The Lower Sub member of the 4th Member of the Shahejie Formation (Es4x) in the Bonan Depression of the Bohai Bay Basin shows a distinctive continental red bed succession characterized by alternating red and gray layers, mainly composed of red stratigraphy. This differs from earlier views of the unit as being uniformly red and indicates a more dynamic depositional environment than previously thought. A better understanding of the sedimentary structure in this interval is essential for reconstructing the early Eocene paleogeographic and paleoclimatic the Bonan Depression, for detailed understanding of its evolution, and for guiding exploration for deep hydrocarbon resources in the Bonan Depression. This study integrates detailed core observations, mud-logging records, laboratory-based geochemical and petrographic analyses supported by high-resolution seismic data and wireline logs to investigate the depositional processes and sedimentary evolution of the lower succession of the Eocene Shahejie Formation. The findings identify a wide range of depositional facies, such as alluvial fans, braided fluvial, floodplains, ephemeral saline lakes, perennial lacustrine, nearshore subaqueous fans, and storm-influenced lacustrine deposits. These facies indicate diverse hydrodynamic conditions and a dynamic interaction among climate, tectonics, and sediment supply. During predominantly arid climatic periods, alluvial fan–floodplain–saline lake facies associations developed, with coarse-grained alluvial fan deposits along basin margins transitioning basinward into fine-grained floodplain and ephemeral saline lake sediments. Intermittent humid phases promoted the expansion of perennial lakes, resulting in the deposition of gray-green bar sands and interbar mudstones along gentle slopes, while the northern basin evolved into semi-deep to deep lacustrine environments. Seismic and sedimentological evidence further indicates the development of subaqueous fan systems within incised gullies along steep northern slopes, controlled by syn-depositional tectonism and episodic high-energy events. The refined depositional model highlights the critical role of ephemeral lakes in the formation of continental red beds, providing new insights into early Eocene paleoclimate variability and offering a valuable framework for red-bed reservoir prediction and hydrocarbon exploration in the Bohai Bay Basin.
The Neogene Kechabta foreland basin of northern Tunisia contains well-exposed Messinian outcrops, including the Oued Bel Khedim (OBK) and Chaabet Et Tabbela (CET) formations. Through bed-by-bed logging and comprehensive analyses that combined biostratigraphy, sedimentology, and petrography, the age and depositional environments were established. The Messinian succession of the studied sections includes four distinct lithological units that have produced a rich and well-preserved ostracod fauna. Species such as Cyprideis ruggierii, Callistocythere antoniettae, Dorukella aff. bireticulata, Leptocythere sanmarinensis, and Costa batei collected from the OBK1 and OBK2 units provide biostratigraphic evidence supporting an Early Messinian attribution for the first time. In contrast, the OBK3 unit of the Oued Bel Khedim Formation, along with the overlying CET unit of the continental Chaabet Et Tabbela Formation, exhibits a significant fauna scarcity, tentatively indicating a Late Messinian age for these units. The facies analysis enabled the characterisation of eight distinct facies (F1-F8) organised into four associations (FA1, FA2, FA3, and FA4), representing different depositional environments. FA1 and FA2 include gypsum, clay, as well as oolitic and peloidal limestone mixed with predominantly clay-rich sediments containing both brackish and marine benthic foraminifers (Ammonia tepida and Elphidium, along with ostracods including Cyprideis, Loxoconcha, and Xestoleberis). These sediments likely formed in shallow lagoonal settings that experienced fluctuations between restricted and fully marine conditions. In contrast, FA3 features varved micritic to peloidal limestone and gypsum/clays, with very few bivalves and ostracods, and was formed in a saline lake environment that later shifted to a sebkha setting. Lastly, FA4, composed of conglomerates and coarse-grained sandstones, indicates continental environments dominated by alluvial fan and fluvial deposits. The correlation of lithostratigraphic units and their associated facies across the Kechabta foreland basin, located in front of the Maghrebian chain, allows reconstruction of the depositional history and emphasises the main local tectonic and regional eustatic factors that influenced the evolution of the Mediterranean basin during the Messinian. The revised stratigraphic framework for Messinian deposits in the studied section, when compared with the Gulf of Tunis and the Cap Bon Peninsula in northeast Tunisia, suggests that the three-stage model consisting of Lower Evaporites, Upper Evaporites/Terminal Carbonate Complex, and Lago Mare, which is proposed for the central Mediterranean basin, is applicable in northern Tunisia. This study also shows that the Messinian deposits in the Kechabta foreland basin, despite being formed at the foot of the Maghrebian chain and in a proximal margin, may record several major phases associated with the Messinian salinity crisis across the Mediterranean.
