Journal of Petrology Advance Access published online on January 22, 2009
Journal of Petrology, doi:10.1093/petrology/egn078
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Zircon Hf Isotopic Evidence for Mixing of Crustal and Silicic Mantle-derived Magmas in a Zoned Granite Pluton, Eastern Australia
Gemoc National Key Centre, Department of Earth and Planetary Sciences, Macquarie University, Nsw, 2109, Australia
Received April 16, 2008; Revised typescript accepted December 19, 2008
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONNEW ENGLAND BATHOLITHANALYTICAL PROCEDURESWALCHA ROAD ZONED PLUTONDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAAPPENDIX: ROCK TYPE AND...REFERENCES |
|---|
Zircon Hf isotopic data from a zoned pluton of the Moonbi supersuite, New England batholith, eastern Australia, are consistent with magma mixing between two silicic melts, each derived from isotopically distinct sources. Although zircons from three zones within the Walcha Road pluton give a U–Pb crystallization age of 249 ± 3 Ma, zircon populations from each zone have a range in
Hf . Zircons from the mafic hornblende–biotite monzogranite pluton margin and intermediate zones have Hf 
+5 to +11, whereas those from the more felsic centre of the pluton have Hf +7 to +16, representing a total variation of 11 Hf units. The Lu–Hf depleted mantle model ages range from 650 to 250 Ma, with the younger zircons present only in the felsic pluton centre. The variation in Hf indicates the involvement of silicic melts from at least two sources, one a crustal component with a Neoproterozoic model age and the other a primitive mantle-derived component with model ages similar to the U–Pb crystallization age of the pluton. The zircons reflect the isotopic compositions of the different proportions of crustal-derived silicic melt, relative to mantle-derived silicic melt, between melt generation and final pluton construction. The Walcha Road pluton is considered to have formed by incremental assembly of progressively more felsic melt batches resulting from mixing, replenishment and crystal–melt separation, with final pluton construction involving mechanical concentration as zones of crystal mush. The zoned pluton and, more broadly, the Moonbi supersuite provide examples of magma mixing by which the more silicic units have more juvenile isotopic compositions as a result of increasing proportions of residual melt from basalt fractionation, relative to crustal partial melt. KEY WORDS: Australia; granite magma mixing; zircon; zoned pluton; Hf isotopes
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONNEW ENGLAND BATHOLITHANALYTICAL PROCEDURESWALCHA ROAD ZONED PLUTONDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAAPPENDIX: ROCK TYPE AND...REFERENCES |
|---|
The internal structure of zoned granite plutons is important in understanding the intrusive history and petrogenetic processes occurring in magmas that have already separated from their source rocks. The Tuolumne intrusive suite (Bateman & Chappell, 1979
; Coleman et al., 2004) is probably the best-known example, but others include the Palisade Crest intrusive suite (Sawka et al., 1990), the La Posta pluton, Peninsular Ranges Batholith (Clinkenbeard & Walawender, 1989), the Hall Canyon pluton, Panamint Mountains (Mahood et al., 1996) and the Boggy Plain pluton, Lachlan Fold Belt (LFB) (Wyborn et al., 2001). Models proposed to explain zoned pluton development include side-wall crystallization and movement of evolved boundary-layer liquids upwards (Bateman & Chappell, 1979; Ayuso, 1984; Sawka et al., 1990), inward crystallization of a static magma chamber followed by diapiric uprise (Clinkenbeard & Walawender, 1989), cooling-driven in situ fractionation with loss of residual liquid upward (Mahood et al., 1996) and a regime of convective fractionation (Rice, 1981; Sparks et al., 1984) with back-mixing of the fractionated melt into the main magma chamber (Wyborn et al., 2001).
Precise dating of some plutons suggests a more complex history of pluton construction. Features include incremental pluton growth over extended periods (e.g. Coleman et al., 2004; Walker et al., 2007) and depositional fabrics indicating relative movement of crystals and liquid to produce zones of mechanical crystal accumulation and disruption, the development of crystal-mush zones and multiple magmatic fabrics (e.g. Collins et al., 2006; Bachmann et al., 2007; Zak et al., 2007; Vernon & Paterson, 2008). Vernon & Paterson (2008) considered that these structures are common in granites showing evidence of magma mixing and mingling, possibly formed during or after periodic replenishment. Such structures are particularly evident in the Tuolumne intrusive suite, for which assembly times are around 10 Myr (Coleman et al., 2004). Successive increments of magma production and emplacement of the intrusive suite may well explain the changes in geochemical and isotopic melt compositions, as indicated by a general increase in initial 87Sr/86Sr (Sri) and a decrease in initial 143Nd/144Nd (Ndi) inwards (Kistler et al., 1986).
Examples of mixing and mingling within mafic and silicic-layered intrusions have been well documented (Wiebe et al., 1997, 2004; Bachl et al., 2001; Miller & Miller, 2002; Collins et al., 2006) and are generally agreed to be one of several mechanisms that contribute to the generation of granitic plutons and silicic volcanic rocks. However, the role of mixing between two silicic magmas is less well documented. Hildreth (1981), among others, considered magmatic systems ‘fundamentally basaltic’ in the sense that basaltic magmas are the source of the heat and the source of felsic fractionated melts. A number of studies have focused on experimental data and numerical modelling relating to the primary controls of silicic melt generation, and the ultimate source and level of silicic magma generation (e.g. Bachmann et al., 2007, and references therein). These studies have shown that both felsic residual melt fractionated from injected basalt [hereafter referred to as residual melt following Annen et al. (2006)] and felsic partial melt derived from crust adjacent to the injected basalt can be present under certain conditions in the crust.
In this study, we present new chemical and isotopic evidence for silicic magma mixing in the Walcha Road zoned pluton, based on zircon Hf isotope data, and we discuss evidence relevant to the emplacement and crystallization history of the pluton.
Granites described by the terms I-type and S-type, once considered to identify plutons formed from single-component source rocks of igneous or sedimentary derivation (Chappell & White, 2001, and references therein), have now been shown to have more complex sources (Miller et al., 1990; Collins, 1996; Frost et al., 2001; Kemp et al., 2007). Our usage of those terms is restricted to descriptive petrography as reviewed by Chappell & White (2001), implying only that I-type and S-type granites contain either a significant crustal metaigneous or a metasedimentary source component.
|
NEW ENGLAND BATHOLITH |
|---|
|
TOPABSTRACTINTRODUCTIONNEW ENGLAND BATHOLITHANALYTICAL PROCEDURESWALCHA ROAD ZONED PLUTONDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAAPPENDIX: ROCK TYPE AND...REFERENCES |
|---|
The New England batholith of northeastern New South Wales and southeastern Queensland is of Late Carboniferous to Late Triassic age (Shaw & Flood, 1981
; Bryant et al., 1997). The volumetrically dominant phase of igneous activity occurred in the Late Permian–Early Triassic and consisted of a 300 km NNE–SSW-trending belt of metaluminous and weakly peraluminous plutons, some with associated volcanic rocks (O’Neil et al., 1977; Shaw & Flood, 1981). The granites form: (1) the Clarence River supersuite (Shaw & Flood, 1981; Eggins & Hensen, 1987; Bryant et al., 1997), a group of primitive mainly low-potassium granodiorite plutons; (2) the Uralla supersuite, a central zone of granodiorites and monzogranites that are ilmenite-bearing [in the sense of Ishihara (1977)] and are considered to contain components of metaigneous and metasedimentary source rocks (Shaw & Flood, 1981; Bryant et al., 2004); (3) the Moonbi supersuite, of which the Walcha Road zoned pluton is part, a group of magnetite-bearing monzogranite, monzonite, and granodiorite plutons in the northern and southern parts of the batholith. The Moonbi supersuite is associated with widespread ignimbrite rocks, has relatively juvenile Nd and Sri values, and is considered to have formed from a large ion lithophile element-rich (shoshonitic) crustal source region (Shaw & Flood, 1981). |
ANALYTICAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONNEW ENGLAND BATHOLITHANALYTICAL PROCEDURESWALCHA ROAD ZONED PLUTONDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAAPPENDIX: ROCK TYPE AND...REFERENCES |
|---|
Major and trace elements
All analyses were conducted at the combined facilities of the Department of Earth and Planetary Sciences and the GEMOC National Key Centre, Macquarie University. Whole-rock analyses were determined by X-ray fluorescence (XRF) with a Siemens SRS-1 system: major elements using glass fusion discs with a La heavy absorber; trace elements using pressed powder pellets (Norrish & Chappell, 1977
). Ferrous iron was determined by titration. A Camebax SX50 electron microprobe (EM) was used for mineral analyses on polished rock thin sections, employing an accelerating voltage of 15 kV and a beam current of 20 nA. Rare earth and other minor elements on selected rocks were determined from solution by inductively coupled plasma mass spectrometry (ICP-MS). Rare earth and other elements for apatite, titanite and zircon were analysed in situ on single grains, by laser ablation (LA)-ICP-MS. The NIST 610 standard glass was used as the external calibration standard.
