12 (uvarovite and knorringite) +
½(Ca2Mg)Al2Si3O12 (grossular and pyrope)
42
The xenoliths thermobarometry calculations have been compared to a number of
different data sets to better understand the geotherm under South Australia (Pearson &
O'reilly 1991, Pearson et al. 1991, Pearson et al. 1995, Segui 2010, Tappert et al. 2011).
Individual spot Temperature estimations were produced using the Ellis and Green
(1979) and Krogh’s (1998) Fe2+-Mg garnet-clinopyroxene exchange thermometers.
These estimations can be seen in Table 5, pressure estimations were calculated with
Nimmis and Taylor (2000) CPX barometer. In addition, these pressures were used for
the temperature calculations. Fe microprobe data used in the geothermobarometry
calculations has been assumed as Fe2+ as the microprobe does not distinguish between
ferric (Fe3+) and ferrous (Fe2+) iron. This assumption of all ferrous iron gives a
minimum temperature for both thermometers. Averaged microprobe data of garnets and
single values of adjacent clinopyroxene spot values were used to create temperature
estimations. The garnet values were averaged to give more confidence they were in
equilibrium with the clinopyroxene. The Ellis and Green (1979) thermometer showed
slightly higher temperatures than Krogh (1988) throughout the slides. Except for rocks
with high pressures (≈ >15kbar) and higher Temperatures (≈ >1035oC), Krogh’s (1988)
thermometer estimated higher temperatures (Table 5).
43
Figure 16. Pseudosection calculated for Pit M22 (see table 2) using THERIAK-DOMINO program (De Capitani & Petrakakis 2010), for the geologically realistic chemical system SiO2-Al2O3-FeO-Fe2O3-MgO-CaO-Na2O-K2O-H2O-TiO2 (NCKFMASHTO). The dataset used compiles the following a-x models which incorporate Fe3+ end-member minerals: garnet, biotite and melt (White et al. 2007), orthopyroxene and magnetite (White et al. 2002), amphibole (Diener et al. 2007), clinopyroxene (Green et al. 2007), K-feldspar and plagioclase (Holland & Powell 2003) and ilmenite (White et al. 2000). Mn is not considered for the reasons given by White et al (White et al. 2007). Blue lines represent major introduction of a mineral to the assemblage (amphibole, garnet and plagioclase), arrow represent direction on pseudosection the mineral labled is introduced. The introduction of garnet to the assemblage turns to Gabbroic rock to granulite and the loss of plagioclase turns granulite to eclogite. Blue shaded polygon represents the mineral assemblage seen for Pit M22 and the blue star represents the pressure and temperature estimations for the sample (see table 6).
44
Table 6 Presure and temperature estimates where TEG79 (Ellis & Green 1979), TK88(Krogh 1988) and PNT95(Nimis & Taylor 2000)
Sample TEG79 TK88 PNT95 Min Max Min Max Min Max
Fd rich Granulite Pit M9 933 972 925 971 14.7 15.3
Pit M20 857 895 836 879 13.1 14.1 Pit M23 1035 1144 1031 1164 23.4 24.8 Pit M24 783 851 715 790 12.4 14.5 Pit M25 893 969 878 967 14.8 15.5 PA 7x2 737 889 659 823 7.1 9.4 PA 78x 733 871 630 818 6.5 8.8 PA7x1 751 896 705 877 6.3 9.6 Eclogitic Rock
PA 5x1 946 1054 931 1060 13 14 PA 5x2 941 977 930 971 29.1 29.8 PA 6x2 857 903 832 883 21 22 PA 6x12 955 1048 946 1056 21.8 28.1 Amphibolite Granulite
Pit M26 812 840 779 810 5.9 7.2 JS Kim 897 944 864 922 15.4 15.9 Amphibolite Eclogite
Pit M10 785 825 724 768 9.1 9.5 Pit M22 724 799 629 739 8.9 11.5
The P- T estimations for the El Alamein xenoliths show a minimum temperature range
between 620°C-1200°C and pressures between 5 and 30kbar (Figure 18). This equates to
a geotherm. Pitcairn xenoliths show a minimum temperature range between 620-1120°C
and pressures between 6-24kbar. The Angaston xenoliths (Segui, 2010) displayed very
similar temperature but slightly more xenoliths with an overall range between 11-
30kbar and between 800-1130°C. The Monk Hill estimated Geotherm (Tappert et al.
2011) will be referred to as the Jurassic paleo-geotherm as it represents a steady state,
whereas the xenoliths of Angaston, Pitcairn, Angaston, EMAC (Pearson & O'REILLY
1991) and SEA (O'Relly & Griffin 1985) show variable equilibration towards this
geotherm. The Monk Hill geotherm is calculated to 40mW/m2
45
Figure 17 Pseudosection calculated for Pit M25 (see table 2) using THERIAK-DOMINO program (De Capitani & Petrakakis 2010), for the geologically realistic chemical system SiO2-Al2O3-FeO-Fe2O3-MgO-CaO-Na2O-K2O-H2O-TiO2 (NCKFMASHTO). The Dataused compiles the following a-x models which incorporate Fe3+ end-member minerals: garnet, biotite and melt (White et al. 2007), orthopyroxene and magnetite (White et al. 2002), amphibole (Diener et al. 2007), clinopyroxene (Green et al. 2007), K-feldspar and plagioclase (Holland & Powell 2003) and ilmenite (White et al. 2000). Mn is not considered for the reasons given by white et al (2007). Red lines represents major introductions of mineral to an assemblage (amphibole, garnet and plagioclase), arrows represent direction on pseudosection the mineral in labled is introduced. The Introduction of garnet to the assemblage turns Gabbroic rock to granulite and the loss of plagioclase turns granulite to eclogite. Red shaded polygon represents the mineral assemblage seen for Pit M25 and the red star represents the pressure and temperature estimations for the sample (see table 6).
