Elsevier

Gondwana Research

Volume 98, October 2021, Pages 312-316
Gondwana Research

Comment on “Petrology, geochemistry and Sm-Nd systematics of the Paleoproterozoic Itaguara retroeclogite from Sao Francisco/Congo Craton: One of the oldest records of the modern-style plate tectonics”

https://doi.org/10.1016/j.gr.2020.12.007Get rights and content

Abstract

The recent paper by Chaves and Porcher published in Gondwana Research (2020, v.87, 224–237) asserts the presence of 2.2 Ga eclogite facies rocks in the Itaguara Sequence, São Francisco craton (SFC), Brazil. They present images in which the diagnostic high-pressure minerals either occur on cracks (rutile), or as fragmentary matrix grains with unclear mineral relationships (omphacite and phengite). Here I discuss the sample petrology in order to identify the equilibrium assemblage, and utilise a P-T phase modelling approach, integrating the major mineral compositions, to show that these rocks are not retrogressed eclogites.

The modelling presented here indicates that peak conditions were likely in the amphibolite facies, reaching maximum pressures of no more than 8 kbar and peak temperatures of ca. 700 °C (at 5 kbar). This result is consistent with that found by Massonne (2020, JSAMES, v.99, 102495) who suggested that mica schists from Itaguara reached maximum pressures of 13–14 kbar, and possibly lower than 10 kbar at 600 °C (i.e. within the medium to high-pressure amphibolite facies). This work shows that declarations of eclogite facies conditions, particularly from Paleoproterozoic terrains, should be accompanied by carefully evaluated petrography and thermodynamic investigations based on minerals (or mineral replacement textures) that indicate equilibrium relationships.

Introduction

P-T pseudosections are a forward modelling approach that shows every possible equilibrium assemblage that a rock composition may experience (Holland and Powell, 1998; Powell et al., 1998). This is incredibly useful, particularly when combined with inverse modelling because it is possible to use the P-T pseudosection to interpret the change in observed mineral assemblages or locate intersections of observed equilibrium mineral compositions (i.e. garnet), and define a P-T evolution, or P-T-t evolution if the timing of metamorphic mineral growth can be constrained (e.g. Mottram et al., 2015; Lanari and Engi, 2017; Williams et al., 2017). The method is not without complications– determining the effective equilibrium volume (or reactive bulk composition) of a specific sample can be difficult for a multitude of reasons (i.e. unknown H2O and Fe3+ content, melt loss, porphyroblast growth, local domain reactions, etc.; Rebay et al., 2010; Lanari and Engi, 2017; Tedeschi et al., 2017). This problem can be addressed by creating multiple diagrams investigating the effect of varying compositional components (T-X or P-X sections). This has been also successfully circumvented with thermodynamic calculations of equilibrium based on iterative thermodynamic models allowing the reactive bulk composition of a sample to also be modelled (in terms of fit) in relation to observed mineral assemblages, modes and compositions (Bingo-Antidote; Duesterhoeft and Lanari, 2020). P-T phase modelling relies upon, and is only as good as the sample petrography.

Conventional thermobarometry presumes equilibrium of two (or more) mineral compositions and determines what temperature or pressure these must have formed at. For this assumption to work, the textural relationships of the rock must be well known. However, metamorphic rocks experience a variety of P-T conditions on the path to peak conditions and also on the way back, and different minerals will be reactive to different parts of the P-T evolution. Unless every mineral is covered with compositional analyses and thousands of pressure and temperature estimates are calculated via a variety of thermobarometers (this can be done – see Lanari et al., 2013, Lanari et al., 2014), it is extremely unlikely that a selection of several points in different minerals represents equilibrium.

The objective of this comment is to re-evaluate the P-T conditions of the garnet amphibolite from Itaguara based on observable equilibrium relations and P-T phase diagram modelling. I show that the use of conventional thermobarometry and P-T modelling by Chaves and Porcher (2020) is based on flawed interpretation of the sample petrography and phase diagram (see Table 1). I have used their bulk and mineral compositions to reproduce P-T constraints that are more consistent with the observed mineral relationships and compositions. A detailed description of the methods used and additional diagrams referred to in the text are included in the supplementary material.

Section snippets

Petrography

Based on the images and descriptions presented by Chaves and Porcher (2020), it seems that the Itaguara mafic rocks have experienced two different P-T events. The earlier is responsible for the coarse-grained mineral assemblage including garnet, amphibole, ilmenite, quartz, biotite and relatively calcic plagioclase (XAb = 31–37; see Table 3 of Chaves and Porcher, 2020). Garnet (XAlm = 65–61; XGrs = 20–17; XPyr = 15–12; XSps = 5–4; Table 1 of Chaves and Porcher, 2020), amphibole, ilmenite,

P-T conditions

Additional modelling was done to explore the effect of compositional variation on the modelled P-T results using T-M and P-M sections. For the P-T evolution described above, the XPyr end-member was not included because the XPyr found in the sample plots above 20 kbar on the phase diagram (see supplementary Figs. 1 and 4). There are many reasons why this can happen, one is of course that the bulk rock composition is not appropriate (see below), another possibility is that the garnet selected

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

Ian Buick, Hugo Moreira and Mahyra Tedeschi are acknowledged for discussion and comments that greatly improved the manuscript. The author acknowledges funding from Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ) under the scheme Pós doc Nota 10.

References (30)

  • E.K. Ravna

    Distribution of Fe2+ and Mg between coexisting garnet and hornblende in synthetic and natural systems: an empirical calibration of the garnet–hornblende Fe–Mg geothermometer

    Lithos

    (2000)
  • M. Tedeschi et al.

    Reconstruction of multiple P-T-t stages from retrogressed mafic rocks: Subduction versus collision in the Southern Brasilian orogen (SE Brazil)

    Lithos

    (2017)
  • M. Brown et al.

    Metamorphism and the evolution of subduction on Earth

    Am. Mineral.

    (2019)
  • M.J. Caddick et al.

    Preservation of garnet growth zoning and the duration of prograde metamorphism

    J. Petrol.

    (2010)
  • A.O. Chaves et al.

    Petrology, geochemistry and Sm-Nd systematics of the Paleoproterozoic Itaguara retroeclogite from São Francisco/Congo Craton: one of the oldest records of the modern-style plate tectonics

    Gondwana Res.

    (2020)
  • Cited by (2)

    View full text