Sub-surface structure of La Soufrière of Guadeloupe lava dome deduced from a ground-based magnetic survey

https://doi.org/10.1016/j.jvolgeores.2016.04.037Get rights and content

Highlights

  • We present a new ground-based magnetic survey over La Soufrière of Guadeloupe dome.

  • Topographic anomalies are used to estimate the shallow sub-surface magnetization.

  • A large volume of weakly magnetized material is inferred within and below the dome.

  • It is interpreted to be hydrothermally altered rock that may impact dome stability.

Abstract

In this study, we present the analysis and interpretation of a new ground magnetic survey acquired at the Soufrière volcano on Guadeloupe Island. Observed short-wavelength magnetic anomalies are compared to those predicted assuming a constant magnetization within the sub-surface. The good correlation between modeled and observed data over the summit of the dome indicates that the shallow sub-surface displays relatively constant and high magnetization intensity. In contrast, the poor correlation at the base of the dome suggests that the underlying material is non- to weakly-magnetic, consistent with what is expected for a talus comprised of randomly oriented and highly altered and weathered boulders. The new survey also reveals a dipole anomaly that is not accounted for by a constant magnetization in the sub-surface and suggests the existence of material with decreased magnetization beneath the Soufrière lava dome. We construct simple models to constrain its dimensions and propose that this body corresponds to hydrothermally altered material within and below the dome. The very large inferred volume for such material may have implications on the stability of the dome.

Introduction

The Soufrière volcano is the highest peak of the Guadeloupe Island in the Lesser Antilles subduction Arc (Fig. 1A,B). It is composed of an andesitic dome formed during the most recent magmatic eruption in 1530 A.D. (Boudon et al., 2008) and is located within a larger volcanic complex called La Grande Découverte-Soufrière, whose activity started around ~ 0.2 Ma (Carlut et al., 2000, Samper et al., 2009) and which remained the only active volcanic center in the island over the past 10,000 years (e.g., Komorowski et al., 2005). The formation of the Soufrière dome was preceded by several episodes of edifice collapse and magmatic activity. In particular, a major edifice collapse occurred ~ 3100 BP and formed (or widened) the south-facing horseshoe-shaped Amic crater (see Fig. 1C and Boudon et al., 1988). This edifice collapse led to the emplacement of thick debris avalanche deposits located principally to the southwest of the Soufrière dome. It was followed by the formation inside this crater of the Amic dome and scoria cones of L'Echelle and La Citerne. The Soufrière dome was formed after another edifice collapse that led to the formation of the Soufrière crater and cut into the Amic dome and La Citerne and L'Echelle volcanoes. The Soufrière dome currently displays intense hydrothermal activity evidenced by fumaroles and thermal springs located at its summit and periphery (see Fig. 1C; Komorowski et al., 2005). Acid sulfate fluids, which circulate through faults and fractures across the dome and along detachments beneath the dome associated with old flank collapse (see Fig. 1C), transform the fresh volcanic material into clays and deposit precipitates (Komorowski, 2008, Salaün et al., 2011). This hydrothermal alteration tends to weaken the dome's flanks and facilitate flank collapses. The resulting deposits are mainly composed of intensely hydrothermally altered materials (e.g., Boudon et al., 2007). This hydrothermal activity also induced several violent phreatic eruptions that produced pits and craters at the summit of the dome. Detailed knowledge of the Soufrière hydrothermal system is therefore necessary to provide an assessment of landslide hazards at the volcano.

Geophysical imaging of the Soufrière dome through SP (Spontaneous Potential) survey (Zlotnicki et al., 1994, Brothelande et al., 2014), VLF (Very Low Frequency) survey (Zlotnicki et al., 2006), electric tomography (Nicollin et al., 2006, Lesparre et al., 2014, Brothelande et al., 2014), joint gravity and seismic inversion (Coutant et al., 2012), and muon tomography (Lesparre et al., 2012) suggests a heterogeneous structure with intense ground-water circulation. Low resistivity, density and seismic velocity indicate the occurrence of highly altered or unconsolidated material whereas higher density, resistivity and seismic velocities indicate the presence of massive unaltered volcanic bodies. In addition, Nicollin et al. (2006) and Lesparre et al. (2012) suggest the existence of a high conductivity, low density, and highly altered layer at the base of the dome that may correspond to intense hydrothermal alteration coinciding with scars of flank collapse as suggested by Komorowski (2008) and Salaün et al. (2011). Brothelande et al. (2014) concluded that this low resistivity and altered material is restricted to the dome and its immediate vicinity.

