Journal of Petrology Advance Access originally published online on November 18, 2006
Journal of Petrology 2007 48(2):365-393; doi:10.1093/petrology/egl063
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The Role of Pressure in Producing Compositional Diversity in Intraplate Basaltic Magmas
Department of Geosciences, Stony Brook University, Stony Brook, NY 11794-2100, USA
RECEIVED JANUARY 24, 2006; ACCEPTED SEPTEMBER 29, 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL METHODSEXPERIMENTAL RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Basaltic magmas found in intraplate suites appear to follow more than one differentiation trend. Many ocean island suites follow the ocean island tholeiitic trend, with the basalts differentiating from olivine tholeiite through basaltic andesite, andesite, and dacite to sodic rhyolite. Many continental intraplate magmatic regimes, such as those of the Snake River Plain and the plutonic sequences associated with massif anorthosites, follow the potassic silica-saturated alkalic trend, in which basalt differentiates from olivine tholeiite through ferrobasalt (jotunite or ferrodiorite), Fe-rich intermediate rocks (trachybasalt or monzonite), and trachyte (syenite) to potassic rhyolites and granites. Crystallization experiments on an olivine tholeiite from the Snake River Plain show that the basaltic portions of the ocean island tholeiitic trend and the potassic silica-saturated alkalic trend (which leads to strong alkali, P, Ti, and Fe enrichment and silica depletion) can arise from the same ‘dry’ tholeiitic parental magma. These compositional differences are induced by changes in phase equilibria as a function of pressure, with the ocean island tholeiitic series arising from crystal–liquid differentiation at low pressure and the potassic silica-saturated alkalic series arising via differentiation at elevated pressures.
KEY WORDS: tholeiite differentiation; experimental petrology; phase equilibria; ferrodiorite; ferrobasalt
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL METHODSEXPERIMENTAL RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Magmatic suites associated with hotspot or continental rift activity commonly contain a wide variety of rocks that are spatially and temporally related (i.e. Carmichael, 1964
; Abbott, 1969; Frey et al., 1991; Frost et al., 1993). Nekvasil et al. (2000, 2003) have shown that this compositional diversity comprises several distinct trends. Figure 1a shows the three primary compositional trends of bulk-rocks of silica-saturated [in the sense of Yoder & Tilley (1962)] intraplate magmatic suites (Trends 1–3). Trend 1, defined by lavas of the ocean island tholeiitic series, is exemplified by rocks from Easter Island (Baker et al., 1974; Haase et al., 1997), the Galapagos archipelago (Pinzon Island, Baitis & Lindstrom, 1980; Volcan Alcedo, Geist et al., 1995), and the Thingmuli volcano in Iceland (Carmichael, 1964). Bulk compositions of rocks of this series are subalkalic throughout their entire range of compositions (Fig. 1b), and follow the sequence olivine tholeiite–basaltic andesite–andesite (icelandite)–dacite–sodic rhyolite.
|
Rocks of the potassic silica-saturated alkalic series (Trend 2, Fig. 1a) are found exclusively in continental regimes. This series is exemplified by rocks of hotspot-related large igneous provinces (i.e. Yellowstone–Snake River Plain volcanic province) and the rocks associated with massif anorthosite complexes (excluding the obvious cumulate rocks). This trend is characterized by the tholeiitic nature of the least-evolved mafic rocks, the mildly alkalic nature of its intermediate units, and the potassic nature (i.e. high K2O/Na2O ratio) of the silicic rocks. Volcanic suites from this trend make up the sequence tholeiitic basalt–ferrobasalt–trachybasalt–trachyandesite–tristanite–trachyte–potassic rhyolite (Fig. 1b, Leeman et al., 1976
; Stout & Nicholls, 1977; Leeman, 1982a; Kuntz et al., 1992; Stout et al., 1994; Frost & Frost, 1997; McCurry et al., 1999, 2006), as seen in the Snake River Plain (SRP). Plutonic equivalents, such as from the Laramie Anorthosite Complex (LAC), follow the series olivine gabbro–ferrodiorite (jotunite)–monzodiorite–monzonite–monzosyenite–syenite–potassic granite (Frost et al., 1993; Scoates et al., 1996; see Frost et al., 1999, fig. 1b).
The potassic silica-saturated alkalic trend can be subdivided into two main segments (Fig. 1c): a silica-depletion (but not commonly silica-undersaturated) and a silica-enrichment segment. The silica-depletion segment (segment A, Fig. 1c) of this trend begins in the tholeiitic field with the least-evolved mafic rocks and extends toward low-silica, high-Fe–Ti–P ferrobasaltic rocks with decreasing MgO, crossing the subalkalic–alkalic boundary. The silica-enrichment segment begins primarily with the low-silica ferrobasalts and extends through the trachybasalt and trachyandesite fields to trachyte and potassic rhyolite (segment B, Fig. 1c). [Whitaker et al. (2006) noted an additional silica-enrichment trend emanating from tholeiite and extending to the more evolved ferrobasalts and on to rhyolite (Fig. 1c).] The high-silica portion of this trend that leads to potassic rhyolite has recently been experimentally investigated (Whitaker et al., 2004, 2006). The work presented here focuses specifically on the compositional diversity of the basaltic rocks.
Trend 3, the sodic silica-saturated alkalic trend, reflects the evolution of lavas from mildly alkalic basalt to alkali-rich sodic rhyolite and is exemplified by lavas from Ascension Island (e.g. Harris, 1983) and the Nandewar Volcano, Australia (e.g. Stolz, 1985). Nekvasil et al. (2004) experimentally produced liquids along this trend through crystallization of a mildly alkalic basalt from the Nandewar Volcano at 9·3 kbar with dissolved water content >0·5 wt %. Because of the importance of water to this trend, it will not be discussed further in this paper. Trend 4 is the silica-undersaturated alkalic series, exemplified by rocks from Hawaii (e.g. Frey et al., 1991) and Gough Island (e.g. Leroex, 1985). The rocks of this series have preliminarily been shown to form at higher pressures than those investigated in this study (Filiberto & Nekvasil, 2003), and are the subject of a forthcoming study (Filiberto et al., in preparation).
The liquid line of descent of the basaltic rocks of both the ocean island tholeiitic series and the potassic silica-saturated alkalic series appears to be controlled by crystal–liquid differentiation. Thompson (1975) carried out melting experiments on a tholeiite (6·7 wt % MgO) from the SRP and determined that at pressures between 8 and 12 kbar the liquids produced were similar in composition to the SRP ferrobasalts; Litvin (2002) determined experimentally that the ferrobasalts of the SRP can be reproduced by crystallization of a slightly more primitive tholeiitic magma (8·6 wt % MgO) at lower crustal pressures (9·3 kbar) under nominally anhydrous (
0·13 wt % H2O) conditions. Similarly, Scoates et al. (1999) showed experimentally that crystallization of tholeiitic high-aluminum olivine gabbro from the LAC under these same conditions can give rise to the ferrodiorites ( jotunites) of the anorthosite complexes.