Strontium isotopic analyses
Rubidium and Sr abundances in whole-rock pressed powder pellets were determined by XRF, and unspiked 87Sr/86Sr values were determined by thermal ion mass spectrometry at the Centre of Isotope Studies, Commonwealth Scientific and Industrial Research Organisation. Feldspar Rb, Sr and 87Sr/86Sr were determined on spiked samples following standard HF dissolution and ion exchange column procedures. Errors in 87Rb/86Sr are estimated to be 1%, and errors in 87Sr/86Sr are estimated to be 0.015% at 2SD. The decay constant of 
87Rb used was 1·42 x 10–11/year and the normalizing value of 86Sr/88Sr used was 0·1194.
Zircon Lu–Hf analyses
Zircon grains were separated from three zones representing different parts of the pluton (Fig. 1): (1) pluton margin (sample FS828); (2) intermediate zone (sample FS821); (3) pluton centre (samples FS811 and FS813). The grains were set in an epoxy block and surface polished.
|
In situ Hf isotopes of zircon grains were analysed using a Nu Plasma multi-collector LA-ICP-MS system with a Merchantek LUV266 Nd:YAG laser probe, following the method of Griffin et al. (2000, 2002). Those workers developed an elegant correction procedure for the correction of 176Hf based on the interference of 176Yb and 176Lu even in the presence of significant Yb and Lu. The decay constant of
176Lu used was 1·93 x 10–11/year. Other constants were: 176Hf/177Hf for chondrite = 0·282772 and 176Hf/177Hf for depleted mantle (DM) = 0·283250; 176Lu/176Hf for chondrite = 0·0332 and 176Lu/176Hf for DM = 0·0384. |
WALCHA ROAD ZONED PLUTON |
|---|
|
TOPABSTRACTINTRODUCTIONNEW ENGLAND BATHOLITHANALYTICAL PROCEDURESWALCHA ROAD ZONED PLUTONDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAAPPENDIX: ROCK TYPE AND...REFERENCES |
|---|
The Walcha Road pluton is a high-level intrusion that crops out over 420 km2 (Fig. 1) with contact-parallel thermal metamorphic schists along the western and southern pluton margins. At their highest grade, the schists contain the assemblage garnet–cordierite–biotite. The pluton has a foliation best defined in the field by the preferred orientation of margin-parallel elongate K-feldspar megacrysts, but alignment of tabular plagioclase and hornblende is also strong. The strike of the foliation generally parallels the compositional and textural zonation contours and is steeply dipping, except in the pluton centre. Even near the contact margins, where this foliation is most strongly developed, it is not associated with solid-state deformation features, and so is inferred to result from magmatic flow. Locally, concentrations of K-feldspar megacrysts are aligned in swirl-like foliation patterns that are discordant with the main foliation, in a manner suggestive of mechanical aggregation and flow as crystal mush (Zak et al., 2007
; Vernon & Paterson, 2008).
Adjoining the pluton to the south is a lobate stock, the Back Creek quartz monzodiorite, which is 6 km2 in area (Fig. 1). The contact between the pluton and stock is gradational and suggests that the stock was intruded at an early stage of pluton construction.
Microgranitoid enclaves form a minor part of the pluton and are more abundant in the outer margin of the pluton. The margin and, to a lesser degree, the intermediate zone of the pluton contain abundant clinopyroxene + hornblende ± biotite ± plagioclase aggregates (clots) from 3 mm to 10 mm across, some of which are cored by coarse-grained, radiating idiomorphic clinopyroxene aggregates (Fig. 2a). Compositionally similar, although finer grained, radiating clinopyroxene aggregates are also present in the Back Creek stock. Radiating clinopyroxenes are not present in the enclaves.
|
The pluton has been subdivided nominally into the following three zones, corresponding to variations in mineralogy and texture: (1) the marginal zone of mafic hornblende–biotite monzogranite, containing the majority of enclaves and mafic clots; (2) an intermediate zone of porphyritic K-feldspar–hornblende–biotite monzogranite with sub-equal amounts of megacryst and groundmass K-feldspar; (3) a medium-grained biotite monzogranite to granodiorite central zone with <1% K-feldspar megacrysts and with little or no hornblende (Fig. 1b). The locations of the studied samples are given in the Appendix.
Petrography
The most characteristic features of this pluton, and other plutons of the Moonbi supersuite, are large (30–40 mm) pink K-feldspar megacrysts and orange–brown sphenoids of titanite. Tabular crystals of hornblende (2–8 mm), commonly surrounding cores of clinopyroxene, tend to aggregate with large plates of biotite, titanite, apatite and zircon in the more mafic rocks. Plagioclase, up to 8 mm in length, is commonly rimmed by more sodic compositions, the more sodic plagioclase also forming as interstitial grains. K-feldspar megacrysts invariably show Carlsbad twinning and ghost internal outlines marked by inclusion trails. Accessory minerals are magnetite, apatite, titanite, sparse crystals of yellowish zoned allanite (Fig. 2b) and zircon (Fig. 2c).
Zircon morphology
Zircon grains from the three pluton zones [margin (FS828), intermediate zone (FS821) and pluton centre (FS811 and FS813)] tend to be idiomorphic, and are up to 250 µm in length (locations are shown in Fig. 1 and listed in Table 1). In the pluton centre, zircon aspect ratios vary from 1:1 to 5:1. The dominant face forms for all the zircon grains are {100} and {110} prisms, with {101} and {211} pyramids less well developed. Combined EM back-scatter (BSE) and cathodoluminescence (CL) images indicate considerable oscillatory zoning, with evidence of partial resorption in some inner growth stages and the preferential growth of U-rich bands along pyramid faces, rather than prism faces (Fig. 2c). This form is similar to that described for U-rich zircons by Bernisek & Finger (1993).
|
Modal composition
Selected modal analyses of rock samples are presented in Table 1, with a more comprehensive dataset in Electronic Appendix 1 (available for downloading at http://www.petrology.oxfordjournals.org). The strong mineral zonation of the pluton is displayed on the simplified geological map (Fig. 1a) as contours of colour index, and as K-feldspar megacryst abundance contours in Fig. 1b. The Walcha Road pluton and adjoining Back Creek stock are monzogranite (adamellite) varying to granodiorite and quartz monzodiorite (Fig. 3). In Fig. 4, the modal data have been plotted vs bulk-rock SiO2; as SiO2 values increase reasonably systematically from margin to centre the x-axis (SiO2) direction approximates to distance from pluton margin to pluton centre. The combined modal curves of K-feldspar megacrysts and K-feldspar groundmass show an initial rise inward from the margin, and thereafter a flat but diffuse scatter to
20% total K-feldspar; plagioclase remains reasonably constant at 35% throughout; both hornblende and biotite decrease inward from a total of about 25% to 8%. In Fig. 4, a dotted line at about 66% SiO2 marks a compositional gap (silica gap) between the more mafic marginal rocks and the intermediate and central parts of the pluton. No physical outcrop boundary has been identified and, although the silica gap may be due to sampling bias, it marks a change in abundance of K-feldspar and plagioclase across the boundary.
|
|
Geochemistry
Whole rocks—major and trace elements
Representative major and trace element data for the Walcha Road pluton and the Back Creek stock are given in Table 1, with the full dataset listed in Electronic Appendix 1 (http://www.petrology.oxfordjournals.org).