46
47
Figure 18 Pressure and temperature plot of geothermobarometry estimations for the SEA (O'Relly & Griffin 1985), EMAC (Pearson & O'REILLY 1991), Monk Hill (Tappert et al. 2011), Angaston (Segui 2010) and Pitcairn and El Alamein. Pressure and temperature estimations using garnet-clinopyroxene Fe-Mg thermometer (Ellis & Green 1979, Krogh 1988) and clinopyroxene barometer (Nimis & Taylor 2000). Data for UHP metamorphic rocks Refrence) schematic subduction metamorphic path taken from Agard (2009) and subduction data points taken from numerous sources(Gao 1999, Dale 2003, Janak 2004). Arrows right of Monk Hill Geotherm (Tappert et al. 2011) show the metamorphic path for the South Australian xenoliths.
and is equivalent to a steady-state continental geotherm (Tappert et al. 2011). Pressure
temperature data for all three regions Pitcairn, El Alamein and Angaston plot at
significantly higher T-values than the subduction P-T related eclogites (Gao 1999, Dale
& Holland 2003, Janak 2004) (Figure 18). This emphasises that the eclogite and
granulite were probably unrelated to subduction. The El Alamein sample derived P-T
array geotherm plot as two distinct groups: 1) a shallow group that plot a wide range of
temperatures for pressures, and ;2) a deep group that plot on the Jurassic paleogeotherm
(Tappert et al. 2011). The Pitcairn geotherm plots very similar to the EMAC (Pearson et
al. 1991) pressure and temperature data.
DISCUSSION
The aim for this study was to better understand the protoliths and metamorphic history
of these unique and distinct upper mantle rocks from the eastern margin of the
Australian Craton, close to the Tasman Line (Veevers & Conaghan 1984). This
knowledge has helped to tell us about paleo tectonics and geothermal dynamics near the
eastern margin of the Australian Craton before the Jurassic (~180ma), the emplacement
age of the xenoliths from kimberlite intrusions (Stracke et al. 1979, Tappert et al. 2011).
The core to this study is the interpretation of the origin of the mafic granulite and
eclogite xenoliths transported from the mantle by Jurassic kimberlite dykes that intrude
48
Adelaidean rocks in the Adelaide Fold Belt. The Adelaide Fold Belt is a composed of
rocks from the Neoproterozoic to Early Cambrian passive margin/rift sequences that
were deformed and metamorphosed during the Mid– to Late Cambrian Delamerian
Orogeny. This orogeny resulted from the onset of subduction in this part of the west
Pacific Margin of Gondwana (Foden et al. 2002, 2006). These mafic xenoliths are,
therefore, potentially very important as they may present us with a unique sample set
from the mantle beneath this rifted-orogen.
These are clearly metamorphic rocks, with strong indications that metamorphism
occurred at upper mantle depths (i.e. within the sub-continental lithospheric mantle).
Their equant and well-equilibrated textures indicate passive metamorphism,
unconnected with strain. As we will discuss, their garnet-cpx-rich metamorphic
assemblages in some samples clearly replace prior plagioclase bearing igneous
assemblages and textures. All the indications are that they have undergone
metamorphism during cooling at depth.
Based on their major element geochemical characteristics (see section Whole rock
geochemistry), there are good grounds for concluding that the origins of the mafic
xenoliths are an igneous suite with geochemical affiliations to rift –related parent
magmas of generally MORB –like geochemical character. This is also the conclusion if
their HFSEs (Nb, Zr and Ti) and HREE and Y concentrations and ratios are considered.
However, they are systematically more silica undersaturated then MORB (Figure 12)
and it is clear that many have bulk compositions that fall outside the field of mafic
igneous melts (Figure 13). These samples may result from crystal-liquid sorting (i.e. the
49
bulk compositions of some of the xenoliths is biased towards their crystallizing
minerals, specifically pyroxenes and plagioclase ± spinel). Some preferential
contamination of the xenoliths has been observed and is likely a result of being
transported by kimberlites, particularly affecting Ba and perhaps other LIL and LREE
elements.
It seems increasingly likely that these eclogites and mafic granulites are mafic magmas
emplaced in the upper mantle below the moho. Their Nd-isotope compositions make
them like some of the Late Neoproterozoic rift tholeiites emplaced during the passive
margin stage in this part of S.E. Gondwana in Adelaidean sequences in Tasmania, South
Australia, W. New South Wales and W. Victoria, (Foden et al., 2002). This origin is
similar to that proposed for some petrologically similar suites of xenoliths transported
by Cenozoic alkali basalts in eastern Australia (O'Relly & Griffin 1985, Rudnick et al.
1986). However, it is supposed that the eastern Australian xenolith suite represents
mafic sub-crustal underplating of Cenozoic age (Rudnick et al. 1986) whereas these
South Australian xenoliths probably preserve Late Neoproterozoic to Early Cambrian
mantle events.
If the xenoliths represent magmas that crystallized at high pressures, then they may
have distinctive characteristics that distinguish them from erupted or shallowly intruded
mafic magmas, such as MORB. We have already observed that though there are
suggestions that the suite may have a MORB-like mantle source (based on HFSE and
HREE geochemistry), they are systematically more silica undersaturated than MORB
(Figure 12A). It is a systematic feature that melts derived from increasingly high
50
pressure partial melting of mantle peridotite are increasingly more silica undersaturated
(Takahashi 1983, Hirose 1993, Baker 1995) (Figure 12B). The liquidus phase of mafic
melts typically shifts from being olivine (± plagioclase) at low pressures (< 5 kbar) to
pyroxene (CPX and/ or OPX) + olivine at pressures > 5 kbar, to be joined by garnet
above ~ 11-12 kbar (see fig 2 in Rudnick et al. 1986).
Under these circumstances, the fractionation trends of mafic melts crystallizing at >
Moho depths should be dominated much more by pyroxene than those of shallowly
crystallizing mafic magmas, where early olivine will dominate. As pyroxene has a
crystal-melt distribution Cr >> Ni (but both > 1) , whereas olivine has Ni>>Cr (Ni >>1),
this should mean that high pressure fractionation should lead to more rapid Cr depletion
than in lower pressure fractionation. Figure 19 shows the Ni-Cr variation of the South
Australian mafic xenoliths and the field of MORB melts. This shows that the most melt-
like xenoliths have relatively lower Cr/Ni ratios than MORB. Figure 19 shows melt
fractionation curves for high pyroxene/olivine (high pressure) and lower
pyroxene/olivine trends and indicate that the cumulate compliment to the high pressure
trend falls across the more cumulative xenolith compositions.
