Rhyolitic volcano dynamics in the Southern Andes: Contributions from 17 years of InSAR observations at Cordón Caulle volcano from 2003 to 2020

https://doi.org/10.1016/j.jsames.2020.102841Get rights and content

Highlights

  • InSAR records 17 years of volcanic unrest with pre-eruptive uplift, co-eruptive subsidence and post-eruptive uplift.

  • Pre and post-eruptive uplift signals are indicative of magma injection.

  • Co-eruptive subsidence signals are indicative of tephra and lava effusion during a VEI 4-5 eruption.

  • Source models suggest a plumbing system stable in location and depth and episodically active.

Abstract

In this article I present a review of InSAR observations of ground deformation at Cordón Caulle volcano, whose 2011–2012 VEI 4-5 eruption is the best scientifically observed and instrumentally recorded rhyolitic eruption to date. I document a complete cycle of pre-eruptive uplift, co-eruptive subsidence and post-eruptive uplift with InSAR data between March 2003 and May 2020, and produced by a complex interplay of magmatic processes. Pre-eruptive data show ~0.5 m of ground uplift in three distinct episodes between 2003 and 2011, with uplift rates between ~3 and ~30 cm/yr. The uplift was likely caused by magma injection resulting in pressurization of the magmatic system at depths of 4–9 km. Data spanning the first 3 days of the eruption show ~1.5 m of deflation produced by two distinct sources at 4–6 km depth located 18 km from each other and up to 10 km from the eruptive vent -- suggesting hydraulic connectivity of a large magma mush zone. A third source of deformation was recorded during the rest of the eruption at a depth of ~5 km, resulting in a total subsidence of ~4.2 m during the whole eruption. On a much smaller spatial scale (~25 km2), InSAR-derived digital elevation models recorded ~250 m of uplift in the area of the eruptive vent interpreted as the intrusion of a shallow laccolith during the first 2.5 months of the eruption and time averaged lava discharge rates up to ~150 m3/s. The co-eruptive time series of reservoir pressure drop and extruded volume follow exponential trends that can be explained by a model of magma reservoir depressurization and conduit flow. Since the end of the eruption, the surface of the volcano was uplifted ~1 m in a sequence of three transient episodes of unrest during 2012 and 2019, with uplift rates between 6 and 45 cm/yr and lasting between 0.5 and 3.2 years. These pulses can be modeled by the same source, a sub-horizontal sill at a depth of ~6 km. Viscoelastic relaxation is not significant on these time scales, hence I interpret these uplift signals as being produced by episodic pulses of magma injection in the crystal mush that likely underlies the volcano. The episodic and abrupt changes of the ground deformation suggest a restless trans-lateral magmatic system at depths of 4–9 km, and active across multiple spatial and temporal scales. Finally, I also discuss challenges of the InSAR technology that should be addressed to detect ground deformation on short time scales, particularly under the low coherence conditions of Cordón Caulle.

