Job offers - WIND project

PHD THESIS, POSTDOC, HPC ENGINEER POSITIONS

We currently open three positions which can be attributed either to PhD students (three-year positions) or young researchers (two-year postdoctoral positions). One of them can be also suitable for a HPC engineer (the first one hereafter).

These three positions deal with the following topics:

• Development of a 3D massively parallel frequency-domain FWI code (PhD, Postdoc, HPC engineer). The overall objective is to further develop a 3D frequency-domain FWI code (called FFWI) such that it includes all of the necessary functions to tackle a broad range of applications. These functions include different linear algebra solvers (direct solvers for sparse matrices, hybrid direct/iterative solvers based on domain-decomposition preconditioner), different discretization schemes (finite differences, finite elements, spectral elements), different wave physics (visco-acoustic and visco-elastic) and new optimization schemes aiming at extending the search space of FWI. These developments will have to be validated against 3D large-scale synthetic and real data benchmarks including the application to the OBN datasets from the 2015/2015 Gorgon experiment, offshore Australia.

For more information, download the description of the proposal (French/English).

• Application of elastic time-domain FWI to multi-component ocean bottom seismometer (OBS) datasets (PhD, Postdoc). Two case studies are targeted: (1) Seismic imaging of the crustal structure of the Eastern Nankai trough, offshore Japan, from the 2D SFJ-OBS dataset to characterize the structural factors controlling large subduction earthquakes. This dataset has been already processed by visco-acoustic FWI with quite relevant results. Our goal is to assess which kinds of added-valued from the geological and geophysical viewpoints will be provided by the imaging of the shear wavespeed in addition to the compressional counterpart. (2) Seismic imaging of the Gorgon field, Western Australia, from the OBN data of the 2015/2015 Gorgon experiment, offshore Australia. In this case too, a preliminary application of 3D visco-acoustic VTI FWI was already applied to this dataset. These results will be used as a baseline to assess the added-value provided by elastic FWI. This application will required some developments of the existing code to mitigate the computational burden of 3D elastic FWI such as (reciprocal) source subsampling or frequency encoding.

For more information, download the description of the proposal (French/English).

• Seismic survey design for seabed acquisition (PhD, Posdoc): Sparse ocean-bottom seismic survey are used both by academia and industry. In the framework of sparse acquisition, it is important to develop a methodology that allows one to assess quantitatively the acquisition footprint on FWI results. Once the tools to assess this footprint are vailable, we need to recast the survey design as an optimization problem in order to find the optimal geometry that will mnimize the acquisition footprint during FWI. Formulating and testing this optimization problem is the goal of this proposal.

For more information, download the description of the proposal (French/English).

If you are interested in one of this position, please contact Stéphane Operto (email: operto@geoazur.unice.fr)