Elsevier

Chemical Geology

Volume 448, 5 January 2017, Pages 43-70
Chemical Geology

Invited research article
Primitive arc magma diversity: New geochemical insights in the Cascade Arc

https://doi.org/10.1016/j.chemgeo.2016.11.006Get rights and content

Highlights

  • High-precision geochemical data for 53 Cascade Arc primitive magmas

  • Calc-alkaline & tholeiitic groups (CABs, HAOTs) define common isotopic trends

  • Isotopic end-members are MORB-type depleted background mantle and subducting sediment

  • Additional oceanic crust input in CABs

  • Intraplate basalts (Garibaldi Belt, Adams-Simcoe) tap isotopically distinct enriched mantle

Abstract

The origin of the large range of compositional diversity encompassed by primitive arc magmas and the respective roles of mantle heterogeneity versus slab-derived contributions are the center of ongoing debate. The Cascade Arc of western North America is a global endmember hot subduction zone and an ideal setting in which to examine fundamental questions about the origin of arc magmas. Cascade Arc primitive magmas define several distinct compositional endmembers, the most widespread of which are high alumina olivine tholeiite (HAOT), calc-alkaline basalt (CAB), and intraplate-type basalt (IPB). New high precision Sr-Nd-Hf-Pb isotopic and trace element data for 49 of the most primitive magmas from seven major High Cascades volcanic centers are used to assess whether the basalt groups are derived from geochemically distinct mantle sources. Along with recently published data for the Garibaldi Volcanic Belt (GVB), the northern and other major segment of the Cascade Arc, and Mt. Rainier, the new data expand the high-precision isotopic coverage to seventeen volcanic centers spanning the ~ 1300 km length of the arc. In all investigated isotopic systems, the new High Cascades data allow for finer resolution with much less scatter than previous data. HAOT mantle sources are systematically more depleted in incompatible elements than CAB sources, yet High Cascades CABs and HAOTs are isotopically indistinguishable, defining a remarkable linear Pb isotopic array consistent with a single mantle endmember that is similar to the source of Juan de Fuca mid-ocean ridge basalts. This mantle has been variably modified by contributions from a single subducting sediment endmember that is a close isotopic match to average Northern Cascadia basin sediment. Although Astoria Fan sediment was previously used as a proxy for the subducting sediment composition in the Cascadia subduction zone, the isotope ratios of Cascades basalts record no input from Astoria Fan sediment. High Cascades CABs and HAOTs both contain subducting sediment input in the form of a melt, but CABs also contain melts of subducting oceanic crust which is most apparent in the southern Cascades high-Sr/P suite and in the GVB. The least sediment input is recorded by GVB basalts, which also become progressively lower in 208Pb*/206Pb* and Hf isotope ratios to the north, reflecting the influx of an isotopically distinct enriched mantle component that generates IPBs at the northern slab edge. Enriched mantle with a composition trending towards HIMU is also recorded by a distinct isotopic array that is defined by IPBs from the Mt. Adams-Simcoe back arc region. We propose that slab gaps permit the influx of sub-slab asthenospheric mantle in both the northern GVB and Mt. Adams-Simcoe back-arc. Otherwise, western North America is underlain predominantly by isotopically homogeneous, MORB-type depleted mantle from which both the CAB and HAOT basalt groups are derived. Our results from the Cascades imply that most of the compositional heterogeneity encompassed by primitive arc magmas is not related to the presence of multiple mantle components but to variability in the composition and quantity of slab contributions. This study highlights the importance of high precision Sr-Nd-Hf-Pb isotope and trace element data in obtaining new insights on the petrogenesis of arc magmas.

Introduction

Primary magmas in subduction settings encompass a wide range of compositional variability on local to global scales (e.g., Kelemen et al., 2014, Arculus, 1994). Deciphering the factors responsible for this diversity is difficult because these magmas are likely to contain contributions from a compositionally heterogeneous array of fluids and/or melts derived from several reservoirs in the subducting plate during prograde metamorphism. These reservoirs include altered oceanic crust, overlying sediment, and underlying hydrated mantle lithosphere (e.g., Schmidt and Poli, 2014). Pre-existing compositional variabilities within the mantle wedge are an additional potential source of diversity among primary arc magmas. Many studies concur that the sub-arc mantle primarily consists of depleted mantle that is similar in composition to the sources of mid-ocean ridge basalts (MORBs) (e.g., Green, 1976, Pearce and Peate, 1995, Perfit et al., 1980). However, enriched ocean island basalt (OIB) source mantle has been proposed to play an important role in controlling magma compositions in some arcs (Central America: Reagan and Gill, 1989; Mexico: Petrone et al., 2003; Sulu: Macpherson et al., 2010, Castillo et al., 2007; Izu-Bonin: Hochstaedter et al., 2000; Marianas: Woodhead et al., 2012; Aleutians: Morris and Hart, 1983; Fiji: Gill, 1984, Ellam and Hawkesworth, 1988).