The Ediacaran to Early Cambrian sedimentary succession is well-preserved in the Lesser Himalayan sequence of Garhwal, Uttarakhand. The Upper Krol (Kaudiyala Formation) is characterised by phosphatic nodules occurring within calcareous shale, limestone, dolomite, siltstone, and quartz arenite. In contrast, the Lower Tal (Deo ka Tibba Formation) consists of bedded chert interlayered with black shale and phosphatic horizons. Petro-mineralogical and SEM analyses of the phosphate-bearing rocks reveal that apatite is the main phosphate mineral, accompanied by gangue minerals such as calcite, dolomite, quartz, hematite, pyrite, and organic matter. A significant values of P₂O₅ and CaO indicate the presence of apatite, calcite, or gypsum. The inverse relationship between P₂O₅ and major oxides (SiO2, Al2O3, Fe2O3, MgO, TiO2) suggests processes like replacement, leaching, and weathering, during which P₂O₅ fills voids, cavities, and veins. Trace element data show that phosphate-bearing sedimentary rocks are rich in V, Cr, Cu, and Y, and elements like Co, Ga, Nb, and Hf appear in lower concentrations. The high V content may be connected to organic matter because of its strong association with P₂O₅. Vanadium (V) co-occurs with Ni, Cr, Zn, Cu, and Ag, a pattern typically observed in the organic matter of marine sediments. Low Ni/Co ratios (< 5) in phosphate-bearing sedimentary rocks suggest deposition under oxic conditions. This interpretation is further supported by the binary relationship between V/(V + Ni) and Ni/Co. In the present dataset, most samples display V/Cr ratios below 2, and only few exceed this value, implying that the deposition predominantly occurred under oxic conditions with limited influence of reducing environments. Compared to UCC, phosphatic black shale and chert samples show marked enrichment of V and Mo, as well as moderate enrichment of Cr, Cu, and Pb, suggesting highly oxygen-depleted deep waters. These phosphate-bearing sedimentary rocks also exhibit a negative Ce anomaly (average Ce/Ce* = 0.53), indicating the influence of oxidized surface water.
This study evaluates the susceptibility of collapse dolines in the Sahel-Doukkala region using a combination of machine learning (Random Forest) and statistical models (Analytical Hierarchy Process, Shannon Entropy, and Weight of Evidence). The models were built from an inventory of 58 collapse dolines and six conditioning factors relevant to karst development: lithology, distance to faults and underground drainage axes, slope, aspect, and elevation. While this inventory provides a valuable basis for analysis, its relatively small size (n = 58) may influence the robustness and generalizability of the modeling results. Model performance was evaluated using spatial cross-validation, a method that partitions data into geographically independent subsets to reduce spatial bias in model assessment. RF achieved the highest median predictive performance (ROC-AUC ≈ 0.74), followed by WOE (≈ 0.71), SE (≈ 0.69), and AHP (≈ 0.67). These AUC values indicate moderate predictive performance, with RF showing comparatively better discrimination between collapse and non-collapse areas. High-susceptibility zones are mainly concentrated along the Sahel coastal corridor, where highly soluble limestones, dense fault networks, and focused groundwater flow coincide. Factor contributions were quantified using permutation importance and SHAP analysis. To support reproducibility and risk-informed decision-making, the code and datasets are made publicly available, and uncertainty and calibration maps accompany the categorical susceptibility outputs.
