Combined migrations and time-depth conversions: first results
- 1University of Calabria, Dipartimento di Ingegneria dell'Ambiente e del Territorio, Rende (CS), Italy (raffaele.persico@unical.it)
- 2Geostudi Astier s.r.l.
- 3Institute for the Electromagnetic Sensing of the Environment IREA-CNR
Commonly exploited migration algorithms or also well-established linear inverse scattering algorithms [1] for the focusing of GPR data are often based on the hypothesis of a homogeneous soil. However, this assumption is not valid always, and it provides deformed results when it is applied to image scenarios where it is not valid. More complex models of the scattering can afford the situation of a stratified medium, but only if the layers are flat and parallel to each other these model assumes analytic forms. In any case, commonly available commercial codes do not allow to implement the reflections and refraction of the waves associated to these cases [2]. More recently, time reverse migration algorithms have been introduced. They can account efficiently of non-homogeneous soils, but their performances in case of large and strong scattering targets are not yet completely established and they make use of forward numerical solvers, not all the times available and user friendly. At the conference, we will describe a strategy based on suitable combination of migration results achieved from different homogeneous media, accompanied by a time-depth conversion accounting for the occurrence of different values of the wave propagation velocity in the investigated domain. We will show how an improvement of the imaging result is achieved even in the lack of a correct mathematical model of the scattering phenomenon. Last but not least, the proposed strategy exploits software routines easy to be implemented.
How to cite: Persico, R., Morelli, G., Esposito, G., and Catapano, I.: Combined migrations and time-depth conversions: first results, EGU General Assembly 2023, Vienna, Austria, 24–28 Apr 2023, EGU23-2106, https://doi.org/10.5194/egusphere-egu23-2106, 2023.