The Cascade Arc of western North America offers a particularly interesting case in which to examine the factors contributing to arc magma compositional diversity because several genetically distinct primitive magma lineages have been identified on the basis of major and trace elements. The most widespread of these suites are referred to in this paper as high alumina olivine tholeiite (HAOT), calc-alkaline basalt (CAB), and intraplate-type basalt (IPB). Previous studies have concluded that these lineages reflect a sub-arc mantle that is composed of geochemically and mineralogically distinct domains of lithospheric and asthenospheric mantle, ranging in composition from MORB- to OIB-source-type, that have been modified by a spectrum of slab-derived components on multiple occasions (Righter, 2000, Hildreth, 2007, Bacon et al., 1997, Leeman et al., 1990, Leeman et al., 2005, Borg et al., 1997, Borg et al., 2002, Conrey et al., 1997, Grove et al., 2002, Schmidt et al., 2008, Green and Harry, 1999, Hughes, 1990, Baker et al., 1994). The Cascades is also considered an endmember ‘hot’ subduction zone, and the relatively high temperatures predicted for the subducting Juan de Fuca plate (Syracuse et al., 2010, Wada and Wang, 2009) imply nearly complete dehydration of the slab at sub-arc depths (e.g., van Keken et al., 2011, Hacker et al., 2003, Harry and Green, 1999). In turn, several workers have proposed that some Cascade Arc magmas may be little-influenced by modern slab-derived components (e.g., Leeman et al., 1990, Leeman et al., 2004, Leeman et al., 2005, Conrey et al., 1997, Borg et al., 1997, Borg et al., 2002). In this study, we use new high-precision Sr-Nd-Hf-Pb isotope and trace element data to test the hypothesis that the major endmember primary magma suites in the Cascade Arc represent isotopically distinct mantle sources on a regional scale. The high precision geochemical data that have recently become available for subducting sediment on the Juan de Fuca plate (Carpentier et al., 2013, Carpentier et al., 2014, Prytulak et al., 2006) and oceanic crust of the Explorer plate (Cousens et al., in review) further allow us to ascertain which slab-derived components have contributed to Cascade Arc magmas.

Radiogenic isotope ratios can be effective tracers of potential mantle heterogeneities and the fate of slab-derived materials, particularly when measured with modern analytical techniques that offer higher accuracy and precision. For example, multicollector inductively coupled plasma mass spectrometry (MC-ICP-MS) coupled with advances in sample preparation techniques have led to significant improvements over TIMS (thermal ionization mass spectrometry) in accuracy and precision, particularly in the case of lead isotopes, and to the routine analysis of Hf isotopic ratios (Albarède et al., 2004, Blichert-Toft et al., 1997, Weis et al., 2006). In the Garibaldi Volcanic Belt (GVB), the northern segment of the Cascade Arc (Fig. 1), high-precision Sr-Nd-Hf-Pb isotopic studies have demonstrated that IPBs and CABs tap two geochemically distinct mantle sources (Mullen and Weis, 2013, Mullen and Weis, 2015). In the main segment of the arc, the High Cascades (Fig. 1), numerous Sr-Nd-Pb radiogenic isotope studies have been conducted over the past 4 + decades. However, it has not been possible to conclusively resolve whether the Cascade primary magma groups originate from isotopically distinct mantle domains on a whole-arc scale because few of the data are associated with the requisite precision. Published high-precision Pb isotope data are limited to the Mt. Rainier area (Sisson et al., 2014), and Hf isotope data, which can be particularly useful in arc settings because Hf is mobilized to a lesser extent in slab-derived fluids than are Sr, Nd and Pb (Woodhead et al., 2012), are available only for Lassen Peak and Mt. Adams (Borg et al., 2002, Jicha et al., 2009a).

