Elemental mixing systematics and Sr–Nd isotope geochemistry of mélange formation: Obstacles to identification of fluid sources to arc volcanics

https://doi.org/10.1016/j.epsl.2006.03.053Get rights and content

Abstract

We present major and trace element concentrations in conjunction with Sr–Nd isotope ratios to investigate the geochemical characteristics of mélange formation along the subduction zone slab–mantle interface. Mélange matrix of the Catalina Schist formed within an active subduction zone of the southern California borderland in Cretaceous time. Mélange formed through the synergistic effects of deformation and metasomatic fluid flow affecting peridotite, basaltic, and sedimentary protoliths to form hybridized bulk compositions not typical of seafloor “input” lithologies. In general, all elemental concentrations primarily reflect mechanical mixing processes, while fluid flow mediates all elemental systematics to a varying extent that is largely a function of inferred “mobility” for a particular element or the stability of suitable mineral hosts. Elemental data reveal that mineral stabilities defined by the evolution of bulk composition within mélange zones are probably the most important control of solid, liquid, or fluid geochemistry within the subduction system. Sr–Nd isotope ratios are highly variable and reflect contributions of mélange protoliths to varying extents. A weak mechanical mixing array present in Sr isotope data is strongly overprinted by a fluid signal that dominates mélange Sr systematics. Nd isotope data suggest that Nd is more conservative during metamorphism and is largely controlled by mechanical mixing. We argue that mélange formation is an intrinsic process to all subduction zones and that the geochemistry of mélange will impart the strongest control on the geochemistry of metasomatic agents (hydrous fluids, silicate melts, or miscible supercritical liquids) progressing to arc magmatic source regions in the mantle wedge. Mélange formation processes suggest that comparisons of subduction “inputs” to arc volcanic “outputs” as a means to infer recycling at subduction zones dangerously over-simplify the physics of the mass transfer in subduction zones, as subducted mass is consistently redistributed into novel bulk compositions. Such mélange zones along the slab–mantle interface simultaneously bear characteristic elemental or isotopic signals of several distinct input lithologies, while experiencing phase equilibria not typical of any input. We recommend that future studies explore the phase equilibria of hybridized systems and mineral trace element residency, as these processes provide for a physical baseline from which it will be possible to follow the path of subducted mass through the system.

Introduction

Analyses of arc volcanic rocks have long recognized the importance of metamorphic reactions and processes to arc magmatic petrogenesis and recycling to the mantle. However, despite igneous constraints, linking exact metamorphic processes to the development of distinct magmatic geochemical compositions has been ambiguous, especially since the geochemistry of metamorphic rocks beneath arcs is a combination of conjecture based on pre-subduction seafloor compositions (i.e. [1], [2], [3]) and the comparatively limited data for subduction-zone metamorphic rocks [4], [5], [6], [7], [8], [9], [10], [11], [12], [13]. Even more problematic is that data do not exist for the entire metamorphic suite produced within subduction zones, so our appreciation of the full spectrum of petrologic processes and their geochemical effects is incomplete. Most conspicuous among the missing data is a description of the geochemical ramifications of mélange formation.

Tectonic mélange is one of the most characteristic by-products of the subduction cycle, occurring in most orogenic belts. Styles and petrologic settings of mélange can range from highly deformed metasedimentary collages typical of incipient metamorphism within accretionary prisms (e.g. [14]) to chaotic hybridized mixtures of peridotite, basalt, and sediment produced at blueschist-, amphibolite-, or eclogite-facies conditions in forearc to sub-arc regions (e.g. [5], [6], [7], [15]). In all instances, mélange formation appears intimately linked to reactive fluid flow, indicating metasomatic processes are important in addition to mechanical deformation.

The extent and volumetric significance of mélange formation within deeper portions of subduction zones are debated, although anomalous geophysical results indicating deep subduction of hydrated material can be interpreted as measurement of mélange. Abers [16] reported seismic evidence for 1–7 km thick zones of anomalously slow material, interpreted by Abers [16] as hydrated oceanic crust along the surface of the subducting slab, in subduction zones worldwide. An intriguing alternative interpretation is that these seismically slow waveguides represent hydrated mélange zones dominated by chlorite + talc [7], a high-variance assemblage that will persist to the appropriate ∼ 250 km depths [4], [17]. This alternative interpretation suggests the possibility that mélange is an intrinsic feature of the slab–mantle interface and may play an important role in buffering fluid or melt compositions ultimately transferred to arc magma source regions in the mantle wedge.

