Journal of Petrology Volume 41 Number 12 Pages 1759-1776 2000
© Oxford University Press 2000
Origin of the Charnockites of the Louis Lake Batholith, Wind River Range, Wyoming
DEPARTMENT OF GEOLOGY AND GEOPHYSICS, UNIVERSITY OF WYOMING, LARAMIE, WY 82071, USA
Received August 3, 1999; Revised typescript accepted May 3, 2000
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONREGIONAL GEOLOGYPETROLOGYWHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYINTENSIVE PARAMETERSFLUID AND SOLID INCLUSIONSDISCUSSIONCONCLUSIONS: THE CONTROLS ON...REFERENCES |
|---|
The 2·63 Ga Louis Lake batholith, a calc-alkalic pluton exposed in Wind River Range of western Wyoming, consists of minor diorite, quartz diorite, granodiorite, and granite. At shallow structural levels the batholith is pyroxene free, but at deeper levels, all units of the batholith contain pyroxenes. On its northern margin the batholith was emplaced at P = 5–6 kbar, T = 775–800°C, fO2 at FMQ (fayalite–magnetite–quartz) + 1·5 to FMQ + 1·8, and aH2O
0·1. Along the southern margin of the batholith the emplacement pressure was 3 kbar. The batholith includes two compositional series. The peraluminous high-REE series is rich in K2O, and displays larger FeOt/(FeOt + MgO). The metaluminous low-REE series has less K2O and smaller FeOt/(FeOt + MgO). In the Boulder Canyon area, most of the rocks belong to the low-REE series. The Nd, Pb, and Sr initial isotopic compositions of both series vary and do not correlate with SiO2 abundance. Relatively constant Fe/Mg ratios over a wide range of SiO2 and the large variation in initial isotopic compositions indicate that the sources for the Louis Lake batholith include older Wyoming province crust- and mantle-derived mafic melts. The calc-alkalic nature of the Boulder Canyon charnockites indicates that there is no special C-type magma. Instead, charnockites will form in any rock type, if the water activity is low enough. KEY WORDS: Archean geology; charnockite; granite; Wind River Range; Wyoming province
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGYPETROLOGYWHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYINTENSIVE PARAMETERSFLUID AND SOLID INCLUSIONSDISCUSSIONCONCLUSIONS: THE CONTROLS ON...REFERENCES |
|---|
Charnockite was originally defined as an orthopyroxene (Opx)-bearing granite (Holland, 1900
). Rock names for Opx- and fayalite (Fay)-bearing felsic igneous rocks include opdalite (Opx granodiorite), enderbite (Opx tonalite), and mangerite (Opx monzonite, Le Maitre et al., 1989). Many of these rock names are not commonly used; the term charnockite (sensu lato) is typically applied to any Opx-bearing granitoid, regardless of its bulk composition. For example, the charnockite suite of south India includes enderbites, opdalites, and charnockites (Howie, 1955), the Nigerian suite includes charnockite and opdalite (Olarewaju, 1987), and that of Lofoten is associated with mangerites (Malm & Ormaasen, 1978). Because the terms enderbite and opdalite are so rarely used, in this paper we use the term ‘charnockite’ for all Opx-bearing granitic rocks. Where it is important to describe a specific rock type, we use Opx as a varietal modifer to the rock name (e.g. Opx granodiorite instead of opdalite).
In the 1980s workers in southern India recognized areas where biotite-bearing granitic gneiss had been converted to pyroxene-bearing gneiss along fluid pathways (Janardhan et al., 1982; Friend, 1985). This led to the term ‘charnockitization’ (Srikantappa et al., 1985) and the concept that charnockite is a metamorphic, rather than an igneous rock. As a result, ‘charnockite’ has been used incorrectly by some workers as a synonym for granulite. Kilpatrick & Ellis (1992) postulated that igneous charnockites form from a distinct, ‘C-type’ magma rich in TiO2, P2O5, and K2O and display an iron-enrichment trend intermediate between that of A-type granites and typical calc-alkalic granites.
In this paper, we describe charnockites that are undisputedly igneous in origin, but that do not share the chemical characteristics of C-type magma as defined by Kilpatrick & Ellis (1992). These rocks are part of a calc-alkalic, Late Archean batholith that shows a transition from pyroxene-bearing granitic rocks at deep structural levels to pyroxene-free ones at shallower levels.
|
REGIONAL GEOLOGY |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGYPETROLOGYWHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYINTENSIVE PARAMETERSFLUID AND SOLID INCLUSIONSDISCUSSIONCONCLUSIONS: THE CONTROLS ON...REFERENCES |
|---|
The Wind River Range is a northwest-trending uplift in western Wyoming that exposes >10 000 km2 of Precambrian rocks. The range consists almost entirely of high-grade Archean gneisses and granites that were thrust to the west over Paleozoic and Mesozoic sedimentary rocks during the Laramide orogeny, 40–80 my ago. The range exposes a cross-section through 10 km of the upper crust (Frost et al., 2000
). The Archean rocks are dominated by Late Archean granitic rocks, including the 2·8 Ga Native Lake gneiss, the 2·67 Ga Bridger batholith, the 2·63 Ga Louis Lake batholith, and small granitic plutons, including the Bears Ears pluton, that were emplaced at 2·54 Ga (Frost et al., 1998). The Louis Lake batholith, which is the largest igneous body in the Wind River Range, crops out over an area of >2000 km2 in the southern portion of the range (Fig. 1). The batholith contains two major units: a medium-grained granodiorite to diorite and a coarse-grained granite. In the northern portion of the batholith, the granodiorite yielded a U–Pb zircon age of 2629·2 ± 2·8 Ma and the granite gives an age of 2629·5 ± 1·5 Ma (Frost et al., 1998), both of which are within uncertainty of the 2642 ± 13 Ma U–Pb zircon age determined for granodiorite from the southern portion of the Louis Lake batholith (Naylor et al., 1970). Over most of its outcrop area the Louis Lake batholith consists of typical biotite- and hornblende-bearing granitic rocks, but along its northeastern margin it contains pyroxene (Fig. 2).
