Skip to content
Log in

Similar papers

Open access Jul 2026

Characterization of stratigraphic interfaces and embedded limestone cavities using integrated geophysical and geological data

Accurate identification of stratigraphic interfaces and embedded cavities is critical for geological modeling, engineering design, and infrastructure maintenance. However, integrating geophysical and geological data remains challenging due to discrepancies in spatial resolution, signal-to-noise ratio, and geological representation. This study presents a stratigraphic U-Net model that combines geophysical and geological information to improve subsurface interpretation. An initial U-Net model is trained using a preliminary P-wave velocity model derived from conventional seismic inversion. The predicted interfaces, however, show noticeable discrepancies with borehole observations because of manual velocity picking and limited borehole data. To address these limitations, the stratigraphic U-Net model is retrained using the initial predictions together with both drilled and synthetic boreholes. This iterative strategy significantly improves the identification of stratigraphic interfaces, particularly at greater depths. The stratigraphic U-Net model is further integrated with an opening detection model to identify embedded cavities within geological layers. Model evaluation shows that geological diversity captured by borehole data contributes more to prediction accuracy than simply increasing the borehole number. A field case study demonstrates that the proposed framework accurately identifies stratigraphic interfaces and subsurface cavities, providing a robust workflow for reducing geological uncertainty and improving the spatial coverage and reliability of subsurface characterization.

Wenzhao Meng, J. Chong, Wei Wu · 0 citations
Open access Aug 2026

INTEGRATED GEOMECHANICS AND SEISMIC INVERSION WORKFLOW FOR PORE-PRESSURE PREDICTION AND RESERVOIR CHARACTERISATION USING TREND-KRIGING IN DEEPWATER BASINS

Geological complexity, limited well control, and the inability to accurately predict pore pressure and characterise reservoirs using traditional stand-alone seismic and/or well-log techniques continue to pose challenges in deep-water basins. In this study, an integrated geomechanics and seismic inversion workflow has been developed for pore-pressure prediction and reservoir characterisation, using the Bonga Field of the deepwater Niger Delta as the study area and incorporating the trend-kriging technique. The methodology integrated well-log analysis, Eaton and Bowers pore-pressure prediction models, seismic velocity inversion and trend-kriging interpolation in a structurally constrained geostatistical framework. The results indicated the mean absolute error of the Eaton method was 0.15 SG in the upper overpressure zone, while the Bowers method yielded better accuracy (0.09 SG) in the deeper unloading-dominated zone. Trend-kriging showed good predictive performance (R = 0.87; MAE = ±65 m/s), providing a very good integration of well and seismic data in terms of spatial prediction. The study shows that the recommended workflow can help to improve subsurface characterisation, increase drilling safety, and aid in optimised reservoir development. It suggests increasing the size of calibration seismograms, using hybrid machine learning algorithms and including 4D geomechanical monitoring to further enhance predictive quality and reservoir management in deep water.

Tamunosiki Dieokuma, Lawson-Jack Osaki · 0 citations
Aug 2026

Autotie: A partial automation of the seismic to well tie with matching region estimation

Tying seismic data to well logs is a critical step in seismic amplitude processing and interpretation. This process is often repeated throughout the exploration phase to enhance the understanding of well characteristics, such as the time-depth relationship. However, the presence of data noise and inherent uncertainties makes the well-tie task time-consuming and complex. It requires the estimation and careful control of several parameters to ensure reliable results. We propose a workflow based on segmented global optimization that partially automates the well-seismic tying process by estimating the matching position between the synthetic and seismic traces and automatically aligning them. The matching position is estimated using Time-Weighted Dynamic Time Warping (TWDTW), which accounts for the maximum allowable perturbation in the velocity log. In conjunction with TWDTW, a global Bayesian optimizer estimates the tie-related parameters. Following this step, the alignment is performed through an optimization-based method that incorporates segmentation to estimate both the velocity log perturbation and wavelet phase, constrained by a predefined tolerance. The well logs and seismic traces are segmented, and each segment is aligned using constrained Dynamic Time Warping (DTW). This segmentation increases the number of knots available for interpolating the velocity perturbation curve, resulting in a more detailed and refined alignment. The proposed partial automation improved the correlation in all six wells evaluated, with four wells achieving correlation values above 70%, without introducing physically unrealistic distortions.

