Michael Nickel
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Michael Nickel has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Alejandro Escalona
SLB, Innovation Factori, Oslo, Norway
1 shared publication - Cagil Karakas
SLB, Innovation Factori, Oslo, Norway
1 shared publication - Cristiano Tarchiani
ENI S.p.A
1 shared publication - Daniele Blancone
SLB, Innovation Factori, Oslo, Norway
1 shared publication - Fabio Ciabarri
Edison RD&I, Milan, Italy.;.
1 shared publication - Filippo Chinellato
ENI S.p.A
1 shared publication
Latest Publications
- 4 November 2026
- Published by Society of Exploration Geophysicists in Interpretation
ML-assisted intra-salt interpretation of the Zechstein Group in the Norwegian North Sea: Implications for salt caverns
Salt formations often consist of layered evaporite sequences (LES) with mechanically distinct lithologies, which generate complex internal structures during deformation. Interpreting these structures in seismic data is challenging because of poor imaging, discontinuous reflectors, and strong velocity contrasts. While machine learning (ML) has been increasingly applied to salt boundary detection, its potential for resolving intra-salt structures remains underexplored. This study evaluates the performance of two supervised ML tools—fault prediction and multi-horizon interpretation—across three 3D seismic surveys from the Norwegian North Sea, where the Zechstein salt ranges from bedded salt to salt pillows and diapirs. Results show that ML performance is strongly dependent on seismic image quality and the degree of deformation: bedded salt allows accurate tracking of top, base, and intra-salt horizons, whereas pillows and diapirs exhibit degraded reflector continuity and lower interpretation confidence. Fault prediction not only detects faults but highlights intra-salt discontinuities such as rotated stringers. Confidence maps provide uncertainty-aware interpretations that delineate the zones of higher risk. In the context of subsurface hydrogen storage, the ML-derived confidence maps can guide salt cavern placement and reduce geomechanical risk.
- 4 November 2026
- Published by Society of Exploration Geophysicists in Interpretation
ML-assisted intra-salt interpretation of the Zechstein Group in the Norwegian North Sea: Implications for salt caverns
Salt formations often consist of layered evaporite sequences (LES) with mechanically distinct lithologies, which generate complex internal structures during deformation. Interpreting these structures in seismic data is challenging because of poor imaging, discontinuous reflectors, and strong velocity contrasts. While machine learning (ML) has been increasingly applied to salt boundary detection, its potential for resolving intra-salt structures remains underexplored. This study evaluates the performance of two supervised ML tools—fault prediction and multi-horizon interpretation—across three 3D seismic surveys from the Norwegian North Sea, where the Zechstein salt ranges from bedded salt to salt pillows and diapirs. Results show that ML performance is strongly dependent on seismic image quality and the degree of deformation: bedded salt allows accurate tracking of top, base, and intra-salt horizons, whereas pillows and diapirs exhibit degraded reflector continuity and lower interpretation confidence. Fault prediction not only detects faults but highlights intra-salt discontinuities such as rotated stringers. Confidence maps provide uncertainty-aware interpretations that delineate the zones of higher risk. In the context of subsurface hydrogen storage, the ML-derived confidence maps can guide salt cavern placement and reduce geomechanical risk.