The major and trace element characteristics of rocks of ocean island tholeiitic suites can be accounted for by crystallization of observed phenocryst phases (Baitis & Lindstrom, 1980; Geist et al., 1995). Experimental crystallization of a Kilauea basalt at low pressure (1 kbar) by Spulber & Rutherford (1983) produced a liquid line of descent with similar characteristics to the trend in major element compositions found at Thingmuli. Even in cases where there has been some contamination by crustal material (i.e. Macdonald et al., 1987), the evolutionary trends in major and trace element composition have been shown to be consistent with crystal–liquid differentiation.
A potential link
Many petrologists maintain that the processes or sources that have produced the ocean island tholeiitic series and the potassic silica-saturated alkalic series (Trends 1 and 2, Fig. 1a) must be fundamentally different (i.e. Geist et al., 1995; Frost & Frost, 1997) because the rocks from these suite types are clearly different. The least-evolved members of both suite types, however, are very similar in major element composition (Fig. 1). This similarity, when combined with the experimental and chemical evidence that the rocks within each suite type are related by differentiation, raises the question of whether the same type of parent magma crystallizing under different conditions can produce the major element characteristics of the basaltic rocks of both suite types.
This paper presents new experimental data evaluating the extent of compositional diversity in basaltic liquids that can be produced by crystallization of olivine tholeiite magma at various pressures under ‘anhydrous’ conditions. Although the primary variable under consideration in this paper is pressure some consideration must be given to oxygen fugacity. In natural rocks of the two suite types there is a difference in oxygen fugacity, with the ocean island tholeiite showing more oxidized compositions. The experiments mimic this difference; the higher pressure experiments were conducted at lower oxygen fugacity than the lower pressure experiments.
Experimental strategy
Experiments were designed to determine the liquid lines of descent of an ‘anhydrous’ intraplate olivine tholeiite as it ponds and undergoes equilibrium crystallization at various levels within the crust, producing liquids that eventually separate from the crystallized phases before further crystallization, eruption, or final emplacement. The pressures for these experiments were chosen to simulate ponding and crystallization in magma chambers at the base of the average continental crust (9·3 kbar, equivalent to 30 km), at mid-crustal levels (4·3 kbar, 15 km), and at sub-volcanic levels (‘0 kbar’). Also, intermediate pressures between these were chosen (6·8 and 2·8 kbar), corresponding to the top and bottom of a mid-crustal high-density layer at 10–20 km depth along the axis of the Snake River Plain (Sparlin et al., 1982; Peng & Humphreys, 1998).
|
EXPERIMENTAL METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL METHODSEXPERIMENTAL RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Starting material
The starting material for the experiments was a natural olivine tholeiite from the eastern Snake River Plain (ICPP123 260, abbreviated hereafter as I260; see Table 1), selected because of its high mg-number, high MgO content, low Ni content (87 ppm), lack of abundant phenocrysts, and minor amounts of alteration. The low bulk-rock Ni content and the compositional similarity of the groundmass and phenocrystic olivine suggest that the high Mg content of sample I260 is not the result of accumulated olivine, and that it could represent a liquid.
|
Sample I260 does not fall on the same Ni vs Mg trend as the vast majority of other SRP basalts; however, most basaltic rocks from the SRP have abundant olivine phenocrysts. Because the olivine–melt partition coefficients for these two elements are vastly different (Duke, 1976
; Takahashi, 1978), accumulation of olivine would alter their relative proportions in the bulk-rock, and thus may explain why sample I260, which contains very little phenocrystic olivine, falls off of this compositional trend, but still probably represents a liquid.
Sample I260 is significantly less evolved (10·62 wt % MgO) than the samples used in previous experimental studies and typical of the tholeiite (non-primitive) commonly found in these types of suites. Sample I260 is a mildly vesicular rock containing 2 vol. % phenocrystic olivine, and lesser amounts of plagioclase phenocrysts. The groundmass consists primarily of plagioclase, clinopyroxene and olivine, with minor amounts of Fe–Ti oxides, quenched glass, iddingsite, and trace amounts of apatite. Additional information on this rock has been given by Reed et al. (1997; sample ICPP123-79).
The rock sample was crushed and ground to yield a fine homogeneous powder with an average grain size of a few microns. This powder, when fused at 1320°C and 9·3 kbar in a graphite capsule, contains 0·35 wt % H2O as determined by micro-Fourier Transform Infrared Spectroscopy (FTIR). For the ‘anhydrous’ experiments reported in this paper, as much of this water as possible was removed, as described below.
Piston-cylinder experiments
For each experiment, the powdered rock was dried after loading into a graphite capsule by placing the capsule into a silica glass tube with an Fe° oxygen ‘getter’ and glass spacer to keep them separated. This assembly was evacuated and heated for 20 min in a vertical furnace. The drying geometry ensured that the capsule was heated to 800°C while the Fe° ‘getter’ was at 600°C, thereby preventing oxidation of the sample material during drying. After cooling, the sample was removed and placed into the cell assembly, loaded into the 1/2 inch piston-cylinder apparatus, and pressurized to prevent re-adsorption of water. The bulk water content of the dried sample was found by FTIR analysis of a fused sample to be 0·05 wt %.
The cell assembly for the piston-cylinder experiments consisted of a low-porosity (14 %) BaCO3 outer sleeve, into which a graphite furnace was inserted along with dried mullite spacers and crushable MgO discs. The graphite capsule in this assembly was in direct contact with the graphite sleeve, thereby effectively becoming part of the furnace assembly. An insulating alumina disc separated the thermocouple from the graphite capsule. The temperature of the experiment was both controlled and monitored by a Pt–Pt90Rh10 thermocouple. This assembly has a temperature gradient of +12°C from the thermocouple to the center of the capsule at 9·3 kbar, and a gradient of +14°C at 4·3 kbar. The temperatures reported here are those at the hotspot of the sample. Pressures reported include a –0·7 kbar correction based on the subsolidus reaction Mg-cordierite = Sapphirine + Quartz (Newton et al., 1974) at 7 kbar.
Because the experimental charges were housed in graphite, their fO2 was partially controlled by the graphite–CO–CO2 (GCO) buffer. However, because there was no vapor phase evident in our experiments, the GCO buffer provides only an upper limit to fO2 (Eugster & Skippen, 1967). The actual fO2 is affected by the bulk composition of the sample (especially the Fe3+/Fe2+ ratio) within the graphite capsule. Whenever an appropriate assemblage was present, the fO2 was calculated using QUILF assemblage of olivine, pyroxene, and ilmenite and the technique of Andersen et al. (1993). For the piston-cylinder experiments, the fO2 of the system was computed (using the QUILF) to be 2·5 log units below fayalite–magnetite–quartz (FMQ) buffer at all pressures. However, an appropriate assemblage was available only at low temperatures for this starting material. To assess the fO2 at higher temperatures, calibration experiments were conducted at 1200°C by adding sufficient spinel (Usp95Mt5) and ilmenite (Hem20Ilm80) to I260 powder to ensure that two oxides remained after melting and crystallization. The resulting oxide pairs yielded fO2 values 1·5 log units (9·3 kbar) and 2·3 log units (4·3 kbar) below the FMQ buffer. These calibration experiments provide, however, only an upper limit to the fO2, because the bulk oxide compositions were reduced during reaction, indicating that they effectively added oxygen to the sample, raising its initial Fe3+/Fe2+ ratio.