Introduction

Volcanic eruptions are one of the most spectacular geological processes observed on Earth. These events are produced by the ascent and extrusion of magma, molten rock composed of melt, crystals, and gases. The occurrence, duration, and style (either explosive or effusive) of the resulting eruption depend on a complex interplay of factors. These include the magma volume, ascent rate, composition, volatile content, and physicochemical transformations that the magma undergoes as it ascends and depressurizes from its storage area in a shallow reservoir through either a narrow conduit or a sill to the surface (Wilson et al., 1980; Tait et al., 1989; Jaupart and Tait, 1990; Jaupart, 2000; Edmonds and Wallace, 2017; Tait and Taisne, 2012; Dufek et al., 2012; Gonnermann and Manga, 2012; Gonnermann, 2015). The complexity of volcanic processes has been recently highlighted by a review article which stated that the first grand challenge in volcano science is to “forecast the onset, size, duration, and hazard of eruptions by integrating observations with quantitative models of magma dynamics” (National Academies of Sciences and Medicine, 2017). Fortunately, eruptions and/or the emplacement of magma in the upper crust are typically preceded by several signs of unrest including changes in ground deformation (e.g., Pinel et al., 2014), temperature (Reath et al., 2019), seismicity (e.g., Chouet and Matoza, 2013), and degassing (e.g., Carn et al., 2016) that can provide insights into their dynamics, and potentially forecast them (Sparks et al., 2012). However, there are still basic volcanological questions that remain unanswered. These include: 1. How is magma stored and transported in the crust? 2. What triggers eruptions? 3. What controls the duration and magnitude of eruptions? 4. What unrest signals are evidence of an imminent eruption? (National Academies of Sciences and Medicine, 2017; Wilson, 2017). Even in the best monitored volcanoes on Earth, eruption forecasting can be very challenging (Thelen et al., 2017; Peltier et al., 2018). Our understanding of the dynamics of these systems is still incomplete because we are inherently limited by the low resolving power of observations made from the Earth's surface (Bachmann and Huber, 2016) rather than in the actual reservoirs where magma is stored (Lowenstern et al., 2017).

Ground deformation data is a useful tool for volcano monitoring because the ascent of magma and the resulting eruption are usually coeval with displacement on the Earth's surface (Sparks et al., 2012). Thereby deformation allows us to potentially forecast and better understand volcanic processes. Volcano geodesy has traditionally relied on ground measurements including tiltmeters and continuous GPS but it has been revolutionized by Interferometric Synthetic Aperture Radar (InSAR), providing new insights on a variety of volcanic processes like eruption dynamics (Dzurisin and Lu, 2007; Pinel et al., 2014; Lu and Dzurisin, 2014; Dumont et al., 2018; Dzurisin et al., 2019). The key advantage of InSAR is that it is the only geodetic method that can measure ground deformation over large areas (> 40 × 40 km2) with repeat periods of a few days and with small uncertainties (~5 cm per interferogram). Compilations of satellite observations show a wide diversity of InSAR-derived deformation signals on volcanoes in Latin America (Reath et al., 2019) and elsewhere (e.g., Lu and Dzurisin, 2014), but the relation between deformation and eruption is not always clear (Biggs et al., 2014; Biggs and Pritchard, 2017; Delgado et al., 2017; Reath et al., 2019).

In this review article I present a summary of 17 years of InSAR observations at Cordón Caulle volcano (Fig. 1) in the Southern Volcanic Zone (SVZ) of the Chilean and Argentinian Andes (Stern, 2004) and the magmatic processes that can be unravelled with InSAR observations. Although InSAR data have contributed to key observations of volcanic processes in the SVZ, particularly during the VEI4-5 2008–2009 Chaitén (Wicks et al., 2011) and 2015 Calbuco (Nikkhoo et al., 2016; Delgado et al., 2017) eruptions and a sequence of unrest at Laguna del Maule volcano (Feigl et al., 2014; Le Mével et al., 2015; Novoa et al., 2019), in no other volcano in the SVZ than at Cordón Caulle it has shed light about a wide variety of volcanic processes (Pritchard and Simons, 2004; Fournier et al., 2010; Jay et al., 2014; Bignami et al., 2014; Delgado et al., 2016, 2018, 2019; Castro et al., 2016; Wendt et al., 2017; Euillades et al., 2017). These include a sequence of transient pre-eruptive pulses of magma injection, co-eruptive subsidence, lava flow extrusion and shallow laccolith intrusion, lava flow subsidence, and episodic post-eruptive magma injection that can lead to a potential new eruption (Fig. 2). At the time of writing (July 2020), no other subduction volcano in America except for Okmok in the Aleutians (Lu and Dzurisin, 2014) displays the wide variety of signals due to magmatic and superficial processes that can be observed with InSAR. The discovery of ground deformation at Cordón Caulle has been directly related to improvements in the SAR civilian platforms. Therefore, in this review I rely on multiplatform InSAR data and relate them to other geological observations (e.g., Castro et al., 2013, 2016; Bonadonna et al., 2015) only when they are relevant for the scope of this study.