- 1 June 2023
- Published by Society of Petroleum Engineers (SPE) in SPE Reservoir Evaluation & Engineering
- Vol. 26 (04) , 1224-1241
Real-Time Rock-Properties Estimation for Geosteering: Statistical Rock-Physics-Driven Inversion of Seismic Acoustic Impedance and LWD Ultradeep Azimuthal Resistivity
Summary: This work describes a statistical rock-physics-driven inversion of seismic acoustic impedance (AI) and ultradeep azimuthal resistivity (UDAR) log data, acquired while drilling, to estimate porosity, water saturation, and facies classes around the wellbore. Despite their limited resolution, seismic data integrated with electromagnetic resistivity log measurements improve the description of rock properties by considering the coupled effects of pore space and fluid saturation in the joint acoustic and electrical domains. The proposed inversion does not explicitly use a forward model, rather the correlation between the petrophysical properties and the resulting geophysical responses is inferred probabilistically from a training data set. The training set is generated by combining available borehole information with a statistical rock-physics modeling approach. In the inversion process, given colocated measurements of seismic AI and logging-while-drilling (LWD) electromagnetic resistivity data, the pointwise probability distribution of rock properties is derived directly from the training data set by applying the kernel density estimation (KDE) algorithm. A nonparametric statistical approach is used to approximate nonsymmetric volumetric distributions of petrophysical properties and to consider the characteristic nonlinear relationship linking water saturation with resistivity. Given an a priori facies classification template for the samples in the training set, it is possible to model the multimodal, facies-dependent behavior of the petrophysical properties, together with their distinctive correlation patterns. A facies-dependent parameterization allows the effect of lithology on acoustic and resistivity responses to be implicitly considered, even though the target properties of inversion are only porosity and saturation. To provide a realistic uncertainty quantification of the estimated rock properties, a plain Bayesian framework is described to account for rock-physics modeling error and to propagate seismic and resistivity data uncertainties to the inversion results. In this respect, the uncertainty related to the scale difference among the well-log data and seismic is addressed by adopting a scale reconciliation strategy. The main feature of the described inversion lies in its fast implementation based on a look-up table that allows rock properties, with their associated uncertainty, to be estimated in real time following the acquisition and inversion of UDAR data. This gives a robust, straightforward, and fast approach that can be effortlessly integrated into existing workflows to support geosteering operations. The inversion is validated on a clastic oil-bearing reservoir, where geosteering was used to guide the placement of a horizontal appraisal well in a complex structural setting. The results show that the proposed methodology provides realistic estimates of the rock-property distributions around the wellbore to depths of investigation of 50 m. These constitute useful information to drive geosteering decisions and can also be used, post-drilling, to update or optimize existing reservoir models.
- 1 June 2023
- Published by Society of Petroleum Engineers (SPE) in SPE Reservoir Evaluation & Engineering
- Vol. 26 (04) , 1224-1241
Real-Time Rock-Properties Estimation for Geosteering: Statistical Rock-Physics-Driven Inversion of Seismic Acoustic Impedance and LWD Ultradeep Azimuthal Resistivity
Summary: This work describes a statistical rock-physics-driven inversion of seismic acoustic impedance (AI) and ultradeep azimuthal resistivity (UDAR) log data, acquired while drilling, to estimate porosity, water saturation, and facies classes around the wellbore. Despite their limited resolution, seismic data integrated with electromagnetic resistivity log measurements improve the description of rock properties by considering the coupled effects of pore space and fluid saturation in the joint acoustic and electrical domains. The proposed inversion does not explicitly use a forward model, rather the correlation between the petrophysical properties and the resulting geophysical responses is inferred probabilistically from a training data set. The training set is generated by combining available borehole information with a statistical rock-physics modeling approach. In the inversion process, given colocated measurements of seismic AI and logging-while-drilling (LWD) electromagnetic resistivity data, the pointwise probability distribution of rock properties is derived directly from the training data set by applying the kernel density estimation (KDE) algorithm. A nonparametric statistical approach is used to approximate nonsymmetric volumetric distributions of petrophysical properties and to consider the characteristic nonlinear relationship linking water saturation with resistivity. Given an a priori facies classification template for the samples in the training set, it is possible to model the multimodal, facies-dependent behavior of the petrophysical properties, together with their distinctive correlation patterns. A facies-dependent parameterization allows the effect of lithology on acoustic and resistivity responses to be implicitly considered, even though the target properties of inversion are only porosity and saturation. To provide a realistic uncertainty quantification of the estimated rock properties, a plain Bayesian framework is described to account for rock-physics modeling error and to propagate seismic and resistivity data uncertainties to the inversion results. In this respect, the uncertainty related to the scale difference among the well-log data and seismic is addressed by adopting a scale reconciliation strategy. The main feature of the described inversion lies in its fast implementation based on a look-up table that allows rock properties, with their associated uncertainty, to be estimated in real time following the acquisition and inversion of UDAR data. This gives a robust, straightforward, and fast approach that can be effortlessly integrated into existing workflows to support geosteering operations. The inversion is validated on a clastic oil-bearing reservoir, where geosteering was used to guide the placement of a horizontal appraisal well in a complex structural setting. The results show that the proposed methodology provides realistic estimates of the rock-property distributions around the wellbore to depths of investigation of 50 m. These constitute useful information to drive geosteering decisions and can also be used, post-drilling, to update or optimize existing reservoir models.