Granites of the pluton and stock range from about 59% to 74% SiO2, but group around maxima at 63% and 70% SiO2. Plots of TiO2, MgO and Cr define reasonably inverse linear trends against SiO2 (Fig. 5); Y and Zr are less well correlated. Plots of P2O5, Sr and Ba have inverse linear trends, although scattered at the low-SiO2 end; and plots of K2O and Rb are relatively flat and scattered throughout. The trend of Ba follows that of Sr, rather than K2O and Rb, suggesting that Ba has acted as a compatible element in early formed plagioclase and biotite.
|
Whole rocks—rare earth and other elements
The rare earth element (REE) abundances and selected trace elements for seven whole-rock samples were determined by ICP-MS solution chemistry (Table 2) and the REE elements plotted as standard chondrite-normalized patterns (Fig. 6). All samples form a relatively tight group characterized by light REE (LREE) abundances
80–100 times chondrite, a slight dip in the middle REE (MREE), overall moderate slopes (La/Lun 13·0) and only minor negative Eu anomalies (Eu/Eu* 0·8). For other trace elements (Table 2) there are only modest changes in abundances with increasing SiO2, the more obvious being a decrease in Sc, V, Cr, Ni, Sr, Nb and Ba, and rather flat trends for Zn, Ga, Rb, Y, Zr, Th, U and Cs. An overall decrease in total REE and trace element chemistry correlates with increasing SiO2 (Table 2; Fig. 6).
|
|
Mineral chemistry
A comprehensive table of mineral compositions is available in Electronic Appendix 2 (http://www.petrology.oxfordjournals.org).
Clinopyroxene occurs only within the marginal zone of the pluton as euhedral grains pseudomorphed in part by actinolite and surrounded by hornblende. The cores of these pyroxenes have an mg-number [100Mg/(Mg + Fetot)] of 77–74 and composition En43–38Fs13Wo44–49. Clinopyroxene also occurs within the mafic clots, some as radiating glomerocrysts (Fig. 2a), with mg-number 87–78 and composition En47–42Fs8–12Wo45–46, that are significantly more Mg-rich than those of the host-rock clinopyroxenes. Texturally identical radiating clinopryoxenes in the Back Creek stock have a similar range of mg-number. Euhedral crystals of magnesiohornblende occur in both the marginal and intermediate zones of the pluton and have an mg-number of 68–63. Biotite composition varies from mg-number 55–34 from pluton margin to centre. The compositions of plagioclase cores vary from An40 to An25, whereas plagioclase rims around larger grains and interstitial plagioclase grains, less than 3 mm in size, average around An20, irrespective of position within the pluton.
Zircon
Trace element and REE data for three zircon populations (pluton margin, pluton intermediate zone and pluton centre) are presented in Table 3, with additional data in Electronic Appendix 3 (http://www.petrology.oxfordjournals.org).
|
The chondrite-normalized REE patterns of zircon (Fig. 7a) are characterized by HREE enrichment, a systematic increase in total REE from the margin to the pluton centre, and significant negative Eu anomalies (Eu/Eu*
0·3). Ce anomalies vary from zero in the pluton centre to slightly positive in the pluton margin and intermediate zone. The erratic behaviour of Ca, Ti and Fe in zircon of the pluton centre, sample FS811 (Table 3), is considered an artefact of metamict alteration, rendering this zircon unacceptable for Ti-in-zircon thermometry measurements. The pattern of REE behaviour may be interpreted in the light of the experimental studies of zircon–melt partition coefficients (Kd) in which Rubatto & Hermann (2007) showed a dramatic increase in Kd (zircon–melt) for the MREE and LREE with decreasing temperature. Similar increases occur for Th and U; that is, the preferential REE enrichment of those elements and Th and U in zircon towards the pluton centre is inferred to reflect the higher Kd values at lower temperatures of crystallization.
|
Titanite
Chondrite-normalized REE patterns for titanite (Electronic Appendix 4; http://www.petrology.oxfordjournals.org) are plotted in Fig. 7b. Titanite compositions fall into two distinct patterns, with those from the pluton margin and intermediate zones having higher LREE relative to heavy REE (HREE; La/Lun
5·0) than titanite from the pluton centre (La/Lun 0·10). All patterns have significant negative Eu anomalies (Eu/Eu* 0·3).
Apatite
Single-crystal ICP-MS trace element analyses of apatite from the pluton margin and intermediate zone are listed in Electronic Appendix 5 (http://www.petrology.oxfordjournals.org). Although apatite is present in the pluton centre, the fine grain size did not allow for successful separation and laser ablation analysis. Chondrite-normalized apatite REE data (Fig. 7c) show that total REE decreases slightly from pluton margin to intermediate zone and that there is a significant negative Eu anomaly (Eu/Eu* 0·3).
Isotope data
Rb–Sr of rocks and minerals
A 17-point isochron of rock and mineral data from the zoned pluton (Table 4) yields a Model 3 age solution (Ludwig, 2001) of 247 ± 16 Ma, an Sri of 0·70470 ± 0·00037 (2SD) and a mean square of weighted deviates (MSWD) = 34. A MSWD of unity indicates that all errors are within experimental error. A Model 3 isochron (McIntyre et al., 1966) indicates that data of differing initial ratios have been combined into a single isochron. The Model 3 age, despite having a high MSWD, still falls within the error of the 238U/206Pb single-zircon (n = 26) age of 249 ± 3 Ma (Jackson et al., 2004). Adopting the U/Pb age of 249 Ma for the pluton, Sri ratios indicate a small but significant difference between the average of four samples from the pluton centre (0·70436) and the average of six samples from the marginal and intermediate zones (0·70492) (Table 4).
|
Lu–Hf zircon
Single-grain zircon isotopic data are presented in Table 5, including
Hf values at 250 Ma, and Lu–Hf TDM (depleted mantle model ages). Zircon grains from the pluton show considerable variation in Hf isotopic composition, initial 176Hf/177Hf (Hfi) varying from 0·282755 to 0·283073 and Hf varying from +5 to +16. Zircon Hfi for the three zircon populations separated from the Walcha Road pluton are plotted in Fig. 8 with error bars of 1SD. Frequency distribution diagrams of model ages are presented in Fig. 9. The oldest TDM model ages are minimum values only, as they reflect the 176Lu/177Hf ratios of the host magma from which the zircon grains crystallized, rather than the source rocks from which the partial melts were derived (Griffin et al., 2002). The procedure for determining 176Hf/177Hf by ICP-MS includes correcting for 176Yb (Griffin et al., 2002), and where Yb is significant the degree of uncertainty in the correction for inter-element overlap increases. Griffin et al. considered that high precision could be achieved only where 176Yb/177Hf in zircon is less than 0·2. Because in only four instances is this value exceeded (Table 5), we judge that the observed trend in 176Hf/177Hf (Fig. 8) has not been affected by a correction bias. The robustness of the correction procedure has recently been tested using zircon standards 61308A and 61307B and using two different ICP-MS collector configurations that allow for independent measurement of Yb mass bias. The results (Pearson et al., 2008) show no systematic bias between the two independent methods and further justify the use of the Yb correction procedure of Griffin et al. (2002).