|
|
In mapping the Wind River Range, the US Geological Survey geologists recognized two post-tectonic igneous units: a medium-grained, equigranular granodiorite, and a coarse-grained granite. The former they mapped as the Louis Lake batholith and the latter as Bears Ears pluton (Pearson et al., 1971; Worl et al., 1986). In the Boulder Canyon area, we found gradational contacts between the granite and granodiorite. This confirms our U–Pb zircon age results, which indicate that in this area the two rock types are contemporaneous. We conclude that the Louis Lake batholith consists of both granite and granodiorite, at least in the western portion of the Wind River Range, which explains the difference between our map of the Louis Lake batholith (Fig. 1) and that of Worl et al. (1986).
Along its northwestern margin, in the structurally deepest portion of the Wind River uplift, both the granodioritic and granitic portions of the Louis Lake batholith contain pyroxenes (Fig. 2). Both phases grade from pyroxene bearing to pyroxene absent from the west to the east, that is, from deeper to shallower levels. The transition takes place over a distance of several kilometers: pyroxene-free areas appear first as bands and patches that become progressively larger as one moves to the east. Farther to the east the rusty-weathering pyroxene-bearing areas become smaller and then disappear.
|
PETROLOGY |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGYPETROLOGYWHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYINTENSIVE PARAMETERSFLUID AND SOLID INCLUSIONSDISCUSSIONCONCLUSIONS: THE CONTROLS ON...REFERENCES |
|---|
Although the rocks of the Louis Lake batholith range in silica content from 52 wt % to >75 wt %, all have similar textures. They consist of frameworks of euhedral to subhedral feldspars (plagioclase predominant in the mafic rocks and K-feldspar in the felsic ones) that contain interstitial orthopyroxene (Opx), clinopyroxene (Cpx), biotite, hornblende, and quartz. In outcrop, Opx-bearing rocks weather brown but are greenish on fresh surfaces, whereas Opx-free rocks are white to pink on both weathered and fresh surfaces.
In thin section the pyroxene grains are euhedral to anhedral. Most of the latter are surrounded by intergrowths of biotite + quartz or hornblende + quartz, which probably formed by hydration reactions during late magmatic or submagmatic conditions. Hornblende and biotite grains show two distinct textures: clearly magmatic ones are euhedral, whereas those that formed by late-magmatic and subsolidus reactions are fine grained and intergrown with quartz (Fig. 3). In the pyroxene-bearing rocks the K-feldspar is orthoclase, whereas in pyroxene-absent rocks it is either microcline or a mixture of orthoclase and microcline. The accessory minerals include titanomagnetite and hemoilmenite, zircon, apatite, and, in a few of the pyroxene-free rocks, titanite.
|
In areas where both pyroxene-bearing and pyroxene-absent granitic rocks occur, field relations show a complex history of intrusion, hydration, and dehydration. These relations are particularly well exposed in the glaciated slabs along the north side of the Boulder Creek Canyon. There it is common to find Opx-bearing granite adjacent to contacts with the main body of the Opx granodiorite or in halos around inclusions of Opx granodiorite (location 1 in Fig. 4). These relations indicate that at this level of exposure the early phases of the granite that crystallized against the wall rock or inclusions of granodiorite were Opx bearing. As crystallization proceeded, however, the magma became more hydrous, stabilizing biotite at the expense of Opx. Numerous granitic and pegmatitic dikes cut the rocks in this area. Some are Opx free; where these cut Opx-bearing rocks they are surrounded by hydration halos (location 2 in Fig. 4). Where Opx-bearing dikes cut the pyroxene-free rocks they are surrounded by dehydration zones (location 3 in Fig. 4). Similar dehydration zones are found around charnockite dikes in the country rock (Koesterer et al., 1987).
|
|
WHOLE-ROCK CHEMISTRY |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGYPETROLOGYWHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYINTENSIVE PARAMETERSFLUID AND SOLID INCLUSIONSDISCUSSIONCONCLUSIONS: THE CONTROLS ON...REFERENCES |
|---|
Major elements
Chemical analyses were obtained for 45 samples of the Louis Lake batholith from the Boulder Canyon area (Table 1). The locations of these, as well as samples analyzed by Stuckless (1989)
and Frost et al. (1998), are shown in Fig. 5. The samples analyzed included 21 Opx + Cpx-bearing granitic rocks, 13 pyroxene-absent granitic rocks, five granitic rocks that contained Cpx as the only pyroxene, and six that contained only Opx.
|
|
) and those reported by Stuckless (1989)
and gray squares).