51
Figure 19 Ni-Cr (ppm) variation of the South Australian mafic xenoliths (Segui 2010) (red circles) and MORB data (Jenner & O'Neill 2012) (green circles). Trends on this Figure show melt fractionation curves for high pyroxene/olivene (high pressure) (blue line) and lower pyroxene/olivene trends (black line with yellow triangles) and the complimentary cumulate trend for a high pressure (black line with orange circles). Trends created using MELTS (Ghiorso & Sack 1995). The specific chosen starting basalt used was an olivene tholeiite from the Adelaidean Smithon basin in N.W. Tasmania. This was chosen as it clearly had experienced no crustal contamination (John Foden, per comms)
Another piece of very good evidence for high pressure crystallization of the xenolith
pre-cursor magmas is provided by the orthopyroxenite xenolith from El Alamein (EA
08 6). This plots as a relatively high MgO, low CaO point on the MgO-CaO variation
diagram (Figure 13) and aligns with some other samples at higher Cao and lower MgO,
inferring a phase of orthopyroxene crystallization. To produce a mono-mineralic
52
cumulate figure 13 shows that the P-T conditions must be on the mafic melt liquidus
and implies high-pressure crystallization.
To test if the compositions of the xenoliths can be modelled by high pressure
crystallization of an olivine tholeiite using one of the specific Neoproterozoic basalts
from this part of S.E Gondwana, calculations were made using the MELTS2 software.
The specific chosen starting basalt used was an olivine tholeiite from the Adelaidean
Smithton basin in N.W. Tasmania. This was chosen as it had clearly experienced no
crustal contamination.
This composition was run with low water content and an oxygen fugacity of QFM +1.
Figure 20 plots the temperature versus melt percentage for the most favourable run
made at a pressure of 8.5 kbar, and indicates that this composition has orthopyroxene on
its liquidus at this pressure. At pressures less than 5kbar it does, indeed, have olivine on
its liquids, while at pressures > 10 kbar it crystallizes significant amounts of garnet later
in its cooling history (note that there is no geochemical evidence that the xenoliths had a
magmatic history of garnet crystallization and accumulation).
2 MELTS is a software package designed to facilitate thermodynamic modeling of phase equilibria in magmatic systems. It provides the ability to compute equilibrium phase relations for igneous systems over the temperature range 500-2000 °C and the pressure range 0-2 GPa. Ghiorso, Mark S., and Sack, Richard O. (1995) Chemical Mass Transfer in Magmatic Processes. IV. A Revised and Internally Consistent Thermodynamic Model for the Interpolation and Extrapolation of Liquid-Solid Equilibria in Magmatic Systems at Elevated Temperatures and Pressures. Contributions to Mineralogy and Petrology, 119, 197-212 Gualda G.A.R., Ghiorso M.S., Lemons R.V., Carley T.L. (submitted) Rhyolite-MELTS: A modified calibration of MELTS optimized
53
Figure 20. percentage of melt remaining vs Temperature showing melt evolution path (red crossed) and the complimentary solid cumulate path (black dashes) created from the results of MELTS (Ghiorso & Sack 1995) modelling on the specific chosen starting basalt, an olivene tholeiite from the Adelaidean smithton basin in N.W. Tasmania. This was chosen as it clearly had experienced no crustal contamination (John Foden, per comms). The most favourable run made at pressure 8.5 kbar, low water content and oxygen fugacity of QFM + 1as seen in the MELTS list. Minerals crystallised orthopyroxene (OPX), clinopyroxene (CPX), spinel (SP) and plagioclase (PLAG)
The results of the MELTS modelling showing the melt evolution path (blue crosses) and
the complimentary solid cumulate path (black dashes), is shown for MgO v CaO and
SiO2 v MgO on Figure 21A and B. The model maps the early OPX-driven trend of the
melt with its prominent inflection in the MgO-CaO variation brought about by the onset
of CPX and plagioclase crystallization. It also predicts an initial liquidus OPX that is
virtually identical to that in El Alamein xenolith EA 08 6.
54
In summary, there seems excellent evidence that the mafic xenolith suite formed as a
sub-Moho depth, mafic underplate during Neoproterozoic rifting.
Figure 21 MELTS (Ghiorso & Sack 1995) modelling of the south Australian xenoliths on Wt% MgO vs Wt% CaO (A) and SiO2 (B). Melt evolution path (blue crosses) and the complimentary solid cumulate path (black dashes) are shown on the diagram. Starting compostion is shown to be the orthopyroxenite (sample EA08 6).
The Sm-Nd isochron calculated from whole rock geochemistry gives the protoliths for
the xenoliths an age of 656±92Ma (Figure 15B). This age corresponds to the
Neoproterozoic to Early Cambrian passive margin/rift sequences found at the south
eastern margin of Gondwana. Magmatic underplating has been explained as the
plausible origin for these rocks, this crystallisation saw the magmatic melts rise through
the mantle and then along the adiabat (Figure 18) to crystallise at high pressures. The
rocks then proceeded to be metamorphosed and now show re-equilibration textures, as
noted in the “Petrology” section before.
A B
55
Metamorphism of the xenoliths saw them evolve from plagioclase + pyroxene ± olivine
cumulates/melts to the garnet granulites and eclogites that make up the samples (Figure
3A). As observed in the “whole rock geochemistry” section and above in Figure 20, the
starting rock assemblage is assumed to be an igneous plagioclase + olivine + pyroxene
cumulate, possibly similar to the gabbroic rock seen in Figure 3C. These samples
showed no primary garnet crystallisation. I suggest that these rock cooled near
isobarically as the EMAC xenoliths did (Pearson & O'reilly 1991), to temperatures of
620-1200°C from ~1300°C at pressures between 5-30kbar (geothermobarometry
estimations). Once cooling crystallised the xenoliths, they continued to cool to the
geothermal gradient. This initiated the metamorphism and activated the plagioclase
pyroxene reaction to form garnet + quartz carona rims (Equations 1 and 2). The
Angaston xenoliths experienced a very similar event except these xenoliths
recrystallised Kyanite in their metamorphic reactions much like the EMAC suite
(Pearson & O'reilly 1991, Segui 2010). These reactions went on the colder the rocks
became, until the plagioclase was removed from the assemblage to create eclogites or
they reached the geotherm.