We have measured new high-precision Sr-Nd-Hf-Pb isotopic and trace element data for magmas from seven High Cascades volcanic fields. The samples analyzed from Lassen Peak, Medicine Lake, Crater Lake, Mt. Shasta, Mt. Adams, and Simcoe (Fig. 1) were identified by Bacon et al. (1997) as representative of the most primitive magmas in the High Cascades. We also analyzed a suite of the distinctive North Sister-type basaltic andesites from the Three Sisters region (Fig. 1) to examine the isotopic effects of crustal contamination because these magmas record pervasive assimilation of mafic deep crust (Schmidt and Grunder, 2011, Schmidt et al., 2013). We also present several new analyses of GVB magmas to augment the existing high-precision isotopic database (Mullen and Weis, 2013, Mullen and Weis, 2015, Mullen and McCallum, 2014) and to strengthen our comparison between the GVB and High Cascades segments. The new GVB dataset includes the first published geochemical data for the Silverthrone volcanic field, which was proposed to be the northernmost volcano in the Cascades by Green et al. (1988) but has been excluded from the arc by others (e.g., Hildreth, 2007).

Our new isotopic data differ significantly from previously published data on the same samples, particularly in Pb isotopes, which are associated with a ten-fold increase in analytical precision. Most importantly, the new data show that all CABs and HAOTs from the High Cascades are derived from a single isotopic mantle endmember that has been variably modified by subduction components. Factors other than mantle heterogeneity must be responsible for the differences between HAOTs and CABs. In contrast, IPBs sample isotopically distinct mantle components.

Section snippets

Cascadia subduction system: High Cascades and Garibaldi Volcanic Belt

The active Cascade Arc extends ~ 1300 km between southwestern British Columbia and northern California (Fig. 1). Magmatism initiated at ~ 4 Ma (du Bray and John, 2011) and is a consequence of northeasterly subduction of the Juan de Fuca plate system beneath the North American plate. Rates of convergence increase northward along the trench from 30 to 45 mm/year (Wilson, 2002). The southern portion of the Juan de Fuca plate, known as the Gorda sub-plate, has been internally deforming for the past ~ 3 Ma

Cascade Arc primitive magmas

Primary partial melts of the mantle, which have undergone no differentiation since separating from their sources, would ideally be used to investigate mantle sources but such melts are rare, especially in arc settings. The magmas most closely approaching primary compositions (i.e., most magnesian) must therefore be used instead, and we refer here to such magmas as “primitive” (e.g., Kelemen et al., 2014). Previous studies have established that the most primitive magmas in the Cascades have Mg#

Samples analyzed

In an effort to evaluate the numerous hypotheses proposed for the origin of Cascade primitive magma diversity, we analyzed 53 samples from 11 Quaternary volcanic centers for Sr-Nd-Hf-Pb isotope ratios and trace elements. Although the samples are assigned the name of the nearest major composite volcano, most of the magmas were erupted from peripheral vents as is typical for basalts in the Cascades (Hildreth, 2007). Thirty-six of the samples are re-analyzed splits of the same sample powders used

Analytical techniques

Sr-Nd-Hf-Pb isotopic and trace element analyses were conducted at the Pacific Centre for Isotope and Geochemical Research (PCIGR) at The University of British Columbia. All chemical separations and mass spectrometric analyses were carried out in Class 100 and 10,000 clean laboratories. Sample powders (~ 0.1 g) were digested in 15 mL screw-top Savillex beakers on a hotplate at 120–130 °C for 2–3 days in a 10:1 mixture of concentrated sub-boiled HF and HNO3, with periodic ultrasonication. Samples were

Trace elements

Trace element data are listed in Table 1, Table 2. The data are compared to previous analyses of the same samples by Bacon et al. (1997) in Fig. A1 and Table A4. Most REEs are within 10% of the previous values that were measured by isotope dilution (Fig. A1; Table A4). Pb concentrations are uniformly higher in the new dataset, although Bacon et al. previously analyzed Pb by isotope dilution. For elements previously measured by XRF, ICP-AES, or INAA, the new data are typically within ± 30%, with

New vs. previous isotope data

The significant difference between our new Pb isotope ratios and those previously published is attributable to the nearly 10 times better reproducibility of MC-ICP-MS analyses as compared to the previous TIMS measurements (~ 0.1% external 2σ reproducibility), as well as the employment of acid leaching in this study. In the case of Nd isotopes, although MC-ICP-MS and TIMS data can, in principle, be associated with similar external reproducibilities, our new MC-ICP-MS data are outside the 2σ

Conclusions

New high precision Pb-Sr-Nd-Hf isotope and trace element data for magmas from seven volcanic centers in the High Cascade segment of the Cascade Arc, coupled with new and published high precision isotopic data from the GVB segment, permit a re-evaluation of models proposed for the petrogenesis of compositionally diverse Cascade Arc primitive magmas from three endmember groups, CABs, HAOTs, and IPBs. The most widespread basalts in the High Cascades, CABs and HAOTs, are distinct in major and trace

Acknowledgements

We thank T. Horscroft for encouraging us to submit this invited paper and C. Chauvel for efficient editorial handling. The comments of G. Jacques and an anonymous reviewer improved the presentation and helped us to clarify our scientific arguments. C. Bacon, A. Grunder, G. Woodsworth, N. Vigouroux, and P. Adam are acknowledged for generously providing the samples from the High Cascades, North Sister, Silverthrone, Mt. Garibaldi, and the Bridge River Cones and Elaho Valley areas, respectively.