The granitic rocks of Boulder Canyon do not form a single, well-defined trend on a Harker diagram (Fig. 6). This dispersion is best seen in K2O, for which compositions range from medium to ultra-high K, but it is also present in Na2O, Al2O3, and, to a lesser extent, TiO2 and MgO. There is no simple relation between major element composition and the presence or absence of pyroxene (Fig. 6). Although the low-SiO2 rocks are most likely to contain pyroxene, both pyroxene-bearing and pyroxene-free granitic rocks exhibit large ranges of SiO2 contents, from 52 wt % to >75 wt %. Most of the granitic rocks with the highest K2O are pyroxene free, but several of the high-K granites are pyroxene bearing. Conversely, although most of the medium-K rocks are pyroxene bearing, some are pyroxene free. The rocks of Boulder Canyon display relatively constant FeOt/(FeOt + MgO) ratios of 0·7–0·8 (Fig. 7a). They are dominantly metaluminous, although some of the most siliceous rocks are weakly peraluminous (Fig. 7b). There seems to be no chemical difference between rocks with only a single pyroxene and those with two pyroxenes. However, the rocks that contain Cpx as the only pyroxene tend to have lower SiO2 and lower A/CNK ratio than those that contain only Opx.
|
|
Analyses by Stuckless (1989)
indicate that the Louis Lake batholith contains a high-REE and a low-REE series (Fig. 8). The high-REE series is peraluminous at SiO2 > 68 wt % and has FeOt/(FeOt + MgO) ratios >0·75 (Fig. 7b). These rocks are most common in the southern Louis Lake batholith, particularly in the Prospect Mountain area (PRM samples, Fig. 5). The rocks from Boulder Canyon fall mostly into the low-REE series, although one K-rich sample has Fe enrichment typical of the high-REE series (Fig. 7a).
|
Rare earth elements
REE analyses were obtained from six samples of rocks that range from 55 wt % to >70 wt % SiO2 (Table 2). These samples have uniformly low total REE contents, although the most siliceous one may fit into either the low- or high-REE series (Fig. 8). The rocks show a slightly wider range of REE abundance than do the rocks of the low-REE series of Stuckless (1989) (Fig. 9), probably because the samples with the lowest REEs are diorites, a composition not represented in the suites of granodiorites and granites analyzed by Stuckless (1989). The rocks are depleted in light REEs. A negative Eu anomaly increases with wt % SiO2. The single sample from the high-REE series is strongly depleted in light REE. It is similar in composition to some of the more siliceous members of the high-REE granitic rocks of Stuckless (1989).
|
|
Radiogenic isotopes
Along with geochemical heterogeneity, the Louis Lake batholith exhibits Pb, Sr, and Nd isotopic heterogeneity (Frost et al., 1998). Initial 
Nd and 87Sr/86Sr range from values typical of contemporary depleted mantle to ones that resemble those of pre-existing crustal rocks in the Wind River Range (Frost et al., 1998). There is no correlation between initial Pb or Sr isotopic composition and initial Nd, nor between these values and any other compositional feature of the batholith. These results led Frost et al. (1998) to conclude that the batholith was derived from multiple sources and was poorly homogenized.
|
MINERAL CHEMISTRY |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGYPETROLOGYWHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYINTENSIVE PARAMETERSFLUID AND SOLID INCLUSIONSDISCUSSIONCONCLUSIONS: THE CONTROLS ON...REFERENCES |
|---|
Mineral analyses were obtained from Cameca Camebax and JEOL 8900 Superprobe electron microprobes using a 15 kV accelerating voltage and sample current of 20 nA for all minerals except feldspars, which were analyzed with a sample current of 10 nA. Both natural and synthetic silicates and oxides were used as standards. Analyses were corrected using the ZAF procedure. Errors are less than 1% of the measured values of major elements, and increase greatly for elements with abundances below 1 wt % oxide.
Pyroxenes
The Opx and Cpx from the Louis Lake batholith are uniform in composition (Table 3). Augite ranges in composition only from Wo43En38Fs19 to Wo44En39Fs17 whereas orthopyroxene ranges from Wo01En56Fs42 to Wo02En52Fs45 (Fig. 10). XFe changes by only 0·04 in rocks varying from 56 wt % to 63 wt % SiO2. The most Fe-rich Opx is from the least siliceous rock whereas the most magnesian comes from the most siliceous rock.
|
|
Hornblende
As with the pyroxenes, hornblende shows a narrow range of Fe/(Fe + Mg) ratios (0·45–0·50) (Table 4). It has 1–2 wt % TiO2 and moderate amounts of Al2O3 (1·6–1·8 atoms per formula unit). The amphiboles fall into the edenitic hornblende field of Leake (1978) and have Al contents typical of hornblende from other mid-crustal-level granitic plutons (Anderson & Smith, 1995). Contents of F and Cl are low, generally <0·2 wt %, with Cl > F.
|
Biotite
Biotite displays 2–4 wt % TiO2 with XFe ranging from 0·38 to 0·5 (Table 4). The most Mg-rich biotite comes from the most SiO2-rich rock. Halogens are low to moderate; F ranges from 0·15 to 0·55 wt % and Cl from <0·1 to >0·2 wt %. In rocks where there are two textural types of biotite both types of biotite display similar TiO2 contents (Hulsebosch, 1993), suggesting that the secondary biotite formed at magmatic temperatures.
