Compositional evolution of magma from Parícutin Volcano, Mexico: The tephra record

https://doi.org/10.1016/j.jvolgeores.2009.09.015Get rights and content

Abstract

The birth of Parícutin Volcano, Mexico, in 1943 provides an unprecedented opportunity to document the development of a monogenetic cinder cone and its associated lava flows and tephra blanket. Three ‘type’ sections provide a complete tephra record for the eruption, which is placed in a temporal framework by comparing both bulk tephra and olivine phenocryst compositions to dated samples of lava and tephra. Our data support the hypothesis of Luhr (2001) that the first four months of activity were fed by a magma batch (Phase 1) that was distinct from the magma that supplied the subsequent eight years of activity. We further suggest that the earliest erupted (vanguard) magma records evidence of temporary residence at shallow levels prior to eruption, suggesting early development of a dike and sill complex beneath the vent. Depletion of this early batch led to diminished eruptive activity in June and July of 1943, while arrival of the second magma batch (Phase 2) reinvigorated activity in late July. Phase 2 fed explosive activity from mid-1943 through 1946, although most of the tephra was deposited by the end of 1945. Phase 3 of the eruption began in mid-1947 with rapid evolution of magma compositions from basaltic andesite to andesite and dominance of lava effusion. The combined physical and chemical characteristics of the erupted material present a new interpretation of the physical conditions that led to compositional evolution of the magma. We believe that syn-eruptive assimilation of wall rock in a shallow complex of dikes and sills is more likely than pre-eruptive assimilation within a large magma chamber, as previously assumed. We further suggest that waning rates of magma supply from the deep feeder system allowed evolved, shallowly stored magma to enter the conduit in 1947, thus triggering the rapid observed change in the erupted magma composition. This physical model predicts that assimilation should be observable in other monogenetic eruptions, particularly those with low pressure melt inclusions and with eruption durations of months to years.

Introduction

Cinder cones are the most common volcanic landform on Earth. Cinder cone fields commonly contain hundreds of individual vents, many of which have associated lava flows. Individual cones and associated flows and tephra blankets are assumed to be monogenetic, formed during a single eruptive episode from the ascent and eruption of a single magma batch. However, detailed studies indicate that some eruptions are fed by multiple magma batches (e.g., Strong and Wolff, 2003, Cervantes and Wallace, 2003a, Johnson et al., 2008), that individual magma batches may evolve through both fractionation and assimilation (e.g., Wilcox, 1954, McBirney et al., 1987, Johnson et al., 2008), and that some cones are re-occupied, feeding multiple eruptions over decades to centuries (McKnight and Williams, 1997). Also variable are the styles of activity that comprise a single eruptive episode, which may include Hawaiian, Strombolian, violent Strombolian, Vulcanian and purely effusive phases (e.g., Valentine and Krogh, 2006, Valentine et al., 2006, Valentine et al., 2007, Valentine and Gregg, 2008). The combined evidence for multiple magma batches and variations in eruptive style raises questions about the extent to which the nature of the physical activity reflects changes in both deep- and shallow-sourced variations in magma supply rate or composition.

The link between physical and compositional characteristics is of particular interest in determining the origin of violent Strombolian eruptions, perhaps the most distinctive of all of the forms of eruptive activity associated with cinder cone fields. The term violent Strombolian was introduced by Macdonald (1972) to describe the energetic ‘cineritic’ phase of Parícutin's activity, which involved strongly pulsatory explosions of moderate size (eruption columns ≤7–8 km high) and simultaneous lava effusion from lateral vents (typically located either within or at the base of the cinder cone). This type of activity appears to be characteristic of mafic eruptions when mass eruption rates are ~105 kg/s and the magma is sufficiently hydrous to produce separated flow conditions in the conduit (e.g., Pioli et al., 2008, Pioli et al., 2009). However, although the high H2O content of mafic subduction zone and intraplate magmas is now broadly recognized (e.g., Roggensack et al., 1997, Nicholis and Rutherford, 2004, Spilliaert et al., 2006, Johnson et al., 2008, Pioli et al., 2008), the extent to which magmatic vs. phreatic water is responsible for both eruptive vigor and extensive ash production accompanying violent Strombolian activity remains a topic of debate (e.g., Valentine et al., 2005, Martin and Nemeth, 2006).