Karst aquifers are highly susceptible to contamination and the application of methodologies that identify these critical areas is an excellent strategy for promoting their conservation. With this in mind, this study aims to assess the applicability of two aquifer vulnerability classification methods, COP and EPIK, in the Terra Ronca State Park (PETeR) and Serra Geral APA conservation units, located in Central Brazil. The chosen methods assess the physical conditions of the environment (relief, lithology, soils, karst features, vegetation) to determine the areas naturally vulnerable to contamination. The analyses were carried out in a GIS environment. The results for the COP method showed a predominance of areas with low (43.08
Tight sandstone hydrocarbon resources are strategically important during the global energy transition. However, their small pore-throat sizes, complex pore-throat architectures, and strong heterogeneity produce large variations in seepage capacity and make efficient development difficult. The Chang 6 reservoir in the Ordos Basin is a representative low-permeability tight sandstone reservoir that requires detailed pore-structure characterization. In this study, mercury intrusion porosimetry (MIP), scanning electron microscopy (SEM), and nuclear magnetic resonance (NMR) were integrated to characterize pore morphology, pore-throat distribution, and fluid occurrence. Compared with studies that mainly combine MIP, SEM, and NMR qualitatively, this work establishes a quantitatively coupled classification framework using K-means clustering and fractal descriptors, thereby enabling objective pore-type division and cross-method consistency analysis. Multi-scale characterization indicates that the Chang 6 tight sandstones can be classified into four pore-structure types. Type IV, defined as the high displacement pressure-fine throat type, is dominant in the study area. The NMR T₂ spectra show a bimodal distribution dominated by small pores, and movable fluid saturation is generally lower than 50
Karst aquifers are critical freshwater resources developed in soluble rocks such as limestone and dolomite, characterized by complex hydrological features including conduits, fractures, and sinkholes. Over the past decade, numerous modeling approaches have been applied to investigate different aspects of karst aquifer behavior. This review synthesizes recent advances in modeling techniques used to describe karstification processes, recharge, groundwater flow, storage, and discharge, while highlighting the challenges posed by heterogeneous porosity and fracture–conduit systems. A wide range of modeling approaches is reviewed, including conceptual, laboratory-scale, analytical, numerical, and data-driven models. These approaches aim to represent preferential flow paths, groundwater movement, and contaminant transport under varying hydrological conditions. Numerical studies employing both functional and physically based models are critically evaluated in terms of their ability to simulate karst hydrodynamics using geological, hydraulic, and hydrochemical parameters. Model strengths and limitations are discussed with respect to data availability, spatial scale, and dominant flow processes. Recent developments in machine learning and advanced data analytics are reviewed as complementary tools rather than standalone solutions, with particular emphasis on hybrid and theory-guided modeling frameworks. These approaches show promise in improving predictive capability while preserving physical interpretability. This review highlights current challenges and emerging opportunities in karst aquifer modeling and provides guidance for selecting appropriate modeling strategies to support groundwater management, contamination risk assessment, and sustainable use of karst water resources.
The Cambrian carbonate rocks exhibit widespread distribution in the Yiyuan area, constituting a prominent strontium–enriched mineral water zone in central Shandong Province (strontium concentrations up to 35.22 mg/L). To clarify the hydrogeochemical characteristics and the formation mechanisms of the Sr-rich mineral water, this study systematically elucidated them by integrating hydrogeological surveys, geochemical analyses, hydrogeochemical modeling, ion ratio analysis, and dynamic monitoring datasets. Major hydrogeological units, including the Sancha and Tumen units, were delineated. Results indicate that groundwater in the study area is predominantly fresh (mean total dissolved solids (TDS) 531.05 mg/L) with neutral to slightly alkaline pH (mean 7.67 ± 2.45
This study investigates karst collapse induced by water inflow in the Chaoyang Tunnel, Guizhou Province. Through field investigations and tracer tests, the study ascertains the hydrogeological characteristics of the tunnel and its surrounding terrain, analyzing the impact of sudden water inflow in karst tunnels on the development of surface collapse. The following findings include: (1) confirmed direct hydraulic connectivity between the tunnel and the Dongdiu collapse zone, with abrupt water inrush causing rapid groundwater decline and altering hydrodynamic conditions in karst conduits; (2) multi-peak tracer curves reveales complex internal conduit networks and localised water storage structures, providing new insights into subsurface flow paths; (3) karst strata and fault structures serve as the primary internal controls on collapse development, while extreme rainfall acts as a major external driver that accelerates overburden erosion through rapid groundwater fluctuations, linear conduit-induced hydraulic scouring, and cyclic desiccation-saturation under dual water table disparitie; (4) the collapses can be classified as a drainage-precipitation type governed by subsurface erosion. The study innovatively combines tracer tests and hydrogeological modeling to quantify conduit structures and their role in collapse initiation, providing a predictive framework for karst tunnel hazards.