Feldspars
Plagioclase from the dioritic rocks is normally zoned. Cores range to An60, whereas rims are around An35 (Table 5). This diffuse zoning is less pronounced in the more siliceous rocks: compositions are in the range An30–40 in the granodiorites and An25–30 in the granites. K-feldspars show a wide range of compositions. Orthoclase has a composition of Or81–89 whereas coexisting microcline is Or90–95. The feldspar compositions clearly reflect sub-solidus re-equilibration, because the K-feldspars are too poor in CaO and the plagioclase too poor in K2O to be igneous compositions (Fig. 11).
|
|
|
INTENSIVE PARAMETERS |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGYPETROLOGYWHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYINTENSIVE PARAMETERSFLUID AND SOLID INCLUSIONSDISCUSSIONCONCLUSIONS: THE CONTROLS ON...REFERENCES |
|---|
Pressure
The pressure of the northern contact of the Louis Lake batholith is based on an inclusion of pelitic hornfels (FL85-2A) from the northwestern margin of the batholith. This sample contains the assemblage quartz–plagioclase–garnet–orthopyroxene–cordierite. The multiple equilibria in this assemblage intersect at pressures of 5–7 kbar and temperatures around 800°C (Frost et al., 2000
). Pressure along the southwestern margin of the Louis Lake batholith was 3 kbar or less, as estimated from the coexistence of andalusite and K-feldspar in the contact aureole (Frost et al., 2000). These data suggest that there is about a 10 km difference in crustal depth between southwestern and northwestern contacts of the batholith (Frost et al., 2000). Based upon the down-plunge projection of the Wind River Range (Mitra & Frost, 1981), a similar pressure gradient exists between the western and eastern margins of the batholith, indicating that the deepest exposures in the Wind River Range are in the west–central part of the uplift.
Temperature
Pyroxene thermometry yields temperatures from 775 to 800°C, as calculated by the QUILF program of Andersen et al. (1993). The uncertainties of 1–15°C (Fig. 10) refer to the precision of the mathematical fit to the Cpx–Opx solvus and Fe–Mg exchange thermometers. The accuracy of this estimate is unknown because one cannot estimate the accuracy of the empirical projection from multicomponent pyroxene composition space into the system CaO–FeO–MgO–SiO2, but it is probably on the order of ±25°C. The most siliceous rock (88HL15) equilibrated at the highest temperature (804°C), whereas the most mafic rock (88HM1) equilibrated at the lowest temperature (775°C). It is likely that this range is less than the uncertainty inherent in the pyroxene thermometry.
This temperature estimate of 775–800°C is consistent with that of 800 ± 50°C obtained from the mafic hornfels (Frost et al., 2000) and with feldspar thermometry from sample 88SC1. Sample 88SC1 contains no primary K-feldspar; the only K-feldspar in these rocks has formed by exsolution from plagioclase. The temperature at which the plagioclase in 88SC1 would have been saturated in K-feldspar would be 780°C (Elkins & Grove, 1990), indicating that these rocks crystallized at temperatures above 780°C.
These temperatures probably represent the conditions at which the melt solidified, i.e. when fluids were no longer available to flux cation diffusion. Many charnockitic plutons show evidence of extremely high temperature crystallization (900°C or higher) (Kilpatrick & Ellis, 1992). The Boulder Canyon charnockites may have also experienced such high temperatures, but because the rocks are so magnesian, pigeonite, the presence of which is necessary to record the extreme temperatures, never formed.
Oxygen fugacity
Oxygen fugacity is calculated from the assemblage orthopyroxene–magnetite–ilmenite–quartz (Lindsley & Frost, 1992) using the QUILF program (Andersen et al., 1993). The rocks crystallized at oxygen fugacities that were 1·5–1·8 log units above FMQ (fayalite–magnetite–quartz) (Fig. 10), which is similar to the range typical of calc-alkalic magmas (Frost & Lindsley, 1992). The calculated ilmenite in this assemblage would have 10–18 mol % of the Fe2O3 component, which is consistent with the widespread occurrence of hemoilmenite in these samples.
|
FLUID AND SOLID INCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGYPETROLOGYWHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYINTENSIVE PARAMETERSFLUID AND SOLID INCLUSIONSDISCUSSIONCONCLUSIONS: THE CONTROLS ON...REFERENCES |
|---|
Fluid inclusions
Samples from the Boulder Canyon area contain quartz with abundant solid (Fig. 12a and b) and fluid inclusions (Fig. 12c–e). Both types of fluid inclusions commonly contain daughter minerals; the CO2-bearing ones have calcite (Fig. 12c), the aqueous ones, halite. The CO2-bearing inclusions inferred to be primary occur as rare isolated individuals, and as isolated groups. Those inferred to be secondary occurs as planar arrays. Some of these arrays contain inclusions that have negative crystal to rounded forms (Fig. 12d), but others contain irregularly shaped inclusions that show variable degrees of necking and decrepitation. The aqueous inclusions occur as rare isolated groups and as trails that cut planes of CO2 inclusions (Fig. 12e).
|
CO2 inclusions
These fluid inclusions have homogenization temperatures that range from -5° to
30°C. The lower homogenization temperatures (-5 to 5°C) are those of the primary-appearing inclusions, whereas higher temperatures are displayed by necked or decrepitated inclusions (Fig. 13a). The homogenization temperature of primary fluids indicates densities from 0·9 to 0·95 g/cm3. At 800°C, the approximate crystallization temperature of the rocks of Boulder Canyon, fluids of this density would have been trapped at pressures of 4·7–5·7 kbar. This is somewhat lower than the pressure estimates for FL85-2A, the pelitic hornfels inclusion. Because the fluid inclusion sample (SC85-1) lies 10 km to the east of FL85-2A (Fig. 2), and hence occupies a higher structural position, the densest CO2 fluid inclusions are probably of magmatic origin. Texturally later CO2 inclusions show lower densities, which indicates that the rocks probably underwent decompression during cooling (see bold arrow in Fig. 13b); inclusions trapped during isobaric cooling would have been denser.
