Elastic Seismic Imaging and Probabilistic AVO Inversion: Application to CO2 Storage at Sleipner
Master thesis
Åpne
Permanent lenke
https://hdl.handle.net/11250/3071470Utgivelsesdato
2023-06-01Metadata
Vis full innførselSamlinger
- Department of Earth Science [1034]
Sammendrag
In order to reduce the effect of global warming, carbon capture and storage (CCS) is considered to be important as it reduces the amount of CO2 in the atmosphere. Geophysical monitoring is essential for safe storage of CO2. This requires accurate seismic inversion techniques so that the subsurface can be mapped in detail. However, the inverted model is not unique. Therefore, estimating the uncertainty of the model can be valuable. As a result, the development of probabilistic seismic inversion techniques is a key part of CCS. In this thesis, I present and compare inversion techniques. A deterministic and two probabilistic inversion techniques are applied to amplitude versus offset (AVO) data. The probabilistic inversion techniques are gradient-based and Ensemble Smoother for Multiple Data Assimilation (ES-MDA). In addition to these methods, I use elastic imaging using the block-diagonal of the Hessian and then compare these different methods. These inversion techniques are applied to two elastic models, where the reservoir consists of alternating thin layers of CO2-saturated sandstone and shale. The first model is 1D, and the second model is 2D, with the reservoir layers being slight anticlines. These models mimic CO2 storage at Sleipner in the North Sea. ES-MDA gives the most accurate inversion results for the 1D model. However, imaging is more useful for the 2D model and to obtain large-scale realistic models of the subsurface. Imaging is, therefore, preferred over AVO inversion for reservoir characterization and CO2 monitoring.
Beskrivelse
Postponed access: the file will be accessible after 2025-05-31