Publications in NVA | Google Scholar.
Tags:
Geology,
Geophysics
Publications
-
Oliveira, Lorenna Sávilla Brito; Lucas, Luiza Cavalcante Vinhas; Iacopini, David; Balsamo, Fabrizio; Torabi, Anita & Alaei, Behzad
[Show all 11 contributors for this article]
(2026).
Structural control and geometric characterization of a hydrothermal vent complex using seismic data, Potiguar Basin – Brazil.
Journal of South American Earth Sciences.
ISSN 0895-9811.
184.
doi:
10.1016/j.jsames.2026.106255.
Full text in Research Archive
Show summary
Hydrothermal vent complexes in sedimentary basins represent structural and thermal perturbations driven by fluid flow, leading to forced folding, hydrocarbon maturation, and fluid remobilization. Although their genesis has been widely studied, the factors controlling their geometry and evolution remain debated. This study analyzes 3D seismic data from the onshore Potiguar Basin in Brazil to identify hydrothermal vent structures. Two representative vents were extracted as 3D geobodies, using seismic attribute extrusion, enabling detailed characterization of their internal geometry. Results indicate that all identified vents are structurally controlled by regional-scale faults, which enhance hydrothermal fluid flow. Seismic attributes, including variance and dip illumination, proved effective for identifying vent structures, their relationship with faults, and associated fluid pathways, providing insights into their spatial distribution and geometry. In particular, the cosine of the phase attribute reveals that the geometry of hydrothermal vents varies as the conduits propagate through different sedimentary units within the basin. These findings highlight the role of fault zones in the development of hydrothermal vent complexes and improve our understanding of silicification processes in sedimentary reservoirs.
-
Basa, Abhisek; Alaei, Behzad & Torabi, Anita
(2025).
Detailed 3D characterization of fault geometric attributes: Insights from deep learning-based fault imaging in seismic data.
Journal of Structural Geology.
ISSN 0191-8141.
203.
doi:
10.1016/j.jsg.2025.105569.
Full text in Research Archive
Show summary
Generation of 3D fault network in conventional seismic interpretation involves creation of structural models by manually tracking the same discontinuity across seismic horizons on several adjacent vertical profiles. This process is often time-consuming, susceptible to human bias and introduces uncertainties in the characterization of fault geometric attributes - such as length, width and throw - that cannot be reliably quantified. In this study, we adopt a data-driven approach and extract the geometric attributes of the individual segments constituting the 3D seismic fault zone architecture in remarkable detail, that provides insights beyond the capabilities of conventional seismic attributes. This is achieved through the application of 3D Deep Learning (DL) Networks trained on 3D synthetic seismic datasets on a seismic survey from the eastern flank of Polhem Subplatform, SW of Norwegian Barents Sea to automatically create probabilistic fault volumes. Detailed extraction of individual fault segments from the fault probability volume allows us to investigate the seismic fault zone architecture at multiple scales. Our results show that the maximum segment length of all the studied faults is present near the upper tips and reduces towards the lower tips. The fault zone widths of individual segments measured on E-W-oriented vertical scanplanes orthogonal to the strike of the dominant fault set mostly vary between 10 and 20m. They exhibit higher values at locations where fault segments link laterally/vertically or interact with antithetic fault sets with similar strike of the dominant westward dipping fault set. Throw estimates and the total accumulated fault zone widths are higher towards the lower tips of the faults. Thus, the application of Deep Learning enables data-driven, high-resolution visualization of the seismic fault zone architecture and facilitates comprehensive extraction of fault geometric attributes, providing a more robust complement to traditional interpretation techniques.
-
Holden, Nora; Skurtveit, Elin; Alaei, Behzad; Midtkandal, Ivar & Braathen, Alvar
(2025).
Influence of growth fault complexities on fault seal assessments in the Smeaheia CO2 storage site, northern North Sea.
Journal of Structural Geology.
ISSN 0191-8141.
200.
doi:
10.1016/j.jsg.2025.105501.