The 1943–1952 eruption of Parícutin is the type example of a violent Strombolian eruption. For volcanologists, the abundance of tephra produced by the eruption (~ 2/3 of the total mass erupted), the strength of the tephra-producing explosions, and the fine average grain size of the tephra deposit have become defining characteristics of this eruptive style. For petrologists, Parícutin provided one of the first well-documented examples of assimilation (Wilcox, 1954). For these reasons, we have returned to Parícutin to re-examine both the physical volcanology of the eruption (Pioli et al., 2008) and the magma evolution through a compositional and textural study of the pyroclastic deposits (this study). We demonstrate that: (1) early tephra samples provide important clues to the nature of the initial magma that are not preserved in the classic lava flow samples from this eruption (e.g., Wilcox, 1954, McBirney et al., 1987); (2) the tephra records all of the bulk geochemical changes present in the lava flow sequence, thus providing a detailed view of both deep and shallow magmatic processes (e.g., Johnson et al., 2008, 2010b; (3) combined evidence of shallow (<100 MPa) magma storage and early melting of granite basement support a model of syn-eruptive crustal assimilation within a shallow magma storage system; and (4) preserved relationships among magma eruption rate, compositional change, and eruptive style provide insight into both the longevity of individual cinder cone eruptions and the range of activity observed at volcanoes of basaltic andesite composition (including current activity at Kluchevskoy, Llaima, and Tungurahua).

Section snippets

Geological setting of Parícutin volcano, Mexico

Parícutin is located in the Michoacán-Guanajuato Volcanic Field (MGVF) of the Trans-Mexican Volcanic Belt (TMVB), which trends east–west across Mexico. Volcanism in the TMVB is related to subduction of the Cocos and Rivera Plates beneath Mexico at the Middle America Trench (Fig. 1). The MGVF is made up of about 1000 small eruptive vents and 378 medium-sized volcanoes over a 40,000-km2 area (Hasenaka, 1994). The medium-sized volcanoes are dominantly shield volcanoes (Hasenaka et al., 1994),

Field investigations

Pioli et al. (2008) sampled and characterized tephra deposits from four locations that were chosen to encompass the full range of explosive activity at Parícutin, and that we use in our petrologic study (Fig. 4). Sample site A is close to the cone, on top of late 1946 lava flows. This section thus represents tephra accumulated from the start of 1947 onwards and was selected to encompass the compositional change during the last five years of the eruption. Site B, 5.5 m thick and located about 2 km

Field observations

Our field investigations, made >50 years after the end of the eruption, show that excavated tephra sections preserve original depositional textures and deposit thicknesses in most proximal locations (<5 km from the vent; Fig. 4). Even in sections where the uppermost part of the tephra sequence has been disturbed, the basal sequence is often well preserved (Fig. 5A), particularly in areas to the south and west of the cone that received most of the tephra fall during early phases of the eruption (

Discussion

The close observed correlation between tephra and lava compositions, mineral assemblages, and textures supports physical models that have a common source for these two eruptive products throughout the eruption (e.g., Krauskopf, 1948, Pioli et al., 2008) and points to the importance of considering both lava and tephra in dynamical studies of violent Strombolian activity. More importantly, our ability to place the tephra deposits within a temporal framework allows us to use the wealth of

Conclusions

The data presented above show that tephra deposits preserve a wealth of information about the compositional and textural evolution of magma during cinder cone eruptions and provide a record that is complementary to the lava record that has been preferred, traditionally, for petrologic studies. The tephra analyses presented here show a greater compositional range of material than previous studies, particularly in primitive basal samples. At the same time, combined petrologic analysis of tephra

Acknowledgements

We would first like to recognize the inspiration and support for this work that Jim Luhr provided—we miss him a lot. We also gratefully acknowledge an insightful review by Greg Valentine, particularly for suggesting that the precursory activity may reflect initial formation of shallow sills. We also thank Robert Trumbull for his careful editorial comments, Anne Peslier for acting as Associate Editor, and Joop Varenkamp, for assembling this tribute volume in Jim Luhr's memory. This work was

References (52)

  • L. Pioli et al.

    Controls on the explosivity of scoria cone eruptions: magma segregation at conduit junctions

    Journal of Volcanology and Geothermal Research

    (2009)
  • J. Taddeucci et al.

    Conduit processes during the July–August 2001 explosive activity of Mt. Etna (Italy): inferences from glass chemistry and crystal size distribution of ash particles

    Journal of Volcanology and Geothermal Research

    (2004)
  • G.A. Valentine et al.

    Emplacement of shallow dikes and sills beneath a small basaltic volcanic center—the role of pre-existing structure (Paiute Ridge, southern Nevada, USA)

    Earth and Planetary Science Letters

    (2006)
  • G. Valentine et al.