Accurate horizontal-stress prediction in carbonate reservoirs is challenging because mechanical properties commonly vary over short vertical intervals as a result of lithologic heterogeneity, diagenesis, pore development, and fracture distribution. This study develops a facies-controlled geomechanical workflow for the Lower Cambrian Canglangpu Formation in the Sichuan Basin, using integrated mineralogical, petrographic, image-based, acoustic, strength, Kaiser-effect, and well-log data from four wells. The formation was partitioned into three mechanically meaningful facies because the laboratory and image-derived data demonstrate that it is mechanically nonuniform and cannot be reliably represented by a single linear elastic transform or one interval-wide model. The defined facies include a carbonate-dolomite compact to moderately porous facies, a mixed sandy-dolomitic dissolution facies, and an argillaceous tight mixed facies, each reflecting distinct local mechanical behavior. Facies-specific static properties and dynamic-to-static transforms were assigned from triaxial and uniaxial compression data where available, while Kaiser-effect stress measurements were used to calibrate horizontal strains within a poroelastic framework. The calibrated model was then extended continuously through conventional well logs to predict maximum and minimum horizontal stresses. Results show that facies partitioning provides a geologically defensible means of capturing local stiffness and stress variations in a heterogeneous carbonate interval, especially where laboratory data are unevenly distributed among facies. Sensitivity analysis further indicates that variations in the assumed Biot coefficient affect stress magnitudes but do not change the overall facies-controlled stress architecture. The proposed workflow therefore offers a practical approach for stress evaluation and borehole-stability screening in mechanically heterogeneous carbonate reservoirs.
Dissolved inorganic carbon (DIC) is a key concentration term in hydrochemical-runoff estimates of karst carbon sinks, but dense and accurate DIC observations are difficult and costly to obtain. In this study, 692 groundwater and spring samples from the karst region of Guangxi, China, were combined with multi-source environmental predictors to develop a machine-learning framework for regional DIC mapping. Five machine-learning models were evaluated, and the three best-performing models (XGBoost, BRT, and RF) were further integrated using a stacking ensemble. The stacked model achieved a test-set R2 of 0.83, with an RMSE of 0.45 mmol/L, MAE of 0.30 mmol/L, and MSE of 0.20. Predicted DIC showed a persistent spatial pattern. The northwestern regions revealed high values with a maximum of 3.83 mmol/L in 2015, while the southeastern region was low with a minimum of 3.48 mmol/L in 2004. The distribution of DIC concentration is governed by several factors, with elevated concentrations in the southwest region, forested areas, and higher altitudes. The proposed model provides a reliable basis for regional DIC mapping and can support subsequent karst carbon sink estimation when combined with hydrological flux data.
