Our analysis of the long-term volcanic record of Nevado de Toluca volcano in Central Mexico reveals that:
1) Dacite and andesite petrogenesis is dominated through the entire life cycle of the volcano by hybridization processes of deep basaltic-andesite recharge magma and an upper crustal silicic mush. This process is captured in compositional zoning of all major mineral phases.
2) The compositional tuning to preferentially erupted dacite compositions (‘petrologic mode’) is a primordial and resilient features throughout the volcanoes history.
3) Minor long-term temporal trends towards more silicic compositions are observed from bulk-rock compositions and mineral chemistry and show that these data are best
explained by increasing interaction of mafic recharge magma with growing volumes of silicic mush through time.
4) Very mafic clino- and orthopyroxene compositions indicate that Nevado de Toluca is fed from 2 spatially separated magma sources either in the upper mantle and/or lower crust, at least in the younger history (<59 ka) of the volcano.
5) A comparison of petrologically monotonous stratovolcanoes in the TMVB shows that the mode of the preferentially erupted composition in the main volcano and in peripheral cones varies in concert. This points towards an important role of recharge magma composition and/or temperature in controlling the compositional mode of Mexican stratovolcanoes.
3 D ETERMINING THE CURRENT SIZE AND STATE OF
SUBVOLCANIC MAGMA RESERVOIRS
Abstract
Determining the state of magma reservoirs is essential to mitigate volcanic hazards. However, geophysical methods lack the spatial resolution to quantify the volume of eruptible magma present in the system, and the study of the eruptive history of a volcano does not constrain the current state of the magma reservoir. Here, we apply a novel approach to Nevado de Toluca volcano (Mexico) to tightly constrain the rate of magma input and accumulation in the subvolcanic reservoir. We show that only a few percent of the supplied magma erupted and a melt volume of up to 350 km3 is currently stored under the volcano. If magma input resumes, the volcano can reawake from multi-millennial dormancy within a few years and produce a large eruption, due to the thermal maturity of the system. Our approach is widely applicable and provides essential quantitative information to better assess the state and hazard potential of volcanoes.
Gregor Weber1, Luca Caricchi1, José Luis Arce2, Axel K Schmitt3
1 Department of Earth Sciences, University of Geneva, Geneva, Switzerland.
2 Instituto de Geología, Universidad Nacional Autónoma de México, Coyoacán, México.
3 Institut für Geowissenschaften, Universität Heidelberg, Heidelberg, Deutschland
Submitted to Nature Communications on the 24 April 2020
3.1 Introduction
Understanding the temporal evolution, size and physico-chemical state of igneous plumbing systems beneath volcanoes is of paramount importance to develop prognostic models for volcanic activity (Sparks et al., 2019). It is now well established that most volcano plumbing systems are built over prolonged periods of time by pulsed magma injection (Glazner et al., 2004; Annen et al., 2006; de Saint Blanquat et al., 2011; Edmonds et al., 2019) and that magma reservoirs are complex with transiently interconnected and vertically extensive storage regions in the Earth’s crust (Delph et al., 2017; Karakas et al., 2019; Cashman et al., 2017). To unravel such complexity and quantify parameters such as the volume of potentially eruptible magma, which are essential for hazard assessment, we rely on geophysical methods and the record of past eruptions.
Geophysical surveys typically estimate melt fractions <20% in crustal reservoirs
beneath active volcanoes (Pritchard and Gregg, 2016; Magee et al., 2018), which is significantly lower than the crystallinity of almost all observed erupted products (Marsh, 1981; Takeuchi, 2011). However, geophysical estimates represent average melt fractions because of the limited spatial resolutions of such techniques (Lowenstern et al., 2017; Magee et al., 2018). Geological and geochemical data provide information on the structure, dynamics and timescales operating within volcanic plumbing systems (Caricchi and Blundy, 2015; Cashman, et al., 2017; Sparks et al., 2019; Bachmann and Huber, 2019; Edmonds et al., 2019; Costa et al., 2020). Zircon crystals have proven particularly useful to trace the temporal evolution and physico-chemical state of crustal magma bodies (Schmitt, 2011). Crystallization age spectra and trace element
geochemistry of igneous zircon have been used to constrain thermal histories of magmatic systems (Kent and Cooper, 2018; Cooper, 2019; Claiborne et al., 2010; Klemetti and Clynne, 2014; Barboni et al., 2016; Szymanowski et al., 2017) and diffusion chronometry has been used to identify the timescales of pre-eruptive magmatic processes (e.g. Costa et al., 2020 and
references therein). However, while the analysis of geological and geochemical record provides quantitative information on the processes and timescales associated with past eruptive activity, it does not provide information on the current status of a magmatic system.