References (138)

  • S.M. Eggins et al.

    A simple method for the precise determination of ≥ 40 trace elements in geological samples by ICPMS using enriched isotope internal standardisation

    Chem. Geol.

    (1997)
  • H. Gao et al.

    Crust and lithosphere structure of the northwestern US with ambient noise tomography: terrane accretion and Cascade arc development

    Earth Planet. Sci. Lett.

    (2011)
  • J.B. Gill

    Sr Pb Nd isotopic evidence that both MORB and OIB sources contribute to oceanic island arc magmas in Fiji

    Earth Planet. Sci. Lett.

    (1984)
  • D.W. Graham et al.

    Mantle source provinces beneath the northwestern USA delimited by helium isotopes in young basalts

    J. Volcanol. Geotherm. Res.

    (2009)
  • N.L. Green

    Influence of slab thermal structure on basalt source regions and melting conditions: REE and HFSE constraints from the Garibaldi volcanic belt, northern Cascadia subduction system

    Lithos

    (2006)
  • N.L. Green et al.

    On the relationship between subducted slab age and arc basalt petrogenesis, Cascadia subduction system, North America

    Earth Planet. Sci. Lett.

    (1999)
  • T.L. Grove et al.

    The influence of H2O on mantle wedge melting

    Earth Planet. Sci. Lett.

    (2006)
  • D.L. Harry et al.

    Slab dehydration and basalt petrogenesis in subduction systems involving very young oceanic lithosphere

    Chem. Geol.

    (1999)
  • W.K. Hart

    Chemical and isotopic evidence for mixing between depleted and enriched mantle, northwestern U.S.A.

    Geochim. Cosmochim. Acta

    (1985)
  • J. Hermann et al.

    Accessory phase control on the trace element signature of sediment melts in subduction zones

    Chem. Geol.

    (2009)
  • B.R. Jicha et al.

    Discriminating assimilants and decoupling deep-vs. shallow-level crystal records at Mount Adams using 238U–230Th disequilibria and Os isotopes

    Earth Planet. Sci. Lett.

    (2009)
  • P.B. Kelemen et al.

    4.21 – one view of the geochemistry of subduction-related magmatic arcs, with an emphasis on primitive andesite and lower crust

  • S. Labanieh et al.

    Isotopic hyperbolas constrain sources and processes under the Lesser Antilles arc

    Earth Planet. Sci. Lett.

    (2010)
  • C.T.A. Lee et al.

    Constraints on the depths and temperatures of basaltic magma generation on Earth and other terrestrial planets using new thermobarometers for mafic magmas

    Earth Planet. Sci. Lett.

    (2009)
  • W.P. Leeman et al.

    Boron and lithium isotopic variations in a hot subduction zone—the southern Washington Cascades

    Chem. Geol.

    (2004)
  • W.P. Leeman et al.

    Petrologic constraints on the thermal structure of the Cascades arc

    J. Volcanol. Geotherm. Res.

    (2005)
  • C.G. Macpherson et al.

    Plio-Pleistocene intra-plate magmatism from the southern Sulu Arc, Semporna peninsula, Sabah, Borneo: implications for high-Nb basalt in subduction zones

    J. Volcanol. Geotherm. Res.

    (2010)
  • M. Martindale et al.

    High pressure phase relations of subducted volcaniclastic sediments from the west pacific and their implications for the geochemistry of Mariana arc magmas

    Chem. Geol.

    (2013)
  • W.F. McDonough et al.

    The composition of the Earth

    Chem. Geol.

    (1995)
  • J.D. Morris et al.

    Isotopic and incompatible element constraints on the genesis of island arc volcanics from Cold Bay and Amak Island, Aleutians, and implications for mantle structure

    Geochim. Cosmochim. Acta

    (1983)
  • E.K. Mullen et al.

    Evidence for trench-parallel mantle flow in the northern Cascade Arc from basalt geochemistry

    Earth Planet. Sci. Lett.

    (2015)
  • M.R. Perfit et al.

    Chemical characteristics of island-arc basalts: implications for mantle sources

    Chem. Geol.

