Modeling heterogeneous geological reservoirs using multiple-point statistics simulation

• Main Authors: Ro Comunian, Julien Straubhaar, Olivier Besson
• Other Authors: The Pennsylvania State University CiteSeerX Archives
• Format: Text
• Language: English
• Subjects: method to perform multiple-points geostatistical simulations; Water; lena river
• Online Access: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.510.6081; http://geoscience-meeting.scnatweb.ch/sgm2009/sgm09_abstracts/powerpoints/comunian_sgm09_poster.pdf

Landsat project); e) simulated delta structure (size 1500 x 900) based on the TI in a) and its auxiliary variable map b), and conditioned to c) auxiliary variable (distance to the coast) for the SG, and d) orientation zones map for the SG. Fig 2: a) outcrop sections used to deduce the training structures and as conditioning data; b) large scale structures interpolated with ordinary Kriging; c) one of the training images used for the MPS simulation of a single layer of b); d) one realization of the full 3D geometry obtained with the hierarchical approach. Conclusion MPS technique provides realistic and highly heterogeneous models accounting for: conceptual model (TI), conditioning data, non stationarity. 1s t case study: Lena River 2D example, simulation of a delta: 3 facies (channels, lakes, land), non stationarity handled with auxiliary variable (distance to coast) and zones for rotations, conditioning data in sea (in the SG). 2n d case study: Herten aquifer analog Reconstruction of the 3D structure of highly heterogeneous sedimentary deposits: Herten dataset [1]: six parallel outcrops (Fig.2a) of a quarry in the Rhine basin close to Basel, Ordinary Kriging for simulating large scale structures (Fig.2b), then, MPS for simulating small scale heterogeneities (Fig.2d).