.png){kind=icon}
.png){kind=icon}
Return
Pattern-Feature Correlation Interpolation Method With Horizon Constraints
.png){kind=icon}
OA
AI
DOI：10.1109/ACCESS.2025.3642295.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Low-frequency model construction plays a crucial role in seismic inversion by providing the essential initial model. Traditional two-point geostatistical modeling, which relies on variograms, is often insufficient for capturing complex geological structures. While multipoint geostatistics can better address lateral discontinuities, these methods typically depend on training images, which are from geological interpretation and seldom available in practice. The pattern-feature correlation method circumvents this limitation by employing seismic profiles as an effective substitute for training images. However, this approach tends to produce smoothed results in areas distant from well control. To overcome this drawback, we have developed a novel horizon-constrained multipoint pattern-feature correlation interpolation technique. This method utilizes interpreted horizons as structural constraints to partition the modeling domain into geologically consistent subregions. Within each subregion, multipoint pattern-feature correlation interpolation is applied, significantly improving the accuracy and lateral continuity of the reconstructed models. Synthetic and field data tests demonstrate that the incorporation of horizon constraints yields initial models that are not only more accurate but also better aligned with the actual subsurface geology. Furthermore, we validate the effectiveness of our method by applying the reconstructed models to Bayesian linearized inversion. The results confirm the high accuracy of both the interpolation and subsequent inversion, highlighting the superior performance of the proposed technique.
Keywords:
Initial model
interpolation
pattern-feature
horizon constraints
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
3.6
Papers:
9.8W
Citations:
29.4W