Joint inversion ofP-wave velocity and density, application to La Soufrière of Guadeloupe hydrothermal system

Abstract

SUMMARY We present the result of a 3‐D gravity and P‐wave traveltime joint inversion applied to the hydrothermal system of La Soufrière of Guadeloupe. The joint inversion process is used here to overcome the different resolution limitations attached to the two data sets. P‐wave traveltimes were obtained from three active seismic surveys that were conducted from 2001 to 2007. Gravity data collected during a microgravity campaign is described in a companion paper. We use a joint inversion process based on a Bayesian formulation and a deterministic iterative approach. The coupling between slowness and density is introduced through a supplementary constraint in the misfit function that tries to minimize the distance between parameter values and a theoretical relationship. This relationship is derived from measurements on samples representative of Mt Pelée of Martinique and La Soufrière volcanoes. We chose a grid discretization that leads to an under‐determined problem that we regularize using spatial exponential covariance between the nodes parameters. Our results are compared to geophysical electromagnetic results obtained using resistivity and VLF surveys. They confirm the presence of highly contrasted dense/fast and light/slow zones in La Soufrière dome and crater basement. Our images suggest however that some non‐conductive zones may be massive andesite bodies rather than argilized zones, and that these bodies may have deeper roots than hypothesized.

Density muon radiography of La Soufrière of Guadeloupe volcano: comparison with geological, electrical resistivity and gravity data

“… Horizontal cross‐sections (1350, 1300 and 1250 m a.s.l.) of the density model derived by Coutant et al (2012) from the gravity data by Gunawan (2005). The thick red curves separate the well‐resolved regions (resolution >0.9, in bright colours) from the poorly resolved ones (in dim colours).…”

“…The thick red curves separate the well‐resolved regions (resolution >0.9, in bright colours) from the poorly resolved ones (in dim colours). For more details, see Coutant et al (2012).…”

“…The gravity measurements from Gunawan (2005) have been inverted by Coutant et al (2012) to derive a density model of the lava dome. The cross sections in Fig.…”

“…The thick red curves separate the well-resolved regions (resolution >0.9, in bright colours) from the poorly resolved ones (in dim colours). For more details, see Coutant et al (2012). dense body seen in the gravity model but, instead, are limited to the low-density volume (L1) located in the southwestern quarter of the dome (Figs 9A and B). This corresponds to the low-density region RS1 observed in the upper part of the Ravine Sud radiography.…”

3-D density structure and geological evolution of Stromboli volcano (Aeolian Islands, Italy) inferred from land-based and sea-surface gravity data

“…They definitively helped to improve the understanding of structural settings and their control on volcanic activity. Gravity inversions are typically performed using finely discretized models and leastsquare methods that seek smooth property variations (e.g., Coutant et al, 2012) or methods that seek the appropriate location, shape and volume of anomalies with predefined density contrasts Montesinos et al, 2005). The first category refers to linear inverse problems that are easily solved, but the resulting models have smoothly varying property variations that make the identification of geological contacts difficult.…”