    (1980)
  • C.M. Petrone et al.

    Unusual coexistence of subduction-related and intraplate-type magmatism: Sr, Nd and Pb isotope and trace element data from the magmatism of the San Pedro–Ceboruco graben (Nayarit, Mexico)

    Chem. Geol.

    (2003)
  • T. Plank

    4.17 – the chemical composition of subducting sediments

  • R.W. Porritt et al.

    Investigation of Cascadia segmentation with ambient noise tomography

    Earth Planet. Sci. Lett.

    (2011)
  • J. Prytulak et al.

    Astoria fan sediments, DSDP site 174, Cascadia Basin: Hf–Nd–Pb constraints on provenance and outburst flooding

    Chem. Geol.

    (2006)
  • P. Audet et al.

    Morphology of the explorer–Juan de Fuca slab edge in northern Cascadia: imaging plate capture at a ridge-trench-transform triple junction

    Geology

    (2008)
  • C.R. Bacon

    Calc-alkaline, shoshonitic, and primitive tholeiitic lavas from monogenetic volcanoes near Crater Lake, Oregon

    J. Petrol.

    (1990)
  • C.R. Bacon et al.

    Eruptive history and geochronology of Mount Mazama and the Crater Lake region, Oregon

    Geol. Soc. Am. Bull.

    (2006)
  • C.R. Bacon et al.

    Multiple isotopic components in Quaternary volcanic rocks of the Cascade arc near Crater Lake, Oregon

    J. Petrol.

    (1994)
  • C.R. Bacon et al.

    Primitive magmas at five Cascades volcanic fields: melts from hot, heterogeneous sub-arc mantle

    Can. Mineral.

    (1997)
  • D.G. Bailey et al.

    Common parent magma for Miocene to Holocene mafic volcanism in the northwestern United States

    Geology

    (1992)
  • M.B. Baker et al.

    Origin of compositional zonation (high-alumina basalt to basaltic andesite) in the giant crater lava field, Medicine Lake Volcano, northern California

    J. Geophys. Res. Solid Earth

    (1991)
  • M.B. Baker et al.

    Primitive basalts and andesite from the Mt. Shasta region N. Califomia: products of varying melt fraction and water content

    Contrib. Mineral. Petrol.

    (1994)
  • K. Bartels et al.

    High pressure phase relations of primitive high alumina basalts from Medicine Lake volcano, northern California

    Contrib. Mineral. Petrol.

    (1991)
  • J. Blichert-Toft et al.

    Separation of Hf and Lu for high-precision isotope analysis of rock samples by magnetic sector-multiple collector ICP-MS

    Contrib. Mineral. Petrol.

    (1997)
  • L.E. Borg et al.

    The petrogenesis of felsic calc-alkaline magmas from the southernmost Cascades, California: origin by partial melting of basaltic lower crust

    J. Petrol.

    (1998)
  • L.E. Borg et al.

    The variable role of slab-derived fluids in the generation of a suite of primitive calc-alkaline lavas from the southernmost Cascades, Calilornia

    Can. Mineral.

    (1997)
  • L.E. Borg et al.

    Ancient and modern subduction zone contributions to the mantle sources of lavas from the Lassen region of California inferred from Lu–Hf isotopic systematics

    J. Petrol.

    (2002)
  • R.W. Carlson et al.

    Crustal genesis on the Oregon Plateau

    J. Geophys. Res.

    (1987)
  • Cited by (37)

    • Sediment and ocean crust both melt at subduction zones

      2022, Earth and Planetary Science Letters
    • Two distinct mantle sources beneath the Garibaldi Volcanic Belt: Insight from olivine-hosted melt inclusions

      2020, Chemical Geology
      Citation Excerpt :

      Volcanism within the GVB has resulted in more than 2300 distinct vents and 22 major volcanic structures, including Glacier Peak, Mount Baker and the volcanic fields of Mount Garibaldi, Garibaldi Lake, Mount Cayley, Mount Meager Volcanic Complex, Salal Glacier, Bridge River and Silverthrone (Wilson and Russell, 2018; Hildreth, 2007). Recently, the study by Mullen et al. (2017) classifies Glacier Peak as the northern termination of the High Cascades segment. The dominant eruptive products of the GVB are intermediate calc-alkaline suites with basaltic eruptions comprising a small volume fraction in the south and a large volume fraction in the north (Bridge River to Salal Glacier; Wilson and Russell, 2018).

    View all citing articles on Scopus
    1

    Present address: Lorax Environmental Services, 2289 Burrard Street, Vancouver, British Columbia V6J 3H9, Canada.

    View full text