Multivariate Temporal Analysis of Muography Data at La Soufrière de Guadeloupe Volcano

We analyzed a muography dataset acquired at La Soufrière de Guadeloupe volcano, spanning more than three years from 2022 to the present, to investigate subsurface mass variations within the active volcanic system. Muography is a passive geophysical technique that exploits the attenuation of cosmic-ray muons to estimate the opacity of large geological structures. Muons are subatomic particles capable of traversing large amounts of matter. The flux of muons is measured along many distinct axes of observation, each corresponding to a specific trajectory through the subsurface. Because muons are absorbed according to the amount of matter they encounter, changes in the measured flux along each axis can be interpreted as variations in subsurface mass over time. This setup allows a single muon detector to investigate multiple regions of the volcanic edifice simultaneously, providing spatially and temporally resolved information on subsurface mass distribution.  A key aspect in analyzing muon time series is deciding how to group the signals from different trajectories to calculate the flux through distinct regions, since combining trajectories that are not coherent could mask meaningful variations. To address this, we applied a PCA-based multivariate analysis to jointly analyze the time series from all trajectories and identify spatially coherent regions characterized by common temporal behavior. This study demonstrates how muography, combined with multivariate statistical analysis, can be used to investigate the spatial organization and temporal variability of subsurface mass in active volcanoes.