SIGMA (Seismic Imaging for Geomodel Analysis) is a collection of programs to close the loop between seismic imaging/modeling on the one hand, and seismic interpretation and structural modeling on the other hand. The long term goal of the library is to develop tools needed for reducing structural uncertainties by waveform inversion...

SIGMA is written in C++ and makes also a heavy use of CUDA for GPU parallelization. It compiles and runs on the linux platform. It inlcudes a set of executables controlled by options and parameter files, and a Qt user interface (OpenGL 4.5) for visualization of results.

Seismic Modeling

SIGMA includes a GPU-accelerated finite-difference acoustic simulator, which makes it possible to forward simulate seismic waveforms and generate synthetic surveys from geomodels.The numerical technique is a staggered finite differences (Virieux, 1984), with perfectly matched layer boundary conditions. 

Seismic imaging

To generate seismic images from seismic surveys, SIGMA implements many classical seismic processing and imaging methods: smoothing, filtering, velocity analysis, creation of common image gathers, migration. Details about the methods and their implementation are given in the PhD thesis of Modeste Irakarama and in the training.

Implicit structural modeling

SIGMA also contains a new method for implicit structural modeling to generate a series of conformable surfaces based on a finite difference formulation (Irakarama et al, EAGE 2018; Math. Geosciences, in rev)


Modeste Irakarama

Main references

Appraising structural interpretations using seismic data — Theoretical elements

Finite Difference Implicit Modeling of Geological Structures

Download and tutorial (Sponsors only)