Ardiansyah Koeshidayatullah

The Jenkyns Event on the Arabian Plate caused distinct environmental and sedimentological shifts compared to deeper Tethyan basins. While deeper basins faced oxygen depletion and organic-rich facies, the shallow Marrat Formation recorded an influx of deltaic terrigenous sediments (red beds) and high-energy shoreline facies during the negative carbon isotope excursion, characteristic of the Jenkyns Event. This event drove a transition from a densely bioturbated carbonate platform with diverse benthic macroinvertebrates and microfossil assemblages (Lower Marrat Member) to mixed wave- and river-dominated deltaic deposits (Middle Marrat Member) with low-diversity benthic communities (macroinvertebrates) and opportunistic monospecific ichnoassemblages dominated by suspension feeders ( Arenicolites and Skolithos ). Post-event, the carbonate platform recovered but retained lower biodiversity than pre-event levels for macroinvertebrates and microfossils. Tectonic stability of the Arabian Plate and an equatorial position primarily controlled this unique regional response.

A bstract: The Eocene Epoch is characterised by multiple hyperthermals, which promoted extensive carbonate sediment deposition and dolomitisation, particularly across the Arabian Plate. The Dammam Formation, a key Eocene aquifer and hydrocarbon‐bearing unit in the Arabian platform, is extensively dolomitised. Several models have been proposed to explain its dolomitisation, many of which invoke sea water or its derivatives as the dolomitising fluid. However, the specific sea water‐driven mechanisms remain poorly understood. More broadly, the effectiveness of commonly used diagenetic proxies (e.g. stable isotopes, stoichiometry and trace elements) in distinguishing between closely related sea water‐driven dolomitisation processes has not been systematically assessed. This study addresses these issues by integrating petrographic, mineralogical, chemical and stable isotopic (δ ¹³ C and δ ¹⁸ O) analyses of a shallow subsurface core from the Dammam Formation. Two distinct dolomite types were identified: Type I (TI), predominantly subhedral and Type II (TII), exhibiting euhedral and subhedral textures. Both types are pervasive, fabric non‐selective and non‐mimetic. The dolomite pre‐dates all other diagenetic processes, such as the formation of palygorskite, calcite, pyrite and stylolite, indicating early, shallow emplacement. Mineralogical data reveal that TI dolomite is near‐stoichiometric (51.4 mol%CaCO ₃ ), and well‐ordered, while TII is non‐stoichiometric (56.3 mol%CaCO ₃ ) and poorly ordered. Trace element concentrations (Fe, Mn, Sr) align with some previously reported sea water dolomites. The δ ¹⁸ O values (−2.59‰ to 0.95‰) and temperatures (24–40°C) suggest formation within Eocene marine conditions, with potential evidence of warmer fluids or recrystallisation. The δ ¹³ C values (0.89–1.83‰) indicate buffering during dolomitisation, consistent with Eocene sea water. Contrary to previous interpretations suggesting sabkha mixing or evaporated sea water, this study proposes that the Saudi Dammam dolomite formed primarily through normal‐to‐mesohaline sea water dolomitisation, driven most probably by reflux of mesohaline fluids. This study adds to the understanding of sea water‐driven dolomitisation, emphasising that recrystallisation can make distinction between mesohaline reflux and syn‐depositional sea water‐driven mechanism, in shallow marine dolomitised carbonates, challenging.

Accurate geological modeling is critical for reservoir characterization, yet traditional methods struggle with complex subsurface heterogeneity, and they have problems with conditioning to observed data. This study introduces Pix2Geomodel, a novel conditional generative adversarial network (cGAN) framework based on Pix2Pix, designed to predict reservoir properties (facies, porosity, permeability, and water saturation) from the Rotliegend reservoir of the Groningen gas field. Utilizing a 7.6 million-cell dataset from the Nederlandse Aardolie Maatschappij, accessed via EPOS-NL, the methodology included data preprocessing, augmentation to generate 2,350 images per property, and training with a U-Net generator and PatchGAN discriminator over 19,000 steps. Evaluation metrics include pixel accuracy (PA), mean intersection over union (mIoU), frequency weighted intersection over union (FWIoU), and visualizations assessed performance in masked property prediction and property-to-property translation tasks. Results demonstrated high accuracy for facies (PA 0.88, FWIoU 0.85) and water saturation (PA 0.96, FWIoU 0.95), with moderate success for porosity (PA 0.70, FWIoU 0.55) and permeability (PA 0.74, FWIoU 0.60), and robust translation performance (e.g., facies-to-facies PA 0.98, FWIoU 0.97). The framework captured spatial variability and geological realism, as validated by variogram analysis, and calculated the training loss curves for the generator and discriminator for each property. Compared to traditional methods, Pix2Geomodel offers enhanced fidelity in direct property mapping. Limitations include challenges with microstructural variability and 2D constraints, suggesting future integration of multi-modal data and 3D modeling (Pix2Geomodel v2.0). This study advances the application of generative AI in geoscience, supporting improved reservoir management and open science initiatives.

Recent advancements in computer vision and deep learning have made a paradigm shift in seismic facies interpretation. Various supervised and unsupervised deep learning-based seismic facies re-search using Convolutional Neural Network (CNN) as the baseline architecture have been proposed and achieved promising results. However, these approaches are often limited by the lack of available high-quality labelled dataset and minimum explainability of the model. This is further compounded by the fact that these models do not allow user-guided interaction, which limits the ability of seismic interpreters to perform localized seismic facies segmentation through prompts. To optimize the already existing approach, we present a time-efficient and high precision alternative by reformulating seismic facies segmentation as a Segment All or Segment One (SASO) task. In this paper, we propose FaciesSAM, based on Fast Segment Anything Model, for seismic facies segmentation. For the first time, our paper highlights that FaciesSAM can help to decouple seismic facies identification into all-instance segmentation (segment all) and prompt-guided selection (segment one) for broad and localized facies interpretation, respectively. The effectiveness of our proposed method was evaluated on the benchmark F3 geological dataset. Experimental results shows that our method is effective, even when trained on fewer seismic images, achieving a mAP0.5 of 83.3% with 3.1% increase in pixel accuracy and ~4.6% increase in mean class accuracy when compared to the CNN benchmark results. Furthermore, our method meets real-time processing speed of 1.14ms per seismic sec-tion. These results underscore the capacity of prompt-based CNN detectors to solving specific geological challenges such as seismic interpretation accuracy and processing speed.