Full text in Research Archive
Show summary
Growth faults exhibit great variability in their structural and sedimentological architecture. While these faults create structural traps that are critical for effective CO2 storage, their complexities are often overlooked in the integrity assessment of storage sites. This study focuses on the Vette Fault Zone in the Smeaheia CO2 storage site, northern North Sea. Here, the intended Upper Jurassic storage complex in the footwall juxtaposes predominately Lower Cretaceous syn-rift strata in the hanging wall. The primary seal in the footwall shows signs of degradation, with increased erosion occurring in the fault zone's eastward concave sections. The hanging wall features transverse folding of pre-rift units, overlain by thick syn-rift strata sourced from the immediate fault scarp and catchments beyond the fault block crest. These structures and deposits are not captured in nearby wells and therefore insufficiently considered when assessing fault rock composition and sealing potential. Incorporating these features, we find that the level of detail used to define the syn-rift layering has the most significant influence, whereas restoring the footwall to its pre-erosional state has only a minor effect. Not considering these structural complexities likely leads to underestimating the Vette Fault Zone's sealing potential. For fault seal assessments in settings with complex growth faults, we stress the importance of a detailed understanding of the structural development, defining confidence in host rock composition, utilizing a scenario-based approach for fault seal assessment, and presenting related uncertainties.
-
Basa, Abhisek; Torabi, Anita; Jiménez-Millán, Juan; Alaei, Behzad & García-Tortosa, Francisco Juan
(2025).
Unraveling the key factors controlling active faulting in Tertiary and Quaternary sequences.
Earth and Planetary Science Letters.
ISSN 0012-821X.
656.
doi:
10.1016/j.epsl.2025.119268.
Full text in Research Archive
Show summary
The effects of across-fault variation of mechanical properties and along-fault lithological heterogeneity on the seismicity of active faults is poorly understood. Moreover, the role of chemical processes on the mechanical behavior of active faults has not yet received sufficient attention. Such phenomena have more relevance, especially when faults juxtapose chemically reactive carbonate-dominated host rocks. In this study, we examine the key controls on active faulting by utilizing excellent outcrops of an intrabasinal strike-slip fault cutting across Tertiary and Quaternary sequences. We integrate structural, geomechanical and mineralogical investigations of faults and adjacent host rocks geographically exposed at different structural positions along the length of a 30 km-long, NE-SW-striking Galera Fault (GF) in Guadix-Baza basin, SE Spain. Our results show that fault and host rocks along the GF become carbonate-rich predominantly consisting of calcite towards NE. The thickness of sedimentary cover increases towards SW, where gypsum and phyllosilicate minerals are dominant. Calcite precipitation in fault rocks leads to reduction of porosity and significant enhancement of P-wave velocity (Vp), relative to the adjacent host rocks, and makes the fault zone more prone to seismic slip. On the contrary, phyllosilicates including clay mineral assemblages reduce the strength of the fault rocks and promote aseismic creep. The relationship between Vp and modified estimate of Brittleness Index used in this study indicate that the mineral assemblage of the fault as well as the adjacent host rocks influence seismicity. Our novel approach predicts more seismic activity towards the NE-part of GF due to dominance of calcite-rich rocks and proximity to the basement. This is validated by earthquake records in the studied area, which document shallow earthquakes concentrated towards the NE-part of GF. Thus, our study highlights the importance of adopting interdisciplinary approaches and understanding of the basin architecture controlling the distribution of host rocks to comprehend complex processes associated with active faulting.
-
Torabi, Anita; Alaei, Behzad & Smith, Adrian
(2024).
Fault characteristics in exhumed basement rocks; implications for understanding subsurface basement faults.
Tectonophysics.
ISSN 0040-1951.
887.
doi:
10.1016/j.tecto.2024.230445.
Full text in Research Archive
Show summary
We studied fault core geometry and mechanical properties of exhumed basement rocks at two localities (Storskora and Lislaskora) in Sotra Island, western Norway. We combine outcrop studies with in-situ measurements of the rock stiffness (Young's modulus) to characterize the faults. Faults were investigated both along and across strike using multiple 1D scanlines on the outcrop. Our results show that both fault core thickness and stiffness values vary along the faults. Thicker fault cores (up to 0.8 m and 1.9 m in Storskora and Lislaskora loaclities, respectively) show higher values of stiffness (Young's modulus) up to 70 GPa. The stiffness values of the fault core are generally higher than those measured on the damage zone of the faults in this area. The presence of epidote and compacted fault gouge in the fault core can cause the increase in estimated fault core stiffness. In contrast, fractures are dominant in the damage zones causing local reductions in the stiffness. A map of recent seismic events in this area shows potential seismic activities along some of the major exposed faults in the Sotra Island (e.g. Rustefjorden Fault). Based on the evidence from outcrop, inferred displacements, and interpretation of an available reflection seismic section, we found that the exposed faults could be secondary and part of the damage zone of the Øygarden Fault Complex in the east margin of the rift system in the North Sea. The results of this study could be utilized to predict the architecture and changes in rock stiffness of basement-involved faults in the subsurface.