A comprehensive understanding of hydrogeochemical characteristics, controlling mechanisms and quality status of mine water is critical for its safe utilization in arid mining areas. In this study, 26 mine water samples from underground dewatering boreholes and 2 surface water samples were collected in July 2024 from the Jinjie Coal Mine, Yushen Coalfield, Northwest China. Descriptive statistics, correlation analysis, Piper diagram, Gibbs diagram, Chadha plot and ion ratio methods were integrated to investigate hydrochemical characteristics, water-rock interactions and water quality suitability. The findings indicated that the pH varied from 8.07 to 8.49, with a mean of 8.29, and the electrical conductivity (EC) ranged from 279.00 µS/cm to 352.00 µS/cm, with an average of 316.85 µS/cm. In addition, the total hardness (TH) was in the range of 115.78-160.53 mg/L, and the total dissolved solids (TDS) was between 220.54 mg/L and 292.61 mg/L. The anion concentration was determined as follows: HCO3−> SO42−> Cl− while the cation content was ranked as Ca2+> Na++K+> Mg2+. The main hydrochemical type was HCO3-Ca. The dissolution of silicate and carbonate minerals, and the cation exchange process were the main processes regulating the hydrochemistry of the mine water. It was suitable for agricultural irrigation but not for human consumption due to its high NO3− concentration (which ranged from 7.47 to 15.86 mg/L, with a mean of 12.32 mg/L, and was greater than 10 mg/L according to the Chinese groundwater standard (GB/T 14848–2017)). The findings are significant for the better management and more effective utilization of mine water in China and in other arid and semi-arid mining areas worldwide.
The Sirban Dolomite of the Hazara Basin, Pakistan, records a complex diagenetic evolution that strongly controls its reservoir quality. This study integrates field observations, petrography, stable isotope geochemistry (δ¹⁸O, δ¹³C), X-ray diffraction (XRD), and petrophysical analysis (JMicrovision) to reconstruct dolomitization history and porosity evolution. Data are derived from measured stratigraphic sections of 250 m and 240 m, 40 thin sections, and multiple isotope samples. Field observations reveal bedding-parallel stylolites indicative of burial compaction, varied chert associated with silica-rich conditions and syn-sedimentary exposure, algal stromatolites reflecting biological activity, brecciation linked to tectonic episodes, hematitic beds and quartzites suggesting subaerial exposure, and upwelled phosphatic horizons redeposited on shallow marine shelves. Petrographic analysis identifies five matrix dolomite phases (MD-1 to MD-5), ranging from early cryptocrystalline dolomicrite (MD-1), typically associated with stromatolites, chert, and phosphates in intertidal to supratidal settings, to coarse crystalline and anhedral dolomite (MD-5) formed during deep burial. Intermediate phases include fabric-retentive dolomites (MD-2) and transitional recrystallized dolomites (MD-3), preserving ghost textures such as peloids. In addition, five generations of dolomite cement (including zoned and saddle dolomites), along with detrital quartz, syntaxial calcite, and telogenetic calcite, are recognized. XRD analysis confirms dolomite as the dominant phase with minor calcite, quartz, and gypsum, showing variations in stoichiometry and ordering that reflect early to shallow burial dolomitization. Fabric-retentive and coarse dolomites indicate a progression from near-surface to intermediate burial diagenetic environments. Stable isotope data indicate progressive fluid evolution. Early dolomicrite (MD-1; δ¹⁸O: − 5.71 to − 3.23‰ VPDB; mean − 3.85‰) preserves a Cambrian marine signature, suggesting formation in restricted lagoonal to tidal-flat environments via reflux-seepage processes. Fabric-retentive dolomite (MD-2; δ¹⁸O: − 6.84 to − 5.65‰ VPDB) overlaps with marine values, indicating shallow burial diagenesis. Transitional dolomite (MD-3; δ¹⁸O ≈ − 5.46‰ VPDB) reflects recrystallization of earlier phases by Mg-rich pore fluids. Later dolomite phases and cements exhibit more depleted values (δ¹⁸O down to − 9.37‰; δ¹³C: − 3.87 to − 0.39‰ VPDB), consistent with burial-modified Mg-rich basinal brines. XRD analysis indicates dolomite stoichiometry ranging from 48.43 to 54.66