|
H2O inclusions
Aqueous inclusions have a cubic daughter mineral inferred to be halite, with a relative volume indicating that the primary fluid had a salinity of
30% NaCl. These fluid inclusions homogenize at temperatures of 110–190°C. Pure H2O with this homogenization temperature ranges in density from 0·95 to 0·87 g/cm3 (Fisher, 1976). Isochores with these densities lie at temperatures that are too low for entrapment at igneous temperatures (dashed lines in Fig. 13b). Saline fluids would have isochores with a flatter slope (Crawford, 1981) but the exact slope of such isochores is unconstrained. Whether the isochores of the saline fluids in the inclusions are sufficiently shallow to allow them to intersect the P–T conditions of crystallization is not known, but the lack of low-temperature alteration in the rocks with aqueous fluid inclusions suggests that they were trapped at or near magmatic temperatures. Probably these are fluids that were present during the crystallization of the latest water-rich melts.
Water activity
The assemblage orthopyroxene–biotite–K-feldspar–quartz monitors water activity through the equilibrium
 |
, an ideal two-site model for orthopyroxene, the Indares & Martignole (1985) model for biotite, and the Elkins & Grove (1990) model for K-feldspar, we calculate a water activity of 0·11–0·13 for reaction (1) at 5000 bars and 775–800°C. In a H2O–CO2 mixture this corresponds to XH2O 0·1. This much water cannot be seen in a fluid inclusion; Roedder (1984) noted that less than 20% water in a mixed-vapor fluid inclusion is invisible. These calculations indicate that even in rocks where Opx has completely hydrated to biotite, water abundance in the fluid phase may be so low that it may not be visible in the fluid inclusion. This explains why CO2-rich fluid inclusions are found in both the pyroxene-bearing and pyroxene-free rocks.
Solid inclusions
In addition to fluid inclusions, carbonate inclusions are present in both feldspar and quartz grains (Fig. 12a and b). The inclusions appear as greenish spheroids that range from 2 to 20 µm and occur as isolated inclusions, small isolated groups of inclusions, and as planar arrays that cross-cut grain boundaries (Fig. 12d and e). Microprobe analyses and Raman spectra indicate that these inclusions are calcite with small (0·8 wt %) amounts of FeCO3, and no recognizable NaCO3. Inclusions are found in all Opx-bearing samples, and are rare to absent in Opx-absent rocks.
Origin of the solid inclusions
Textural relations strongly suggest that the solid inclusions are primary and that the original magma consisted of a silicate melt, a carbonate melt, and a CO2-rich vapor. Supporting evidence includes: (1) the isolated inclusions are similar in appearance to primary CO2 fluid inclusions; (2) the primary CO2 fluid inclusions contain carbonate inclusions; (3) the solid inclusions occur in rocks that lack the greenschist retrogression that is usually associated with secondary carbonate mineralization.
There are three problems with ascribing the solid inclusions to a carbonate melt. First, carbonate solid inclusions have not been reported previously from felsic plutonic rocks, although they are found in some ultramafic and mafic igneous rocks (Bogoch & Magaritz, 1983; Schiano et al., 1994). Second, carbonate melts are typically associated with alkalic rocks, rather than calc-alkalic ones. Third, the carbonate inclusions lack NaCO3, and experiments indicate that in the system Na2O–CaO–Al2O3–SiO2–CO2 a carbonate melt in equilibrium with a sodium-bearing silicate melt should contain some measurable NaCO3 (Kjarsgaard & Hamilton, 1989). This is a common problem. Carbonatites are generally considered to form by immiscibility, but most lack NaCO3 (Bailey, 1993). Although recognizing these problems, we conclude that the best interpretation is that the solid carbonate inclusions originated from a carbonate melt.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGYPETROLOGYWHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYINTENSIVE PARAMETERSFLUID AND SOLID INCLUSIONSDISCUSSIONCONCLUSIONS: THE CONTROLS ON...REFERENCES |
|---|
Origin of the Boulder Canyon batholith
The calc-alkalic composition of the Louis Lake batholith together with the geographic distribution of rocks of this age in the Wyoming province indicate that the batholith formed in a convergent margin setting similar to Phanerozoic destructive margins (Frost et al., 1998
, 2000). Rb–Sr, Sm–Nd, and U–Pb isotopic data indicate that the Louis Lake batholith was not derived from a single, isotopically homogeneous source, nor did it form from a simple mixture of two sources. Rather, it formed from multiple sources, one of which had isotopic values close to that of contemporary depleted mantle (Frost et al., 1998). The non-linear variation in major, minor, and trace elements on Harker diagrams (Fig. 6) also suggests multiple sources. This evidence includes the wide range in K2O and Na2O contents and the presence of the high- and low-REE series with distinct FeOt/(FeOt + MgO) and A/CNK ratios. The nearly constant Fe/(Fe + Mg) ratio of the ferromagnesian silicates over the whole range of silica content confirms that differentiation was not a major cause for the chemical variation among the different rocks of the batholith.
Like the isotopic data the compositional variation in Louis Lake batholith is best explained by mixing of variable amounts of mafic, mantle-derived melt with melts derived from at least two different crustal sources. The mafic melt may contribute the heat to melt compositionally heterogeneous crust as well as most of the CaO, FeO and MgO. Because mafic melts can efficiently transport CO2 (see Frost & Frost, 1987), most of the CO2 in the resulting mixture probably came from this mafic magma. The crustal melts would contribute most of the K2O, Na2O, Al2O3, and SiO2, as well as most of the H2O to the mixed magma. The crustal magmas include at least one peraluminous and one or more metaluminous melts.