Figure 18 shows the pressure and temperature estimations for the mafic eclogites and
granulites along with the adiabat and subduction metamorphic paths. Rocks that have
metamorphosed through subduction are characterised by ultra high pressures and low
temperature gradient. They also display a clockwise metamorphic path due to thermal
lag. If the xenoliths were metamorphosed due to subduction you would expect there to
be some evidence of this within the rocks, such as up temperature and up pressure
metamorphic textures. However I believe these rocks however I believe have evidence
56
for a metamorphic path that reflects isobaric cooling from the adiabat towards the stable
cratonic geotherm gradient, which in the Jurassic was defined by the Monk Hill
Geotherm (Tappert et al. 2011) as mentioned above.
Two groups of pressure and temperature estimations were noted in the xenoliths data; a
low pressure group with varied temperatures and a high pressure group (Figure 18). The
high pressure group has estimations that reach the Jurassic geotherm (Monk Hill
geotherm) (Tappert et al. 2011), while the low pressure group is much more diverse in
temperature and no samples have reached the geotherm. A simple answer for this
difference in location due to depth (30-90km) is the amount of cooling needed to reach
the geotherm and to equilibrate according to pressure and temperature conditions.
Pseudosections created using the THERIAK-DOMINO program (De Capitani &
Petrakakis 2010) for samples Pit M22 (Amphibole Eclogite) (Figure 16) and Pit M25
(Fd-rich Granulite) (Figure 17) give further evidence to isobarically cooling
metamorphic reactions that were observed in the “mineral relationship” section by
Equation 1 and 2 and Figure 3 A, B and C. Tracing metamorphic paths under isobaric
cooling, the pseudosections of Pit M22 and Pit M25 (Figure 16 and 17) show that the
xenoliths, depending on their original pressure will produce the assemblages and rock
types we have sampled from Angaston (Segui 2010), El Alamein and Pitcairn. Segui
(2010) created a pseudosection for a granulite xenolith from Angaston and found
similar mineral relationships, mainly that an isobarically cooled magma will form
granulite and eclogite rock types.
57
The comparison of the two pseudosections (Figure 16 and 17) gives evidence for the
transition of granulite to eclogite, to not just represent differences in pressure and
therefore depth, but also differences in bulk composition. Wood (1987) showed that a
quartz-tholeiite composition, which experienced Isobaric cooling (IBC) at lower crustal
pressures, would not produce eclogites; even on a steady state geotherm (40mW/m2),
but would remain a granulite. The rock type transition zones, granulite → eclogite (loss
of plagioclase from the assemblage) and gabbroic rock → granulite (introduction of
metamorphic garnet) depicted by blue lines in Pit M22 and red lines in Pit M25 are at
much lower temperatures for Pit M22. For a given temperature of 600oC the granulite to
eclogite transition (loss of plagioclase from the mineral assemblage) is at 14.7kbars for
Pit M25 and 11.6kbars for Pit M22 this shows that it is more difficult for Pit M25 to
reach the eclogite field (no Plagioclase). This difference in transition temperatures at the
same pressures, is possibly due to the high amount of aluminium needed to form
Plagioclase. Once in the eclogite P-T field, the granulites are at a much higher
temperature; thus a lot more energy, in the form of heat or pressure, is required to from
the cpx and garnet mineralogy of eclogites.
The geochemistry section describes that garnet was not a primary mineral in
crystallisation from the original melt. The MELTS software was used to calculate the
melt and cumulate trends associated with these xenoliths (Figure 20 and 21) and it
illustrated that garnet was a part of the primary mineral trends. Therefore, I suggest the
original emplacement onto the bottom of the Moho was no deeper than 9kbar or 30km.
Using the geothermobarometry estimations we see that some of the estimates reach
pressures of 25-30kbar, which represents depths of 80 to 90km (Table 6, Figure 18).
58
The cessation of the Delamerian Orogeny has been inferred to have occurred due to a
buoyancy control (i.e. delamination) (Foden et al. 2006). Modelled delamination of
eclogite has shown that some of the delaminated material can be left behind (Percival &
Pysklywec 2007) and not fully delaminated into the mantle. This delamination is
inferred to take ~12myr but the pressure increase is instantaneous and may result in the
higher pressure rocks observed in the suite (Figure 18).
In conclusion these xenoliths show metamorphic signatures indicative of near isobaric
cooling towards the stable geotherm and converted gabbroic rocks to granulites and
eclogites due to their original emplacement depth (30km) and their original bulk
composition higher pressures were then reached by some xenoliths possibly due to
delamination. There is also evidence that the metamorphic paths were not controlled by
subduction.
CONCLUSIONS
The major aim of this thesis was to further constrain the origin and metamorphic
processes under the eastern margin of the Australian craton. Kimberlitic xenoliths from
Pitcairn, Angaston and El Alamein suites are dominated by mafic granulite and eclogite
xenoliths. Modal proportions of the minerals garnet, clinopyroxene and plagioclase
were used, along with Al(6)/Al(4) ratios of clinopyroxenes, to distinguish between the
different rock types. Geochemically all three regions of xenoliths were very similar.
However, modally the Pitcairn xenoliths showed a larger percent of feldspar and a more
varied composition of feldspar. Whole rock geochemistry gave evidence that the mafic
xenolith suite formed as a sub-Moho depth, mafic underplate during Neoproterozoic
rifting with a more than likely MORB style melt high pressure cumulates and melts
59
Geothermobarometry estimates were varied but all fell between the EMAC geotherm
(Pearson et al. 1991) and the Monk Hill garnet peridotite geotherm (Tappert et al.
2011). This has been identified as the signature of the sub lithospheric mantle cooling
from underplating mafic intrusions from the adiabat towards the stable geotherm (Monk
Hill geotherm) and not subduction related.
The end of the Delamerian Orogeny has been suggested to have occurred due to a
delamination event around the middle Ordovician under the Adelaidean Fold Belt
(Foden et al. 2006). The xenoliths of this study may, therefore, represent the
delaminated remains of the eclogite that caused the cessation of the Delamerian
Orogeny. Further work could be done to justify this hypothesis through dating the
metamorphic age of these rocks to further constrain the process.
Acknowledgments.
