English   español  
Please use this identifier to cite or link to this item: http://hdl.handle.net/10261/171783
logo share SHARE   Add this article to your Mendeley library MendeleyBASE
Visualizar otros formatos: MARC | Dublin Core | RDF | ORE | MODS | METS | DIDL | DATACITE
Exportar a otros formatos:


Feasibility analysis of travel-time tomography of downward continued streamer data followed by full waveform inversion in limited-low frequency recordings

AuthorsGras Andreu, Claudia ; Sallarès, Valentí ; Dagnino, D. ; Jiménez-Tejero, C.E. ; Meléndez, Adrià ; Ranero, César R.
Issue Date17-May-2016
Citation17th International Seismix Symposium (2016)
AbstractSeismic tomography methods are useful techniques to retrieve the physical properties of the subsurface. In particular, adjoint-state full waveform inversion (FWI) of controlled source data is considered to be one of the most powerful tools to obtain accurate, high-resolution models. However one of its main drawbacks is the strong non-linearity of the problem, which makes the solution strongly dependent on the initial model and on the low frequency content of the data set. A common strategy to mitigate these issues is to combine the robustness of Travel Time Tomography (TTT) to obtain an appropriate reference model that is subsequently refined by FWI. The combination of TTT and FWI is often used for long-offset acquisition geometries, where refracted waves are present as first arrivals. Conversely, its application to streamer-type multichannel seismic (MCS) data is infrequent, because these data are intrinsically short offset so the presence of refractions is very limited. In this case, Downward Continuation (DC) or redatuming of the MCS data to simulate an ocean bottom experiment allow recovering the refracted waves as first arrivals, so it is possible to perform TTT. In this work we use synthetic data to show that such TTT-based velocity models are suitable initial models for FWI even when data lack frequencies below 4 Hz; a realistic value in field data recordings. In summary, the strategy proposed, implemented and tested consists of several processing and inversion steps. First we compute the downward continued wavefield using a finite difference solution of the acoustic wave equation in time domain. The solver used for the propagation, developed at BCSI, is parallelized and incorporates a multi-shooting strategy to back-propagate the wavefield. The chosen datum level is the sea bottom. Then we pick the first arrivals of the DC data and we perform the TTT. We show that, in contrast to other possible choices for the initial model, the TTT solution is kinetically correct to start FWI at 4 Hz, in the sense that it allows to overcome cycle-skipping. Ongoing work is the application of this strategy to field data acquired in the Alboran basin with a 6 km-long streamer (TOPOMED experiment). The goal is to characterize the shallow structure in a tectonically active area the east Alboran basin volcanic arc. This includes the geometry of the sedimentary layers and basement outcrops and mapping large active faults that may represent a regional earthquake hazard
Description17th International Seismix Symposium (Seismix 2016), 15-20 May 2016, Aviemore, Scotland
Publisher version (URL)https://www.abdn.ac.uk/geosciences/events/downloads-1073.php
Appears in Collections:(ICM) Comunicaciones congresos
Files in This Item:
There are no files associated with this item.
Show full item record
Review this work

WARNING: Items in Digital.CSIC are protected by copyright, with all rights reserved, unless otherwise indicated.