Deciphering the source and contamination history of peraluminous magmas using δ¹⁸O of accessory minerals: examples from garnet-bearing plutons of the Sierra Nevada batholith

Abstract

Peraluminous granitoids provide critical insight as to the amount and kinds of supracrustal material recycled in the central Sierra Nevada batholith, California. Major element concentrations indicate Sierran peraluminous granitoids are high-SiO₂ (68.9–76.9) and slightly peraluminous (average molar Al₂O₃/(CaO + Na₂O + K₂O)=1.06). Both major and trace element trends mimic those of other high-silica Sierran plutons. Garnet (Grt) in the peraluminous plutons is almandine–spessartine-rich and of magmatic origin. Low grossular contents are consistent with shallow (<4 kbar) depths of garnet crystallization. Metasediments of the Kings Sequence commonly occur as wallrocks associated with the plutons, including biotite schists that are highly peraluminous (A/CNK=2.25) and have high whole rock (WR) δ¹⁸O values (9.6–21.8‰, average=14.5±2.9‰, n=26). Ultramafic wallrocks of the Kings–Kaweah ophiolite have lower average δ¹⁸O (7.1±1.3‰, n=9). The δ¹⁸O(WR) of the Kings Sequence is variable from west to east. Higher δ¹⁸O values occur in the west, where quartz in schists is derived from marine chert; values decrease eastward as the proportion of quartz from igneous and metamorphic sources increases. Peraluminous plutons have high δ¹⁸O(WR) values (9.5–13‰) consistent with supracrustal enrichment of their sources. However, relatively low initial ⁸⁷Sr/⁸⁶Sr values (0.705–0.708) indicate that the supracrustal component in the source of peraluminous magmas was dominantly altered ocean crust and/or greywacke. Also, plutons lack or have very low abundances (<1% of grains) of inherited zircon (Zrc) cores. Average δ¹⁸O(Zrc) is 7.9‰ in peraluminous plutons, a higher value than in coeval metaluminous plutons (6–7‰). Diorites associated with peraluminous plutons also have high δ¹⁸O(Zrc), 7.4–8.3‰, which is consistent with the diorites being derived from a similar source. Magmatic garnet has variable δ¹⁸O (6.6–10.5‰, avg.=7.9‰) due to complex contamination and crystallization histories, evidenced by multiple garnet populations in some rocks. Comparison of δ¹⁸O(Zrc) and δ¹⁸O(Grt) commonly reveals disequilibrium, which documents evolving magma composition. Minor (5–7%) contamination by high δ¹⁸O wallrocks occurred in the middle and upper crust in some cases, although low δ¹⁸O wallrock may have been a contaminant in one case. Overall, oxygen isotope analysis of minerals having slow oxygen diffusion and different times of crystallization (e.g., zircon and garnet), together with detailed textural analysis, can be used to monitor assimilation in peraluminous magmas. Moreover, oxygen isotope studies are a valuable way to identify magmatic versus xenocrystic minerals in igneous rocks.

Appendices

Appendix 1

Oxygen isotope fractionation factors

Comparison of the δ¹⁸O values of mineral pairs to equilibrium fractionation factors is used to test if minerals are in isotopic equilibrium. Isotopic fractionation factors are expressed as:

$$\Delta_{i-j} = \delta^{18}\hbox{O (mineral-i)}- \delta^{18}\hbox{O (mineral-j)} \approx 1000\,\hbox{ln}\,\alpha(i-j).$$

Equilibrium fractionation factors are calculated for a particular temperature using the expression:

$$\Delta\hbox{(mineral-i--mineral-j)} \approx {A}_{i-j}\times10^{6}/{T} ^{2} + {B}_{i-j}\times10^{3}/{T} + {C}_{i-j}.$$

In the expression, T is temperature (K) and A, B, and C are experimentally or empirically determined coefficients; if no coefficient is given for A, B, or C, then its value is 0.0. The following factors are used for oxygen isotopes: quartz–zircon A=2.64 (Valley et al. 2003); quartz–almandine A=2.71 (Valley et al. 2003); quartz–spessartine A=2.83 (Lichtenstein and Hoernes 1992); quartz–grossular A=3.03 (Matthews 1994); quartz–sillimanite A=2.25 (Sharp 1995); and quartz–H₂O A=2.51, C= −1.46 (Clayton et al. 1972). Combining the oxygen isotope equilibrium fractionation factors for quartz–almandine yields an almandine–aluminosilicate fractionation “A” factor of 0.46.

Garnet cation chemistry and δ¹⁸O

Garnet cation chemistry was compared to δ¹⁸O to determine if compositional effects correlate to the variations Δ¹⁸O of garnet and other minerals. For example, Δ¹⁸O(Qtz–Grt) can vary up to 0.8‰ in andradite and grossular-rich garnets at magmatic (>700°C) temperatures (Kohn and Valley 1998; Valley et al. 2003). Thus dependence of Δ¹⁸O on garnet composition must be considered in estimates of contamination. Grossular concentrations in SNB garnets are too low (X _grs<0.04; see Supplementary Data Table) for corresponding changes in Δ¹⁸O (0–0.01‰) to be detectable by laser fluorination. Magnesium content is likewise low and would not considerably affect Δ¹⁸O of garnet and other minerals. Because Mn and Fe content varies considerably in the garnets studied, the effect of these cations on δ¹⁸O was evaluated. The equilibration fractionation factors for Δ¹⁸O(Qtz–Grt) of spessartine (Lichtenstein and Hoernes 1992) and almandine (Valley et al. 2003) are very close, therefore the effect of large compositional differences on Δ¹⁸O(Qtz–Grt) is small and garnet chemistry doesn’t markedly affect Δ¹⁸O in these rocks. For instance, at magmatic temperatures, the calculated Δ¹⁸O(Qtz–Grt) for the most almandine- and spessartine-rich garnets studied (1S38, X _Alm=0.87, X _Sps=0.09 and 1S128, X _Alm=0.39, X _Sps=0.55) shifts only 0.05‰. The above calculation of sensitivity to Mn and Fe content in garnet assumes a magmatic temperature and therefore the correlation of high-Mn, low-δ¹⁸O garnets may result from Mn stabilizing garnet at lower temperatures (Green 1977; Miller and Stoddard 1981). Reverse Mn zoning in some garnet phenocrysts suggests this process (e.g., Fig. 7b).