I would like to thank Kevin Wills for his extensive knowledge of the kimberlites located
in South Australia and for his help while conducting the field work. Thanks must go to
Phd student, Alec Walsh for helping me to make my two pseudosections, which would
not have been completed without his help. I would also like to thank David Bruce, John
Stanley and Katie Howard for introductions on the sample handling and preparing
equipment in the Mawson Laboratories. I would also like to thank Angus Netting, Ben
Wad and Adelaide microscopy for the use and help with their instruments. But my
60
biggest thanks must go to my supervisor John Foden for his countless talks and hours he
put into to help me understand and finish this thesis.
REFERENCES
AGARD P., YAMATO, P, JOLIVET, L, BUROV, E 2009. Exhumation of oceanic blueshcists and eclogites in subduction zones: Timing and mechanisms. Earth-Science Reviews 92, 53-79. BAKER M. B. A. S., E 1995. Compositions of near-‐solidus peridotite melts from experiments and thermodynamic calculations. Nature 375, 309-‐311. BRANSON J., MOSS F. & TAYLOR F. 1968. Deep crustal reflection seismic test survey, Mildura, Victoria and Broken Hill. NSW. CARR M. 2002. IGPET for Windows. Somerset, NJ: Terra Softa Inc. COLCHESTER D. M. 1972. A perliminary note on kimberlite occurences in South Australia. Journal of the Geological Society of Australia 19, 383-‐386. COLEMAN R. G., LEE D. E., BEATTY L. B. & BRANNOCK W. W. 1965. ECLOGITES AND ECLOGITES -‐ THEIR DIFFERENCES AND SIMILARITIES. Geological Society of America Bulletin 76, 483-‐&. DALE J. & HOLLAND T. 2003. Geothermobarometry, P–T paths and metamorphic field gradients of high-‐pressure rocks from the Adula Nappe, Central Alps. Journal of Metamorphic Geology 21, 813-‐829. DALE J. A. H., T.J.B. 2003. Geothermobarometry, P–T paths and metamorphic field gradients of high-‐pressure rocks from the Adula Nappe, Central Alps. Journal of Metamorphic Geology 21, 813-‐829. DE CAPITANI C. & PETRAKAKIS K. 2010. The computation of equilibrium assemblage diagrams with Theriak/Domino software. American Mineralogist 95, 1006-‐1016. DIENER J. F. A., POWELL R., WHITE R. W. & HOLLAND T. J. B. 2007. A new thermodynamic model for clino-‐ and orthoamphiboles in the system Na2O–CaO–FeO–MgO–Al2O3–SiO2–H2O–O. Journal of Metamorphic Geology 25, 631-‐656. ELLIS D. J. & GREEN D. H. 1979. An experimental study of the effect of Ca upon garnet-‐clinopyroxene Fe-‐Mg exchange equilibria. Contributions to Mineralogy and Petrology 71, 13-‐22. FERGUSON J. & SHERATON J. 1979. Petrogenesis of kimberlitic rocks and associated xenoliths of southeastern Australia, pp. 140-‐160. FLEET M. & ZUSSMAN J. 2003. Rock-‐forming minerals (Vol. 3). Geological Society of America. FODEN J., ELBURG M. A., DOUGHERTY-‐PAGE J. & BURTT A. 2006. The timing and duration of the Delamerian orogeny: Correlation with the Ross Orogen and implications for Gondwana assembly. Journal of Geology 114, 189-‐210. FODEN J., SONG S. H., TURNER S., ELBURG M., SMITH P., VAN DER STELDT B. & VAN PENGLIS D. 2002. Geochemical evolution of lithospheric mantle beneath SE South Australia. Chemical Geology 182, 663-‐695.
61
GAO J., KLEMD, R., ZHANG, L., WANG, Z. AND XIAO, X. 1999. P–T path of high-‐pressure/low-‐temperature rocks and tectonic implications in the western Tianshan Mountains, NW China. Journal of Metamorphic Geology 17, 621-‐636. GHIORSO M. S. & SACK R. O. 1995. Chemical mass transfer in magmatic processes IV. A revised and internally consistent thermodynamic model for the interpolation and extrapolation of liquid-‐solid equilibria in magmatic systems at elevated temperatures and pressures. Contributions to Mineralogy and Petrology 119, 197-‐212. GLEADOW J. & EDWARDS A. 1978. Mineralogical note: Fission-‐track age of a basic inclusion from the Kayrunnera kimberlitic breccia pipe. Journal of the Geological Society of Australia 25, 359-‐359. GOLDSTEIN S. L. O. N. R. K. H. P. J. 1984. A Sm-‐Nd isotopic study study of atmospheric dust and particulates from major river systems. Earth and Planetary Science Letters 70, 221-‐236. GREEN E., HOLLAND T. & POWELL R. 2007. An order-‐disorder model for omphacitic pyroxenes in the system jadeite-‐diopside-‐hedenbergite-‐acmite, with applications to eclogitic rocks. American Mineralogist 92, 1181-‐1189. GRIFFIN W. L. & O'REILLY S. Y. 2007. Cratonic lithospheric mantle: is anything subducted? Episodes-‐Newsmagazine of the International Union of Geological Sciences 30, 43-‐53. HAWTHORNE F. C., KATO A., KISCH H. J., KRIVOVICHEV V. G., LINTHOUT K., LAIRD J., MARESCH W. V., SCHUMACHER J. C., STEPHENSON N. C. N. & WHITTAKER E. J. W. 1997. Nomenclature of amphiboles: report of the subcommittee on amphiboles of the International Mineralogical Association, Commission on New Minerals and Mineral Names. The Canadian Mineralogist 35, 219-‐246. HIROSE K. A. K., I. 1993. Partial melting of dry peridotites at high pressures: determination of compositions of melts segregated from peridotite using aggregates of diamond. Earth and Planetary Science Letters 114, 477-‐489. HOLLAND T. & POWELL R. 2003. Activity–composition relations for phases in petrological calculations: an asymmetric multicomponent formulation. Contributions to Mineralogy and Petrology 145, 492-‐501. HOLLAND T. J. B. & POWELL R. 1998. An internally consistent thermodynamic data set for phases of petrological interest. Journal of Metamorphic Geology 16, 309-‐343. JACOB D. 2004. Nature and origin of eclogite xenoliths from kimberlites. Lithos 77, 295-‐316. JANAK M., FROITZHEIM, N., LUPTAK, B., VRABEC, M., KROGH RAVNA, E.J. 2004. First evidence for ultrahigh-‐pressure metamorphism of eclogites in Pohorje, Slovenia: Tracing deep continental subduction in the Eastern Alps. Tectonics 23. JENNER F. E. & O'NEILL H. S. C. 2012. Analysis of 60 elements in 616 ocean floor basaltic glasses. Geochemistry Geophysics Geosystems 13, Q02005. KROGH E. J. 1988. The garnet-‐clinopyroxene Fe-‐Mg geothermometer -‐ a reinterpretation of existing experimental data. Contributions to Mineralogy and Petrology 99, 44-‐48. LUDWIG K. 2003. Isoplot/Ex Version 3.00: a geological toolkit for Microsoft Excel. Berkeley Geochronology Center Special Publication, 70pp 4.