Field relations shown schematically in Fig. 3 indicate that during the evolution of the Louis Lake batholith there were repeated injections of CO2-rich melts, of which at least one was granodiorite in composition and several were granitic. Initially these melts crystallized pyroxene-bearing assemblages, but as crystallization continued, water activity increased, eventually stabilizing biotite and hornblende. The fact that even some of the latest pegmatites carried CO2-rich fluids indicates that mafic magmas were being emplaced into the system throughout the history of the batholith.
There is no single C-type granite composition
The compositions of the charnockites of the Wind River Range differ significantly from the relatively iron-enriched C-type magmas of Kilpatrick & Ellis (1992). This is best seen on a plot of FeOt/(FeOt + MgO), which distinguishes a tholeiitic vs calc-alkalic magma affinity. Many charnockites are of tholeiitic affinity (i.e. have A-type characteristics), but some, including those from south India and the Louis Lake batholith, have calc-alkalic affinity. Because charnockites form from magmas that have such a broad range in composition there can be no bulk chemical control on the occurrence of Opx (or Fay) in true granites (Fig. 14).
|
The reason why most charnockites form from iron-enriched magmas can be shown by considering the effect of variation of µ(MgFe-1) on the breakdown of biotite to Opx (or fayalite) + K-feldspar (Fig. 15). This diagram is calculated for the system K2O–FeO–MgO–Al2O3–SiO2 using the data of Berman (1988) and assuming that Opx is an ideal two-site solution and that biotite is an ideal three-site solution. The calculations also assume that the tetrahedral site in biotite contains only one atom of Al per formula unit. The temperatures and water activities shown in Fig. 15 cannot be easily applied to natural rocks, because the solution of Ti, Fe3+, Na, and F into biotite will tend to stabilize it to higher temperatures, and the solution of Na into K-feldspar is probably not enough to offset this effect. Despite this limitation, the diagram shows that because biotite is always richer in Mg than Opx or Fay, biotite will be more stable in Mg-rich melts than in Fe-rich ones. There is a 200°C difference between the breakdown of phlogopite and that of annite. This means that at a fixed temperature, the assemblage biotite–Opx–K-feldspar–quartz may coexist over a wide range of water activities, depending on the XFe of the biotite. At 800°C, for example, this assemblage (containing a biotite with XFe = 0·45) will coexist with a water activity of 0·7, whereas that with a biotite of XFe = 0·7 will occur with a water activity near unity.
|
|
CONCLUSIONS: THE CONTROLS ON THE FORMATION OF CHARNOCKITES |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGYPETROLOGYWHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYINTENSIVE PARAMETERSFLUID AND SOLID INCLUSIONSDISCUSSIONCONCLUSIONS: THE CONTROLS ON...REFERENCES |
|---|
The first-order control over the occurrence of pyroxene in granitic rocks is the composition of the coexisting fluid phase. If the fluid phase has a low enough water activity (i.e. is sufficiently enriched in CO2), Opx forms in a granitoid melt of almost any composition. However, aegirine will form in strongly peralkaline melts, and garnet in peraluminous melts, instead of Opx. The influence of fluid composition on the presence of Opx is indicated by the common occurrence of these phases in calc-alkaline lavas and their comparative scarcity in plutonic rocks of similar compositions (Frost & Lindsley, 1992
). Evidently the venting of fluid during the eruption of volcanic rocks preserves the early-formed pyroxene that would have otherwise reacted to biotite or hornblende. The main question regarding the preservation of pyroxenes in plutonic rocks is how the granitoid melt can become enriched in CO2. One important source is through mantle degassing (Frost & Frost, 1987), but the participation of CO2 evolved from carbonates during metamorphism cannot be ruled out (Glassley, 1983).
Three additional factors favor the crystallization and formation of Opx in granitic rocks. The first is a high Fe/(Fe + Mg) ratio in the melt. As noted above, because biotite is less stable in rocks with high Fe/(Fe + Mg), Opx should more commonly form from iron-enriched melts than from more Mg-enriched ones. This explains why most charnockites, such as those of Lofoten (Malm & Ormaasen, 1978), Farsund (Middlemost, 1968), Nigeria (Olarewaju, 1987), Ardery (Kilpatrick & Ellis, 1992), and Kleivan (Petersen, 1980), are of the A-type association. However, the calc-alkalic nature of the charnockites from the Wind River Range and south India indicate that charnockites are not restricted to A-type granites.
A second contributing factor is the presence of mafic melts as a source of CO2. Leucogranitic magmas that formed simply by crustal melting are likely to be water enriched and therefore will not crystallize as charnockites, whereas A-type granites, which may form by melting or differentiation of anhydrous tholeiitic rocks, commonly have low water activity (Frost & Frost, 1997).
Finally, deep crustal levels favor the crystallization of pyroxene-bearing granites. The solubility of CO2 in silicate melts is strongly pressure dependent so that most of the CO2 in a melt will be exsolved at lower- or mid-crustal depths. At shallower crustal levels, the fluid in a magma is more likely to be dominated by H2O released both by crystallization of the magma itself and by dehydration of the wallrock (Frost et al., 1989). Moreover, early-formed pyroxene-bearing cumulates are likely to be found at the base of a pluton. In addition to the Wind River Range, charnockites are found in the deeper structural levels of the south Indian subcontinent and the Sierra Nevada batholith (Ross, 1985).