I start this review with a summary of InSAR and volcano geodesy studies in the Southern Andes, highlighting the value of the method with respect to other geodetic techniques, particularly for the scope of this special issue on New advances on SAR Interferometry in South America. I then describe a complete cycle of pre-eruptive uplift, co-eruptive subsidence and post-eruptive uplift at Cordón Caulle imaged with InSAR. Then, I describe volcanological aspects where InSAR has made a leap forward in our understanding of rhyolitic dynamics. I finalize with a discussion on the challenges and opportunities for a better use of InSAR in the Southern Andes, including a qualitative comparison of different SAR data sets for the environmental conditions of Cordón Caulle. The time period of this study starts in March 2003 and ends in January 2020. It spans since the beginning of the routine phase of ENVISAT which was the first platform to systematically acquire data in the area to the current COSMO-SkyMED, TerraSAR-X, Sentinel-1, ALOS-2 and RADARSAT-2 acquiring several hundreds of SAR images per year. Therefore, significant changes in the InSAR technology have resulted in a much faster discovery and better understanding of magmatic processes than before.

Section snippets

Volcano geodesy in the Southern Andes

The Southern Volcanic Zone (SVZ) (Fig. 1) is one of the most active volcanic segments in the Andean volcanic arc of South America (Stern, 2004), with a time-averaged eruption rate of ~0.5 events/year during the 20th century (Dzierma and Wehrmann, 2012). The rate increased to ~1.3 events/year between 2008 and 2016 (Llaima January 01 2008, Chaitén May 02 2008–2009, Llaima April 03 2009, Cordón Caulle June 2011, Peteroa 2010–2011, Hudson October 26 2011, Copahue December 2012, Villarrica 03 March

Cordón Caulle geological background

Cordón Caulle is a long-lived system made up of a graben bounded by two sets of NW-SE trending fissures and the central volcano of a NW-SE volcanic range made up by Cordillera Nevada caldera to the NW and Puyehue volcano to the SE (Fig. 1, Lara et al., 2004, 2006a; 2006b). These three volcanoes have chemically distinct evolutions, with Cordón Caulle erupting only rhyolitic and rhyodacitic lavas in the Holocene (Singer et al., 2008). The lava flows erupted in 1921–1922 and 1960 were sourced from

InSAR methods

In this review I have included observations from almost every SAR mission available since 2003, which include ENVISAT, ALOS-1, TerraSAR-X/TanDEM-X (TSX/TDX), COSMO-SkyMED (CSK), RADARSAT-2 (RS2), UAVSAR, Sentinel-1 (S1) and ALOS-2 (Table 1). Data from ERS-1/2 are only briefly described due to its low quality (Pritchard and Simons, 2004). SAR data from the legacy JERS and RADARSAT-1 and from the newer SAOCOM-1 and PAZ missions are not available.

The first InSAR studies at Cordón Caulle (Pritchard

InSAR observations

In this section I describe all the deformation signals observed between 2003 and 2020 at Cordón Caulle.

Discussion

Here I discuss some interesting observations and lessons learned from the Cordón Caulle InSAR data and models. These include mainly the long term evolution of the plumbing system of the volcano, the triggering mechanism and the temporal evolution of the 2011–2012 eruption.

InSAR technological challenges

Although the InSAR data availability at Cordón Caulle has increased exponentially in the past decade, significant technical challenges remain in this area for a wider use of InSAR. These challenges are the limited and non-ideal data coverage and systematic coherence loss.