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONNEW ENGLAND BATHOLITHANALYTICAL PROCEDURESWALCHA ROAD ZONED PLUTONDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAAPPENDIX: ROCK TYPE AND...REFERENCES |
|---|
Silicic magma generation in deep crustal hot zones
A recent series of papers has focused on how crustal lithologies respond when basaltic magma is emplaced at different levels and at different rates into the crust (Petford & Gallagher, 2001
; Annen & Sparks, 2002; Sisson et al., 2005; Bachmann et al., 2007). These papers have used numerical modelling to predict the relative abundance of evolved magmas formed from the crystallization of basaltic magma and the partial melting of pre-existing crust, depending on tectonic setting. Annen et al. (2006), who referred to the site of basalt emplacement and melt generation in the lower crust as a ‘deep crustal hot zone’, have focused on subduction-related magmatism, for which heat and H2O from the crystallization of basalt produces both a fractionated residual silicic melt and a partial melt from the adjacent pre-existing crust. Annen et al. (2006, 2008) argued that the first melts to form following partial crystallization of the basalt sills are H2O-rich, low-viscosity, low-density residual silicic differentiates of mantle-derived melts, followed by partial melts of pre-existing crustal rocks after the solidus of the crust is exceeded. The melt components generated are dependent on the fertility of the crust, the temperature and H2O content of the basaltic magma, the sill emplacement geometry and the depth at which the basalt magmas are emplaced. Of importance to the Annen et al. model is the sill emplacement geometry, which determines the relative proportions of residual melt and crustal partial melt for the different crustal lithologies. Three styles of emplacement geometry were modelled: (1) sill overaccretion with the most recent sill uppermost; (2) sill underaccretion with the most recent sill lowermost; (3) random sill emplacement. For each of those geometries, partial melt generation was modelled for the two common crustal lithologies: low-fertility amphibolite with nominal 50% plagioclase, 50% amphibole (Petford & Gallagher, 2001), and high-fertility muscovite–biotite pelite (Clemens & Vielzeuf, 1987).
We examine the results of melt generation for the amphibolite and pelite lithologies numerically modelled for sill emplacement by overaccretion and by random emplacement. Our preference for a probable crustal source rock of the Moonbi supersuite is a K-rich amphibolite (meta-shoshonite), which would be more fertile than the Petford & Gallagher (2001) amphibolite and a necessary composition if a high-K magma is to be generated (Roberts & Clemens, 1993), as is the case for the Walcha Road pluton.
For overaccretion, the model predicts that the total volume of residual melt is a function of the intrusion rate, whereas the generation of the crustal partial melt is controlled by heat conduction. For the example we have chosen, the modelled conditions were mafic magma injected at a depth of 30 km, an injection temperature of 1285°C, 2·5% H2O content and a sill emplacement rate of 5 mm/year (Annen et al., 2006). Following an incubation period of 0·1 Myr for the formation of the first residual melt, partial melting of amphibolite and, if present, pelite would occur and, as the ratio of partial melt to residual melt generation is normally less than unity, the volume of residual melt dominates. This is shown in Fig. 10a for which the ratio residual melt/partial melt increases with time. The selection of 30 km depth as the initial depth to the Moho is in accord with estimates of depths existing in the Early Triassic along the Gondwanan margin (Mantle & Collins, 2008). The calculated rates of melt generation are within the range for typical magma productivity rates found in arcs (Jicha et al., 2006).
|
For random mafic sill emplacement, the sills were modelled as being emplaced between 29·5 km and the Moho with the Moho displaced downward with successive sill intrusions. Other conditions were an injection temperature of 1346°C, 0·3% H2O content and a sill emplacement rate of 5 mm/year (Annen et al., 2008
; C. Annen, personal communication). Compared with the overaccretion model, the total volume of residual and partial melts is reduced, as the result of less focused heat advection response and also because the sills are scattered among crustal screens where melting occurs both above and below the sills. Model simulations predict that the generation of crustal partial melts would dominate over the residual melt for periods up to 3 Myr (Fig. 10b). Because of the different proportions of residual and partial melt fractions in the random sill emplacement model, the probability of magmas being modified by assimilation and fractional crystallization (AFC) (DePaolo, 1981), melting, assimilation, storage and homogenization (MASH) (Hildreth & Moorbath, 1988) and chemical transfer (Weibe, 1993) would be greatly enhanced as melt fractions rise through and interact with interlayered crustal screens.
The production of two isotopically different silicic melts within a deep crustal hot zone, one crustal and the other a residual silicic melt fractionated from the basaltic magma heat source, suggests that some degree of mixing between the two melts is inevitable. The sill emplacement geometry of the overaccretion model of Annen predicts that the residual mantle-derived silicic melt should dominate over the crustal partial melt with time. As shown in Fig. 10a, the curves of partial melt production, plotted as thickness (y-axis) relative to the total thickness of intruded basalt magma or time (x-axis), clearly show the relationships between melt generation and fertility, with crustal partial melt thicknesses increasing from amphibolite to more fertile pelite. For periods of crustal melting less than 1 Myr (5 km basalt sill emplacement), pelite partial melt production, depending upon composition, increases rapidly and is accompanied by approximately similar thicknesses of residual silicic melt. For extended periods of melting, the rates of melt production change, although the general relationship of fertility to melt thickness is maintained. After 3 Myr, equivalent to a 15 km thickness of emplaced basaltic magma, the thickness of partial melt production from amphibolite is 1200 m, and from pelite 2000 m. These thicknesses compare with a residual silicic melt production of 5200 m. With a K-rich shoshonitic crustal component source as proposed for the Walcha Road pluton and the Moonbi supersuite, the partial melt/residual melt ratio would be higher than the modelled Petford & Gallagher (2001) amphibolite.
Although trench-complex metasedimentary rocks are major components of the New England Fold Belt (Fergusson, 1985), we propose that the level of melt generation for the Walcha Road pluton lies beneath the Palaeozoic subduction complex metasediments (mixed volcaniclastic–aluminous-metasediment), as most granite Sri values are low, 
18O values are low (O’Neil et al., 1977) and compositions are strongly metaluminous except for the most felsic rocks.
The Moonbi supersuite includes other zoned plutons in the northern part of the batholith around Stanthorpe, Queensland, with leucocratic monzogranites forming a volumetrically significant part of the pluton complex. From the low 18O of these granites, O’Neil et al. (1977) and Shaw & Flood (1981) considered them to be of primitive I-type association and this has been confirmed by the high 7Li values of associated granites (Bryant et al., 2004) and recent analyses of zircon 176Hf/177Hf (S. E. Shaw, unpublished data) with primitive Hf values up to + 16. Thus the trend from more mafic, more crustal compositions towards more silicic, more isotopically primitive magmas appears to be a characteristic of the Moonbi supersuite.
Magma mixing within large-scale zoned complexes, such as the Tuolumne intrusive suite, is indicated by co-variation of Sr and of Nd isotopes between intrusive units (Kistler et al., 1986) with a trend towards more felsic magmas with more crustal isotopic signatures. This has been confirmed by recent analyses of extracted zircon from the collected samples of Bateman & Chappell (1979) from the Tuolumne intrusive suite. Zircon 176Hf/177Hf ratios (S. E. Shaw, unpublished data; samples courtesy B. W. Chappell) substantiate the trend of Sri and Ndi with variations in zircon Hf from + 3·0 (quartz diorite of May Lake) to –2·7 (Johnson granite porphyry). Although not the only possible explanation for the Hf variation, the isotopic trend is consistent with random sill emplacement geometry (Fig. 10b) where thicknesses of isotopically evolved crustal partial melt dominate over isotopically primitive residual melt for periods of up to 3 Myr.