62
MCCULLOCH M. T. A., ARCULUS, R.J.A, CHAPPELL, B.W.B, FERGUSON, J. 1982. Isotopic and geochemical studies of nodules in kimberlite have implications for the lower continental crust. Nature 300. MEZGER K., ESSENE E. J. & HALLIDAY A. 1992. Closure temperatures of the Sm—Nd system in metamorphic garnets. Earth and Planetary Science Letters 113, 397-‐409. NIMIS P. & TAYLOR W. R. 2000. Single clinopyroxene thermobarometry for garnet peridotites. Part I. Calibration and testing of a Cr-‐in-‐Cpx barometer and an enstatite-‐in-‐Cpx thermometer. Contributions to Mineralogy and Petrology 139, 541-‐554. O'RELLY S. & GRIFFIN W. 1985. A xenolith-‐derived geotherm for southeastern Australia and its geophysical implications. Tectonophysics 111, 41-‐63. PEARSON N. & O'REILLY S. 1991. Thermobarometry and P–T–t paths: the granulite to eclogite transition in lower crustal xenoliths from eastern Australia. Journal of Metamorphic Geology 9, 349-‐359. PEARSON N., O'REILLY S. & GRIFFIN W. 1995. The crust-‐mantle boundary beneath cratons and craton margins: a transect across the south-‐west margin of the Kaapvaal craton. Lithos 36, 257-‐287. PEARSON N. J., O'REILLY S. Y. & GRIFFIN W. L. 1991. The granulite to eclogite transition beneath the eastern margin of the Australian craton. European Journal of Mineralogy 3, 293-‐322. PERCIVAL J. A. & PYSKLYWEC R. N. 2007. Are Archean lithospheric keels inverted? Earth and Planetary Science Letters 254, 393-‐403. RUDNICK R., MCDONOUGH W., MCCULLOCH M. & TAYLOR S. 1986. Lower crustal xenoliths from Queensland, Australia: evidence for deep crustal assimilation and fractionation of continental basalts. Geochimica et Cosmochimica Acta 50, 1099-‐1115. SEGUI D. 2010. A study of kimberlitic eclogites and mafic granulites fromt the southern Adelaide fold belt, Geology Honours. SONG S. H. 1994. Geochemical evolution of Phanerozoic Lithospheric mantle beneath S.E. South Australia. SPARKS R. S. J., BAKER L., BROWN R. J., FIELD M., SCHUMACHER J., STRIPP G. & WALTERS A. 2006. Dynamical constraints on kimberlite volcanism. Journal of Volcanology and Geothermal Research 155, 18-‐48. STRACKE K., FERGUSON J. & BLACK L. 1979. Structural setting of kimberlites in southeastern Australia. Kimberlites, diatremes, and diamonds: their geology, petrology, and geochemistry 1, 71. SUTHERLAND F. L. & HOLLIS J. D. 1982. MANTLE LOWER CRUST PETROLOGY FROM INCLUSIONS IN BASALTIC ROCKS IN EASTERN AUSTRALIA -‐ AN OUTLINE. Journal of Volcanology and Geothermal Research 14, 1-‐29. TAKAHASHI E. A. K., I 1983. Melting of a dry peridotite at high pressures and basalt magma genesis. American Mineralogist 68, 859-‐879. TAPPERT R., FODEN J., MUEHLENBACHS K. & WILLS K. 2011. Garnet Peridotite Xenoliths and Xenocrysts from the Monk Hill Kimberlite, South Australia: Insights into the Lithospheric Mantle beneath the Adelaide Fold Belt. Journal of Petrology 52, 1965-‐1986.
63
TAPPERT R., FODEN J., STACHEL T., MUEHLENBACHS K., TAPPERT M. & WILLS K. 2009. Deep mantle diamonds from South Australia: A record of Pacific subduction at the Gondwanan margin. Geology 37, 43-‐46. VEEVERS J. & CONAGHAN P. 1984. Phanerozoic earth history of Australia (Vol. 2). Oxford University Press, USA. WHITE R. W., POWELL R. & CLARKE G. L. 2002. The interpretation of reaction textures in Fe-‐rich metapelitic granulites of the Musgrave Block, central Australia: constraints from mineral equilibria calculations in the system K2O-‐FeO-‐MgO-‐Al2O3-‐SiO2-‐H2O-‐TiO2-‐Fe2O3. Journal of Metamorphic Geology 20, 41-‐55. WHITE R. W., POWELL R. & HOLLAND T. J. B. 2007. Progress relating to calculation of partial melting equilibria for metapelites. Journal of Metamorphic Geology 25, 511-‐527. WHITE R. W., POWELL R., HOLLAND T. J. B. & WORLEY B. A. 2000. The effect of TiO2 and Fe2O3 on metapelitic assemblages at greenschist and amphibolite facies conditions: mineral equilibria calculations in the system K2O-‐FeO-‐MgO-‐Al2O3-‐SiO2-‐H2O-‐TiO2-‐Fe2O3. Journal of Metamorphic Geology 18, 497-‐511
64
APPENDIX A: METHODS
Sample Preparation
Whole rock and mineral separate preparation for radiogenic isotope and XRF work was carried
out in the Mawson Laboratories at Adelaide University. Sample preparation involved the removal
of weathered surfaces from samples using a diamond saw to produce 5-7 cm3 blocks to be
crushed. The crushing was completed in a stainless steel jaw crusher which was cleaned in-
between each sample. The fine and coarse material produced from crushing was then divided
into thirds with care taken to ensure each fraction contained representative amounts. Two of
these fractions were further processes for whole rock analysis and isotope work while the other
fraction was set aside for future work. Cleaning was very extensive using all machines and
implements between samples.