-
-
Alaei, Behzad & Torabi, Anita
(2024).
Fault asperity and roughness, insight from faults in 3D reflection seismic data.
Marine and Petroleum Geology.
ISSN 0264-8172.
170.
doi:
10.1016/j.marpetgeo.2024.107145.
Full text in Research Archive
Show summary
The complexity of faults and the bias in the gathered fault data from different sources such as outcrops, reflection seismic data and earthquakes causes uncertainty to the understanding of fault plane geometry at the seismic scale, among other fault geometric attributes. Fault plane roughness and irregularities at different scales are important geometric attributes caused by asperities, which are linked to the fault frictional behaviour and mechanics. Asperities are localized areas with higher strength on the fault plane, which resist the applied stress to a certain limit before the fault ruptures. Imaging faults’ real geometry can help us to gain more insight into the fault roughness and asperities, thereby enhancing our understanding of the mechanics of faulting and earthquakes. In this study, for the first time, we map fault asperities and roughness on seismic scale faults using their 3D structures extracted from reflection seismic data. We have studied 21 normal faults ranging in depth from about 0.5 to 3 km with clastic dominated lithology and some carbonates by extracting and characterizing their true fault plane geometry on 3D reflection seismic data using Gaussian filtered coherency attributes and measuring throw over the entire fault plane. The outline of the mapped asperities is the envelope of internal tip-points between the segments on the fault plane 3D structure. Our results show that in most cases, at the boundary of the asperities, the strike of fault internal segments changes and the throw increases. We have introduced two new attributes characterizing fault plane roughness at seismic scale, notably 2D roughness attribute and segment mean roughness. The variations of these two attributes, i.e. higher values near the asperity boundary, have close link to the locations of asperities on the fault planes.
-
Holden, Nora; Alaei, Behzad; Skurtveit, Elin & Braathen, Alvar
(2024).
Implications of depth conversion on fault geometries and fault-risk assessment in the Smeaheia CO2 storage site, northern North Sea.
Geoenergy.
2(1).
doi:
10.1144/geoenergy2024-006.
Full text in Research Archive
Show summary
Assessments of fault geometries and fault-risk parameters are crucial when evaluating the integrity of a structurally controlled CO2 storage site. To perform these assessments, seismic data, recorded in time, must be converted to depth. The velocity models used for this time to depth conversion influence the final depth image and, consequently, the geometry of the interpreted faults. Against this background, we created four velocity models for depth conversion, assessed the impact on fault throw, dip and thickness of the primary seal, and, subsequently, a fault-risk assessment of the Vette Fault Zone in the Smeaheia CO2 storage site. We found that depth conversion had a larger influence on fault throw and thickness of the primary seal than on fault dip. In contrast, the overall assessment of the presence of a membrane seal and geomechanical integrity showed less sensitivity to the depth conversion process. Consequently, we suggest that a relatively robust fault-risk assessment can be made with a variety of velocity model designs and data input. Nevertheless, we found a mean difference of 2% in the shale gouge ratio, 4% in the slip tendency and 9% in the dilation tendency for the Vette Fault Zone, emphasizing the importance of accounting for the influence of depth conversion in optimizing structural assessments in potential CO2 storage sites.
-
Bönke, Wiktor; Alaei, Behzad; Torabi, Anita & Oikonomou, Dimitrios
(2024).
Data augmentation for 3D seismic fault interpretation using deep learning.
Marine and Petroleum Geology.
ISSN 0264-8172.
162.
doi:
10.1016/j.marpetgeo.2024.106706.