Crystallization experiments were conducted using the piston-out method in which the pressure was slowly raised to a value approximately 2 kbar greater than the desired pressure while still at room temperature, then the sample was heated to the melting temperature (1300–1320°C) before decreasing the overpressure. The sample remained at this melting temperature and pressure for 2 h to ensure complete fusion, then was rapidly cooled to the final crystallization temperature of the experiment, where the sample remained for 2–3 days. Each experiment was quenched by turning off the power, resulting in an initial cooling rate of 100°C/s.
Experiments at ‘0 kbar’
Low-pressure experiments were conducted in a Pt-wound vertical quench furnace using Au80Pd20 capsules inserted into silica glass tubes. To compensate for potential Fe loss to the capsule, 6 relative wt % Fe° sponge (the amount of Fe lost to the capsule in a test run) was added to the rock powder. The Fe°-enriched I260 powder was loaded into each capsule and the top was crimped shut. The sample was then dried by placing it into a silica glass tube with a glass spacer, drawing out a capillary just above the spacer, placing a piece of Fe° oxygen ‘getter’ into the tube above the capillary, and drying under the conditions described for the piston-cylinder experiments. After drying, and with the sample still under vacuum, the silica-glass capillary was sealed, severed, and worked into a loop, producing an evacuated ampoule that could be hung in the vertical quench furnace.
The Pt-wound quench furnaces used for these experiments were restricted to temperatures of 1250°C or below. As the liquidus temperature of the I260 starting material is higher than this maximum, the evacuated ampoule containing the sample material was first placed in a Deltech furnace at 1280°C for 2–3 h to ensure complete melting, and then air-quenched. The ampoule was then suspended in the quench furnace by a fine Pt wire, heated to 1250°C, and then allowed to cool to its final crystallization temperature at a rate of 1°C/min, where it remained for 2–3 days. To quench the charge, the fine Pt wire holding it in the furnace was fused, causing the sample to drop into a cold water bath. The integrity of the ampoule after the experiment was tested by comparing pre- and post-experiment weights, and by heating the end of the ampoule near the spacer to check the quality of the vacuum. Only experiments that passed both the weighing and vacuum tests are reported here.
At the temperature of each experiment, the pressure in the tube must equal the sum of the vapor pressures of the phases. This sum is not known, but must be less than the tensile strength of the silica-glass tubes (3 bar). These experiments are referred to here as ‘0 kbar’ experiments because of this uncertainty. The fO2 of these experiments was assessed using the technique of Andersen et al. (1993) whenever the QUILF mineral assemblage olivine + pyroxene + spinel ± ilmenite was present. Such calculations yielded values of 1–2 log units above FMQ . The fO2 of these experiments is not greatly different from that found in the rocks of Thingmuli (Carmichael, 1964), which Frost & Lindsley (1992) calculated to be 1 log unit below to 1 log unit above FMQ .
Analytical techniques
The major element compositions of all phases present in experimental run products were determined by electron microprobe. Analytical conditions were 15 kV and 10 nA, with an 11 µm x 11 µm rastered beam whenever grain size permitted. Counting was for 30 s or to 40 000 counts for all elements except Si (40 s or 111 111 counts) and Al (40 s or 62 500 counts), and the PAP correction routine was used. Standards were microcline for K and Si, anorthite for Ca and Al, ilmenite for Fe and Ti, albite for Na, forsterite for Mg, apatite for P, and rhodonite for Mn. Care was taken to avoid alkali loss (especially from glass) by analyzing Na and K first and exposing the analyzed area to the beam for as short a time period as possible prior to analysis. Alkali loss from glass compositions was signaled by the appearance of normative corundum (Co); the absence of Co in all glass compositions reported here suggests that alkali loss has been minimal. All pyroxene, olivine and oxide analyses were projected through QUILF (Andersen et al., 1993) and checked for equilibrium between these phases, and the fO2 of the system was calculated when the assemblages permitted.
The IgPet Program Suite (Carr, 2002) was used to calculate molar norms of all residual liquids and pyroxenes as well as to conduct mass-balance calculations using a least-squares routine. These least-squares calculations yielded phase proportions (in wt %) and helped ensure that no phases were overlooked during microprobe analysis. Only those combinations of phases in a single experiment that yielded sums of the squares of the residuals less than 0·15 are reported here.
The water contents of glasses were measured by analyzing doubly polished wafers of the experimental run products with micro-IR spectroscopy using the Nicolet 20SXB FTIR spectrometer and Spectra Tech IR Plan microscope at the American Museum of Natural History. The density of the glasses and the total water concentrations were calculated using the methods of Dixon et al. (1995) and Mandeville et al. (2002).
|
EXPERIMENTAL RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL METHODSEXPERIMENTAL RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The run products of experiments contained variable amounts of crystalline phases and glass. There was little to no compositional zoning present in the crystals with the exception of augite, which commonly displayed significant variation in non-quadrilateral components, especially Ti and Al. Only experiments that produced homogenous residual glass are reported here.
Phase relations
The assemblages crystallizing from dry olivine tholeiite are strongly pressure-dependent (Tables 2–6, Fig. 2). Figure 3 shows the computed abundances of phases present in the experiments. At all pressures, olivine is the liquidus phase; at ‘0 kbar’ it is joined by a spinel phase that persists over the entire range of temperatures explored. At 9·3 kbar, sub-calcic augite is the next phase to appear, followed by plagioclase and eventually ilmenite in the lowest-temperature experiment. At pressures below 9·3 kbar, plagioclase joins the crystallizing assemblage before augite (not sub-calcic); ilmenite is absent at 6·8 and 2·8 kbar even in the lowest-temperature experiments. Plagioclase becomes the dominant phase in the assemblage at lower temperatures at all pressures except 9·3 kbar where it is in equal abundance with augite.