Here, we explore the petrologic and geochemical diversity of mélange formed during active subduction in the Catalina Schist subduction complex exposed on Santa Catalina Island, CA. We report major element, trace element, and Sr–Nd isotope geochemistry for a large number of mélange samples spanning ranges in bulk composition and metamorphic grade. Our current dataset expands upon the more limited dataset of Bebout and Barton [7] for the Catalina Schist amphibolite-facies mélange by including a samples recording lawsonite–albite and lawsonite blueschist metamorphic conditions reminiscent of existing models for subduction-zone geothermal gradients. The focus of the current contribution is an investigation of geochemical effects stemming from mélange formation, as opposed to the petrogenetic modeling of Bebout and Barton [7].

Section snippets

Samples and analytical methods

The Catalina Schist has well-documented ranges in both bulk composition and metamorphic grade [4], [5], [6], [7], [18], [19]. Due to these factors and the expected variability stemming from mélange formation, we comprehensively analyzed a large population of mélange matrix samples formed at three metamorphic grades for major and trace elements (n = 58), while a representative subset were analyzed for Sr–Nd isotope ratios (n = 29). Samples of mélange were collected from zones of lawsonite–albite

Phenomenological model of mélange formation

Previous studies of the Catalina Schist have developed a model to explain the formation of mélange via the synergistic effects of fluid flow, metamorphism, and deformation occurring amongst the more “coherent” sedimentary, basaltic, and ultramafic blocks in the subduction complex; we briefly review this model here. Within the Catalina Schist, abundant subduction related fluid flow is evident, with the greatest extents of fluid–rock interaction and stable isotope homogenization occurring along

Major element geochemistry

Major element compositions for mélange matrix of the Catalina Schist reflect protolith contributions inferred from field relations. Variations for selected major elements, plotted as a function of MgO (Fig. 2), indicate fundamental distinctions between the dominantly binary ultramafic–mafic AM-facies mélange mixture and ternary protolith mixtures expected for LB- and LA-facies mélange. We choose to present major element variations as a function of MgO in that the predominance of an ultramafic

Sr–Nd isotope geochemistry

In this contribution we report Sr–Nd isotope ratios as age-corrected initial ratios based upon existing white mica Ar–Ar closure ages [18]. Unique ages for each sample are not available, so we have chosen to present maximum and minimum age corrections based on the range in Ar–Ar ages for each unit [18]. The range in age corrections spans 15 Ma in all cases.

When viewed on a εNd vs. 87Sr/86Sr isotope variation diagram, no firm array in the mélange data is evident (Fig. 7). However, the isotope

Mélange and the slab–mantle interface

Prior to any discussion of the geochemistry of the mélange zones of the Catalina Schist, it is worthwhile to briefly discuss why mélange will form in all subduction zones, and the physical role these mélange zones will have in the overall subduction system. Mélange must intrinsically form along the slab–mantle interface to form the chemical bridge between the depleted peridotites of the mantle wedge and the evolved components of the subducting slab. Across this interface, severe contrasts exist

The role of mélange in the subduction system

The dominant approach many utilize to resolve geochemical recycling during subduction is a comparison of subduction inputs vs. subduction outputs, usually to resolve what subducted lithologies are important contributors to the arc magmatic system or to constrain what proportion of subducted material is returned to the deep mantle (e.g. [1], [2], [3], [38], [56], [57], [58], [59], [60], [61]). Many of these studies are limited by our inadequate understanding of the physical processes

Conclusions

We have investigated the geochemical ramifications of subduction-zone mélange formation as recorded by the mélange matrix of the Catalina Schist, CA. The bulk compositions of these samples are consistent with their demonstrated petrogenesis and similar mélange zones are probably ubiquitous to subduction zones worldwide. Our major conclusions are:

  • 1.

