JpGU-AGU Joint Meeting 2020

Presentation information

[E] Poster

S (Solid Earth Sciences ) » S-MP Mineralogy & Petrology

[S-MP38] Deformed rocks, Metamorphic rocks and Tectonics

convener:Yoshihiro Nakamura(Geological Survey of Japan, National Institute of Advanced Industrial Science and Technology), Yumiko Harigane(Geological Survey of Japan, National Institute of Advanced Industrial Science and Technology (AIST))

[SMP38-P08] Multiple timings of high-grade garnet-forming metamorphism from Sør Rondane Mountains, East Antarctica

*Fumiko Higashino1, Tetsuo Kawakami2, Shuhei Sakata3, Takafumi Hirata3 (1.Faculty of Science, Okayama University of Science, 2.Kyoto University, 3.The University of Tokyo)

Keywords:REE, LA-ICPMS, high-temperature metamorphic rock

In high-grade metamorphic rocks, U-Pb geochronology of zircon is a powerful tool to understand metamorphic events. This is because zircon represents relatively high closure temperature of U-Pb system and its atomic structure remains stable over long periods of geological time. Since zircon could record various metamorphic stages of high-grade metamorphism (e.g., Harley et al., 2007), the appropriate interpretation of their ages is significant. The REE pattern of distribution coefficient between zircon and garnet (DREE(z/g)) has been used as a useful tool to establish the link between zircon geochronology and timing of garnet formation. A considerable issue is that the REE patterns of DREE(z/g) have been reported to become positive with decreasing temperature, increasing pressure, or increasing XCa in garnet (e.g., Harley et al., 2001; Rubatto, 2002; Hokada and Harley, 2004; Buick et al., 2006; Rubatto and Hermann, 2007). In spite of these uncertainty, traditional REE patterns of DREE(z/g) have been used to understand whether the garnet was formed in equilibrium with zircon or not. The typical way that experimental DREE data sets are used to evaluate empirical data, is to overlay them on natural data sets and gauge their similarity. The flat REE pattern of DREE(z/g) is often interpreted to represent equilibrium between these two phases.
In order to handle empirical data sets in a practical manner, Taylor et al. (2017) proposed to use array plot whose slope from DYb(z/g)/DGd(z/g) is plotted against DYb(z/g) values. An advantage of the new array plot is usages of two key parameters which together can represent the slope of the REE pattern and REE abundance of the traditional REE partitioning plot. It is noted that more than one garnet and zircon generation may be formed by recrystallization and a degree of chemical re-equilibration varies in terranes that have undergone polymetamorphism.
In the Sør Rondane Mountains (SRM), East Antarctica, Late Proterozoic to Cambrian granulites are widely exposed (e.g., Jacobs and Thomas, 2004; Shiraishi et al., 2008; Osanai et al., 2013; Kitano et al., 2016). In the SRM, each metamorphic event has been correlated to the different zircon age groups, mainly using Th/U ratio of the dated zircon domains (e.g., Shiraishi et al., 2008). Limited number of previous studies have investigated REE patterns of zircon and garnet in the SRM (e.g., Hokada et al., 2013). Therefore, in this study, in situ U-Pb dating of zircon included in garnet and the DREE(z/g) values are evaluated using the array plot, in order to determine the timing(s) of garnet-forming metamorphism in selected regions of the SRM.
In situ U-Pb dating of zircon and quantitative analysis of REE in garnet and zircon have been performed using LA-ICPMS on seven mafic and pelitic gneisses collected from Perlebandet, Brattnipene, Pilten, Mefjell, and Balchenfjella in the SRM. The rims of zircon inclusions in garnet from seven samples gave U-Pb ages of ~560 Ma (Pilten and Balchenfjella), ~580 Ma (Perlebandet and Mefjell) and ~610-620 Ma (Balchenfjella and Brattnipene). Each sample gave a single peak age distribution for the zircon rim dating. Using the traditional DREE(z/g) pattern plots and the array plot, five samples are evaluated to be in equilibrium between host garnet and the dated zircon domain. Zircon rims included in garnet from the other two samples, on the other hand, were not in equilibrium with the host garnet. In summary, the five equilibrium samples could reveal the multiple timing of garnet formation in different regions of the SRM; Perlebandet (~580 Ma), Brattnipene (~610-620 Ma), Pilten (~580 Ma), and Balchenfjella (~560 Ma and ~610 Ma). The details of garnet-forming metamorphism will be discussed.