Version française

(Secretary of D2SET)
PhD student in geophysics
You can write to me adam.gosselet@geophy.ensmp.fr

Project:Thesis
Place: École des Mines de Paris
Date:10/2000 - 10/2003
Summary:During my PhD I work on a tomography (see previous projects) method designed at the GEOPHY laboratory of the Ecole des Mines de Paris (Paris School of Mines) and named Stereotomography. Thanks to this method, a velocity model of the subsoil can be built by using not only the seismic waves travel times but also their emerging orientations (slopes). The use of the slopes gives better constraint of the inversion process. The waves modelisation is performed by using ray theory (high frequencies approximation). The aim of my PhD is to use the method not only with surface seimic data (obtained by marine acquisition campaign for example) but also with wells data (surface to well or well to well seismic, depending on the available data). The inverse problem becomes a multi-data one, which will bring difficulties. The final purpose is to inverse anisotropy (velocity depending on the orientation of rays).

Project:Geophysics engineering
Place: TOTALFINA Scientific Centre
Date:09-1999 / 08-2000
Summary:To build a velocity model (see below), tomography computes synthetic traveltimes by ray-tracing in a a-priori model. The difference between synthetic and real (recorded during seismic acquisition) traveltimes indicates how to change the model to make it better. An iterative process provides a final model, able to explain observed traveltimes. Nowadays, geophysicists try to use compression waves traveltimes and also shear waves traveltimes which were previously neglected.

Project:Engineer training period
Place: TOTALFINA Scientific Centre
Date:01-1999 / 08-1999
Summary:Imaging is the last stage of seismic processing, which allows a time to depth conversion. For that, a subsoil velocity model is needed. Different ways are used to build such a model. During this training period, I worked on a very detailed velocity analysis in order to improve imaging quality.

Project:Bac+5 training period
Place: C.G.G. Processing Center, Massy
Date:07-1998 / 12-1998
Summary:The aim of seismic acquisition is to generate compression waves on the ground and to record reflections or refractions caused by the different layers in the substratum. In the petroleum and mining industry, large seismic campaigns are led in order to discover oilfields and veins. These campaigns generate a huge quantity of data which must be processed (noises and multiple reflections filtering, signal deconvolution, geometrical effects correction...). These processings lead to a structural image of the subsoil that geologists can interpret.

Project:Bac+4 training period
Place: C.E.A. (French atomic energy agency),.Saclay
Date:07-1997 / 09-1997
Summary:During this trainig period, I coded a F77 binary Cu-Pd alloy simulation using a Monte-Carlo method. Firstly, the aim was to simulate an order/disorder transition using the mean field approximation of the atomic interaction. Secondly, the aim was to simulate a simple step/double step transition on sloped alloy's surface. such a transition had been observed during X rays diffraction experiments.

Project:Bac+3 numerical project
Place: Paris VII University
Date:
Summary:When a compass is placed in an oscillating magnetic field, his behaviour can become chaotic, depending on the initial conditions. This numerical simulation allowed us to display the strange attractor of the system, to make Poincaré's slices in it and to compute the Feigenbaum's constant.