Full text in Research Archive
Show summary
Manual seismic interpretation of faults is a tedious and complicated process, which is prone to human error and bias. A semi-automatic approach for interpreting faults on seismic data is to use attributes that highlight discontinuities. These methods need to be optimized by the interpreter, thus constantly rely on the interpreter's knowledge. Recently, Machine Learning (ML) techniques in general and Convolutional Neural Networks (CNN) as part of Deep Neural Networks (DNN), in particular, have been used to detect and image faults on seismic data in order to make the process more automated and accurate. Supervised CNN learns and evolves from manually annotated or labeled fault interpretations. In this study, we have applied supervised 2D CNN to image faults on seismic data through binary segmentation. The study was performed on 3D marine seismic surveys in off-shore Norway collected from three separate locations along the Norwegian Continental Shelf, utilizing Efficient UNET CNN architecture. We applied data augmentation (geometric transformations) and hyperparameter tuning to improve the learning process and performance of the deep learning algorithm. We used noise content, acquisition type, imaging type, and fault scale as the main criteria to select seismic surveys. The application of data augmentation to the training and testing data generally led to improvement in the performance of 2D CNN on fault predictions, although the amount of improvement varied with respect to different surveys. The initial 2D CNN fault prediction improvement mainly relied on the quality, and size of faults present in the 3D seismic volumes. Further, improvement was achieved by the adjustment of certain hyperparameters affecting the training and testing process of the 2D CNN. However, we found little to no improvement on one seismic volume containing high levels of noise.
-
Torabi, Anita; Rudnicki, John; Alaei, Behzad & Buscarnera, Giuseppe
(2023).
Envisioning faults beyond the framework of fracture mechanics.
Earth-Science Reviews.
ISSN 0012-8252.
238.
doi:
10.1016/j.earscirev.2023.104358.
Full text in Research Archive
Show summary
Faults are complex structures that substantially influence the mechanical behavior and hydraulic connectivity of rock formations. Therefore, studying faults is important for a variety of disciplines such as geoscience, civil, geotechnical, reservoir engineering, and material science among others. Researchers from these disciplines have considered different aspects of faults, namely geometry, petrophysical properties and mechanics. Until now, these studies have evolved separately and at different scales, making it difficult to connect the geometric development of fault structure to its mechanics. The current understanding of fault geometry and growth is based on fracture mechanics and on many qualitative and quantitative studies on outcrop and seismic reflection surveys among other datasets. The application of fracture mechanics theory is mostly confined to simple geometries: elliptical models for a single fault plane and uniform properties. These applications predict the maximum displacement at the center of the fault, which is not in agreement with the new findings from 3D seismic and outcrop studies. These fracture mechanics models emphasize fault propagation along strike (in 2D). Although they can include the presence of a process zone at the fault tip, the models fail to explain the development of cross-fault damage zones and localization within the fault core as well as fault segmentation and displacement partitioning. Therefore, it is timely to revise the existing applications of fracture mechanics to simple fault geometries and to develop a data-driven fault mechanics possessing closer agreement with real, observed subsurface heterogeneity. This would allow better prediction of fault geometry, propagation, and growth in 3D. We suggest recent advances in non-destructive numerical characterization of faults and application of Deep Neural Networks (DNN) to map fault geometry and predict its properties from seismic data enable us for the first time to extract simultaneously faults' geometrical and mechanical properties at an unprecedented speed and accuracy, thus resolving the 3D fault shape and properties in ways that were unthinkable just a decade ago.
-
-
Libak, Audun; Torabi, Anita & Alaei, Behzad
(2022).
Normal fault geometric attribute variations with lithology: examples from the Norwegian Barents Sea.
Geological Society Special Publication.
ISSN 0305-8719.
495(1),
p. 7–27.
doi:
10.1144/SP495-2018-164.
Full text in Research Archive
Show summary
The geometry of faults exposed in the field is at least partly controlled by host-rock lithology. However, little work has been published on the effect of lithology on the geometry of very large faults imaged in seismic data. This paper presents such a study for seismic-scale faults in the Norwegian Barents Sea. Gamma-ray data are used to extract clay volumes of the host rock. These clay volume logs are used to populate 3D seismic cubes with lithology information. Finally, the clay volume cubes are co-rendered with seismic coherence cubes using opacity blending. This makes it possible to visualize both...
-
-
Haines, Emma Michie; Alaei, Behzad & Braathen, Alvar
(2021).
Assessing the Accuracy of Fault Interpretation using Machine Learning Techniques when Risking Faults for CO2 Storage Site Assessment.