From the above discussion it is clear that charnockites (sensu lato) are not restricted to a particular geologic period, a specific chemical composition, nor a certain tectonic environment. Charnockites simply represent granitic rocks that have crystallized under low water activities.
|
ACKNOWLEDGEMENTS |
|---|
Much of this work was conducted by Tom Hulsebosch as part of his Ph.D. dissertation. Tom was killed in an auto accident in September 1996; this paper is dedicated to his memory. This study was partially supported by NSF grants EAR-8408357 and EAR-8707296. We wish to thank Jacques Touret at the Free University of Amsterdam for helping Tom obtain Raman spectroscopy of the carbonate inclusions. Helpful reviews of this manuscript were provided by Tom Chacko, Darrell Henry, and Sorena Sorensen.
|
FOOTNOTES |
|---|
*Corresponding author. e-mail: rfrost{at}uwyo.edu
Deceased. |
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGYPETROLOGYWHOLE-ROCK CHEMISTRYMINERAL CHEMISTRYINTENSIVE PARAMETERSFLUID AND SOLID INCLUSIONSDISCUSSIONCONCLUSIONS: THE CONTROLS ON...REFERENCES |
|---|
Andersen, D. J., Lindsley, D. H. & Davidson, P. M. (1993). QUILF: a Pascal program to assess equilibria among Fe–Mg–Mn–Ti oxides, pyroxenes, olivine, and quartz. Computers and Geosciences 19, 1333–1350.
Anderson, J. L. & Smith, D. R. (1995). The effects of temperature and FO2 on the Al-in-hornblende barometer. American Mineralogist 80, 549–559.[Abstract]
Bailey, D. K. (1993). Carbonate magmas. Journal of the Geological Society, London 150, 637–651.
Berman, R. G. (1988). Internally-consistent thermodynamic data for minerals in the system Na2O–K2O–CaO–MgO–FeO–Fe2O3–Al2O3–SiO2–TiO2–H2O–CO2. Journal of Petrology 29, 445–522.
Bogoch, R. & Magaritz, M. (1983). Immiscible silicate–carbonate liquids as evidenced from ocellar diabase dykes, Southeast Sinai. Contributions to Mineralogy and Petrology 83, 227–239.
Crawford, M. L. (1981). Phase equilibria in aqueous fluid inclusions. In: Hollister, L. S. & Crawford, M. L. (eds) Fluid Inclusions, Application to Petrology. Mineralogical Association of Canada Short Course Handbook 6, pp. 75–100.
Elkins, L. T. & Grove, T. L. (1990). Ternary feldspar experiments and thermodynamic models. American Mineralogist 73, 544–559.
Fisher, J. R. (1976). The volumetric properties of H2O—a geographical portrayal. Journal of Research of the US Geological Survey 4, 189–193.[Web of Science]
Friend, C. R. L. (1985). Evidence for fluid pathways through Archean crust and the generation of the Closepet granite, Karnataka, South India. Precambrian Research 27, 229–250.
Frost, B. R. & Frost, C. D. (1987). CO2, melts, and granulite metamorphism. Nature 327, 503–506.
Frost, B. R. & Lindsley, D. H. (1992). Equilibria among Fe–Ti oxides, pyroxenes, olivine, and quartz: Part II. Application. American Mineralogist 77, 1004–1020.[Abstract]
Frost, B. R., Frost, C. D. & Touret, J. L. R. (1989). Magmas as a source of heat and fluids in granulite metamorphism. In: Bridgwater, D. (ed.) Fluid Movements—Element Transport and the Composition of the Deep Crust. NATO ASI Series, C-281. Dordrecht: Kluwer Academic, pp. 1–18.
Frost, B. R., Chamberlain, K. R., Swapp, S. M., Frost, C. D. & Hulsebosch, T. P. (2000). Late Archean structural and metamorphic history of the Wind River Range: evidence for a long-lived active margin on the Archean Wyoming craton. Geological Society of America Bulletin 112, 564–578.
Frost, C. D. & Frost, B. R. (1997). Reduced rapakivi-type granites: the tholeiite connection. Geology 25, 647–650.
Frost, C. D., Frost, B. R., Chamberlain, K. R. & Hulsebosch, T. P. (1998). The Late Archean history of the Wyoming province as recorded by granitic magmatism in the Wind River Range, Wyoming. Precambrian Research 89, 145–173.
Glassley, W. E. (1983). Deep crustal carbonates as CO2 fluid sources: evidence from metasomatic reaction zones. Contributions to Mineralogy and Petrology 84, 15–24.
Holland, T. H. (1900). The charnockite series, a group of Archean hypersthenic rocks in Peninsular India. Memoirs of the Geological Survey of India 28, 119–249.
Howie, R. A. (1955). The geochemistry of the charnockite series of Madras, India. Transactions of the Royal Society of Edinburgh 82, 725–769.
Hulsebosch, T. P. (1993). Aspects of crustal evolution in the Archean Wind River Range, Wyoming. Ph.D. thesis, University of Wyoming, Laramie, 113 pp.
Indares, A. & Martignole, J. (1985). Biotite–garnet geothermometry in the granulite facies: the influence of Ti and Al in biotite. American Mineralogist 70, 272–278.[Abstract]
Janardhan, A. S., Newton, R. S. & Hansen, E. C. (1982). The transformation of amphibolite facies gneiss to charnockite in southern Karnataka and northern Tamil Nadu, India. Contributions to Mineralogy and Petrology 79, 130–149.