Because hydrothermal alteration of volcanic rocks often induces changes in their magnetization, magnetic surveys can also be used to map and evaluate buried hydrothermal alteration over volcanoes (e.g., Finn et al., 2001, Finn et al., 2007, Bouligand et al., 2014). Fresh andesites from the Soufrière volcano contain primary magnetite and host a normal polarity magnetization. Hydrothermal alteration partially transforms magnetite into weakly-magnetic minerals such as hematite, pyrite, or montmorillonite (Salaün et al., 2011) and induces therefore a decrease of the rock magnetization. Hydrothermal alteration and degree of demagnetization may however vary with depth due to changes in temperature and redox conditions (oxidizing conditions at the surface and reducing conditions at depth; e.g. Komorowski, 2008, Salaün et al., 2011).

An aeromagnetic survey was flown in 1975 over the Guadeloupe Island at a constant elevation of 1800 m and provided a regional delineation of the long-wavelength anomalies across the island (Le Mouël et al., 1979, Gailler et al., 2013). In this study, we acquired a new ground magnetic survey at a constant height of ~ 2 m above the ground in order to focus on magnetic anomalies with very short wavelengths (from a few meters to a few hundred meters) in order to search for contrasts of magnetizations within the Soufrière dome and detect possible areas of hydrothermal alteration. In the following, we present these new data, compare observed anomalies with anomalies expected assuming a constant magnetization in the sub-surface below topography and use these anomalies to estimate the magnetization in the shallow sub-surface. We finally derive two simple models that account for a prominent dipole anomaly observed around the dome.

Section snippets

Data

In December 2009, a ground-based magnetic survey was conducted around and over the top of the Soufrière dome using a GEM System overhauser magnetometer (GSM19-W) carried at ~ 2 m above the ground. Because of the dense vegetation around the dome, measurements were limited to accessible trails (Figs. 2a and 3a). Magnetic data were recorded with a sampling rate of 0.5 s while walking at normal to slow pace (resulting in ~ 0.5 m sampling interval) depending on the slope along the trail. Data acquisition

Magnetic properties

The interpretation of observed magnetic anomalies requires some a priori knowledge on the magnetic properties of the volcanic rocks and in particular on the magnetization intensity and direction. The magnetic survey was mostly carried out on and around the Soufrière dome but also overlapped older volcanoes and lava flows from the Grande Découverte-Soufrière volcanic complex (e.g., Boudon et al., 1988; see also Fig. 1C). We relied on the study of Carlut et al. (2000) for magnetic properties of

Terrain effect

Before interpreting the observed anomalies in terms of contrasts of magnetization within the sub-surface, we need first to investigate the influence of topography on observed anomalies. For this purpose, we compare observed anomalies with those calculated when assuming a constant magnetization within the sub-surface. For each measurement, we evaluated the magnetic anomaly due to topography by assuming a constant magnetization within the sub-surface. The topography was approximated using a mesh

Magnetization

The correlation between observed and calculated anomalies at the top of the dome suggests that topography can be used to estimate the magnetization of the shallow sub-surface in a way relatively similar to Nettleton (1939) for estimating the density of terrain or to Honsho et al. (2009) for estimating the magnetization intensity of the seafloor. To this end, we compare observed and predicted profiles within moving windows shifted along the survey lines. Two examples of profiles are shown on

Model

Because of the non-uniqueness problem in potential-field studies, complicated here by the uneven distribution of data, we did not seek to reproduce observed anomalies very accurately but instead searched for the simplest model that could account for the general pattern of anomalies. We showed previously that the dipole anomaly composed of a magnetic high to the north and a magnetic low to the south of the dome cannot be accounted for assuming a constant magnetization within the sub-surface. The

Conclusion

In this study, we collected a new ground magnetic survey around and above the Soufrière dome. This survey displays many large-amplitude short-wavelength anomalies but also prominent long-wavelength anomalies and in particular a negative anomaly located at the summit of the dome and a dipole anomaly composed of a magnetic high and a magnetic low located respectively to the north and to the south of the dome.