Conclusions

In this golden era of InSAR it is impressive how much the technology and the resulting amounts of data have evolved since the ENVISAT mission in the early 2000's leading to outstanding discoveries like the aforementioned sequence of unrest between 2003 and 2020. Many aspects remain to be elucidated from the Cordón Caulle dynamics and here is where I envision that the wealth of InSAR data that will be provided by the next generation InSAR missions (NISAR, Sentinel-1C/D, TanDEM-L, RADARSAT

Data availability

ENVISAT images were provided by the European Space Agency (ESA) and Level 1 SLC images are available at the ESA Online Dissemination site. ALOS-1 data are property of Japanese Aerospace Exploration Agency (JAXA) and the Japanese Ministerium of Trade and Commerce and are available at the Alaska Satellite Facility by NASA. Sentinel-1A/B SLC data were provided by ESA and distributed by Alaska Satellite Facility and CNES PEPS. TerraSAR-X data were provided by Deutsches Zentrum für Luft- und

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.

Acknowledgementsd

This review is the result of discussions and collaborations with many colleagues in the world in the past six years. I thank the French Centre National d'Études Spatiales (CNES) for a postdoctoral fellowship, Editor Andr's Folguera for hditor Pablo Euillades for the invitation to participate in this special issue, and many colleagues who in one way or the other have contributed with either data, software, ideas or comments to a better understanding of the dynamics of Cordón Caulle. These

References (166)

  • L.E. Lara et al.

    Rhyodcacitic fissure eruption in southern Andes (Cordon Caulle; 40.5 degrees S) after the 1960 (Mw : 9.5) Chilean earthquake: a structural interpretation

    J. Volcanol. Geoth. Res.

    (2004)
  • L.E. Lara et al.

    Magmatic evolution of the Puyehue-Cordon Caulle volcanic complex (40 degrees S), Southern Andean Volcanic Zone: from shield to unusual rhyolitic fissure volcanism

    J. Volcanol. Geoth. Res.

    (2006)
  • L.E. Lara et al.

    Structural controls of volcanism in transversal chains: resheared faults and neotectonics in the Cordon Caulle-Puyehue area (40.5 degrees S), Southern Andes

    J. Volcanol. Geoth. Res.

    (2006)
  • F. Albino et al.

    Stress transfer between magma bodies: influence of intrusions prior to 2010 eruptions at Eyjafjallajokull volcano, Iceland

    J. Geophys. Res. Solid Earth

    (2014)
  • F. Albino et al.

    Influence of surface load variations on eruption likelihood: application to two Icelandic subglacial volcanoes, Grimsvotn and Katla

    Geophys. J. Int.

    (2010)
  • B.V. Alloway et al.

    Multiple melt bodies fed the AD 2011 eruption of Puyehue-Cordon Caulle, Chile

    Sci. Rep.

    (2015)
  • K. Anderson et al.

    Physics-based models of ground deformation and extrusion rate at effusively erupting volcanoes

    J. Geophys. Res. Solid Earth

    (2011)
  • K. Anderson et al.

    Bayesian inversion of data from effusive volcanic eruptions using physics-based models: application to Mount St. Helens 2004-2008

    J. Geophys. Res. Solid Earth

    (2013)
  • A. Astort et al.

    Unrest at Domuyo volcano, Argentina, detected by geophysical and geodetic data and morphometric analysis

    Rem. Sens.

    (2019)
  • O. Bachmann et al.

    The magma reservoirs that feed supereruptions

    Elements

    (2008)
  • O. Bachmann et al.

    Gas percolation in upper-crustal silicic crystal mushes as a mechanism for upward heat advection and rejuvenation of near-solidus magma bodies

    J. Volcanol. Geoth. Res.

    (2006)
  • O. Bachmann et al.

    Silicic magma reservoirs in the Earth's crust

    Am. Mineral.

    (2016)
  • S.E. Barrientos

    Large thrust earthquakes and volcanic eruptions

    Pure Appl. Geophys.

    (1994)
  • H. Bathke et al.