Samples that were selected for whole rock XRF work were then milled down to a fine powder in a
tungsten carbide mill which ran for between 2 to 3 minutes.
Major Element Geochemistry
Major and minor element analysis was undertaken in situ on individual mineral grains in thin
section. The Cameca SX51 microprobe located at Adelaide University was used to collect major
and minor element results. It produced these results using a focused beam produced by a 15kV
accelerating voltage and a 20nA beam current. Calibration of the machine was carried out during
and before analysis of samples.
Whole rock major element analysis involved the production of fused glass disks for XRF work. To
remove the water from the sample 6g of the sample powder was heated at 130oC for 4 hours.
Samples were then put into ceramic crucibles and weighed in a Toledo balance. The loss of
ignition was calculated once the sample was ignited at 960oC for 3 hours.1g of this powder was
then mixed with 4 g of lithium borate flux and fused in Pt/Au crucibles and moulds producing
fused discs.
65
Trace Element Geochemistry Whole rock trace element geochemistry was acquired by XRF analyses carried out on pressed
pellets created from milled sample using the same procedure as the major elements. Pressed
pellets were created by mixing 0.8mL of EtOH/PVA binder with 6g of whole rock powder and then
put into the hydraulic press found in the Mawson Laboratories.
Radiogenic Isotope work Radiogenic isotope work was carried to obtain Nd-Sm values of samples so they could be
radiometrically dated. It was undertaken for 7 samples. The standard used was a weighed and
spiked into cleaned 15mL Teflonware PFA Vials using a Mettler Toledo AT201 balance. The
samples were spiked using Nd-Sm spike F (calculated at the additions of 0.2 Sm-Nd spike F per
1µg Nd) at estimations of 4 ppm for the whole rocks and 10ppm for the mineral separates with
the standard known to be 25ppm spiked accordingly.
The samples are dissolved through a process involving 2 lots of HF, and a final dissolution in
HCL with 7M HNO3 being added at crucial stages to retard the formation of insoluble fluorides.
Samples are then centrifuged and loaded onto a Biorad Poly Prep separating columns (2mL
AG50W X8 200-400 mesh Biorad cation exchange resin) for initial REE separation and then
loaded onto the Sm-Nd separating columns (2mL Teflon powder impregnated with HDEHP)
separating the Nd and Sm. Second dissolution involved the addition of 15M HNO3 along with 2mL
.01 µg/g H3BO3 in 6M HCL with the samples being capped and boiled for 3 days after no aqua
regia was observed. Ultra sounding was also undertaken in an attempt to ensure complete
dissolution of samples.
Nd and Sm were subsequently dried and located onto double Re filaments for Thermal Ionisation
Mass Spectrometer (TIMS) analysis. Nd and Sm analysis was carried out on a Finigan Mat 262
TIMS at Adelaide University using dynamic measurement for 143Nd/144Nd and static measurement
for 150Nd/144Nd, 147Sm/149Sm and 152Sm/149Sm. The blank failed contained <200pg 150Nd/144Nd
and <150pg Sm. The international standard JNDi-1 produced 144Nd/ 143Nd measurements of
.512074 ± 27 (n=2) and the BCR2 basalt standard produced results of 144Nd/ 143Nd .512662 ± 41
(n=1). 143Nd/144Nd and 147Sm/144Nd values were calculated using depleted mantle values taken
from Goldstein et al (1984).
66
APPENDIX B AVERAGE GARNET AND CLINOPYROXENE DATA (SEE EXTENDED APPENDIX FOR ALL DATA) Garnet
Sample Rock Type SiO2 TiO2 Al2O3 Cr2O3 ∑FeO MnO MgO CaO Na2O K2O Sum PA 5x1 ER 38.6922 0.1099 21.8583 0.0072 20.3850 0.8484 9.1744 8.3938 0.0426 0.0083 99.52 PA 5x2 ER 38.7040 0.0696 21.1053 0.0101 21.6789 0.4797 6.7896 9.5448 0.0384 0.0103 98.43 PA 6x2 ER 38.2422 0.0353 21.6609 0.0212 0.4386 22.7472 7.0507 9.2672 0.0208 0.0060 99.49 PA 6x12 ER 37.5902 0.1116 21.8501 0.0221 19.5057 0.4637 8.9194 9.4045 0.0450 0.0065 97.92 PA 78x Fd 38.5500 0.0423 21.0820 0.0605 22.2608 0.3594 11.3838 5.7355 0.1990 0.2742 99.95 PA 7x1 Fd 38.0804 0.0318 21.1569 0.0105 24.4398 0.6274 8.1536 8.4626 0.0207 0.0323 101.02 PA 7x2 Fd 38.8425 0.0255 20.9259 0.0189 18.9980 0.3404 11.6534 5.7642 0.3288 0.6839 97.58 Pit M9 Fd 38.8691 0.0605 22.4463 0.0226 17.1976 0.3970 9.7796 11.1657 0.0166 0.0085 99.96 Pit M20 Fd 38.8420 0.0443 22.6527 0.0199 16.5478 0.3961 10.7405 10.2007 0.0165 0.0063 99.47 Pit M23 Fd 38.2208 0.1030 23.5829 0.0203 14.6148 0.1915 9.0226 13.4335 0.0344 0.0174 99.24 Pit m25 Fd 39.6510 0.0485 22.3432 0.0164 16.9428 0.4451 9.2149 11.4013 0.0810 0.0230 100.17 Pit M24 Fd 40.2964 0.0222 22.2574 0.0387 17.6654 0.4461 12.2032 5.8706 0.0656 0.2464 99.11 JS Kim AM G 37.8135 0.0478 21.8043 0.0115 22.5739 0.4605 8.5213 7.8107 0.0184 0.0075 99.07 Pit M26 AM G 37.6813 0.2297 21.5027 0.0270 23.2270 0.4806 7.4231 8.5601 0.0064 0.0212 99.16 Pit M10 AM E 39.8328 0.0000 22.3251 0.0207 21.5280 0.5797 10.0807 6.4044 0.0185 0.0116 100.80 Pit M22 AM E 39.1188 0.0344 22.2700 0.0275 21.5163 0.5668 10.1163 6.3541 0.0344 0.0142 100.05