Tectonic fractures serve as critical storage and flow pathways in carbonate reservoirs, with their development directly governing hydrocarbon enrichment and fluid migration. However, prediction accuracy is often limited by the simplified treatment inherent in traditional homogenized models, hindering the exploration of deep carbonate reservoirs. To fill this gap, this study innovatively constructs a 3D heterogeneous rock mechanics model coupled with an adaptive boundary constraint algorithm, and applies it to the Ordovician fractured reservoirs in the deep coverage area of the Tahe Oilfield, Tarim Basin. High-precision paleo-stress field simulations were conducted for three key tectonic periods (Middle Caledonian, Early Hercynian, and Neotectonic), significantly improving the accuracy and objectivity of stress field characterization. Fracture development intensity is quantitatively predicted based on rock failure criteria and validated against core observations and production data. The results demonstrate that: (1) The heterogeneous rock mechanical model coupled with adaptive boundary conditions significantly improves the reliability of stress field simulations. (2) Fracture development exhibits a strong correlation with the paleo‑stress field, with fractures predominantly concentrated around fault peripheries, tips, and intersections. (3) Fracture intensity shows a pronounced positive correlation with both the horizontal stress difference (∆σh) and the stress heterogeneity coefficient (k) under the geological conditions of the study area; this correlation statistically tends to strengthen once these parameters exceed certain thresholds identified by segmented regression. (4) Predicted fracture linear density matches well with cumulative oil production from individual wells, confirming that the fracture system is a key controlling factor for hydrocarbon enrichment in deeply buried carbonate reservoirs. This study provides a robust methodological framework and practical guidance for quantitative fracture prediction and efficient exploration in carbonate reservoirs. This study provides a robust, repeatable method for quantitative fracture prediction and offers important guidance for the efficient exploration and development of deep carbonate reservoirs worldwide.
The fractures developed during the strike-slip faulting activity represent an important type of carbonate reservoir, playing a significant role in ultra-deep oil and gas exploration. At present, the scale and distribution pattern of fractures associated with strike-slip fault in the Shunnan area are unclear. This study focuses on the SB16 fault in the Tarim Basin as the research target, conducting finite element numerical simulations based on 3D seismic interpretation and combined with rock mechanics experiments to clarify the stress field distribution characteristics. Using stress-strain as a bridge, explore the influencing factors and scope of fracture development associated with strike-slip faults, and predict the distribution pattern of fractures. The research results show that in the middle of the Caledonian, the direction of σH is NE18°–NE34°, and the value of σH is 59–66 MPa. In the late Caledonian early to Hercynian period, the direction of σH is NW16°–NW32°, and the value of σH is 80–88 MPa. The development of fractures in carbonate reservoirs exhibits significant heterogeneity, distributed along strike-slip faults and widely distributed in fault overlap areas. The fractured layers exhibit characteristics of low Young’s modulus and high Poisson’s ratio. The fractures have the largest distribution range in pull-apart segment, the tensile fracture system is conducive to the formation of large-scale reservoirs through fluid transformation. The distribution range of fractures in the uplift segment is second among the three segments, with the fracture system primarily developing along the fault plane. The translation segment has the smallest fracture distribution range, with the fault system developing along the fault plane or its lateral sides.