The random basalt sill emplacement model may also explain some aspects of the Lachlan Fold Belt granites, where McCulloch & Chappell (1982) and Collins (1996) have shown that some granites have Sri, Nd and 18O values that change with increasing SiO2 towards more crustal values, essentially the same as the Tuolumne intrusive series.
Evidence of magma mixing
Magma mixing models imply that chemical variation develops either by mixing of different proportions of contrasting end-members or by fractional crystallization of a hybrid parental magma (Collins et al., 2006). The Walcha Road pluton is a prime example of magma mixing with zircon Hf preserving a record of that process.
A number of features indicate that fractionation from a single magma was not the dominant mechanism in producing the compositional diversity within the Walcha Road pluton.
- Whole-rock REE patterns and the magnitude of the Eu anomaly across the SiO2 spectrum remain relatively constant. Removal of hornblende would tend to deplete the melt in MREE and HREE, and the removal of plagioclase would increase the magnitude of the Eu anomaly. These effects are not observed.
- Total REE abundances for the granites decrease with increasing SiO2. As all major fractionating phases other than hornblende have most REE partition coefficients less than unity, the fractionated melt should increase rather than decrease in total REE.
- Zircon REE patterns from the pluton margin and, to a lesser degree, from the intermediate zone have positive Ce anomalies. These zircon patterns differ significantly from those of the pluton centre, and are inconsistent with the gradual changes expected of fractionation.
- There are no significant changes in Rb/Sr, Cs/Ca or Ba/Sr ratios between the intermediate zone and pluton centre, suggesting little or no removal of plagioclase as a fractionating phase.
- The trend of K2O, although scattered, does not increase with increasing SiO2. As a result, the modal abundance of K-feldspar, and more importantly the K-feldspar/plagioclase ratio, decreases towards the pluton centre (Table 1) in the direction of granodiorite rather than towards the ternary minimum composition of monzogranite (adamellite).
This leads us to conclude that the observed compositional changes are more readily explained by magma mixing involving a hornblende–K-feldspar-rich magma that is isotopically evolved and a K-feldspar-poor magma that is isotopically primitive.
Calculated zircon saturation temperatures (Watson & Harrison, 1983) for the Walcha Road pluton vary from 765°C to 740°C, assuming that the melt composition from which the zircon crystallized is representative of the granites from which the zircon grains have been separated. Resititic zircons are all but absent from the Walcha Road pluton, indicating that melting temperatures were above zircon saturation. The partial melting of amphibolite according to the modelled conditions of Fig. 10a is around 1100°C (Annen et al., 2006), well above the zircon saturation temperature, so that all restite phases might be expected to be resorbed. Zircon grains form only after the zircon saturation temperature is reached, the isotopic composition reflecting that of the local melt at the time of crystallization. Mixing of melts and the dispersal of zircon that crystallized at different crustal levels and from different batches of melt is most effective at low crystallinity (<40 vol.% crystals; Bachmann et al., 2007), diminishing rapidly at high crystallinity (>60 vol.% crystals) to the point of ‘viscous death’ (Bachmann et al., 2007). Crystallization of zircon will continue in the high-crystallinity regime, but unless new melt is supplied, final crystallization will be from the interstitial melt.
A key feature of the Hf isotopic data (Fig. 8) is that Hf values in the pluton span 11 units. Such variations can only be reconciled with open-system behaviour by which the 176Hf/177Hf of the end-member proportions in the hybrid melt changes, with each ratio in Fig. 8 recording the local 176Hf/177Hf of the hybrid melt at the time of crystallization. Zircon frequency distribution diagrams of TDM model ages for each zone of the pluton (Fig. 9a–c) reflect the preserved range of model ages along a two-end-member mixing curve, the ends of which represent the model ages of the crustal and mantle-derived source rocks. For the crustal end-member, we have selected an age 650 Ma as it corresponds to oldest zircon TDM ages (Table 5) and to the oldest age of the New England Fold Belt, considered by Fergusson (1985) to be Neoproterozoic to early Proterozoic. For the younger end-member age we have selected 250 Ma, the age of the Walcha Road pluton (Jackson et al., 2004). For the Walcha Road data (Table 5), the median values of zircon TDM ages from each of the three zircon populations give model ages with the range 600–450 Ma. These ages, based on simple end-member mixing that assumes similar abundances of Hf in the end-member melts, give estimates of the mantle-derived component in the pluton of between 15 and 50%. As the mass proportions of Hf probably differ between the end-members, the proportions of the mantle-derived component are considered to be minimum values, as discussed in the next section.
Large variations in zircon Hf values have also been reported in other granite provinces. For example, Griffin et al. (2002) reported a variation in zircon Hf of 15 units in a single granite sample from China and Kemp et al. (2007) reported a variation in zircon Hf of 10 units in a single sample of I-type granite from the Lachlan Fold Belt. In both these examples, magma mixing between different sources has been used to explain the range of Hf values.
Supporting evidence from Nd and Sri, Moonbi supersuite
Neodymium and Sr in the New England batholith have relatively restricted ranges of isotopic composition, compared with the more extensive isotopic array of S- and I-type granites from the Lachlan Fold Belt (McCulloch & Chappell, 1982; Fig. 11). The fundamental difference between the two fold belts is that in New England the crust is inferred to have been derived from a Neoproterozoic–early Palaeozoic basaltic arc system (Fergusson, 1985), whereas the Lachlan Fold Belt reflects source-rocks for the granites that are inferred to have model ages up to 1500 Ma (McCulloch & Chappell, 1982).
|
Although the Walcha Road pluton has not been analysed sufficiently well to provide an extensive
Nd–Sri array, several plutons of the Moonbi supersuite do provide a range of isotopic compositions that we believe reflect mixtures of residual melt and partial melts derived from a Neoproterozoic, shoshonitic crustal source (Shaw & Flood, 1981). Our choice of a suitable crustal source component is a shoshonitic arc basalt of sufficiently high Sr/Rb ratio that in the 400 Myr period between crustal emplacement and a partial melting event in the late Palaeozoic Sri ratios evolve to no more than 0·7060. We have selected a shoshonite from Carr (1998; his sample R8068) as being appropriate in terms of Sm–Nd and Rb–Sr. For calculation purposes only, the age of the crustal shoshonite is taken to be 650 Ma.
For the mantle-derived end-member, we have selected the isotopic composition of a tonalite from the Clarence River supersuite (Bryant et al., 1997; their sample R70203
[GenBank]
). This sample is among the most isotopically primitive granites of eastern Australia and is of a suitable age for a mantle-derived end-member component. However, as noted by Bryant et al. (1997), indications are that other isotopic sources may have been involved, and that R70203
[GenBank]
can only be considered an approximation of the mantle-derived end-member.
A compilation of Sm–Nd and Rb–Sr isotopic data for the Moonbi supersuite (Hensel et al., 1985; Honma et al., 1988; Landenberger & Collins, 1996; Landenberger et al., 2000; S. E. Shaw, unpublished data) given in Electronic Appendix 6 (http://www.petrology.oxfordjournals.org) plots as a covariant array with a weak inverse correlation between Nd and Sri along a mixing curve joining the end-member compositions (Fig. 11). Compared with the spread of the Lachlan Fold Belt granites (McCulloch & Chappell, 1982), the Moonbi supersuite granites reflect relatively juvenile source rocks.