|
|
Figure 4 shows olivine and pyroxene compositions as projected into the pyroxene quadrilateral through QUILF (Andersen et al., 1993
) using the method of Lindsley & Anderson (1983, pp. A894–A895). The compositions of the olivines are only slightly influenced by the pressure at which they crystallized, with the liquidus olivine being slightly more magnesian at low pressure. In constrast, the compositions of the pyroxenes are strongly dependent upon pressure, particularly the wollastonite component, as pyroxenes formed at higher pressures have higher Al2O3 and Na2O, and lower CaO than those formed at lower pressures. At 9·3 kbar, the augite compositions lie in the range Wo20–Wo25 (increasing with decreasing temperature; Fig. 4a), wherease at lower pressures the Wo content of the pyroxenes is significantly higher (Fig. 4b–e). The augites that crystallized at 9·3 kbar (Fig. 4a) lie in the augite + pigeonite region of the pyroxene quadrilateral (Lindsley, 1983, fig. 9c, p. 485). This is a computational artifact arising from the high alumina content of the pyroxene (8–9 wt % Al2O3). The projection scheme of Lindsley & Anderson (1983) ‘over-corrects’ for the presence of Al in the pyroxene, thereby driving the calculated Wo content of the 9·3 kbar pyroxenes into that two-phase region.
|
The relationships between coexisting feldspars and liquids are shown in Fig. 5. For this figure, the molar normative feldspar components (An, Ab, Or) were normalized to unity for the residual liquids, feldspars, and pyroxenes. (The pyroxenes were included in this projection to meet mass-balance constraints.) The compositions of the feldspars formed during these experiments are highly dependent upon the pressure of crystallization, which is in good agreement with the results of Fram & Longhi (1992
). The anorthite content of the first-formed plagioclase increases with decreasing pressure, from An62 at 9·3 kbar to An78 at ‘0 kbar’. The higher-pressure feldspars are also more Or-rich (Or2–5 at 9·3 kbar) than the lower-pressure feldspars (Or<1–1 at ‘0 kbar’).
|
|
|
|
|
|
Liquid evolution
The compositions of residual liquids produced by crystallization of I260 are highly dependent upon pressure (Tables 7–11
and Figs 6–12). Residual liquids produced at 9·3 kbar become strongly depleted in silica and enriched in total alkalis, crossing the boundary into the alkalic field (Table 7, Fig. 6). The strong silica depletion of the residual liquids is accompanied by depletion in MgO and CaO, and strong enrichment in FeOT, TiO2, and P2O5, thereby producing ferrobasaltic ( jotunitic) compositional characteristics (Fig. 7). Both Al2O3 and Na2O initially increase strongly, but eventually exhibit marked decreases as plagioclase joins the crystallizing assemblage. K2O steadily increases nearly incompatibly throughout the crystallization sequence.
|
|
|
|
|
|
|
|
|
|
|
|
The residual liquids produced at 6·8 kbar show similar, but less pronounced, silica depletion and alkali enrichment; the residual liquids still move from the tholeiitic to the alkalic field (Table 8, Fig. 6). Figure 8 shows that there is sufficient depletion in MgO accompanied by enrichment in FeOT, TiO2, and P2O5, to again lead to the production of ferrobasaltic liquids. Liquids produced by crystallization of I260 at 4·3 kbar show only mild silica depletion during crystallization and less enrichment in FeOT, TiO2, and P2O5 than at higher crystallization pressure; the liquids still cross into the alkalic field (Table 9, Figs 6 and 9). The decrease in Na2O in the lowest temperature residual liquids, although still apparent, is much less pronounced than that observed at both 9·3 and 6·8 kbar. When crystallizing at 2·8 kbar, this olivine tholeiite magma evolves from tholeiitic to mildly alkalic at nearly constant silica content (Table 10, Fig. 6). The FeOT, TiO2, and P2O5 enrichment is less pronounced than at higher pressures, and the decrease in Na2O that was observed in the residual liquids of the lowest temperature experiments at higher pressures is not exhibited (Fig. 10).
Residual liquids produced at ‘0 kbar’ trend directly toward silica enrichment (Table 11, Fig. 6). Although alkalis are also enriched along this liquid line of descent, the liquid never crosses the subalkalic–alkalic boundary of Irvine & Baragar (1971). The melts remain tholeiitic (sub-alkalic) as the magma evolves to basaltic andesite composition with early increases in CaO, TiO2, FeOT, and Al2O3; these elements become depleted as crystallization continues, with Al2O3 leveling off again at the highest degrees of crystallinity.
In experiments where plagioclase was crystallizing, the crystallization pressure had a strong effect on the liquid line of descent (Fig. 5). At low pressure, the high abundance of plagioclase, together with its higher An content, results in greater An depletion and Ab enrichment with increasing differentiation. The incremental change in liquid compositions produced by the change in mineral assemblage at the five experimental pressures can be seen in Figs 6 and 12.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL METHODSEXPERIMENTAL RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Implications for the potassic silica-saturated alkalic series
Snake River Plain
By simulating the process whereby an olivine tholeiite magma would pond and undergo equilibrium crystallization at various levels within the crust, liquid compositions were produced that span much of the diversity in major element bulk-rock composition of basaltic rocks found in the natural suites from the SRP. The experimental mineral assemblages that give rise to these residual liquids can also be used for comparison with the natural assemblages. However, such comparison must be made with caution. The bulk lava compositions are considered here to represent residual liquid removed from the assemblage crystallizing at higher pressure. Upon ascent, this liquid will probably crystallize to some extent; therefore, the existing mineral assemblages in the rock are likely to represent primarily the phases crystallizing from these residual liquids at the final ponding or crystallization pressure prior to eruption rather than those crystallizing at higher pressure. Comparison of the natural and experimental mineral assemblages, however, should provide insight into this final pressure of crystallization of the SRP magmas.
Olivine tholeiites and related basalts from the SRP are commonly reported to contain up to 20 vol. % olivine phenocrysts with less abundant phenocrystic plagioclase and rare clinopyroxene. The cores of olivine phenocrysts in these basalts range from Fo85 to Fo60, and plagioclase phenocrysts range in composition from An80 to An60 (Leeman, 1982b). The more evolved basaltic and ferrobasaltic rocks from the Craters of the Moon (COM) lava field in the SRP commonly contain phenocrysts of olivine and plagioclase with minor amounts of pyroxene locally. The compositions of the olivine phenocrysts range from Fo72 in the basalts to Fo5 in the ‘ferrolatites’ (trachybasalts–trachyandesites), and the plagioclase phenocrysts range from An63 to An30 over the range in bulk-rock composition (Leeman et al., 1976).