- 22 December 2021
- Published by Society of Exploration Geophysicists in Interpretation
- Vol. 10 (2) , SC17-SC30
Machine-learning assisted interpretation: Integrated fault prediction and extraction case study from the Groningen gas field, Netherlands
Seismic fault detection using machine-learning techniques, in particular the convolution neural network (CNN), is becoming a widely accepted practice in the field of seismic interpretation. Machine-learning algorithms are trained to mimic the capabilities of an experienced interpreter by recognizing patterns within seismic data and classifying them. Regardless of the method of seismic fault detection, interpretation or extraction of 3D fault representations from edge evidence or fault probability volumes is routine. Extracted fault representations are important to the understanding of the subsurface geology and are a critical input to upstream workflows including structural framework definition, static reservoir and petroleum system modeling, and well planning and derisking activities. Efforts to automate the detection and extraction of geologic features from seismic data have evolved in line with advances in computer algorithms, hardware, and machine-learning techniques. We have developed an assisted fault interpretation workflow for seismic fault detection and extraction, demonstrated through a case study from the Groningen gas field of the Upper Permian, Dutch Rotliegend; a heavily faulted, subsalt gas field located onshore, northeast Netherlands. Supervised using interpreter-led labeling, we apply a 2D multi-CNN to detect faults within a 3D prestack depth migrated seismic data set. After prediction, we apply a geometric evaluation of predicted faults, using a principal component analysis to produce geometric attribute representations (strike azimuth and planarity) of the fault prediction. Strike azimuth and planarity attributes are used to validate and automatically extract consistent 3D fault geometries, providing geologic context to the interpreter and input to dependent workflows more efficiently.
- 22 December 2021
- Published by Society of Exploration Geophysicists in Interpretation
- Vol. 10 (2) , SC17-SC30
Machine-learning assisted interpretation: Integrated fault prediction and extraction case study from the Groningen gas field, Netherlands
Seismic fault detection using machine-learning techniques, in particular the convolution neural network (CNN), is becoming a widely accepted practice in the field of seismic interpretation. Machine-learning algorithms are trained to mimic the capabilities of an experienced interpreter by recognizing patterns within seismic data and classifying them. Regardless of the method of seismic fault detection, interpretation or extraction of 3D fault representations from edge evidence or fault probability volumes is routine. Extracted fault representations are important to the understanding of the subsurface geology and are a critical input to upstream workflows including structural framework definition, static reservoir and petroleum system modeling, and well planning and derisking activities. Efforts to automate the detection and extraction of geologic features from seismic data have evolved in line with advances in computer algorithms, hardware, and machine-learning techniques. We have developed an assisted fault interpretation workflow for seismic fault detection and extraction, demonstrated through a case study from the Groningen gas field of the Upper Permian, Dutch Rotliegend; a heavily faulted, subsalt gas field located onshore, northeast Netherlands. Supervised using interpreter-led labeling, we apply a 2D multi-CNN to detect faults within a 3D prestack depth migrated seismic data set. After prediction, we apply a geometric evaluation of predicted faults, using a principal component analysis to produce geometric attribute representations (strike azimuth and planarity) of the fault prediction. Strike azimuth and planarity attributes are used to validate and automatically extract consistent 3D fault geometries, providing geologic context to the interpreter and input to dependent workflows more efficiently.