Interpretation (Richmond).
ISSN 0020-9643.
10(1),
p. T73–T93.
doi:
10.1190/int-2021-0077.1.
Full text in Research Archive
Show summary
Generating an accurate model of the subsurface for the purpose of assessing the feasibility of a CO 2 storage site is crucial. In particular, how faults are interpreted is likely to influence the predicted capacity and integrity of the reservoir; whether this is through identifying high-risk areas along the fault, where fluid is likely to flow across the fault, or by assessing the reactivation potential of the fault with increased pressure, causing fluid to flow up the fault. New technologies allow users to interpret faults effortlessly, and in much quicker time, using methods such as deep learning (DL). These DL techniques use knowledge from neural networks to allow end users to compute areas where faults are likely to occur. Although these new technologies may be attractive due to reduced interpretation time, it is important to understand the inherent uncertainties in their ability to predict accurate fault geometries. Here, we compare DL fault interpretation versus manual fault interpretation, and we can see distinct differences to those faults where significant ambiguity exists due to poor seismic resolution at the fault; we observe an increased irregularity when DL methods are used over conventional manual interpretation. This can result in significant differences between the resulting analyses, such as fault reactivation potential. Conversely, we observe that well-imaged faults indicate a close similarity between the resulting fault surfaces when DL and manual fault interpretation methods are used; hence, we also observe a close similarity between any attributes and fault analyses made.
-
Haines, Emma Michie; Braathen, Alvar & Alaei, Behzad
(2021).
How Fault Interpretation Method may Influence the Assessment of a Fault-Bound C02 Storage Site.
In Røkke, Nils Anders & Knuutila, Hanna K (Ed.),
TCCS–11. CO2 Capture, Transport and Storage. Trondheim 22nd–23rd June 2021. Short Papers from the 11th International Trondheim CCS Conference.
SINTEF akademisk forlag.
ISSN 9788253617145.
p. 297–303.
Full text in Research Archive
Show summary
Interpretation of faults in the subsurface hinges on utilizing an optimum picking strategy, i.e. the seismic line spacing. Differences in line spacing lead to significant changes in subsequent fault analyses such as fault growth, fault seal and fault stability, all of which are crucial when analyzing a fault-bound CO2 storage site. With the ever-advancing technologies, machine learning techniques, such as Deep Neural Networks (DNN), used for fault extraction are becoming increasingly common, however their limitations and corresponding uncertainty is still largely unknown. Here, we show how fault extraction using DNN compares with faults that have been picked manually, and with using different line spacing. Uncertainty related to both manual and automated fault extraction methods are heavily reliant on seismic quality. As such, faults that are well-imaged show a closer similarity between those that have been manually picked and automatically extracted. In cases of poorly imaged faults, DNN picking on narrower line spacing creates a fault that is more irregular and with a lower predicted stability than the smoother and simpler fault model created by manual picking. Thereby, DNN creates fault surfaces that are less stable than those that have been picked manually, which is assumed to be associated with the increased irregularity of the fault segments. We conclude that fault picking by DNN without in-depth expertise works for well-imaged faults; poorly imaged faults require additional considerations and quality control for both manually and DNN picked faults.
-
Torabi, Anita; Ellingsen, Tor Saltnes Skram; Johannessen, Magnus Ueland; Alaei, Behzad; Rotevatn, Atle & Chiarella, Domenico
(2020).
Fault zone architecture and its scaling laws: Where does the damage zone start and stop?
Geological Society Special Publication.
ISSN 0305-8719.
496,
p. 99–124.
doi:
10.1144/SP496-2018-151.
Full text in Research Archive
-
-
-
Kolyukhin, Dmitriy; Torabi, Anita; Libak, Audun; Alaei, Behzad & Khachkova, Tatiana
(2018).
Statistical analysis of displacement and length relation for normal faults in the barents sea.
Geosciences.
8(11),
p. 1–10.
doi:
10.3390/geosciences8110421.
Full text in Research Archive
-
Torabi, Anita; Alaei, Behzad & Ellingsen, TTS
(2018).
Faults and fractures in basement rocks, their architecture, petrophysical and mechanical properties.
Journal of Structural Geology.
ISSN 0191-8141.
117,
p. 256–263.
doi:
10.1016/j.jsg.2018.07.001.