The least-evolved phenocrysts in these rocks are nearly identical in composition to the same phases found in the highest temperature experiment at 9·3 kbar that contains both olivine and plagioclase (Fo73 and An62), but there is a greater difference between the assemblages in the more evolved compositions. The major difference between the experimental mineral assemblages that led to the strong Fe–Ti–P enrichment seen in the bulk compositions of the natural rocks and the natural phenocryst assemblages described in SRP–COM basaltic rocks is the presence of abundant clinopyroxene in the former. The lack of abundant pyroxene phenocrysts in the SRP–COM rocks has led previous workers (e.g. Leeman, 1982b) to postulate that these rocks must have formed by differentiation in a magma chamber shallow enough to allow olivine and plagioclase to crystallize without clinopyroxene in the assemblage. Mass-balance calculations to assess the relationship between the bulk compositions of SRP rocks by crystallization have been plagued by this lack of clinopyroxene phenocrysts in the natural samples. Leeman et al. (1976) found it impossible to generate the bulk compositions of the ferrobasalts of the COM suite from the average McKinney basalt composition (Leeman & Vitaliano, 1976) by crystallizing the low-pressure (1 bar) phenocryst assemblage of olivine, plagioclase and spinel. However, crystallization of the mineral assemblage experimentally observed by Thompson (1975) in his 8 kbar experiments on an SRP basalt (olivine, clinopyroxene, and plagioclase) could give liquids that were similar in composition to COM ferrobasalts. Similarly, Stout et al. (1994) also found that they needed a significant amount of clinopyroxene (nearly twice the amount of olivine crystallized) in their calculations to obtain a good match for deriving a COM ferrobasalt from an SRP basalt.
The requirement for significant amounts of clinopyroxene in these mass-balance calculations is consistent with the experimental implications that a higher-pressure history is needed to produce ferrobasaltic liquids from the tholeiite. The absence of high-pressure phenocrystic clinopyroxene crystals in the lavas can be explained simply—they were left behind along with other higher-pressure crystalline phases during ascent of the residual liquid. The greatest similarity in mineral assemblage (relative abundances and compositions) between those produced experimentally and those observed in rocks from the Snake River Plain–Craters of the Moon is is seen in the lower-pressure experimental assemblages. This suggests that following ascent of these residual liquids generated at depth, lower-pressure crystallization took place prior to eruption at the surface. Because the bulk compositions retain their higher-pressure signature, the final magma ascent did not involve the separation of liquid from the low-pressure crystalline assemblage.
It is worth noting here that the results obtained in our 9·3 kbar experiments are very different from those obtained by Villiger et al. (2004), who conducted equilibrium and fractional ‘crystallization’ experiments on a tholeiitic basalt at 1· 0 GPa (10 kbar). Whereas we find significant silica depletion in our residual liquids at this pressure, the residual liquids in their experiments showed silica enrichment. One possible reason for this disparity lies in the differences in experimental approach; our experiments were true crystallization experiments in which the bulk material was melted completely and then allowed to crystallize. The experiments of Villiger et al. (2004) were melting experiments in which the charges were taken directly to the desired crystallization temperature. In the former experiments, both using natural rock powder and melting the starting material above, but close to, the liquidus temperature facilitated nucleation. The latter approach involves melting a metastable assemblage of oxides only at the ‘crystallization’ temperature (Villiger et al., 2004). This may have led to the nucleation problems that were encountered (involving plagioclase and even olivine in some experiments).
Plagioclase crystallization plays an important role in driving residual liquids toward silica depletion. Inhibition of plagioclase nucleation at this pressure (10 kbar) would contribute to the silica enrichment exhibited by the residual liquids of Villiger et al. (2004), or at the very least strongly decrease the amount of silica depletion that might occur. Another possible reason for the discrepancy in residual melt compositions lies in the possible presence of water; the experiments Villiger et al. conducted were referred to as ‘nominally anhydrous,’ but water content was not specified. As Whitaker et al. (2006) determined, the addition of even small amounts of water to a tholeiitic magma at these pressures can cause the residual liquids to evolve toward silica enrichment rather than depletion. Silica enrichment would also be enhanced by the loss of Fe to the Pt lining of the graphite capsule reported by Villiger et al. (2004).
Villiger et al. (2004) compared their experimental results with pMELTS calculations (Ghiorso et al., 2002) and found that there was very poor agreement between the two. Our 9·3 kbar experiments, on the other hand, are in good agreement with the results of pMELTS calculations conducted using I260 at 10 kbar, which also show a strong silica depletion, Fe–Ti–P-enrichment trend in the residual liquid. MELTS calculations (Ghiorso & Sack, 1995) are not appropriate at this pressure.
Because our experiments were equilibrium crystallization experiments, the liquids produced should be identical to those produced by equilibrium partial melting under the same conditions of pressure, f O2, and bulk water content. None of these experiments were able to produce liquids close to rhyolite in composition; therefore, these results do not support generation of the low-silica rhyolites at low melt fractions (<10 %) by partial melting of essentially anhydrous gabbro at any pressure within the crust (Christiansen & McCurry, 2006).
The liquids produced experimentally at 9·3 kbar are in good agreement with the ferrobasalts found in the COM suite (and some ferrodiorites from the Laramie Anorthosite Complex), whereas the liquids produced at pressures from 6·8 to 2·8 kbar are better matches for the basaltic rocks found elsewhere in the SRP. Separation of residual liquids following extensive crystallization within the crust would leave behind a large quantity of mafic minerals. Perhaps the high-density layers found at 10–20 km depth in the SRP (Sparlin et al., 1982; Peng & Humphreys, 1998) represent accumulated crystals left behind during ascent of residual liquid following crystallization, or acted as barriers to upwelling magmas during ascent and facilitated ponding at elevated pressure. When compared with the natural rocks (Figs 7–10
), it appears that the compositional diversity between mafic rocks in large igneous provinces such as the SRP could be caused by tapping of residual liquids formed by crystal–liquid differentiation at a variety of pressures.
Laramie Anorthosite Complex
Whereas much of the compositional diversity observed in the bulk compositions of the gabbroic and dioritic rocks of the LAC has been reproduced in the residual liquids of these experiments over the range of intra-crustal pressures explored, the experimental crystallization path at 9·3 kbar (Figs 6–7) leads to the strong silica depletion and Fe–Ti–P enrichment of residual liquids as they evolve from gabbroic to ferrodioritic (jotunitic) compositions. Liquids strikingly similar to some of the high-aluminum gabbros (HAGs) generally considered to be parental to anorthosite formation (Mitchell et al., 1995) are produced by crystallization of dry olivine tholeiite at 9·3 kbar. The natural gabbros, however, span a large range in alumina content. Those that are higher in alumina than the experimental residual liquids have trace element signatures suggesting excess plagioclase component, perhaps owing to the resorption of accumulated plagioclase during ascent (Mitchell et al., 1995). In general, however, the high alumina contents of our experimental liquids suggest that resorption of accumulated plagioclase is not required to produce many of the observed HAGs.