67
CPX Sample Rock Type SiO2 TiO2 Al2O3 Cr2O3 ∑FeO MnO MgO CaO Na2O K2O Sum
PA7x2 Fd 52.7823 0.1949 3.4947 0.0472 7.3557 0.0539 17.2495 15.9123 1.5538 0.0071 98.6513 PA78x Fd 51.4913 0.2102 4.1645 0.0666 6.2070 0.0446 13.5394 20.0805 2.0399 0.4344 98.2785 PA7x1 Fd 50.2455 0.3142 5.0639 0.0141 8.1499 0.0510 11.5260 19.3984 2.4359 0.0391 97.2380 PA6x2 ER 52.5858 0.2801 8.0752 0.0356 6.4136 0.0228 10.0207 17.0738 4.2285 0.0060 98.7421 PA5x2 ER 52.9142 0.2887 9.8585 0.0061 6.1128 0.0510 8.7463 14.9760 5.0787 0.0123 98.0445 PA5x1 ER 52.5080 0.3849 5.4640 0.0124 7.6880 0.0944 11.5366 18.1390 3.4283 0.0060 99.2615 PA6x12 ER 51.5812 0.5177 8.8141 0.0257 5.8899 0.0514 10.4128 16.1474 4.6852 0.0145 98.1399
Pit M9 Fd 50.1396 0.5055 8.6426 0.0400 4.8288 0.0397 11.8841 20.5608 2.2584 0.0118 98.9113 Pit M20 Fd 50.7554 0.4262 7.6067 0.0452 4.0693 0.0402 12.7513 20.9767 2.0893 0.0086 98.7689 Pit M23 Fd 45.3256 1.1680 13.3129 0.1294 4.1291 0.0293 9.6284 18.3168 2.9883 0.0695 95.0972 Pit M24 Fd 52.9777 0.2728 5.6205 0.0405 4.6481 0.0417 13.3488 19.7133 2.2784 0.0186 98.9603 Pit M25 Fd 51.2061 0.5082 8.6218 0.0386 4.8902 0.0398 11.7049 20.5890 2.2109 0.0062 99.8158 JS Kim AMG 50.8395 0.3951 7.0936 0.0217 7.5194 0.0468 10.7914 18.2690 3.2661 0.0092 98.2517 Pit M26 AM G 50.9573 0.3424 5.2170 0.0222 7.6806 0.0418 11.8716 20.1112 0.0042 2.2741 98.5224 Pit M10 AM E 53.3985 0.3110 4.8223 0.0304 6.1969 0.0601 13.0694 20.3726 2.1216 0.0081 ####### Pit M22 AM E 52.1200 0.3218 4.8368 0.0850 6.0273 0.0616 12.9073 20.0830 2.0993 0.0079 98.5501
68
APPENDIX C HAND SAMPLE DESCRIPTIONS
Lithology Mineralogy Sample names Shape and Size Surface Textures
Fd rich Granulite
garnet + CPX + OPX +plag +quartz
DS012-‐1, DS012-‐2, DS012-‐6, DS012-‐7, EA08-‐3, EA08-‐5, EA08-‐8, PIT-‐M9, PIT-‐M20, PIT-‐M21, PIT-‐M23, PIT-‐M24, PIT-‐M25, PA 7x2, PA 78x, PA 7x1
small (3cm) to medium (8cm) and larger (up to
12cm)
coarse grained light blue green in colour with red orange garnets, black brown Opx and green CPX, possible fabric.weathered plagioclase. Some
with well defined fabrics
Metapelite garnet + qtz + plagioclase +
biotite ± CPX CAX-‐M10, DS012-‐3
large(10cm) well rounded xenoliths
coarse grained smooth light in colour exterios with some
kimberlite still attached. Dark red large (1-‐3cm) garnets very visible
Pyroxenite OPX + CPX ± Qt DS012-‐4, EA08 6 5cm rounded
xenotliths very dark OPX and green cpx
possible foliations
Eclogitic Rock garnet + CPX ± Amph ± Qt DS012-‐5, EA08-‐1, PIT-‐M10, Pit M22 PA 5x1,
PA5x2, PA 6x1, PA 6x12
very small (1cm) and small (3cm) well
rounded spheroidal elipse shaped xenolith
mineralogy dominater by red garnet and dark CPX. Well weathered outside surface
Mafic Gneiss Opx + Calcite + Plagioclase EA08-‐9, RT-‐CAL 2XL10 small (6cm) angular
xenoliths strong fabric defined by OPX
minerals.
Amph Eclogite garnet + CPX + Plag + Amph + Qt Pit M26, JS Kim small to large boulder size and sub-‐rounded
nodules
characterised by red garnets, green cpx and Dark amphibole.
Some plagioclase also seen.
Kimberlite phlogopite + calcite + glass ±
garnet ± serpentine AngasKim, EA02-‐2,
EA02-‐4, EA02-‐7
small to large boulder size and sub-‐rounded nodules of kimberlite
rough green/white well weathered exterior, visible brecciated clasts
69
APPENDIX DTHINSECTION DESCRIPTIONS
lithology Thin Section Major
Mineralogy Minor
Mineralogy Textures
Eclogite Rock PA5x1, PA5x2, PA6x2,
PA6x12 Grt, Cpx, Am, Rt, Qt, llm Triple point equi- to sub granular euhedral texture with coarse grained
rutile
Fd rich Granulite PA7x1, PA78x, PA7x2,Pit M9, Pit M20, Pit M23, Pit
M24, Pit M25
Cpx, Opx, Grt, Pla
Rt, llm, Am Diverse stages of exsolution textures, rutile inclusions within garnet and or clinopyroxene
Amphibole Granulite Pit M26, JS Kim Cpx, Opx, Grt,
Pla, Am Rt, llm, Diverse stages of exsolution textures, rutile inclusions within garnet
and or clinopyroxene
Amphibole Eclogite Pit M10, Pit M22 Grt, Cpx, Am Rt, Qt, llm Triple point equi- to sub granular euhedral texture with coarse grained rutile
Gabbroic Rock Pit 5x9 Pl, Cpx, Opx, Am llm, Rt Doleritic microstructure, lath shaped plagioclase, clinopyroxenes showing an intergrowth with amphibole and minor ilmenite