Résumé
InSAR and GNSS are now the best and most developed techniques in the Earth sciences to track deformation, especially in volcanology. In this study, we assess the reliability and consistency of these two techniques for measuring 3-D ground displacements-and not only the displacement in the direction of the InSAR Line of Sight-on volcanoes during rapid changes. The use of a large amount of satellite data (X, C, L-band as well as right and left looking acquisitions) made it possible to retrieve the 3-D displacement components with an unprecedented accuracy. We carry out this evaluation on the Piton de la Fournaise volcano, where four eruptions occurred in 2015. The comparison between GNSS and InSAR allows us: (i) to describe the deformation pattern associated with these eruptions, (ii) to quantify the discrepancies between InSAR and GNSS, and (iii) to discuss the limits and the complementarities of InSAR and GNSS. The ground deformation patterns associated with the four eruptions of Piton de la Fournaise in 2015 are typical of this volcano, with decimeter ground displacements asymmetrically distributed along the dike path, evidencing a preferential eastward motion, particularly visible thanks to the broad spatial coverage of InSAR Except for the NS component, InSAR and GNSS data are in overall agreement, with most of the GNSS-InSAR residuals <2.5 cm and <5 cm on the EW and vertical component, respectively, i.e. within the error bar of the two methods. Most of the discrepancies on the terminal cone can be attributed to uncorrected atmospheric effects in InSAR. Our study confirms the consistency and the complementarity of the two methods to characterize (i) the 3-D ground deformation distribution in high spatial resolution (InSAR), and (ii) the dynamism (GNSS) associated with eruptive activity.