To constrain the current status of a magmatic system and eventually forecast its future
modelling to identify fundamental parameters controlling the temporal evolution of magmatic systems (Caricchi et al., 2014, 2016; Tierney et al., 2016; Degruyter et al., 2016; Forni et al., 2019; Laumonier et al., 2019; Paulatto et al., 2012). The comparison between models highlights that the thermal evolution of plumbing systems, and the volumes of eruptible magma they contain, depends strongly on the average rate of magma input and the thermal state of the crust (Annen et al., 2006; Menand et al., 2015; Karlstrom et al., 2017; Karakas et al., 2017; Weber et al., 2020). Variations of the modality of magma injection and of the geometry of the magmatic system, or convection play a minor role (Carrigan, 1988; Karakas et al., 2017; Annen et al., 2006, 2015; Caricchi et al., 2014). Howevr, quantitative estimates of average rates of magma input are difficult to obtain from the volcanic record as only a fraction of magma injected into the plumbing system is erupted (White et al., 2006; Tierney, 2016; Karakas et al., 2017). Zircon age distributions in combination with thermal modelling have been shown to provide insights into crustal magma fluxes (Caricchi et al., 2014, 2016; Tierney et al., 2016).
Here, we present a new approach to the study of the thermal evolution of magmatic systems, which provides much higher resolution (factor<2) on the rate of magma input into the system with respect to existing models (about one order of magnitude for Caricchi et al., 2014, 2016). Rather than focusing on the shape of zircon age populations, we focus on the total spread of zircon ages, use trace elements to further constrain our models, and remove from the
calculations all portions of the reservoir that cooled below solidus before eruption. Matching natural data to modelled age and temperature distributions, we determine crustal magma fluxes and the extrusive:intrusive (E:I) ratio at unprecedented resolution for Nevado de Toluca (Fig.
3.1), a dormant dacitic stratovolcano with 1.5 Million years history of explosive and effusive eruptions, situated in the densely populated Central Mexican highlands (Macias et al., 1997;
García-Palomo et al., 2002; Arce et al., 2006; Torres-Orozco et al., 2017). With our approach we determine the size of the subvolcanic reservoir. Our results show that the reservoir still contains magma, and it could be reactivated in few years by renewed magma supply from depth, which is of outmost importance for volcanic hazard assessment in such a highly populated area.
Figure 3.1. Location and studied samples of Nevado de Toluca volcano. Shaded relief map of the Toluca area is shown with sampling locations of 4 eruptions analyzed in this study marked by colored stars. Red star: Upper Toluca Pumice (UTP), purple: Middle Toluca Pumice (MTP), green: Block and Ash Flow (BAF28) and white: White Quarry Pyroclastic Flow (WQ). The inset on the far right shows the long-term 1.5 Ma eruptive history of Nevado de Toluca and division into eruptive episodes based on 40Ar/39Ar-geochronology (Torres-Orozco et al., 2017). The young pyroclastic sequence (PD) is shown as composite stratigraphic column, with indicated the age relations of the eruptions based on radiocarbon dating (Macias et al., 1997; Garcia-Palomo et al., 2002). The studied eruptions are marked by colored stars in the stratigraphic column.
Inset on the top left shows the location of Mexico City (MC) and Guadalajara (Gdl), as well as the position of major volcanic centers (triangles) in the Neogene to recent Trans Mexican Volcanic Belt after Ferrari et al., (2012). Ce: Ceboruco, Tq: Tequila, Co: Colima, MG:
Michoacán-Guanajuato, NT: Nevado de Toluca, P: Popocatépetl, Or: Pico de Orizaba. Dashed lines indicate the inferred depth of the subduction slab(Pardo and Suárez, 1995).