These experiments support the hypothesis of Mitchell et al. (1996) that the ferrodiorites of the LAC represent the liquid complement to the anorthosite and formed by crystallization of a parental magma similar in composition to the LAC Hi-Al gabbros. This support arises from three primary observations. First, experimental liquids residual to tholeiite–HAG crystallization are strikingly similar to the ferrodiorites (Mitchell et al., 1996) with their low silica and alumina, and very high FeOT, TiO2, and P2O5. Second, plagioclase is a volumetrically abundant phase during high-pressure (9·3 kbar) crystallization of this tholeiite. Finally, the composition of the plagioclase crystallizing at the stage when the residual liquid attains a ferrodioritic composition is very similar to that found in the anorthosite bodies of the LAC. Plagioclase in the anorthosites of the LAC generally ranges from An60 to An40 (Frost et al., 1993), which is consistent with the experimental values, An62–An54. (The plagioclase produced in the experiments at lower pressures reaches only the most calcic end of this range at the lowest temperatures explored.) In addition, the plagioclase in the anorthosites has similar Or content to our 9·3 kbar plagioclase. Thus, these experiments also support the formation of the plagioclase of the anorthosite bodies of the LAC at elevated pressure during crystallization of a tholeiitic or HAG-like parental liquid.
Implications for the ocean island tholeiitic series
The experiments carried out on I260 at ‘0 kbar’ produced liquids that are markedly similar in bulk composition to basaltic rocks found in the ocean island tholeiitic series, characterized by a silica-enrichment trend in the residual liquids. There are, however, some deviations in the experimental liquid compositions at ‘0 kbar’ from the natural ocean island tholeiitic series shown. Potassium deviates fairly significantly from the natural trend during crystallization, owing to the relatively high starting K2O content of I260 compared with that of the least-evolved rocks of the ocean island tholeiitic series. TiO2 contents of low-temperature residual liquids lie slightly above the trend, and FeOT plots just barely below the ocean island tholeiitic trend. The deviations in TiO2 and FeOT may result from the relatively high f O2 of the ‘0 kbar’ experiments.
How much of this silica-enrichment trend, however, is an effect of pressure rather than an effect of f O2? Because the experiments conducted at ‘0 kbar’ were conducted at a higher f O2 than those at higher pressures, f O2 represents a variable in addition to pressure that must be considered. Higher f O2 would lead to increased magnetite component in the spinel, which would lead to a decrease in the amount of olivine and pyroxene that can form, resulting in a greater amount of silica enrichment. However, the amount of spinel crystallizing in these experiments is very low. Calculations in which all of the Fe in the spinel is removed to make more olivine still indicated significant silica enrichment of the residual liquid. Therefore, we believe it is primarily a pressure effect that produces the compositional differences between the two trends.
The mineral assemblages produced experimentally at ‘0 kbar’ agree well with the observed phenocryst phases from the natural rocks. Olivine tholeiite, tholeiite and basaltic andesite from Thingmuli (Iceland) contain phenocrysts of plagioclase ranging in composition from bytownite to labradorite, in addition to rare olivine (Fo75) and augite (Wo40En45) phenocrysts (Carmichael, 1964). Pinzon (Galapagos) basaltic rocks contain plagioclase phenocrysts ranging in composition from An76 to An63, olivine phenocrysts ranging from Fo85 to Fo57, phenocrystic augite ranging from Wo43En47 to Wo38En42, and occasional phenocrysts of Fe–Ti oxides (Baitis & Lindstrom, 1980). Phenocryst assemblages in basaltic rocks from Volcan Alcedo (Galapagos) include plagioclase (An84–61), augite (Wo45–41En46–45), olivine (Fo85–63), and small amounts of Ti-magnetite (Geist et al., 1995). The experimental plagioclase in these experiments ranges in composition from An78 to An67, olivines range from Fo87 to Fo71, and augites range from Wo40En49 to Wo38En44. An Fe–Ti oxide phase was also crystallized experimentally.
The similarities between natural bulk lava compositions and experimental residual liquids as well as between experimental and natural phenocryst assemblages support the results of Spulber & Rutherford (1983) suggesting that these rocks form by shallow crystallization of a tholeiitic parental magma.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL METHODSEXPERIMENTAL RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Crystallization of a ‘dry’ olivine tholeiite parental non-primitive melt at various levels within the crust produces markedly different mineral assemblages and residual liquids. Olivine tholeiite of the type found in the Snake River Plain can produce liquid compositions similar to bulk-rock compositions found in both the potassic silica-saturated alkalic series, such as those from the Snake River Plain–Craters of the Moon and Laramie Anorthosite Complex by crystallization at elevated pressures. It can also produce the ocean island tholeiitic series, exemplified by rocks from Iceland and the Galapagos, via crystallization at low pressure.
We suggest here that pressure-induced changes in phase relations—such as early stability of intermediate plagioclase and clinopyroxene at elevated pressure, and early stability of more anorthitic plagioclase at lower pressure as well as late crystallization of pyroxene—are the primary factors that give rise to the compositional diversity in residual liquids of these two suite types. The differences in oxygen fugacity simply cannot account for the differences in liquid evolution observed.
Some variations in the olivine tholeiite compositions that give rise to these two suite types do appear to exist, however. These can be seen by the emanation of the ocean island tholeiitic series from a more sodic tholeiite and the silica-saturated potassic series from a slightly more potassic tholeiite. These differences in olivine tholeiite may arise secondarily, such as through crustal contamination, and may not be the result of differences in source compositions. The primary pressure-dependent liquid evolutionary trends appear robust through this variation.
In addition to the slight compositional variations in the olivine tholeiite of the two suite types, there are differences in oxygen fugacity. However, these differences may also reflect secondary processes, and need not imply different source characteristics. The redox state of ascending magma may change depending upon the nature of the buffering assemblages in the hosting rocks. If ferrous–ferric equilibria dictate the f O2 of the upper mantle and crust (Ballhaus & Frost, 1994), and the melts ascend slowly enough to maintain equilibrium with the hosting rocks, the melts will become more oxidized at shallower levels. Thus, the same parental tholeiite may be more oxidized at shallower depths than at high pressure, and this change in f O2 is brought about not by differences in source region chemistry, but rather by differences in environment.
The results of this study indicate that the broad compositional diversity of dry intraplate basaltic magmas may arise from a single parental magma composition as a result of crystal–liquid differentiation processes at different pressures and does not necessitate source region heterogeneity.
|
ACKNOWLEDGEMENTS |
|---|
The authors would like to gratefully acknowledge the helpful and constructive reviews by Dave Walker, Tony Morse, and Eric Christiansen. Their comments, along with those of editor Ron Frost, helped to greatly improve the quality of this manuscript. The authors acknowledge the financial support of NSF grant EAR 0003443 to D.H.L. and H.N.
*Corresponding author Fax: (631) 632-8240. Email: matthew.whitaker{at}sunysb.edu
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL METHODSEXPERIMENTAL RESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbott MJ. (1969) Petrology of the Nandewar Volcano, N.S.W, Australia. Contributions to Mineralogy and Petrology 20:115–134.[CrossRef]
Andersen DJ, Lindsley DH, Davidson PM. (1993) QUILF—a Pascal program to assess equilibria among Fe–Mg–Mn–Ti oxides, pyroxenes, olivine, and quartz. Computers and Geosciences 19:1333–1350.
