Premium
Vector Field‐Based Simulation of Tree‐Like Nonstationary Geostatistical Models
Author(s) -
Vargas V. L.,
Pesco S.
Publication year - 2020
Publication title -
water resources research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.863
H-Index - 217
eISSN - 1944-7973
pISSN - 0043-1397
DOI - 10.1029/2020wr027542
Subject(s) - field (mathematics) , image (mathematics) , geostatistics , hydrogeology , computer science , point (geometry) , variogram , vector field , novelty , tree (set theory) , training (meteorology) , algorithm , artificial intelligence , mathematics , machine learning , geology , spatial variability , statistics , kriging , geography , geotechnical engineering , geometry , mathematical analysis , philosophy , meteorology , pure mathematics , theology
In this work, a new nonstationary multiple‐point geostatistical methodology called vector field‐based simulation is presented. This method generates multiple realizations of the phenomena being modeled, whose spatial continuity is described by a training image. It can be applied in reservoir characterization and can be used to study the behavior of hydrogeological systems. The main novelty of this work is that nonstationarity is addressed through a vector field inferred from the training image. This field represents the directions in which the image develops and controls the multiple‐point statistics methods (MPS) simulation. Additionally, our simulation is adapted to support hard data conditioning. The method is tested, and the realizations exhibit a similar morphology as the training image but with substantial variability. Furthermore, in the results it can be seen that the connectivity presented in the training image is preserved.