Accounting for Decompaction during 3D Restoration using explicit and implicit Approaches.

Florian Basier and Pauline Durand-Riard and Guillaume Caumon. ( 2009 )
in: Proc. 29th Gocad Meeting, Nancy

Abstract

Structural interpretation is an essential but poorly constrained step in subsurface studies. Balanced restoration can be used to test the consistency of a 3D structural model, which highly contributes to reduce uncertainty and validate geodynamical interpretations. 3D restoration may be turned into a geomechanical problem, solved using a finite element method and setting displacement boundary and fault contact conditions, but decompaction is generally not handled in this process. Our objective is thus to account for decompaction during 3D restoration either on classical structural models, where mesh is conformable to stratigraphic horizons (explicit approach), or on implicit structural models, where surfaces are represented as level sets of a scalar interpolated property (implicit approach). To be able to decompact stratigraphic layers, it is first important to know the different laws that govern compaction. Commonly, isostatic and elasto plastic approaches are used. The isostatic approach considers total compaction for each layer as a function of depth, while the elasto plastic approach tries to faithfully represent compaction phenomena. Elasto plastic decompaction being extremely under-constrained, we propose coupling sequential restoration with isostatic decompaction, which allows for a continuous and time-optimized decompaction. The proposed decompaction algorithm has been adapted to the mesh, particularly to handle faults. Algorithms have also been adapted to the restoration approach (implicit or explicit): whereas explicit surfaces are easy to handle, their mesh generation is time-consuming. On the contrary, the implicit approach makes the mesh generation much easier and faster, but more difficult to consider since horizons are not a mesh interface.

Download / Links

BibTeX Reference

@inproceedings{BasierGM2009,
 abstract = { Structural interpretation is an essential but poorly constrained step in subsurface studies. Balanced restoration can be used to test the consistency of a 3D structural model, which highly contributes to reduce uncertainty and validate geodynamical interpretations. 3D restoration may be turned into a geomechanical problem, solved using a finite element method and setting displacement boundary and fault contact conditions, but decompaction is generally not handled in this process. Our objective is thus to account for decompaction during 3D restoration either on classical structural models, where mesh is conformable to stratigraphic horizons (explicit approach), or on implicit structural models, where surfaces are represented as level sets of a scalar interpolated property (implicit approach). To be able to decompact stratigraphic layers, it is first important to know the different laws that govern compaction. Commonly, isostatic and elasto plastic approaches are used. The isostatic approach considers total compaction for each layer as a function of depth, while the elasto plastic approach tries to faithfully represent compaction phenomena. Elasto plastic decompaction being extremely under-constrained, we propose coupling sequential restoration with isostatic decompaction, which allows for a continuous and time-optimized decompaction. The proposed decompaction algorithm has been adapted to the mesh, particularly to handle faults. Algorithms have also been adapted to the restoration approach (implicit or explicit): whereas explicit surfaces are easy to handle, their mesh generation is time-consuming. On the contrary, the implicit approach makes the mesh generation much easier and faster, but more difficult to consider since horizons are not a mesh interface. },
 author = { Basier, Florian AND Durand-Riard, Pauline AND Caumon, Guillaume },
 booktitle = { Proc. 29th Gocad Meeting, Nancy },
 title = { Accounting for Decompaction during 3D Restoration using explicit and implicit Approaches. },
 year = { 2009 }
}