Full text in Research Archive
-
Klausen, Tore Grane; Torland, Jonas Aas; Eide, Christian Haug; Alaei, Behzad; Olaussen, Snorre & Chiarella, Domenico
(2017).
Clinoform development and topset evolution in a mud-rich delta - the Middle Triassic Kobbe Formation, Norwegian Barents Sea.
Sedimentology.
ISSN 0037-0746.
65(4),
p. 1132–1169.
doi:
10.1111/sed.12417.
Full text in Research Archive
Show summary
Clinoform surfaces are routinely used to mark transitions from shallow waters to deep basins. This concept represents a valuable tool for screening potential reservoir intervals in frontier basins where limited data are available. Variations in the character of clinoform geometries and shoreline and shelf-edge trajectories are indicators of a range of different factors, such as palaeobathymetry, changes in relative sea-level and sediment supply. Applications of conceptual and generalized models might however lead to erroneous assumptions about the supply of coarse-grained material to the delta front and basin when superficial similarities between clinoform geometries are not treated holistically. The present study examines the mudstone-dominated Middle Triassic Kobbe Formation – a potential hydrocarbon reservoir interval in the Barents Sea, where prodeltaic to deltaic deposits can be examined in cores, well logs and two-dimensional and three-dimensional seismic data. Despite pronounced acoustic impedance contrast to the surrounding shale, channel belt networks are not observed close to the platform-edge in seismic datasets, even at maximum regressive stages. However, sub-seismic prodeltaic deposits observed on the shallow platform indicate that prodeltaic deposits were sourced directly from the delta plain. Clinoform surfaces with different geometries and scale are observed basinward of the palaeo platform-edge of underlying progradational sequences, correlative to mudstone-dominated prodeltaic core sections. Results indicate that platform-edge deltas developed at discrete sites in the basin due to normal regression, but the positions of these deltas are not directly relatable to variations in clinoform geometries. Transitions from third-order to fourth-order clinoform geometries record discrete transgressive–regressive cycles but are not necessarily good indicators of sandstone deposition. Because of prolonged periods with high accommodation, channel avulsions were frequent and only very fine-grained sandstone was deposited in heterolithic units at the delta front. Sandstones with good reservoir properties are predominantly found along basin margins.
-
-
Torabi, Anita; Alaei, Behzad; Kolyukhin, Dmitriy; Libak, Audun; Gabrielsen, Roy Helge & Braathen, Alvar
(2016).
Fault geometric and seismic attributes; An integrated study with focus on the Barents Sea.
First Break.
ISSN 0263-5046.
34(5),
p. 51–58.
Full text in Research Archive
-
-
Torabi, Anita; Alaei, Behzad; Kolyukhin, Dmitriy; Libak, Audun; Gabrielsen, Roy Helge & Braathen, Alvar
(2016).
Fault geometry and seismic attributes; an integrated study with focus on the Barents Sea.
First Break.
ISSN 0263-5046.
34(5),
p. 51–58.
Full text in Research Archive
-
Torabi, Anita; Fossen, Haakon & Alaei, Behzad
(2008).
Application of spatial correlation functions in permeability estimation of deformation bands in porous rocks.
Journal of Geophysical Research (JGR).
ISSN 0148-0227.
113.
doi:
10.1029/2007JB005455.
Full text in Research Archive
-
Alaei, Behzad & Petersen, Steen A.
(2007).
Geological modelling and finite difference forward realization of a regional section from the Zagros fold-and-thrust belt.
Petroleum Geoscience.
ISSN 1354-0793.
13(3),
p. 241–251.
Full text in Research Archive
-
-
Alaeimajolan, Behzad & Pajchel, J.
(2006).
Single Arrival Kirchhoff Prestack Depth Migration of complex faulted folds from the Zagros Mountains, Iran.
Recorder.
ISSN 0832-0322.
Jan,
p. 41–48.
Full text in Research Archive
-
Alaeimajolan, Behzad
(2005).
Seismic forward modelling of two fault-related folds from the Dezful embayment of the Iranian Zagros Mountains.
Journal of Seismic Exploration.
ISSN 0963-0651.
14(1),
p. 13–30.
Full text in Research Archive
View all works in NVA
-
Ahmed, Nisar; Cardozo, Nestor; Schulte, Lothar; Alaei, Behzad; Oikonomou, D. & Larsen, E.
