Plenary Lecture - Quantitative inverse wave problem in passive imaging: towards applications to helioseismology

Abstract:We investigate the quantitative reconstruction of physical properties in the context of passive imaging, in which the medium of interest is probed using ambient wavefields, which correspond to wave oscillations in the system generated by stochastic events. In this talk, we focus in particular on local helioseismology, which aims to investigate the solar interior from observed oscillations that are driven by turbulent convection. Considering the data as a superposition of waves generated by stochastic sources, the expected value of the cross-correlation between signals is related to the deterministic Green's function. We focus on time-harmonic wave propagation and employ the Hybridizable Discontinuous Galerkin (HDG) discretization method to efficiently simulate solar Green's functions and subsequently generate synthetic cross-correlation observables. The inverse wave problem in passive imaging is then formulated as an iterative minimization problem using the full cross-correlation data. The gradient of the misfit functional is computed via the adjoint-state method. We carry out synthetic experiments comparing inversions based on full cross-correlation data for passive imaging, with those using direct wavefield measurements in active-source imaging.

Speaker: Florian Faucher (Inria Bordeaux -- Team Makutu)

Date and Time: Tuesday, September 29, 2026 | 09:30 - 10:30
Venue: NTI Lecture Hall