Training courses

The training courses will take place on Monday, September 14 and Tuesday, September 15 in Nancy on the ENSG campus.

When registering, you can indicate your preferences in terms of training courses. The final training program will then be drawn up to meet the expectations of as many participants as possible.

List of training courses:

- Introduction to Radar and visualization tools (M.P Doin + E. Pathier)

Basic notions for beginners but not only, + FLATSIM and INSARVIZ

- Data download platforms (P. Durand, Bryan Raimbault) 

Information session on access to satellite data. Overview of European satellite data access platforms (GEODES, DINAMIS, Copernicus Browser, cartes.gouv.fr, etc.) and associated time-series tools (European Ground Motion Service). Presentation of available features and resources, with application examples using Copernicus Browser.

- On-demand computing services  (F. Provost + E. Pathier) 

The FormaTerre unit aims to facilitate access to data by providing services and tools to discover, process, and analyze the massive data archives currently available. A number of services dedicated to satellite data processing are accessible online, allowing users to measure ground deformation (GDM – Ground Deformation Monitoring services) from Sentinel-1 or Sentinel-2 images, or to generate a Digital Surface Model (DSM – Digital Surface Model service) from Pléiades acquisitions. Ground deformation measurements can thus be computed using InSAR techniques (GDM-SAR-in service) or image correlation methods (GDM-OPT and GDM-SAR-COR services).

The purpose of the training is to present these different services, their specificities, and their limitations through real-world case studies. Participants will become familiar with the steps required to run a calculation using a simple application example. The importance of calculation parameters and the possibilities for post-processing the obtained results will also be discussed.

- AMSTer software (N. d'Oreye + D. Smittarello + D. Derauw)

AMSTer software is a tool dedicated to the processing of Synthetic Aperture Radar (SAR) images. It is mostly designed to allow optimized automatic mass processing of SAR interferometry (InSAR) and the production of 2/3D ground deformation time series. However, AMSTer can also be used for the production of Digital Elevation Models (DEM), coherence maps and amplitude images time series e.g. for land cover or geomorphological changes, flood mapping etc…AMSTer software is able to process nearly all the SAR sensor currently available (ERS1 & 2, EnviSAT, ALOS, ALOS2, RadarSAT 1& 2, CosmoSkyMed, TerraSAR-X, TanDEM-X (incl. bistatic and pursuit mode), Sentinel1, Kompsat5, PAZ, SAOCOM, ICEYE, NISAR...). AMSTer software is made of three components: (1) an InSAR command line processor (the AMSTerEngine), (2) the MSBAS processor (for the computation of 2/3D time series of ground deformation) (3)  and a bunch of shell scripts automatizing all tasks, from data downloading to updated displacement maps and time series and possible automatic display on a dedicated webpage (AMSTer Toolbox). AMSTer software is freely available under the terms of the GNU Affero General Public (AGPL) License.- Optical processing tools (AMES / MicMac)

- AMSTerEngine et Coherence Tracking (D. Derauw)

AMSTerEngine is the command-line InSAR processor at the core of AMSTer Software, but it can also be used independently. The proposed training will present its general operation, main features, and will focus on the latest tool developed: coherence tracking. 

- DEFVOLC Modeling (V. Cayol)

- CSI Modeling (B. Raimbault) 

CSI is a Python package designed to perform slip inversions on faults with variable geometry. It enables the management and formatting of geodetic data (InSAR, GPS, optical) and seismological data to facilitate inversions, while providing a comprehensive framework for handling geodetic data and time series through various tools (profiles, visualization, etc.).

Hands-on session: Data handling and examples of slip inversion on a fault using CSI.

- IA for earth deformation in remote sensing (S. Giffard-Roisin, K. Burrows, J. Bourcier, B. Raimbault):

Understand and build your convolutional neural networks to process remote sensing data (case studies: landslide detection on InSAR/ IA optical image correlator/ and/or scarp height estimation from DEM profiles)

- Earth observation AI foundation models (S. Giffard-Roisin, F. Faure, J. Bourcier):

What is it, for what goal, how to benefit from them? Examples from recent models such as TerraMind and AlphaEarth

- Computation of Green’s functions in a homogeneous elastic medium with topography (T. Ragon) :

In this Python workshop, we will first see how to define a realistic fault geometry based on the Slab2 model, using the gmsh module to generate a triangular mesh. We will then compute the Green’s functions resulting from slip on this fault geometry within a homogeneous elastic medium, taking topography into account, using the cutde module. cutde is a triangular dislocation element (TDE) computation module for displacement and stress.

- Bayesian slip inversion (T. Ragon) :

In this Python workshop, we will illustrate, through a simple example, the principles of Bayesian inversion of slip on a finite fault. We will examine the advantages of this type of optimization compared to more classical approaches, and discuss when each approach is most appropriate. We will compare several algorithms, highlighting their strengths and weaknesses. In particular, we will use the AlTar package, which participants will be able to install beforehand.

- Processing tools optic (?)