Using the highest Sri value (0·7053) for the pluton margin and the lowest Sri value (0·7043) for the pluton centre (Table 4) to constrain the Sri mix proportions along the Nd–Sri array (Fig. 11), the pluton contains between 45 and 80% of the mantle-derived component. However, these figures are at variance with simple end-member mixing calculations using Hf TDM ages from zircon that, from the previous section, indicate a 15–50% mantle-derived component for the pluton. Other than selecting a younger crustal emplacement age for the Nd–Sri array (Fig. 11) that would then increase the calculated crustal component values, particularly those for higher Sri granites of the pluton margin, the discrepancy probably relates to the zircon Hf TDM mixing line calculations (Fig. 9). Simple mixing assumes that the mass fraction of Hf (and by chemical association Zr) is similar in both the residual melt and the crustal partial melt. However, it is known that the abundance of Zr in K-rich shoshonitic rocks is higher than that in low-K basaltic andesites (Gill, 1981) and therefore mixing curves can be constructed using different mass fractions of Hf. If the concentration of Hf in the crustal melt fraction is assumed to be three times that of Hf in the residual primitive melt, the Hf TDM mixing curve for the pluton indicates 30–80% mantle-derived component, as compared with the Nd–Sri estimate of 45–80% mantle-derived component. As mentioned above, the choice of an emplacement age for the crustal end-member is important in the Nd–Sri calculations, and therefore estimates of the proportion of end-member components for the pluton, particularly at the crustal end, are approximate only.
Model of pluton construction
The Walcha Road zoned pluton is best explained as the result of incremental episodes of magma emplacement, the later pulses of magma becoming more silicic yet more isotopically primitive. The process of pluton construction implies open-system behaviour resulting from recharge, possible eruptive evacuation, crystal accumulations, flow and filter pressing, and crystal separation from the interstitial melt. The presence of a magmatic foliation, in most part contact-parallel, although irregular in zones of K-feldspar megacryst concentration, is consistent with final emplacement involving mechanical movement of crystal mushes during pluton construction. The presence of zircon from a single hand-sized sample with a variation of 9 Hf units and a pluton-wide variation of 11 Hf units demonstrate the effectiveness of such a process.
Microstructural evidence for the accumulation and movement of a crystal mush is most apparent in the pluton margin, where the dominant, steeply dipping foliation is disrupted by concentrations of K-feldspar megacrysts as touching frameworks and imbricated patterns. Other local features include scattered white K-feldspar megacrysts among the dominant pink megacrysts, and layers of mafic schlieren aligned in the magmatic foliation. The presence of mafic cumulate clots in the marginal zone and their effective disruption and dispersal within the pluton margin may result from mush-dominated flow. We suggest that magma mixing between the Back Creek magma and a cooler, more viscous magma of the evolving pluton may have been responsible for the growth of more dense crystal cumulate packages along a solidification front (e.g. Marsh, 1988; Sawka et al., 1990; Jerram et al., 2003), with eventual disaggregation and flow as a mix of cumulate clots. The close compositional relationship shown between the radial glomerocrysts of clinopyroxenes in both the Back Creek and the mafic clots implies a genetic association between the two.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONNEW ENGLAND BATHOLITHANALYTICAL PROCEDURESWALCHA ROAD ZONED PLUTONDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAAPPENDIX: ROCK TYPE AND...REFERENCES |
|---|
The general form of the Walcha Road pluton is deceptively simple: a large normally zoned pluton that, from margin to centre, grades from a K-rich hornblende–biotite monzogranite margin to a monzogranite, transitional to a granodiorite, centre. However, closer examination reveals a more complex history and structure. Despite the appearance of regular mineral zonation, the internal structure of the pluton is marked by at least one compositional discontinuity. The general lack of the smooth Harker variation trends normally expected of fractional crystallization also argues against simple fractionation of a single magma body.
The zoned pluton is an example of open-system behaviour and is best explained as resulting from incremental episodes of magma emplacement, each pulse being compositionally more silicic, with the final most felsic pulse being isotopically the most primitive.
An important feature of the zircon Hf data is the spread of Hf values of 11 units. Such variations can be reconciled only with zircon nucleation and growth during magma mixing and/or mixing of crystal mushes that form from a range of hybrid magmas. The variation in Hf indicates the involvement of silicic melts from two sources, one a crustal component with a model Neoproterozoic age and the other a primitive or juvenile mantle-derived component with an age equivalent to the U–Pb crystallization age of the pluton. Our preferred interpretation for the preservation of Hf values in each of the three zircon populations is that the end-member silicic melts with distinct Hf isotopic compositions continued to mix as zircon nucleated and crystallized. The older Hf TDM ages indicate that the lower crust of New England is somewhat older than the oldest rocks exposed at present, but within the range of ages of the underlying lithospheric mantle of around 500–840 Ma (Powell & O’Reilly, 2007).
The magmatic foliation is, for the most part, contact-parallel, although irregular in local zones of K-feldspar cumulate megacryst concentration. Final pluton construction appears to have involved mechanical movement of crystal mushes. The Walcha Road pluton is considered to be a physical mixture of minerals that, at the scale of a hand-sized specimen, do not represent the original composition of the melts from which those minerals crystallized (see Miller & Miller, 2002).
We consider that the melts that formed the Walcha Road pluton were generated in a deep crustal hot zone. Although the melts were silicic, comprising residual melts and isotopically less primitive crustal partial melts, the composition of both became more felsic over time. The silicic nature of those liquids and their isotopic diversity are preserved in the zoned pluton through magma mixing, mingling and perhaps minor fractionation. The probability of these magmas degassing and developing overpressures leading to eruptive evacuation is evidenced by extensive ignimbritic rocks spatially associated with other less deeply eroded plutons of the Moonbi supersuite 200 km to the north. The Annen et al. (2008) model, in which the nature of the sill emplacement geometry in deep crustal hot zones influences the thermal evolution of the crust, provides a mechanism for the isotopic evolution of the derivative melts. The overaccretion sill geometry model predicts that the residual melt component relative to the crustal partial melt component should increase with time, as observed for the Walcha Road pluton and for other plutons of the Moonbi supersuite. The random sill geometry model predicts that with time, the crustal partial melt component would dominate over the residual melt component and that the resulting melts would evolve towards more crustal isotopic signatures, as observed for some of the granites of the Lachlan Fold Belt and for the Tuolumne intrusive suite.
One of the more important conclusions of this study is that the mantle-derived melt component in the felsic centre of the Walcha Road pluton is silicic not mafic. This highlights the problem of always associating mantle-like isotopic values with the direct involvement of mafic magma. Because of the silicic composition of the residual and crustal partial melts, the mixed magma may not produce the major textural and major element variations that result from mixing or mingling of felsic and mafic magmas that are markedly different in temperature and viscosity. Indeed, the mixing of silicic residual melts and crustal partial melts may be difficult to detect within single plutons, except by isotopic studies of zircon and other minerals that exhibit isotopic disequilibrium.
|
SUPPLEMENTARY DATA |
|---|
|
TOPABSTRACTINTRODUCTIONNEW ENGLAND BATHOLITHANALYTICAL PROCEDURESWALCHA ROAD ZONED PLUTONDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAAPPENDIX: ROCK TYPE AND...REFERENCES |
|---|
Supplementary data for this paper are available at Journal of Petrology online.
|
APPENDIX: ROCK TYPE AND LOCATION FOR SAMPLES IN TABLES 1–5
|
|---|
|
TOPABSTRACTINTRODUCTIONNEW ENGLAND BATHOLITHANALYTICAL PROCEDURESWALCHA ROAD ZONED PLUTONDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAAPPENDIX: ROCK TYPE AND...REFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
WGS 84, World Geodetic System 1984.
|
ACKNOWLEDGEMENTS |
|---|
We are grateful to our colleagues at Macquarie University for comment on early drafts of this work. Helpful reviews were made by W. J. Collins, C. F. Miller, R. A. Wiebe and an anonymous referee, and we also acknowledge editorial comments from B. R. Frost and useful suggestions from R. H. Vernon and C. Allen. This research has been supported by Macquarie University Research grants and Australian Research Committee grants. The Centre for Geochemical Evolution and Metallogeny of Continents kindly provided analytical facilities, and the generous assistance of Norm Pearson and Carol Lawson is acknowledged. Judy Davis prepared the figures. This is Contribution 555 from the ARC national Key Centre for Geochemical Evolution and Metallogeny of Continents (GEMOC).