(2026).
Deep learning using FCT-network for petrophysical-property estimation from seismic-angle-stacks: 3D application to the Troll Field, North-Sea.
Earth Doc, European Association of Geoscientists & Engineers.
2026(1),
p. 1–5.
doi:
10.3997/2214-4609.202638039.
Full text in Research Archive
Show summary
This work presents a deep-learning method using a Fully Convolutional Transformer (FCT)-based encoder--decoder network for direct estimation of the petrophysical properties such as porosity, water saturation, and shale volume from seismic angle stacks and interval velocity. The FCT network integrates convolutional operations within Transformer blocks to map both local seismic features and global geological dependencies. Uncertainty in the estimated models is quantified using Monte Carlo dropout and ensemble predictions, and the framework is validated at training and blind well locations and across a full 3D seismic cube in the Troll Field, yielding accurate and geologically consistent petrophysical property estimates. The predicted properties are further verified through unseen-well validation and their consistency with seismic signatures, fault patterns acting as seal, and reservoir-scale geological features.
-
Ahmed, Nisar; Cardozo, Nestor; Oikonomou, Dimitrios; Larsen, Eirik & Alaei, Behzad
(2025).
Reservoir Elastic Properties Prediction using Supervised Deep Learning versus Model-Based Inversion Methods.
Earth Doc, European Association of Geoscientists & Engineers.
2025(1),
p. 1–5.
doi:
10.3997/2214-4609.2025101138.
Full text in Research Archive
Show summary
This work focuses on a 3D supervised deep machine learning method for predicting reservoir elastic properties from partial-angle stack seismic data. Using well-log data from several well locations from Troll field in the North sea and 3D interval velocity volume, we train a deep convolutional neural network to estimate elastic properties such as P and S waves velocities and density from seismic partial stacks. Furthermore, we have also make a comparison between the deep learning predictions with the model-based seismic inversion methods including Bayesian linearized inversion and stochastic nonlinear inversion schemes. The applications tests on well logs and seismic field data demonstrate that deep learning method can effectively predict property volumes with good lateral continuity as compared to traditional model based inversion techniques. Estimation of these elastic properties allows obtaining other rock properties such as porosity, saturation, and shale volume, which are important for reservoir rock characterization.
-
Holden, Nora; Alaei, Behzad; Skurtveit, Elin & Braathen, Alvar
(2024).
The impact of depth conversion on fault geometry and its influence on fault-risk assessment in the Smeaheia storage site.
Full text in Research Archive
-
-
-
-
Holden, Nora; Alaei, Behzad; Skurtveit, Elin & Braathen, Alvar
(2023).
Uncertainty in fault-risk assessment - implications of depth-conversion on fault-risk assessment in the Smeaheia CO2 storage site, northern North Sea.
Full text in Research Archive
-
Holden, Nora; Alaei, Behzad; Skurtveit, Elin & Braathen, Alvar
(2023).
Implications of depth-conversion on fault geometries and fault-risk assessment in the Smeaheia CO2 storage site, northern North Sea.
Full text in Research Archive
-
Holden, Nora; Alaei, Behzad; Skurtveit, Elin & Braathen, Alvar
(2023).
Uncertainty in fault-risking.
Influence of depth-conversion on fault-geometry and fault-risk assessment in the Smeaheia CO2 storage site.
Full text in Research Archive
-
Haines, Emma Michie; Braathen, Alvar & Alaei, Behzad
(2022).
How Fault Interpretation Method May Influence the Assessment of a Fault-bound CO2 Storage Site.
Full text in Research Archive
-
Michie, Emma; Braathen, Alvar & Alaei, Behzad
(2021).
How fault interpretation method may influence the assessment of a fault-bound CO2 storage site.
Full text in Research Archive
-
Botter, Charlotte; Cardozo, Nestor; Tveranger, Jan; Qu, Dongfang & Alaei, Behzad
(2016).
Seismic characterisation of fault models populated using fault-facies techniques.
Full text in Research Archive
-
Torabi, Anita; Fossen, Haakon & Alaei, Behzad
(2007).
Permeability estimation of deformed sandstone using image processing.
Full text in Research Archive
View all works in NVA
Published
June 16, 2021 10:15 AM
- Last modified
June 15, 2026 10:48 AM