We used here a method for fast and accurate computation of magnetic anomalies due to

Acknowledgments

We are grateful to Dominique Gibert, Quentin Gibert, Gaëtan-Thierry Kitou, Alexis Bosson, and the OVSG team for their precious help in the field data collection and to Richard Blakely, Francis Robach, Georges Boudon for helpful discussions and suggestions. This research was funded by ANR (ANR-08-RISKNAT-002-01) (DOMOSCAN Project). We thank Daniel Dzurisin, Jean-François Lénat and an anonymous reviewer for their comments which helped to improve significantly the original manuscript. The Litto3D

References (39)

  • J. Zlotnicki

    Magnetic measurements on La Soufriere volcano, Guadeloupe (Lesser Antilles), 1976–1984: a re-examination of the volcanomagnetic effects observed during the volcanic crisis of 1976–1977

    J. Volcanol. Geotherm. Res.

    (1986)
  • J. Zlotnicki et al.

    State of the hydrothermal activity of Soufrière of Guadeloupe volcano inferred by VLF surveys

    J. Appl. Geophys.

    (2006)
  • R.C. Aster et al.

    Parameter Estimation and Inverse Problems

    (2011)
  • F. Beauducel

    High-Resolution Digital Elevation Model of La Soufrière Lava Dome, OVSG Database, Institut de Physique Du Globe de Paris

    (2001)
  • B.K. Bhattacharyya

    Magnetic anomalies due to prism-shaped bodies with arbitrary polarization

    Geophysics

    (1964)
  • M.H.P. Bott

    Two methods applicable to computers for evaluating magnetic anomalies due to finite three dimensional bodies

    Geophys. Prospect.

    (1963)
  • G. Boudon et al.

    Carte et Notice explicative de la carte géologique à 1/20000e du Massif volcanique de la Soufrière

    (1988)
  • G. Boudon et al.

    Volcano flank instability in the Lesser Antilles arc: diversity of scale, processes, and temporal recurrence

    J. Geophys. Res.

    (2007)
  • C. Bouligand et al.

    Distribution of buried hydrothermal alteration deduced from high-resolution magnetic surveys in Yellowstone National Park

    J. Geophys. Res. Solid Earth

    (2014)
  • Cited by (8)

    • The thermal properties of hydrothermally altered andesites from La Soufrière de Guadeloupe (Eastern Caribbean)

      2022, Journal of Volcanology and Geothermal Research
      Citation Excerpt :

      Volcanic unrest has been increasing at La Soufrière since 1992, manifest as (1) the formation of new fumaroles and an expansion of the outgassing area on top of the lava dome (Brombach et al., 2000; Komorowski et al., 2005; Villemant et al., 2014; Allard et al., 2014; Moretti et al., 2020), (2) the appearance of high-flow-rate thermal acid sulfate-chloride springs on the slopes and at the base of the dome (Villemant et al., 2005, 2014), (3) an increase in the heat output from the dome (Gaudin et al., 2013; Jessop et al., 2021), (4) an increase in seismic activity (hundreds of shallow, low-magnitude earthquakes each month) and the largest felt volcano-tectonic earthquake since 1976–1977 (Moretti et al., 2020), and (5) large flank displacements (Moretti et al., 2020; Heap et al., 2021a). Activity within the shallow hydrothermal system, which has been imaged by a variety of geophysical methods (Nicollin et al., 2006; Brothelande et al., 2014; Bouligand et al., 2016; Rosas-Carbajal et al., 2016), is considered responsible for much of the recent volcanic unrest (Moretti et al., 2020; Heap et al., 2021a). Indeed, La Soufrière has a history of unrest associated with hydrothermal activity.

    • Magnetic structure of Basse-Terre volcanic island (Guadeloupe, Lesser Antilles) inferred from 3D inversion of aeromagnetic data

      2017, Journal of Volcanology and Geothermal Research
      Citation Excerpt :

      The Curie temperature isotherm is, however, likely to be shallower beneath the volcanic island of Basse-Terre and especially below the Soufrière volcano where a magma chamber has been inferred at a depth of ∼ 6 km (e.g. Hirn and Michel, 1979; Pozzi et al., 1979; Boichu et al., 2011; Allard et al., 2014). Indeed, Bouligand et al. (2016) suggested a depth to the Curie temperature isotherm of the order of 5 km, assuming a temperature of ∼ 900° C for the magma chamber and a linear vertical temperature profile. A shallower Curie temperature isotherm is also suggested by the locally very high heat flows (∼ 900–1000 mW/m2) measured by Feuillard (1976) in a shallow borehole near the Soufrière lava dome.

    View all citing articles on Scopus
    View full text