Kilpatrick, J. A. & Ellis, D. J. (1992). C-type magmas: igneous charnockites and their extrusive equivalents. Transactions of the Royal Society of Edinburgh 83, 155–164.
Kjarsgaard, B. A. & Hamilton, D. L. (1989). The genesis of carbonatites by immiscibility. In: Bell, K. (ed.) Carbonatites, Genesis and Evolution. London: Unwin Hyman, pp. 388–404.
Koesterer, M. E., Frost, C. D., Frost, B. R., Hulsebosch, T. P. & Bridgwater, D. (1987). Development of the Archean crust in the Medina Mountain area, Wind River Range, Wyoming (U.S.A.). Precambrian Research 37, 287–304.
Leake, B. E. (1978). Nomenclature of amphiboles. Canadian Mineralogist 16, 501–520.
Le Maitre, R. W., Bateman, P., Dudek, A., Keller, J., Lameyre, J., Le Bas, M. J., Sabine, P. A., Schmid, R., Sørensen, H., Streckeisen, A., Woolley, A. R. & Zanettin, B. (1989). A Classification of Igneous Rocks and Glossary of Terms. London: Blackwell.
Lindsley, D. H. & Frost, B. R. (1992). Equilibria among Fe–Ti oxides, pyroxenes, olivine, and quartz: Part I. Theory. American Mineralogist 77, 987–1003.[Abstract]
Malm, O. A. & Ormaasen, D. E. (1978). Mangerite–charnockite intrusives in the Lofoten–Vesterålen area, North Norway: petrography, chemistry and petrology. Norges Geologiske Undersökelse 338, 38–114.
Middlemost, E. (1968). The granitic rocks of Farsund, South Norway. Norsk Geologisk Tidsskrift 48, 81–99.
Mitra, G. & Frost, B. R. (1981). Mechanisms of deformation within Laramide and Precambrian deformation zones in basement rocks of the Wind River Mountains. University of Wyoming Contributions to Geology 19, 161–173.
Naylor, R. S., Steiger, R. H. & Wasserberg, G. J. (1970). U–Th–Pb and Rb–Sr systematics in a 2700 x 106 year old pluton from the southern Wind River Range, Wyoming. Geochimica et Cosmochimica Acta 34, 1133–1159.
Olarewaju, V. O. (1987). Charnockite–granite association in SW Nigeria: rapakivi granite type and charnockitic plutonism in Nigeria? Journal of African Earth Sciences 6, 67–77.[Web of Science]
Pearson, R. C., Kiilsgaard, T. H. & Patton, L. L. (1971). Mineral resources of the Popo Agie Primitive Area, Fremont and Sublette Counties. US Geological Survey Bulletin 1353B, 55 pp.
Petersen, J. S. (1980). The zoned Kleivan granite—an end member of the anorthosite suite in southwest Norway. Lithos 13, 79–95.
Roedder, E. (1984). Fluid inclusions. Mineralogical Society of America, Reviews in Mineralogy 12, 644 pp.
Ross, D. C. (1985). Mafic gneiss complex (batholithic root?) in the southernmost Sierra Nevada, California. Geology 13, 288–291.
Schiano, P., Clocchiatti, R., Shimizu, N., Weis, D. & Mattielli, N. (1994). Cogenetic silica-rich and carbonate-rich melts trapped in mantle minerals in Kerguelen ultramafic xenoliths: implications for metasomatism in the oceanic upper mantle. Earth and Planetary Science Letters 123, 167–178.
Sheraton, J. W., Black, L. P. & Tindle, A. G. (1992). Petrogenesis of plutonic rocks in a Proterozoic granulite-facies terrane—the Bunger Hills, East Antarctica. Chemical Geology 97, 163–198.
Srikantappa, C., Raith, M. & Spiering, B. (1985). Progressive charnockitization of a leptynite–khondalite suite in southern Kerala, India—evidence for formation of charnockites through decrease in fluid pressure. Journal of the Geological Society of India 26, 849–872.[Web of Science]
Stuckless, J. S. (1989). Petrogenesis of two contrasting, Late Archean granitoids, Wind River Range, Wyoming. US Geological Survey, Professional Papers 1491, 38 pp.
Wen, S. & Nekvasil, H. (1994). SOLVCALC: an interactive graphics program package for calculating the ternary feldspar solvus for two-feldspar geothermometry. Computers and Geosciences 20, 1025–1040.
Worl, R. G., Koesterer, M. E. & Hulsebosch, T. P. (1986). Geologic map of the Bridger Wilderness and the Green–Sweetwater roadless area, Sublette and Fremont counties, Wyoming. US Geological Survey Miscellaneous Field Studies Map, MF-1636-B.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. Pouclet, R. Tchameni, K. Mezger, M. Vidal, E. Nsifa, C. Shang, and J. Penaye Archaean crustal accretion at the northern border of the Congo Craton (South Cameroon). The charnockite-TTG link Bulletin de la Societe Geologique de France, September 1, 2007; 178(5): 331 - 342. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. PERCIVAL and J. K. MORTENSEN Water-deficient Calc-alkaline Plutonic Rocks of Northeastern Superior Province, Canada: Significance of Charnockitic Magmatism J. Petrology, September 1, 2002; 43(9): 1617 - 1650. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. R. FROST, C. G. BARNES, W. J. COLLINS, R. J. ARCULUS, D. J. ELLIS, and C. D. FROST A Geochemical Classification for Granitic Rocks J. Petrology, November 1, 2001; 42(11): 2033 - 2048. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||