A data-driven approach to predict fracture intensity using machine learning for presalt carbonate reservoirs: A feasibility study in the Mero Field, Santos Basin, Brazil

Predicting fracture intensity is essential for optimising reservoir production and mitigating drilling risks in the Brazilian pre-salt layer. However, previous studies rely excessively on conceptual models and typically do not integrate multiple types of data to perform such task. Moreover, to date,...

Full description

Saved in:
Bibliographic Details
Main Authors: Eberton Rodrigues de Oliveira Neto, Fábio Júnior Damasceno Fernandes, Tuany Younis Abdul Fatah, Raquel Macedo Dias, Zoraida Roxana Tejada da Piedade, Antonio Fernando Menezes Freire, Wagner Moreira Lupinacci
Format: Article
Language:English
Published: KeAi Communications Co., Ltd. 2025-06-01
Series:Energy Geoscience
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2666759225000253
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850160322159050752
author Eberton Rodrigues de Oliveira Neto
Fábio Júnior Damasceno Fernandes
Tuany Younis Abdul Fatah
Raquel Macedo Dias
Zoraida Roxana Tejada da Piedade
Antonio Fernando Menezes Freire
Wagner Moreira Lupinacci
author_facet Eberton Rodrigues de Oliveira Neto
Fábio Júnior Damasceno Fernandes
Tuany Younis Abdul Fatah
Raquel Macedo Dias
Zoraida Roxana Tejada da Piedade
Antonio Fernando Menezes Freire
Wagner Moreira Lupinacci
author_sort Eberton Rodrigues de Oliveira Neto
collection DOAJ
description Predicting fracture intensity is essential for optimising reservoir production and mitigating drilling risks in the Brazilian pre-salt layer. However, previous studies rely excessively on conceptual models and typically do not integrate multiple types of data to perform such task. Moreover, to date, no feasibility-like studies have assessed the reasonableness of such approaches. We propose a data-driven approach that utilises upscaled well logs (Young's modulus, Poisson's ratio, and silica content) alongside seismic attributes (curvature, distance to fault) to predict fracture intensity. The distance to fault is measured using the fault probability volume estimated by a pre-trained convolutional neural network (CNN). We evaluate the effectiveness of this data-driven approach employing two tree-ensemble models, eXtreme Gradient Boosting (XGBoost) and Random Forest, to estimate the volumetric fracture intensity (P32) in the wells. Regression and residual analyses indicate that XGBoost outperforms Random Forest. Results from feature importance methods, such as permutation importance and Shapley Additive explanations (SHAP), highlight curvature as the most important feature, followed by distance to fault, Young's modulus (or P-Impedance), silica content, and Poisson's ratio. The approach has been validated with rock sampling information and two blind tests. Consequently, we believe this workflow can be applied to other wells in nearby fields. The study offers a valuable tool for quantitatively estimating fracture intensity in pre-salt reservoirs. Future research may use this study as a reference for estimating fracture intensity within a seismic volume. The predicted fracture intensity estimates can enhance the reliability of reservoir porosity models and serve as a geohazard indicator to mitigate drilling risks.
format Article
id doaj-art-01b46ac87aee44bfa75250edc43059aa
institution OA Journals
issn 2666-7592
language English
publishDate 2025-06-01
publisher KeAi Communications Co., Ltd.
record_format Article
series Energy Geoscience
spelling doaj-art-01b46ac87aee44bfa75250edc43059aa2025-08-20T02:23:11ZengKeAi Communications Co., Ltd.Energy Geoscience2666-75922025-06-016210040410.1016/j.engeos.2025.100404A data-driven approach to predict fracture intensity using machine learning for presalt carbonate reservoirs: A feasibility study in the Mero Field, Santos Basin, BrazilEberton Rodrigues de Oliveira Neto0Fábio Júnior Damasceno Fernandes1Tuany Younis Abdul Fatah2Raquel Macedo Dias3Zoraida Roxana Tejada da Piedade4Antonio Fernando Menezes Freire5Wagner Moreira Lupinacci6GIECAR, Federal Fluminense University, Niterói, Rio de Janeiro, 24210346, Brazil; Corresponding author.GIECAR, Federal Fluminense University, Niterói, Rio de Janeiro, 24210346, BrazilGIECAR, Federal Fluminense University, Niterói, Rio de Janeiro, 24210346, BrazilGIECAR, Federal Fluminense University, Niterói, Rio de Janeiro, 24210346, BrazilGIECAR, Federal Fluminense University, Niterói, Rio de Janeiro, 24210346, BrazilGIECAR, Federal Fluminense University, Niterói, Rio de Janeiro, 24210346, Brazil; National Institute of Science and Technology of Petroleum Geophysics (INCT-GP/CNPQ), Niterói, Rio de Janeiro, 24210346, BrazilGIECAR, Federal Fluminense University, Niterói, Rio de Janeiro, 24210346, Brazil; National Institute of Science and Technology of Petroleum Geophysics (INCT-GP/CNPQ), Niterói, Rio de Janeiro, 24210346, BrazilPredicting fracture intensity is essential for optimising reservoir production and mitigating drilling risks in the Brazilian pre-salt layer. However, previous studies rely excessively on conceptual models and typically do not integrate multiple types of data to perform such task. Moreover, to date, no feasibility-like studies have assessed the reasonableness of such approaches. We propose a data-driven approach that utilises upscaled well logs (Young's modulus, Poisson's ratio, and silica content) alongside seismic attributes (curvature, distance to fault) to predict fracture intensity. The