*Corresponding author. Present address: Department of Earth and Planetary Sciences, Macquarie University, NSW, 2109, Australia. Telephone: + 612 9850 8370. Fax: + 612 9850 6904. E-mail: rflood{at}els.mq.edu.au
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONNEW ENGLAND BATHOLITHANALYTICAL PROCEDURESWALCHA ROAD ZONED PLUTONDISCUSSIONCONCLUSIONSSUPPLEMENTARY DATAAPPENDIX: ROCK TYPE AND...REFERENCES |
|---|
Annen C, Sparks R. SJ. Effects of repetitive emplacement of basaltic intrusions on thermal evolution and melt generation in the crust. Earth and Planetary Science Letters (2002) 203:937–955.[CrossRef][Web of Science]
Annen C, Blundy JD, Sparks R. SJ. The genesis of intermediate and silicic magmas in deep crustal hot zones. Journal of Petrology (2006) 47:505–539.
Annen C, Blundy JD, Sparks R. SJ. The sources of granitic melt in deep hot zones. Transactions of the Royal Society of Edinburgh: Earth Sciences (2008) 97:297–309.[Web of Science]
Ayuso RA. (1984) Compositionally zoned plutons by double-diffusive sidewall fractionation processes; evidence from the Palisade Crest Suite, central Sierra Nevada, California. US Geological Survey Professional Paper, Report 1320.
Bachl CA, Miller CA, Miller JS, Faulds JE. Construction of a pluton: Evidence from an exposed cross section of the Searchlight pluton, Eldorado Mountains, Nevada. Geological Society of America Bulletin (2001) 113:1213–1228.
Bachmann O, Miller CF, de Silva SL. The volcanic–plutonic connection as a stage for understanding crustal magmatism. Journal of Volcanology and Geothermal Research (2007) 167:1–23.[CrossRef][Web of Science]
Bateman PC, Chappell BW. Crystallization, fractionation and solidification of the Tuolumne Series, Yosemite National Park. Geological Society of America Bulletin (1979) 90:465–482.
Bernisek A, Finger F. Factors controlling the development of prism faces in granite zircons: a microprobe study. Contributions to Mineralogy and Petrology (1993) 114:441–451.[CrossRef][Web of Science]
Bryant CJ, Arculus RJ, Chappell BW. Clarence River Supersuite: 250 Ma Cordilleran tonalitic I-type intrusions in eastern Australia. Journal of Petrology (1997) 38:975–1001.
Bryant CJ, Chappell BW, Bennett VC, McCulloch MT. Lithium isotopic compositions of the New England Batholith: correlations with inferred source rock compositions. Transactions of the Royal Society of Edinburgh: Earth Sciences (2004) 95:1–16.[CrossRef][Web of Science]
Carr PF. Subduction-related Late Permian shoshonites of the Sydney Basin, Australia. Mineralogy and Petrology (1998) 63:49–71.[CrossRef][Web of Science]
Chappell BW, White A. JR. Two contrasting granite types; 25 years later. Australian Journal of Earth Sciences (2001) 48:489–499.[CrossRef][Web of Science]
Clemens JD, Vielzeuf D. Constraints on melting and magma production in the crust. Earth and Planetary Science Letters (1987) 86:287–306.[CrossRef][Web of Science]
Clinkenbeard JP, Walawender MJ. Mineralogy of the La Posta pluton: Implications for the origin of zoned plutons in the eastern Peninsular Ranges batholith, southern and Baja California. American Mineralogist (1989) 74:1258–1269.[Abstract]
Coleman DS, Gray W, Glazner AF. Rethinking the emplacement and evolution of zoned plutons: Geochronologic evidence for incremental assembly of the Tuolumne intrusive suite, California. Geology (2004) 32:433–436.
Collins WJ. Lachlan Fold Belt granitoids: products of three-component mixing. Transactions of the Royal Society of Edinburgh: Earth Sciences (1996) 87:171–181.[Web of Science]
Collins WJ, Wiebe RA, Healy B, Richards SW. Replenishment, crystal accumulation and floor aggradation in the megacrystic Kameruka suite, Australia. Journal of Petrology (2006) 47:2073–2104.
Corfu F, Hanchar JM, Hoskin P. WO, Kinny P. Atlas of zircon textures. In: Zircon. Mineralogical Society of America, Reviews in Mineralogy and Geochemistry —Hanchar JM, Hoskin P. WO, eds. (2003) 53:469–500.
DePaolo DJ. Trace-element and isotopic effects of combined wallrock assimilation and fractional crystallisation. Earth and Planetary Science Letters (1981) 53:189–202.[CrossRef][Web of Science]
Eggins S, Hensen BJ. Evolution of mantle-derived, augite–hypersthene granodiorites by crystal–liquid fractionation: Barrington Tops Batholith, eastern Australia. Lithos (1987) 20:295–310.[Web of Science]
Fergusson CL. Trench-floor sedimentary sequences in a Palaeozoic subduction complex, eastern Australia. Sedimentary Geology (1985) 42:181–200.[CrossRef][Web of Science]
Frost BR, Barnes CG, Collins WJ, Arculus RJ, Ellis DJ, Frost CD. A geochemical classification for granitic rocks. Journal of Petrology (2001) 42:2033–2048.
Gill JB. Orogenic Andesites and Plate Tectonics (1981) Berlin: Springer.
Griffin WL, Pearson NJ, Belousova E, Jackson SE, O’Reilly SY, van Acherberg E, Shee SR. The Hf isotope composition of cratonic mantle: LAM-MC-ICPMS analysis of zircon megacrysts in kimberlites. Geochimica et Cosmochimica Acta (2000) 64:133–147.[CrossRef][Web of Science]
Griffin WL, Wang X, Jackson SE, Pearson NJ, O’Reilly SY, Xu X, Zhou X. Zircon chemistry and magma mixing, SE China: In situ analysis of Hf isotopes, Tonglu and Pingtan igneous complexes. Lithos (2002) 61:237–269.[CrossRef][Web of Science]
Hensel HD, McCulloch MT, Chappell BW. The New England Batholith: constraints on its derivation from Nd and Sr isotopic studies of granitoids and country rocks. Geochimica et Cosmochimica Acta (1985) 49:369–384.[CrossRef][Web of Science]
Hildreth W. Gradients in silicic magma chambers: Implications for lithospheric magmatism. Journal of Geophysical Research (1981) 86:10153–10192.
Hildreth W, Moorbath S. Crustal contribution to arc magmatism in the Andes of central Chile. Contributions to Mineralogy and Petrology (1988) 98:455–489.[CrossRef][Web of Science]
Honma H, Iizumi S, Morris PA. Chemistry and Sr- and Nd-isotope ratios of mafic inclusions from the Moonbi and Walcha Road Adamellites, New England Batholith—A preliminary report. Preliminary Report on the Geology of the New England Fold Belt, Australia, Department of Geology, Shimane University, Japan (1988) 1:217–231.
Ishihara S. The magnetite-series and ilmenite-series granitic rocks. Journal of Mineralogy and Geology (1977) 27:293–305.
