Building 3-D Training Images from Historical Satellite Images for Multiple-Point Statistics modeling of Fluvial Reservoirs
Geogaceta · 16.06.2026 · 79 · 67-70
DOI: 10.55407/geogaceta117203 ↗Kurz gefasst
Fachartikel in der Zeitschrift Geogaceta, online erschienen am 16.06.2026. Verfasst von Natalia Puche Polo und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-SA 4.0).
Abstract (Originalsprache)
Geostatistical modeling using multiple-point statistics (MPS) methods simulates complex spatial patterns generated through the use of training images (TIs), enhancing the ability to create geologically realistic models. A key challenge in MPS modeling is the availability of suitable and representative three-dimensional TIs. This is critical in highly heterogeneous reservoirs like fluvial systems, where depositional architecture shows high spatial variability and compartmentalization is significant. This study aims to design and apply a workflow for creating realistic 3-D geological models combining (1) building a 3-D TI based on sedimentological interpretation from historical sequences of satellite images that captures realistic geobody distribution in present-day depositional environments, and (2) MPS simulation using the resulting TI. This 3-D TI successfully captures the fluvial depositional architecture, while the MPS simulation consistently reproduces it in alignment with the sedimentary architecture of a high-sinuosity fluvial reservoir.
Zitieren
APA
Polo, N. P., Díaz, R. P., & Yeste, L. M. (2026). Building 3-D Training Images from Historical Satellite Images for Multiple-Point Statistics modeling of Fluvial Reservoirs. Geogaceta, 79, 67-70. https://doi.org/10.55407/geogaceta117203
Vancouver
Polo NP, Díaz RP, Yeste LM. Building 3-D Training Images from Historical Satellite Images for Multiple-Point Statistics modeling of Fluvial Reservoirs. Geogaceta. 2026;79:67-70. doi:10.55407/geogaceta117203
BibTeX
@article{Polo2026,
author = {Polo, Natalia Puche and Díaz, Ricardo Palomino and Yeste, Luis Miguel},
title = {Building 3-D Training Images from Historical Satellite Images for Multiple-Point Statistics modeling of Fluvial Reservoirs},
journal = {Geogaceta},
year = {2026},
volume = {79},
pages = {67--70},
doi = {10.55407/geogaceta117203}
}
EUVIMED