Understanding lithology and flow unit characterization is essential for effective reservoir evaluation, particularly in complex clastic systems such as the Lower Cretaceous Yageliemu Formation of the Yakela gas condensate field, Kuqa Depression, Tarim Basin, China. This study integrates traditional crossplot techniques with machine learning-based lithology classification and petrophysical evaluation to characterize reservoir heterogeneity. Mineralogical analysis using M-N crossplots identifies a dominantly quartz-rich clastic matrix with subordinate shale content, and this result establishes the compositional basis for lithological interpretation. Lithological classification uses mineralogical constraints and applies diagnostic crossplots and K-means clustering to define four facies, which include clean sandstone, clayey sandstone, shaly sandstone, and shale. This classification provides a framework for evaluating petrophysical variations across reservoir facies. Petrophysical analysis shows significant differences in reservoir quality among facies. Clean sandstone (HFU-01) exhibits the highest reservoir quality with effective porosity of 9.9
Calcite veins are crucial recorders of tectonic-fluid interactions, preserving information on fracture timing and fluid evolution. This study investigates calcite veins within Carboniferous volcanic rocks near Well Pen-1 in the Junggar Basin to decipher the coupling between regional tectonic events, fluid dynamics, and diagenesis. Through comprehensive geochemical and geochronological analyses, including in-situ U-Pb dating, isotopic compositions, and fluid inclusion studies, we identify two distinct stages of calcite precipitation driven by different tectonic regimes. Stage I veins, dated to the Late Permian (257 ± 3.6 Ma), formed contemporaneously with Late Hercynian extensional tectonics. This tectonic setting activated a volcanic-hydrothermal system. The veins’ geochemical signatures—a LREE-enriched pattern with positive Ce and Eu anomalies, high δ¹³C values (− 2.87‰ to 0.39‰), and low ⁸⁷Sr/⁸⁶Sr ratios (0.7033 to 0.7075)—confirm their precipitation from mantle-derived fluids under relatively reducing conditions. This stage documents the direct control of extensional volcanism on the initial fluid activity in the basin. Stage II veins, dated to 137 ± 5.7 Ma (Early Cretaceous), formed under an entirely different tectonic control: the Yanshanian transpressional regime. This intense tectonic activity reactivated fault systems, creating conduits that connected deep crustal fluids with maturing hydrocarbon source rocks. The resulting fluid mixing is recorded in the veins’ distinct geochemistry: a right-inclined REE pattern with a positive Eu but negative Ce anomaly, significantly lower δ¹³C (− 7.08‰ to − 2.14‰) and δ¹⁸O (− 12.96‰ to − 8.54‰) values, and higher ⁸⁷Sr/⁸⁶Sr ratios (0.7045 to 0.7064). These characteristics reflect the incorporation of organic matter and basinal brines, indicating precipitation from hydrocarbon-related fluid systems. In conclusion, the two vein stages provide a direct record of the basin’s tectono-fluid evolution. Stage I captures the signature of Late Hercynian volcanic-hydrothermal activity, while Stage II documents a complex fluid system during the Yanshanian orogeny, where tectonic reactivation facilitated the mixing of deep fluids with migrating hydrocarbons. This study demonstrates that calcite veins serve as an excellent archive for reconstructing how distinct tectonic events govern fluid composition, diagenetic processes, and ultimately, reservoir formation in volcanic basins.
The unique geographical setting of the South China Sea has made it a natural laboratory for dolomite research within reef systems with its extensive reef carbonate sedimentary deposits. Well Xike-1 is a typical example of deep dolomite reservoirs in this region, featuring tight dolomite with low porosity and low permeability. It would be critical to understand the difference between such tight dolomite and conventional dolomite, both theoretically and practically. Multiple approaches were adopted in this study, including core observation, microscopy, scanning electron microscopy (SEM), major and trace element analysis, rare earth element (REE) analysis, and stable isotope (C, O) analysis. The results indicate that the target interval comprises residual bioclast dolomite and crystalline dolomite. These crystals range from silt-sized to fine-grained with subhedral to anhedral forms, and their contacts vary from point and planar to densely packed. Pores are mostly isolated without effective connectivity. The mineral phases mainly include calcite, dolomite and ankerite, with generally low concentrations of Mn and Sr and a relatively high Sr/Ba ratio. The high values and weak correlations of carbon and oxygen isotopes suggest low temperature and high salinity in the diagenetic fluid. The rare earth elements show typical marine patterns, with negative Ce anomalies and weak positive Eu anomalies. It can be concluded that the dolomite in the target interval was developed in a marine-sourced, high-salinity lagoon environment, where the diagenetic setting and processes determine its reservoir properties. Diagenesis of the dolomite in the region occurred in multiple stages under relatively closed conditions. In this process, overdolomitization led to multi-phase alteration by contemporaneous dolomitizing fluids, which filled the pores formed at earlier stages and ultimately led to reservoir densification.