    Inflation and deflation at the steep-sided Llaima stratovolcano (Chile) detected by using InSAR

    Geophys. Res. Lett.

    (2011)
  • M.G. Bato et al.

    Possible deep connection between volcanic systems evidenced by sequential assimilation of geodetic data

    Sci. Rep.

    (2018)
  • D. Bertin et al.

    High effusion rates of the Cordón Caulle 2011–2012 eruption (Southern Andes) and their relation with the quasi-harmonic tremor

    Geophys. Res. Lett.

    (2015)
  • J. Biggs et al.

    The lateral growth and coalesence of magma systems

    Phil. Trans. Math. Phys. Eng. Sci.

    (2019)
  • J. Biggs et al.

    Global volcano monitoring: what does it mean when volcanoes deform?

    Elements

    (2017)
  • J. Biggs et al.

    Magma flux at Okmok Volcano, Alaska, from a joint inversion of continuous GPS, campaign GPS, and interferometric synthetic aperture radar

    J. Geophys. Res. Solid Earth

    (2010)
  • J. Biggs et al.

    Global link between deformation and volcanic eruption quantified by satellite imagery

    Nat. Commun.

    (2014)
  • J. Biggs et al.

    The lateral extent of volcanic interactions during unrest and eruption

    Nat. Geosci.

    (2016)
  • C. Bignami et al.

    Multisensor satellite monitoring of the 2011 Puyehue-Cordon Caulle eruption

    IEEE J. Selected Topics Appl. Earth Observations Rem. Sens.

    (2014)
  • C. Bonadonna et al.

    Dynamics of wind-affected volcanic plumes: the example of the 2011 Cordón Caulle eruption, Chile

    J. Geophys. Res.: Solid Earth

    (2015)
  • J. Browning et al.

    Forecasting magma-chamber rupture at Santorini volcano, Greece

    Sci. Rep.

    (2015)
  • A. Carrara et al.

    Post-emplacement dynamics of andesitic lava flows at Volcán de Colima, Mexico, revealed by radar and optical remote sensing data

    J. Volcanol. Geoth. Res.

    (2019)
  • A. Carrier et al.

    A damage model for volcanic edifices: implications for edifice strength, magma pressure, and eruptive processes

    J. Geophys. Res.: Solid Earth

    (2015)
  • K.V. Cashman et al.

    Vertically extensive and unstable magmatic systems: a unified view of igneous processes

    Science

    (2017)
  • J.M. Castro et al.

    Storage and eruption of near-liquidus rhyolite magma at Cordon Caulle, Chile

    Bull. Volcanol.

    (2013)
  • J.M. Castro et al.

    Rapid laccolith intrusion driven by explosive volcanic eruption

    Nat. Commun.

    (2016)
  • A. Castruccio et al.

    The influence of plumbing system structure on volcano dimensions and topography

    J. Geophys. Res.: Solid Earth

    (2017)
  • W.-L. Chang et al.

    Accelerated uplift and magmatic intrusion of the Yellowstone caldera, 2004 to 2006

    Science

    (2007)
  • E. Chaussard

    Subsidence in the Parícutin lava field: causes and implications for interpretation of deformation fields at volcanoes

    J. Volcanol. Geoth. Res.

    (2016)
  • K.M. Cooper

    What does a magma reservoir look like? The ”crystal’s-eye” view

    Elements

    (2017)
  • K.M. Cooper et al.

    Rapid remobilization of magmatic crystals kept in cold storage

    Nature

    (2014)
  • D. Coppola et al.

    Measuring effusion rates of obsidian lava flows by means of satellite thermal data

    J. Volcanol. Geoth. Res.

    (2017)
  • K.G. DeGrandpre et al.

    High rates of inflation during a noneruptive episode of seismic unrest at semisopochnoi volcano, Alaska in 2014–2015

    G-cubed

    (2019)
  • F. Delgado

    Magma Storage, Transport and Eruption Dynamics in the Southern Andean Volcanic Zone Imaged with InSAR Geodesy

    (2018)
  • F. Delgado et al.