Jackson SE, Pearson NJ, Griffin WL, Belousova EA. The application of laser ablation-inductively coupled plasma-mass spectrometry to in situ U–Pb zircon geochronology. Chemical Geology (2004) 211:47–69.[CrossRef][Web of Science]
Jerram DA, Cheadle MJ, Philpotts AR. Quantifying the building blocks of igneous rocks: are clustered crystal frameworks the foundation? Journal of Petrology (2003) 44:2033–2051.
Jicha BR, Scholl DW, Singer BS, Yogodzinski GM, Kay SM. Revised age of Aleutian island arc formation implies high rate of magma production. Geology (2006) 34:661–664.
Kemp A. IS, Hawkesworth CJ, Foster GL, Paterson BA, Hergt JM, Gray CM, Whitehouse MJ. Magmatic and crustal differentiation history of granitic rocks from Hf–O isotopes in zircon. Science (2007) 315:980–983.
Kistler RW, Chappell BW, Peck DL, Bateman PC. Isotopic variation in the Tuolumne Intrusive Suite, central Sierra Nevada, California. Contributions to Mineralogy and Petrology (1986) 94:205–220.[CrossRef][Web of Science]
Landenberger B, Collins WJ. Derivation of A-type granites from a dehydrated charnockitic lower crust: evidence from the Chaelundi Complex, Eastern Australia. Journal of Petrology (1996) 37:145–170.
Landenberger B, Nakajima T, Collins WJ, Whitford DJ. Mantle input on the character of I-type granitoids of the New England Batholith: Evidence from Sr and Nd isotopes. Centre of Isotope Studies, CSIRO, North Ryde, NSW, Research Report (2000) 1995–2000:15–20.
Ludwig KR. User's Manual for Isoplot/Ex rev. 2.49. Berkeley Geochronology Center Special Publications (2001) 1a.
Mahood GA, Nibler GE, Halliday AN. Zoning patterns and petrologic processes in peraluminous magma chambers: Hall Canyon pluton, Panamint Mountains, California. Geological Society of America Bulletin (1996) 108:437–453.
Mantle GW, Collins WJ. Quantifying crustal thickness variation in evolving orogens: Correlation between arc basalt composition and Moho depth. Geology (2008) 36:87–90.
Marsh BD. Crystal capture, sorting and retention in convecting magma. Geological Society of America Bulletin (1988) 100:1720–1737.
McCulloch MT, Chappell BW. Nd isotopic characteristics of S- and I- type granites. Earth and Planetary Science Letters (1982) 58:51–64.[CrossRef][Web of Science]
McIntyre GA, Brooks C, Compston W, Turek A. The statistical assessment of Rb–Sr isochrons. Journal of Geophysical Research (1966) 71:5459–5468.
Miller CF, Miller JS. Contrasting stratified plutons exposed in tilt blocks, Eldorado Mountains, Colorado River Rift, NV, USA. Lithos (2002) 61:209–224.[CrossRef][Web of Science]
Miller CF, Wooden JF, Bennett VC, Wright JE, Solomon GC, Hurst RW. Petrogenesis of the composite peraluminous–metaluminous Old Woman–Piute Range batholith southeastern California: isotopic constraints. In: The Nature and Origin of Cordilleran Magmatism. Geological Society of America, Memoirs —Anderson JL, ed. (1990) 174:99–109.
Norrish K, Chappell BW. X-ray fluorescence spectrography. In: Physical Methods of Determinative Mineralogy—Zussman J, ed. (1977) London: Academic Press. 201–272.
O’Neil JR, Shaw SE, Flood RH. Oxygen and hydrogen isotope compositions as indicators of granite genesis in the New England Batholith, Australia. Contributions to Mineralogy and Petrology (1977) 62:313–328.[CrossRef][Web of Science]
Pearson NJ, Griffin WL, O’Reilly SY. Mass fractionation correction in laser ablation multiple-collector ICP-MS: precise and accurate in situ isotope ratio measurement. In: Laser Ablation-ICP-MS in the Earth Sciences. Mineralogical Association of Canada Short Course Series —Sylvester P, ed. (2008) 40:93–116.
Petford N, Gallagher K. Partial melting of mafic (amphibolitic) lower crust by periodic influx of basaltic magma. Earth and Planetary Science Letters (2001) 193:483–499.[CrossRef][Web of Science]
Powell W, O’Reilly S. Metasomatism and sulphide mobility in lithospheric mantle beneath eastern Australia: Implications for mantle Re–Os chronology. Lithos (2007) 94:132–147.[CrossRef][Web of Science]
Rice A. Convective fractionation: a mechanism to provide cryptic zoning (macrosegregation), layering crescumulates, banded tuffs and explosive volcanism in igneous processes. Journal of Geophysical Research (1981) 86:405–417.
Roberts MP, Clemens JD. Origin of high-potassium, calc-alkaline, I-type granitoids, Geology (1993) 21:825–828.
Rubatto D, Hermann J. Experimental zircon/melt and zircon/garnet trace element partitioning and implications for the geochronology of crustal rocks. Chemical Geology (2007) 241:38–61.[CrossRef][Web of Science]
Sawka WN, Chappell BW, Kistler RW. Granitoid compositional zoning by sidewall–wall boundary layer differentiation: Evidence from Palisade Crest intrusive suite, Central Sierra Nevada, California. Journal of Petrology (1990) 31:519–553.
Shaw SE, Flood RH. The New England Batholith, eastern Australia: geochemical variations in time and space. Journal of Geophysical Research (1981) 86:10530–10544.
Sisson TW, Ratajcski K, Hankins WB, Glazner AF. Voluminous granitic magmas from common basaltic sources. Contributions to Mineralogy and Petrology (2005) 148:635–661.[CrossRef][Web of Science]
Sparks R. SJ, Huppert HE, Turner F. RS. The fluid dynamics of evolving magma chambers. Philosophical Transactions of the Royal Society of London, Series A (1984) 310:511–534.[CrossRef]
Streckeisen AL. To each rock its proper name. Earth-Science Reviews (1976) 12:1–33.
Vernon RH, Paterson SR. Mesoscopic structures resulting from crystal accumulation and melt movement in granites. Transactions of the Royal Society of Edinburgh: Earth Sciences (2008) 97:369–381.[Web of Science]
Watson EB, Harrison TM. Zircon saturation revisited: temperature and composition effects in a variety of crustal magma types. Earth and Planetary Sciences Letters (1983) 64:295–304.[CrossRef]
Walker BJ, Miller CF, Lowery LE, Wooden JL, Miller JS. Geology and geochronology of the Spirit Mountain batholith, southern Nevada: implications for timescales and physical processes of batholith construction. Journal of Volcanology and Geothermal Research (2007) 167:239–262.[CrossRef][Web of Science]
Wiebe RA. The Pleasant Bay layered gabbro–diorite, coastal Maine; ponding and crystallization of basaltic injections into a silicic magma chamber. Journal of Petrology (1993) 34:461–489.
Wiebe RA, Smith D, Sturm M, King EM, Seckler MS. Enclaves in the Cadillac Mountain Granite (coastal Maine); samples of hybrid magma from the base of the chamber. Journal of Petrology (1997) 38:393–423.
Wiebe RA, Manon MR, Hawkins DP, McDonough WF. Late stage mafic injection and thermal rejuvenation of the Vinalhaven granite, coastal Maine. Journal of Petrology (2004) 45:2133–2153.
Wyborn D, Chappell BW, James M. Examples of convective fractionation in high temperature granites from the Lachlan Fold Belt. Australian Journal of Earth Sciences (2001) 48:531–541.[CrossRef][Web of Science]
Zak J, Paterson SR, Memeti V. Four magmatic fabrics in the Tuolumne batholith, central Sierra Nevada, Califormia (USA): Implications for interpreting fabric patterns in plutons and evolution of magma chambers in the upper crust. Geological Society of America Bulletin (2007) 119:184–201.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||