Baitis HW and Lindstrom MM. (1980) Geology, petrography, and petrology of Pinzon Island, Galapagos archipelago. Contributions to Mineralogy and Petrology 72:367–386.[CrossRef][Web of Science]
Baker PE, Buckley F, Holland JG. (1974) Petrology and geochemistry of Easter Island. Contributions to Mineralogy and Petrology 44:85–100.[CrossRef][Web of Science]
Ballhaus C and Frost BR. (1994) The generation of oxidized CO2-bearing basaltic melts from reduced CH4-bearing upper-mantle sources. Geochimica et Cosmochimica Acta 58:4931–4940.[CrossRef][Web of Science]
Carmichael ISE. (1964) The petrology of Thingmuli, a Tertiary volcano in Eastern Iceland. Journal of Petrology 5:435–460.
Carr MJ. (2002) IgPet for Windows(Terra Softa Inc, Somerset, NJ).
Christiansen EH and McCurry M. (2006) Contrasting origins of Cenozoic silicic volcanic rocks from the Western Cordillera of the United States. Bulletin of Volcanology (in press).
Dixon JE, Stolper EM, Holloway JR. (1995) An experimental study of water and carbon dioxide solubilities in mid-ocean ridge basaltic liquids. Part 1: Calibration and solubility models. Journal of Petrology 36:1607–1631.
Duke JM. (1976) Distribution of the period four transition elements among olivine, calcic clinopyroxene and mafic silicate liquid; experimental results. Journal of Petrology 17:499–521.
Eugster HP and Skippen GB. (1967) Igneous and metamorphic reactions involving gas equilibria. In Abelson PH (Ed.). Researches in Geochemistry(John Wiley, New York) pp. 492–520.
Filiberto J and Nekvasil H. (2003) Linking tholeiites and silica-undersaturated alkalic rocks: an experimental study. Geological Society of America, Abstracts with Programs 35:632.
Fram MS and Longhi J. (1992) Phase equilibria of dikes associated with Proterozoic anorthosite complexes. American Mineralogist 77:605–616.[Abstract]
Frey FA, Garcia MO, Wise WS, Kennedy A, Gurriet P, Albarede F. (1991) The evolution of Mauna-Kea Volcano, Hawaii—petrogenesis of tholeiitic and alkalic basalts. Journal of Geophysical Research—Solid Earth and Planets 96:14347–14375.[CrossRef]
Frost CD and Frost BR. (1997) Reduced rapakivi-type granites: the tholeiite connection. Geology 25:647–650.
Frost BR and Lindsley DH. (1992) Equilibria among Fe–Ti oxides, pyroxenes, olivine, and quartz: Part II. Application. American Mineralogist 77:1004–1020.[Abstract]
Frost BR, Frost CD, Lindsley DH, Scoates JS, Mitchell JN. (1993) The Laramie Anorthosite Complex and Sherman Batholith: geology, evolution, and theories of origin. In: Snoke, A. W, Steidtmann, J. R. & Roberts, S. (eds) Geology of Wyoming. Geological Survey of Wyoming Memoir 5:118–161.
Frost CD, Frost BR, Chamberlain KR, Edwards BR. (1999) Petrogenesis of the 1· 43 Ga Sherman batholith, SE Wyoming, USA: a reduced, rapakivi-type anorogenic granite. Journal of Petrology 40:1771–1802.[CrossRef][Web of Science]
Geist D, Howard KA, Larson P. (1995) The generation of oceanic rhyolites by crystal fractionation—the basalt–rhyolite association at Volcan Alcedo, Galapagos archipelago. Journal of Petrology 36:965–982.
Ghiorso MS and Sack RO. (1995) Chemical mass-transfer in magmatic processes IV. A revised and internally consistent thermodynamic model for the interpolation and extrapolation of liquid–solid equilibria in magmatic systems at elevated temperatures and pressures. Contributions to Mineralogy and Petrology 119:197–212.[Web of Science]
Ghiorso MS, Hirschmann MM, Reiners PW, Kress VC. (2002) The pMELTS: a revision of MELTS for improved calculation of phase relations and major element partitioning related to partial melting of the mantle to 3 GPa. Geochemistry, Geophysics, Geosystems 3:36 doi: 10.1029/2001GC000217.
Haase KM, Stoffers P, Garbe-Schönberg CD. (1997) The petrogenetic evolution of lavas from Easter Island and neighbouring seamounts, near-ridge hotspot volcanoes in the SE Pacific. Journal of Petrology 38:785–813.[CrossRef][Web of Science]
Harris C. (1983) The petrology of lavas and associated plutonic inclusions of Ascension Island. Journal of Petrology 24:424–470.[Web of Science]
Irvine TN and Baragar WRA. (1971) Guide to chemical classification of common volcanic rocks. Canadian Journal of Earth Sciences 8:523–548.
Kuntz MA, Covington HR, Schorr LJ. (1992) An overview of basaltic volcanism of the eastern Snake River plain, Idaho. In Link PK, Kuntz MA, Platt LB (Eds.). Regional Geology of Eastern Idaho and Western Wyoming, Geological Society of America, Memoir 179: pp. 227–267.
Leeman WP. (1982a) In Bonnichsen B and Breckenridge RM (Eds.). Rhyolites of the Snake River Plain—Yellowstone Plateau province, Idaho and Wyoming; a summary of petrogenetic models. Cenozoic Geology of Idaho. Bulletin—Idaho Bureau of Mines and Geology 26:203–212.
Leeman WP. (1982b) In Bonnichsen B and Breckenridge RM (Eds.). Olivine tholeiitic basalts of the Snake River Plain, Idaho. Cenozoic Geology of Idaho. Bulletin—Idaho Bureau of Mines and Geology 26:181–191.
Leeman WP and Vitaliano CJ. (1976) Petrology of McKinney Basalt, Snake River Plain, Idaho. Geological Society of America Bulletin 87:1777–1792.
Leeman WP, Vitaliano CJ, Prinz M. (1976) Evolved lavas from the Snake River Plain—Craters of the Moon National Monument, Idaho. Contributions to Mineralogy and Petrology 56:35–60.[CrossRef][Web of Science]
Leroex AP. (1985) Geochemistry, mineralogy and magmatic evolution of the basaltic and trachytic lavas from Gough Island, South Atlantic. Journal of Petrology 26:149–186.
Lindsley DH. (1983) Pyroxene thermometry. American Mineralogist 68:477–493.[Abstract]
Lindsley DH and Andersen DJ. (1983) Proceedings of the 13th Lunar and Planetary Science ConferenceA two-pyroxene thermometer (Lunar and Planetary InstituteIn Boynton WV and Ahrens TJ (Eds.). , Houston, TX) pp. A887–A906.
Litvin VY. (2002) Application of fractional crystallization to the origin of continental tholeiitic suites(State University of New York at Stony Brook, Department of Geosciences, Stony Brook, NY) pp. 138.
