DeepMagmaWaves is structured in four scientific work packages : WP1 and WP2 aimed at developing the frameworks for the PSVDTs and WP3 and WP4 at their applications. The fifth work package WP5 is dedicated to the coordination of the project.
Main proposal work packages
WP1. Numerical simulation of magmatic processes
A Physics-based Seismo-Volcanic Digital Twin (PSVDT) has three main components : the physical model of the medium, the seismic model of the medium (distribution of seismic velocities and attenuation), and the wavefield simulated from a given distribution of seismic sources.
The first component of a PSVDT is the physical model (distribution of composition, temperature, melt fraction etc) of the sub-volcanic medium. The modeling starts with defining a possible scenario of the evolution of a given volcanic system. This scenario will include such elements as magma supply rates at different levels and their time histories, eruptive histories, magma compositions, temperatures and depths of their storage. After defining a scenario of evolution and respective architecture of a magmatic system, numerical thermo-physical models will be used to compute a distribution of physical properties. The scenarios for real volcanic systems studied in WP4 will be built from the geological, petrological, and geophysical information available in published papers. In WP3, we will also use fully synthetic scenarios to build structures with a-priory imposed melt and magma distributions and properties. Modeling the rapid seismogenic processes within melt-rich parts of the magmatic systems will be implemented in order to obtain physics-based representation of seismic sources.
WP2. Numerical simulation of seismic wave propagation in melt-rich media
The goal of this WP is to develop an “operational” capacity to numerically simulate seismic waves in media with a very strong (more than 100% impedance contrasts) and small-scale (decametric) heterogeneity caused by presence of partially molten magma. This includes : (1) simulating accurate high-frequency seismograms to compare statistically with the observed seismo-volcanic signals and (2) predicting effective “long wave length” properties of the media (i.e., large-scale anisotropy) and propagating waves (i.e., ballistic travel times, surface-wave dispersion curves) needed for typical methods of seismic tomography.
For many seismic imaging methods such as tomography based on ballistic body wave arrival times or dispersion curves of surface waves, the decametric scale heterogeneity is too small in comparison with the wavelengths of observations or theoretical approximations (from several hundreds of meters to tens of kilometers). Therefore, the synthetic models need to be “upscaled”. This will be done using the mathematical formalism of elastic homogenization.
Based on this procedure, the small-scale heterogeneous synthetic distributions of seismic parameters will be converted into their smoothed versions (including apparent seismic anisotropy) that can be quantitatively compared with the long-wavelength travel time and dispersion measurements and tomographic images.
WP3. PSVDT-based experiments to test and to develop seismic methods
A major problem of application of seismological methods to real data is that we do not have the “ground truth” information about the structure and processes at depth. The PSVDTs give us a unique framework for testing the methods of seismic inversion in volcanic environments and, in particular, their ability to recover the quantitative information about the amounts of magma and melts at depth and about their fluxes. Namely, we can simulate seismic wave propagation and generate synthetic seismograms in the complex configurations much closer to the physical reality than simplified nearly transparent media. Moreover, these complex structures will be inherently modeled in terms of magmatic processes and relevant physical parameters such as magma/melt volumes and fluxes. WP3 will develop a set of PSVDT-based numerical experiments in order to simulate volcanic media and related seismic wavefields in different configurations and to quantify the ability of the seismic methods (both previously developed and the new ones) to recover the seismic and physical parameters of the controlled synthetic magmatic systems.
WP4. Studies of real volcanic systems
Five volcanic regions that cover various types and levels of volcano-magmatic activity have been selected because of their representativity for the project :
1. Klyuchevskoy Volcanic Group, Kamchatka, Russia
2. Piton de la Fournaise, la Réunion Island, France
3. Katamai volcanic group, Alasla, USA
4. Toba caldera, Sumatra, Indonesia
5. Massif Central, France
Systems 1, 3, 4 represent subduction volcanism with different sizes and levels of activity. System 2 is a very active oceanic hot spot volcano. System 5 is a province of continental intra-plate volcanism that at present is nearly dormant. The goal of the WP4 is to test in different conditions the methods developed in the project. For this, in all selected regions, there are seismological data available in open access in international data centers.
Mis à jour le 17 novembre 2026
La fédération
Intranet


