Machine learning in reservoir characterization: coupling data resolution-enhancement with hierarchical analysis of 3-D seismic attributes for seismic-facies classification

Papa A. Owusu*, Abdelmoneam Raef

*Corresponding author for this work

Research output: Contribution to journalConference articlepeer-review

1 Citation (Scopus)


Utilizing multiple seismic amplitude attributes as features in various machine learning techniques has enhanced reservoir quality and lithofacies classification. However, the underlying limitations of resolution of thin lithofacies and the associated wavelet interference may adversely impact the utility of seismic attributes in facies classification. Hence, characterizing reservoir thin-lithofacies, seismic responses are still challenging due to the compounding effect of thin-layer interference in the case of multi-member/pay rock formations. In this study, we present a case study evidencing the benefits of coupling seismic resolution enhancement, based on spectral whitening, on one hand, and hierarchical seismic attributes classification and depositional carbonates models, on the other hand, to leverage seismic attributes facies signatures and understand the facies controls on reservoir quality.

Original languageEnglish
Pages (from-to)1344-1348
Number of pages5
JournalSEG Technical Program Expanded Abstracts
Publication statusPublished - Aug 15 2022
Externally publishedYes
Event2nd International Meeting for Applied Geoscience and Energy, IMAGE 2022 - Houston, United States
Duration: Aug 28 2022Sept 1 2022

ASJC Scopus subject areas

  • Geotechnical Engineering and Engineering Geology
  • Geophysics

Cite this