Macdonald R, Sparks RSJ, Sigurdsson H, Mattey DP, McGarvie DW, Smith RL. (1987) The 1875 eruption of Askja Volcano, Iceland—combined fractional crystallization and selective contamination in the generation of rhyolitic magma. Mineralogical Magazine 51:183–202.[Web of Science]
Mandeville CW, Webster JD, Rutherford MJ, Taylor BE, Timbal A, Faure K. (2002) Determination of molar absorptivities for infrared absorption bands of H2O in andesitic glasses. American Mineralogist 87:813–821.
McCurry M, Hackett WR, Hayden K. (1999) Cedar Butte and cogenetic Quaternary rhyolite domes of the Eastern Snake River Plain. In Hughes SS and Thackray GD (Eds.). Guidebook to the Geology of Eastern IdahoIdaho Museum of Natural History. Idaho: Pocatello pp. 169–179.
McCurry M, Hayden KP, Morse LH, Mertzman S. (2006) Genesis of post-hotspot A-type rhyolite of the Eastern Snake River Plain volcanic field by extreme fractional crystallization of olivine tholeiite basalt. Bulletin of Volcanology (in press).
Mitchell JN, Scoates JS, Frost CD. (1995) High-Al gabbros in the Laramie Anorthosite Complex, Wyoming—implications for the composition of melts parental to Proterozoic anorthosite. Contributions to Mineralogy and Petrology 119:166–180.[Web of Science]
Mitchell JN, Scoates JS, Frost CD, Kolker A. (1996) The geochemical evolution of anorthosite residual magmas in the Laramie Anorthosite Complex, Wyoming. Journal of Petrology 37:637–660.
Nekvasil H, Simon A, Lindsley DH. (2000) Crystal fractionation and the evolution of intra-plate hy-normative igneous suites: insights from their feldspars. Journal of Petrology 41:1743–1757.
Nekvasil H, Lindsley DH, Whitaker ML, Filiberto J, DiFrancesco NJ, Rossier L, Horn J. (2003) Tholeiites, anorthosites, potassic granites, sodic trachytes, and tephriphonolites: is there a link? Geological Society of America, Abstracts with Programs 35:395.
Nekvasil H, Dondolini A, Horn J, Filiberto J, Long H, Lindsley DH. (2004) The origin and evolution of silica-saturated alkalic suites: an experimental study. Journal of Petrology 45:693–721.
Newton RC, Charlu TV, Kleppa OJ. (1974) Calorimetric investigation of stability of anhydrous magnesium cordierite with application to granulite facies metamorphism. Contributions to Mineralogy and Petrology 44:295–311.[CrossRef][Web of Science]
Peng XH and Humphreys ED. (1998) Crustal velocity structure across the eastern Snake River Plain and the Yellowstone swell. Journal of Geophysical Research—Solid Earth 103:7171–7186.
Reed MF, Bartholomay RC, Hughes SS. (1997) Geochemistry and stratigraphic correlation of basalt lavas beneath the Idaho Chemical Processing Plant, Idaho National Engineering Laboratory. Environmental Geology 30:108–118.[CrossRef]
Scoates JS, Frost CD, Mitchell JN, Lindsley DH, Frost BR. (1996) Residual-liquid origin for a monzonitic intrusion in a mid-Proterozoic anorthosite complex: the Sybille intrusion, Laramie Anorthosite Complex, Wyoming. Geological Society of America Bulletin 108:1357–1371.
Scoates JS, Lindsley DH, van der Kolk D, Anderson K. (1999) Fractional crystallization experiments on a candidate parental magma to anorthosite. EOS Transactions, American Geophysical Union 80:F1096.
Sparlin MA, Braile LW, Smith RB. (1982) Crustal structure of the Eastern Snake River Plain determined from ray trace modeling of seismic refraction data. Journal of Geophysical Research 87:2619–2633.
Spulber SD and Rutherford MJ. (1983) The origin of rhyolite and plagiogranite in oceanic crust—an experimental study. Journal of Petrology 24:1–25.[Web of Science]
Stolz AJ. (1985) The role of fractional crystallization in the evolution of the Nandewar Volcano, northeastern New South Wales, Australia. Journal of Petrology 26:1002–1026.
Stout MZ and Nicholls J. (1977) Mineralogy and petrology of Quaternary lavas from the Snake River Plain, Idaho. Canadian Journal of Earth Sciences 14:2140–2156.
Stout MZ, Nicholls J, Kuntz MA. (1994) Petrological and mineralogical variations in 2500–2000 yr B.P. lava flows, Craters of the Moon lava field, Idaho. Journal of Petrology 35:1681–1715.
Takahashi E. (1978) Partitioning of Ni2+, Co2+, Fe2+, Mn2+ and Mg2+ between olivine and silicate melts—compositional dependence of partition coefficient. Geochimica et Cosmochimica Acta 42:1829–1844.[CrossRef][Web of Science]
Thompson RN. (1975) Primary basalts and magma genesis II. Snake River Plain, Idaho, USA. Contributions to Mineralogy and Petrology 52:213–232.[CrossRef][Web of Science]
Villiger S, Ulmer P, Muntener O, Thompson AB. (2004) The liquid line of descent of anhydrous, mantle-derived, tholeiitic liquids by fractional and equilibrium crystallization—an experimental study at 1· 0 GPa. Journal of Petrology 45:2369–2388.
Whitaker ML, Nekvasil H, Lindsley DH. (2004) Can fractionation of an olivine tholeiite give rise to potassic rhyolites? Geological Society of America, Abstracts with Programs 36:25.
Whitaker ML, Nekvasil H, Lindsley DH, McCurry M. (2006) Can crystallization of olivine tholeiite give rise to potassic rhyolites? An experimental investigation. Bulletin of Volcanology (in press).
Yoder HS and Tilley CE. (1962) Origin of basalt magmas: an experimental study of natural and synthetic rock systems. Journal of Petrology 3:342–532.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. E. Botcharnikov, R. R. Almeev, J. Koepke, and F. Holtz Phase Relations and Liquid Lines of Descent in Hydrous Ferrobasalt--Implications for the Skaergaard Intrusion and Columbia River Flood Basalts J. Petrology, September 1, 2008; 49(9): 1687 - 1727. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K.-N. Pang, M.-F. Zhou, D. Lindsley, D. Zhao, and J. Malpas Origin of Fe-Ti Oxide Ores in Mafic Intrusions: Evidence from the Panzhihua Intrusion, SW China J. Petrology, February 1, 2008; 49(2): 295 - 313. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. D. Putirka Thermometers and Barometers for Volcanic Systems Reviews in Mineralogy and Geochemistry, January 1, 2008; 69(1): 61 - 120. [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Whitaker, D. H. Lindsley, J. M. K. Whitaker, and H. Nekvasil Carbon is not required during crystallization to produce ferrobasalts/ferrodiorites (FTP rocks) American Mineralogist, October 1, 2007; 92(10): 1750 - 1755. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||