distance to fault is measured using the fault probability volume estimated by a pre-trained convolutional neural network (CNN). We evaluate the effectiveness of this data-driven approach employing two tree-ensemble models, eXtreme Gradient Boosting (XGBoost) and Random Forest, to estimate the volumetric fracture intensity (P32) in the wells. Regression and residual analyses indicate that XGBoost outperforms Random Forest. Results from feature importance methods, such as permutation importance and Shapley Additive explanations (SHAP), highlight curvature as the most important feature, followed by distance to fault, Young's modulus (or P-Impedance), silica content, and Poisson's ratio. The approach has been validated with rock sampling information and two blind tests. Consequently, we believe this workflow can be applied to other wells in nearby fields. The study offers a valuable tool for quantitatively estimating fracture intensity in pre-salt reservoirs. Future research may use this study as a reference for estimating fracture intensity within a seismic volume. The predicted fracture intensity estimates can enhance the reliability of reservoir porosity models and serve as a geohazard indicator to mitigate drilling risks.http://www.sciencedirect.com/science/article/pii/S2666759225000253K1 curvatureNaturally fractured reservoirsP32Machine learningFeature importance
spellingShingle Eberton Rodrigues de Oliveira Neto
Fábio Júnior Damasceno Fernandes
Tuany Younis Abdul Fatah
Raquel Macedo Dias
Zoraida Roxana Tejada da Piedade
Antonio Fernando Menezes Freire
Wagner Moreira Lupinacci
A data-driven approach to predict fracture intensity using machine learning for presalt carbonate reservoirs: A feasibility study in the Mero Field, Santos Basin, Brazil
Energy Geoscience
K1 curvature
Naturally fractured reservoirs
P32
Machine learning
Feature importance
title A data-driven approach to predict fracture intensity using machine learning for presalt carbonate reservoirs: A feasibility study in the Mero Field, Santos Basin, Brazil
title_full A data-driven approach to predict fracture intensity using machine learning for presalt carbonate reservoirs: A feasibility study in the Mero Field, Santos Basin, Brazil
title_fullStr A data-driven approach to predict fracture intensity using machine learning for presalt carbonate reservoirs: A feasibility study in the Mero Field, Santos Basin, Brazil
title_full_unstemmed A data-driven approach to predict fracture intensity using machine learning for presalt carbonate reservoirs: A feasibility study in the Mero Field, Santos Basin, Brazil
title_short A data-driven approach to predict fracture intensity using machine learning for presalt carbonate reservoirs: A feasibility study in the Mero Field, Santos Basin, Brazil
title_sort data driven approach to predict fracture intensity using machine learning for presalt carbonate reservoirs a feasibility study in the mero field santos basin brazil
topic K1 curvature
Naturally fractured reservoirs
P32
Machine learning
Feature importance
url http://www.sciencedirect.com/science/article/pii/S2666759225000253
work_keys_str_mv AT ebertonrodriguesdeoliveiraneto adatadrivenapproachtopredictfractureintensityusingmachinelearningforpresaltcarbonatereservoirsafeasibilitystudyinthemerofieldsantosbasinbrazil
AT fabiojuniordamascenofernandes adatadrivenapproachtopredictfractureintensityusingmachinelearningforpresaltcarbonatereservoirsafeasibilitystudyinthemerofieldsantosbasinbrazil
AT tuanyyounisabdulfatah adatadrivenapproachtopredictfractureintensityusingmachinelearningforpresaltcarbonatereservoirsafeasibilitystudyinthemerofieldsantosbasinbrazil
AT raquelmacedodias adatadrivenapproachtopredictfractureintensityusingmachinelearningforpresaltcarbonatereservoirsafeasibilitystudyinthemerofieldsantosbasinbrazil
AT zoraidaroxanatejadadapiedade adatadrivenapproachtopredictfractureintensityusingmachinelearningforpresaltcarbonatereservoirsafeasibilitystudyinthemerofieldsantosbasinbrazil
AT antoniofernandomenezesfreire adatadrivenapproachtopredictfractureintensityusingmachinelearningforpresaltcarbonatereservoirsafeasibilitystudyinthemerofieldsantosbasinbrazil
AT wagnermoreiralupinacci adatadrivenapproachtopredictfractureintensityusingmachinelearningforpresaltcarbonatereservoirsafeasibilitystudyinthemerofieldsantosbasinbrazil
AT ebertonrodriguesdeoliveiraneto datadrivenapproachtopredictfractureintensityusingmachinelearningforpresaltcarbonatereservoirsafeasibilitystudyinthemerofieldsantosbasinbrazil
AT fabiojuniordamascenofernandes datadrivenapproachtopredictfractureintensityusingmachinelearningforpresaltcarbonatereservoirsafeasibilitystudyinthemerofieldsantosbasinbrazil
AT tuanyyounisabdulfatah datadrivenapproachtopredictfractureintensityusingmachinelearningforpresaltcarbonatereservoirsafeasibilitystudyinthemerofieldsantosbasinbrazil
AT raquelmacedodias datadrivenapproachtopredictfractureintensityusingmachinelearningforpresaltcarbonatereservoirsafeasibilitystudyinthemerofieldsantosbasinbrazil
AT zoraidaroxanatejadadapiedade datadrivenapproachtopredictfractureintensityusingmachinelearningforpresaltcarbonatereservoirsafeasibilitystudyinthemerofieldsantosbasinbrazil
AT antoniofernandomenezesfreire datadrivenapproachtopredictfractureintensityusingmachinelearningforpresaltcarbonatereservoirsafeasibilitystudyinthemerofieldsantosbasinbrazil
AT wagnermoreiralupinacci datadrivenapproachtopredictfractureintensityusingmachinelearningforpresaltcarbonatereservoirsafeasibilitystudyinthemerofieldsantosbasinbrazil