    The 2011 Hudson volcano eruption (Southern Andes, Chile): pre-eruptive inflation and hotspots observed with InSAR and thermal imagery

    Bull. Volcanol.

    (2014)
  • F. Delgado et al.

    Rapid reinflation following the 2011-2012 rhyodacite eruption at Cordon Caulle volcano (Southern Andes) imaged by InSAR: evidence for magma reservoir refill

    Geophys. Res. Lett.

    (2016)
  • F. Delgado et al.

    Renewed post-eruptive uplift following the 2011-2012 rhyolitic eruption of Cordón Caulle (Southern Andes, Chile): evidence for transient episodes of magma reservoir recharge during 2012-2018

    J. Geophys. Res.: Solid Earth

    (2018)
  • Cited by (16)

    • Outgassing through magmatic fractures enables effusive eruption of silicic magma

      2022, Journal of Volcanology and Geothermal Research
      Citation Excerpt :

      The explosive phase continued for ten days at decreasing intensity. Emplacement of a ~ 200 m deep laccolith occurred primarily between June 8 and July 3 at a magma flux of up to 150 m3/s (~106 kg/s) (Castro et al., 2016; Delgado, 2021). The hybrid eruptive phase was marked by the onset of extrusion of mostly outgassed rhyolite lava on June 15th, with an extrusion rate of 40 m3/s (Bertin et al., 2015).

    • Magma buoyancy drives rhyolitic eruptions: A tale from the VEI 5 2008-2009 Chaitén eruption (Chile) from seismological and geodetic data

      2022, Earth and Planetary Science Letters
      Citation Excerpt :

      For example, at least four reservoirs deflated during multiple stages of the Cordón Caulle eruption (Jay et al., 2014; Delgado et al., 2019), as well at the 1912 Novarupta eruption (Hildreth and Fierstein, 2012). How these different sources mechanically interact with each other (Delgado, 2021), and how the pressure drop together with the geometry of the plumbing system (Castro et al., 2013; Delgado et al., 2019) influence the explosive to effusive transitions are aspects not understood. The first geophysically monitored rhyolitic eruption occurred on May 2 2008 at Chaitén volcano in the Andean Southern Volcanic Zone (SVZ) of southern Chile (Fig. 1).

    • The 2011 Cordón Caulle eruption triggered by slip on the Liquiñe-Ofqui fault system

      2022, Earth and Planetary Science Letters
      Citation Excerpt :

      The 2011 eruption at PCCVC was one of the largest sub-aerial eruptions of the 21st century. Despite intensive monitoring by geodetic and seismological data, its primary trigger remains an open issue (Jay et al., 2014; Wendt et al., 2017; Delgado, 2020). The eruption was preceded by an inflation of 22 cm/yr spanning the 2007-2009 period, followed by two years of barely any ground deformation.

    • A century of ongoing silicic volcanism at Cordón Caulle, Chile: New constraints on the magmatic system involved in the 1921–1922, 1960 and 2011–2012 eruptions

      2021, Journal of Volcanology and Geothermal Research
      Citation Excerpt :

      Overall, our results indicate that, both extraction and storage occurred in the shallow crust (<10 km depth) for the last three eruptions at Cordón Caulle. These new data bring additional constraints to our understanding of the architecture of silicic systems, as transcrustal and/or translateral systems (e.g. Kennedy et al., 2018; Biggs and Annen, 2019; Sparks et al., 2019; Delgado, 2021). Current models for the magmatic system operating beneath Cordón Caulle comprise a crystal-rich mush zone between 4 and 9 km depth, with a considerable lateral extent from Cordillera-Nevada to the NW to Puyehue cone to the SE (∼20 km, Jay et al., 2014; Delgado, 2021).

    View all citing articles on Scopus
    View full text