    Continental basaltic volcanoes—processes and problems

    Jounral of Volcanology and Geothermal Research

    (2008)
  • G.A. Valentine et al.

    Eruptive and geomorphic processes at the Lathrop Wells scoria cone volcano

    Journal of Volcanology and Geothermal Research

    (2007)
  • I. Yokoyama et al.

    Precursory earthquakes of the 1943 eruption of Parícutin volcano, Michoacan, Mexico

    Journal of Volcanology and Geothermal Research

    (1990)
  • J. Blundy et al.

    Petrologic reconstruction of magmatic system variables and processes

    In: Minerals Inclusions and Volcanic Processes. Reviews in Mineralogy and Geochemistry

    (2008)
  • J. Blundy et al.

    Magma heating by decompression-driven crystallization beneath andesite volcanoes

    Nature

    (2006)
  • K.V. Cashman et al.

    Cooling and crystallization of lava in open channels, and the transition of Pahoehoe Lava to `a`a

    Bulletin of Volcanology

    (1999)
  • P. Cervantes et al.

    Role of H2O in subduction-zone magmatism: new insights from melt inclusions in high-Mg basalts from central Mexico

    Geology

    (2003)
  • B.R. Edwards et al.

    Time scales of magmatic processes: new insights from dynamic models for magmatic assimilation

    Geology

    (1998)
  • D.H. Eggler

    Water-saturated and undersaturated melting relations in a Paricutin andesite and an estimate of water content in the natural magma

    Contributions to Mineralogy and Petrology

    (1972)
  • W.F. Foshag et al.

    Birth and development of Parícutin volcano Mexico

    Geological Survey Bulletin

    (1956)
  • C.J. Fries

    Volumes and weights of pyroclastic material, lava, and water erupted by Paricutin volcano, Michoacan, Mexico

    American Geophysical Union Transactions

    (1953)
  • T. Hasenaka et al.

    The cinder cones of Michoacán-Guanajuato Central Mexico: petrology and chemistry

    Journal of Petrology

    (1987)
  • T. Hasenaka et al.

    A complication of location, size and geomorphological parameters of volcanoes of the Michoacán Guanajuato Volcanic Field, Central Mexico

    Geofisica International

    (1985)
  • Cited by (73)

    • Andesitic (SiO<inf>2</inf>: ~60 wt%) monogenetic volcanism in the northern Colombian Andes: Crystallisation history of three Quaternary volcanoes

      2021, Journal of Volcanology and Geothermal Research
      Citation Excerpt :

      This volcanism is typically associated with unimpeded magma ascent from upper mantle (Smith and Németh, 2017), which produces mafic eruptive products in diverse eruptive styles (e.g. scoria cones, tuff rings, maars, etc.) (Valentine and Gregg, 2008). However, intermediate to acidic monogenetic volcanism has also been recognised in different tectonic settings and has been attributed to magma residence in crustal reservoirs and long-term magma ascent (Luhr, 2001; Németh et al., 2003; Mordick and Glazner, 2006; Erlund et al., 2009; Miallier et al., 2017). These magmas often produce single effusive monogenetic volcanoes (e.g. Mattioli et al., 2006; Osorio-Ocampo et al., 2018; Deniel et al., 2020) that are in general scarcely mentioned within this type of volcanism around the world (Murcia and Németh, 2020).

    • Understanding the storage conditions and fluctuating eruption style of a young monogenetic volcano: Blue Lake crater (&lt;3 ka), High Cascades, Oregon

      2020, Journal of Volcanology and Geothermal Research
      Citation Excerpt :

      Based on a global compilation of arc magma reservoir depths (Chaussard and Amelung, 2014), the storage depths for the Blue Lake magma would be considered “deep” (>5 km) but are slightly shallower than the range of crystallization depths (9.5–20 km) estimated from corrected melt inclusion data from Cinder Cone in northern California (Walowski et al., 2019). The narrow range of storage depths at Blue Lake, and the lack low pressure melt inclusions indicative of shallow storage, differs dramatically from many mafic monogenetic volcanoes, where melt inclusion volatile data indicate trapping over a wide depth range during ascent (e.g., Pioli et al., 2008; Johnson et al., 2008, 2010; Erlund et al., 2010; Walowski et al., 2019). The Blue Lake melt inclusion H2O and CO2 data suggest that, on average, the magma was stored at ~8.6 km depth (235 MPa) prior to eruption; here we use melt inclusion and olivine geochemistry to elucidate the processes that occurred during pre-eruption storage.

    View all citing articles on Scopus
    View full text