    Mélange formation results in hybridized rock types that are not representative of the incoming subduction section, yet mélange commonly retains

Acknowledgments

We thank the many people who have participated in lively discussion of our data at meetings over the past three years. RLK would like to thank C. Sakaguchi for patience and guidance during TIMS analyses, all the members of the PML for constant support and criticism, and M. Walter for rational discussions of geochemistry and petrogenesis. Formal reviews by Jay Ague, Terry Plank, and the editorial handling of Ken Farley improved the manuscript, although they do not agree with all of our

References (64)

  • S.S. Sorensen et al.

    Enrichment of trace elements in garnet amphibolites from a paleo-subduction zone: Catalina Schist, southern California

    Geochim. Cosmochim. Acta

    (1989)
  • T. Yokoyama et al.

    Evaluation of the coprecipitation of incompatible trace elements with fluoride during silicate rock dissolution by acid digestion

    Chem. Geol.

    (1999)
  • A. Makishima et al.

    Precise measurement of cerium isotope composition in rock samples

    Chem. Geol.

    (1991)
  • S. Weyer et al.

    Nb/Ta, Zr/Hf, and REE in the depleted mantle: implications for the differentiation history of the crust–mantle system

    Earth Planet. Sci. Lett.

    (2003)
  • R.K. Workman et al.

    Major and trace element composition of the depleted MORB mantle (DMM)

    Earth Planet. Sci. Lett.

    (2005)
  • S.M. Peacock

    Large-scale hydration of the lithosphere above subducting slabs

    Chem. Geol.

    (1993)
  • T. John et al.

    Trace element fractionation during fluid-induced eclogitization in a subducting slab: trace element and Lu–Hf–Sm–Nd isotope systematics

    Earth Planet. Sci. Lett.

    (2004)
  • P.D. Noll et al.

    The role of hydrothermal fluids in the production of subduction zone magmas: evidence from siderophile and chalcophile trace elements and boron

    Geochim. Cosmochim. Acta

    (1996)
  • F. Chalot-Prat et al.

    No significant element transfer from the oceanic plate to the mantle wedge during subduction and exhumation of the Tethys lithosphere (western Alps)

    Lithos

    (2003)
  • D. Rumble et al.

    The Qinglongshan oxygen and hydrogen isotope anomaly near Donghai in Jiangsu Province, China

    Geochim. Cosmochim. Acta

    (1998)
  • A.C. Barnicoat et al.

    Focused fluid flow during subduction: oxygen isotope data from high-pressure ophiolites of the western Alps

    Earth Planet. Sci. Lett.

    (1995)
  • J.A. Miller et al.

    An O-isotope profile through the HP-LT Corsican ophiolite, France and its implications for fluid flow during subduction

    Chem. Geol.

    (2001)
  • G.E. Bebout

    The impact of subduction-zone metamorphism on mantle–ocean chemical cycling

    Chem. Geol.

    (1995)
  • T.P. Fischer et al.

    Fluxes and sources of volatiles discharged from Kudryavy, a subduction zone volcano, Kurile Islands

    Earth Planet. Sci. Lett.

    (1998)
  • R. George et al.

    Pressure–temperature–time paths of sediment recycling beneath the Tonga–Kermadec arc

    Earth Planet. Sci. Lett.

    (2005)
  • J.D. Woodhead et al.

    Hafnium isotope evidence for ‘conservative’ element mobility during subduction zone processes

    Earth Planet. Sci. Lett.

    (2001)
  • C. Münker et al.

    Behaviour of high field strength elements in subduction zones: constraints from Kamchatka–Aleutian arc lavas

    Earth Planet. Sci. Lett.

    (2004)
  • C.E. Manning

    The chemistry of subduction-zone fluids

    Earth Planet. Sci. Lett.

    (2004)
  • K.A. Kelley et al.

    Composition of altered oceanic crust at ODP sites 801 and 1149

    Geochem. Geophys. Geosyst.

    (2003)
  • G.E. Bebout

    Field-based evidence for devolatilization in subduction zones: implications for arc magmatism

    Science

    (1991)
  • G.E. Bebout et al.

    Fluid flow and metasomatism in a subduction zone hydrothermal system, Catalina Schist terrane, California

    Geology

    (1989)
  • C. Spandler et al.

    Redistribution of trace elements during prograde metamorphism from lawsonite blueschist to eclogite facies; implications for deep subduction-zone processes

    Contrib. Mineral. Petrol.

    (2003)
  • Cited by (83)

    View all citing articles on Scopus
    View full text