Rafael da Costa Silva, Marcelus Glaucus de Souza Araújo, Luiz Antonio Rozendo et al. · 0 citations
Jul 2026

From relative to absolute: A practical and robust workflow for low-frequency model building

Because seismic data are band-limited and lack low-frequency components, estimation of elastic properties through seismic inversion requires the construction and incorporation of a low-frequency model (LFM). However, building a reliable LFM is often challenging. Consequently, seismic inversion results are sometimes interpreted in relative terms and are strongly influenced by thickness variations and contrasts in overburden and underburden properties. To address this limitation, I propose a simple and transparent workflow for constructing robust LFMs. The method begins with an initial background model (BGM) built from well data and interpreted seismic horizons, followed by correction of the BGM within the reservoir interval through a combination of seismic inversion results and wedge model-derived results. The workflow corrects thickness and background property effects without requiring explicit prior assumptions on reservoir properties such as net-to-gross ratio or fluid. Synthetic tests demonstrate that the method compensates for these effects and improves the prediction of absolute elastic properties by recovering the missing low-frequency component. Field application in the Northern North Sea demonstrates the practical value of the method, showing improved property predictions relative to commonly used deterministic LFM construction methods. Owing to its simplicity and transparency, the proposed workflow complements more complex inversion methods such as probabilistic inversion by offering a practical alternative during exploration and early appraisal stages, while also serving as a useful benchmark for quality control in probabilistic inversion during development and production stages.

T. Yamatani · 0 citations
2026

Physics Prior Constrained ResUNet-BiMamba Network for Prestack AVO Inversion

Prestack seismic data contain rich angle-dependent amplitude variation information, which provides an important basis for the simultaneous inversion of P-wave velocity, SV-wave velocity, and density. However, prestack inversion based on the Aki-Richards approximation is applicable only to relatively small-to-moderate incidence angles (<30°). Although the exact Zoeppritz equation is valid for gathers with relatively large incidence angles, it couples the reflection and transmission of PP- and PS-waves and establishes a strongly nonlinear relationship between the PP-wave reflection coefficient and the three parameters, which is difficult to solve using conventional methods. Building on the exact Zoeppritz equation, this article proposes a nonlinear prestack three-parameter inversion method based on a Residual U-Net (ResUNet)-BiMamba network. A bidirectional Mamba structure is introduced into prestack seismic inversion and combined with the multiscale feature extraction capability of ResUNet, effectively suppressing stripe artifacts in the inverted profiles and improving interface continuity in structurally complex areas. The forward modeling process based on the Zoeppritz equations is embedded into the hybrid network, and a masking function and Huber loss function are introduced to reduce the influence of strong-amplitude anomalies on the inversion results. A semi-supervised joint loss function is constructed by combining a low-frequency consistency term with a well log supervised loss, thereby reducing the dependence on labeled data. Tests on the Marmousi2 model and field data demonstrate that, compared with the Mamba and BiMamba network models, the ResUNet-BiMamba network model produces superior inversion results in terms of stratigraphic boundary characterization, lateral continuity, and identification accuracy.

Ying-Chia Huang, Dongyong Zhou, Jun-Yi Liao et al. · 0 citations
#software testing Review Open access Oct 2026

3D models from geological maps: strengths and weaknesses from the Pasubio Massif (Southern Alps)

3D geological models provide digital representations of subsurface architecture and are increasingly applied in both academic research and industry. Their construction can follow implicit or explicit approaches, depending on data availability and final modelling applications. While 3D models based on subsurface data are traditionally employed in several fields of study, models derived from outcrop data are increasingly being developed. In this study, newly acquired geological mapping data from the Pasubio Massif (Southern Alps, northern Italy), covering an area of ~36 km² within the geographic extent of the CARG sheet 081 “Rovereto”, were used to develop and test an iterative explicit workflow for 3D geological modelling using Move software. The workflow is based on the construction of a structured grid of geological cross-sections, followed by the interpolation of stratigraphic horizons and fault surfaces through Ordinary Kriging. Field observations indicate that the lithostratigraphic units cropping out in the study area form a gently NW-dipping monoclinal structure, exhibit overall constant thicknesses and are affected by the Schio-Vicenza fault system. Model validation first involved a qualitative comparison between mapped geological boundaries and faults with those obtained from the intersection of the modelled surfaces with the topography (DEM), followed by thickness maps evaluation as an internal consistency check. Where inconsistencies emerged, cross-sections were refined and the model was iteratively updated until geological geometries and thickness trends became consistent with field-mapped evidence. A final quantitative assessment was then performed by analysing thickness deviations from mean unit thicknesses and by measuring the spatial overlap between mapped and model-interpolated geological boundaries and fault traces. Comparison between the preliminary and validated models, supported by thickness deviation statistics, indicates that most residual discrepancies are constrained within ±10–20% of expected thickness values. The results highlight the critical role of iterative validation in ensuring geologically robust 3D models, emphasising common sources of uncertainty in explicit geomodelling workflows. This study provides a methodological framework for producing reliable 3D geological models in data-poor regions, complementing the recently published ISPRA guidelines for the organisation and standardisation of model datasets and supporting future applications in academic research and regional or national geological surveys.

Niccolò Coccia, F. Carboni, M. Marini et al. · 0 citations