Laminae play a critical role in controlling the enrichment and high yield of shale oil through the pores within individual laminae and the microfractures between different laminae. Due to the lack of effective methods, research on the contribution of laminar fractures to shale reservoirs has been delayed, primarily relying on indirect evaluation through comparisons of the reservoir properties of laminar shale and massive shale. This has hindered a comprehensive understanding of the role of laminar fractures in the storage and migration of shale oil. Using Shahejie shale from the Jiyang Depression as a case study, this research employs various imaging technologies, including scanning electron microscopy (SEM), Modular Automated Processing System (MAPS), quantitative mineral evaluation using scanning electron microscopy (QEMSCAN), and focused ion beam scanning electron microscopy (FIB-SEM), to conduct both two-dimensional and three-dimensional imaging-based quantitative assessment of the pore structure within individual laminae. The surface porosity in various laminae is ranked as follows: felsic lamination > granular calcite lamination > mixed lamination > clay-organic matter lamination > dolomite lamination > sparry calcite lamination > organic matter lamination. A similar trend in pore size, porosity, and pore connectivity within laminae was observed using SEM and FIB-SEM imaging. Matrix porosity was quantified through binary image processing, and the storage capacity of laminar fractures was assessed by combining this with the measured porosity. Laminated fractures contribute approximately 36.03% to shale storage capacity under laboratory conditions. However, under in-situ conditions, the porosity of laminated fractures decreases from 1.51% to 0.44%, reducing their contribution to total porosity to 10.61%. This study, through extensive image analysis and comparisons, fills the gap in directly quantifying the contribution of laminar fractures to storage capacity. It provides valuable insights into why laminar shales are viable exploration and development targets, offering new perspectives on the differences in hydrocarbon enrichment across various types of laminar shales.

The deep Ordovician reservoirs within the Tuoputai area, Tarim Basin, exhibit a complex geological setting involving multi-stage hydrocarbon generation and charging. Traditional methods for reconstructing the hydrocarbon accumulation process, such as fluid inclusion analysis combined with basin modeling, are often unreliable due to controversial burial and thermal history. Recent advances in in-situ calcite U-Pb dating and crude oil Re-Os dating have provided precise chronological constraints on hydrocarbon charging and generation. In this study, we synergistically combined traditional fluid inclusion analysis with advanced in-situ calcite U-Pb dating and crude oil Re-Os dating, focusing on crude oil samples and calcite fabrics hosting oil inclusion, to quantitatively constrain the timing of oil charging and generation and then reconstruct the hydrocarbon accumulation process. Integration of fluid inclusion data with basin modeling revealed three distinct charging episodes: Late Caledonian, Late Hercynian, and Himalayan orogeny. Calcite U-Pb dating constrained the timing of the initial two charging events at approximately 433.0 Ma and ca. 278.9 Ma, respectively. Re-Os dating of northern heavy oils and southern medium/light oils provided absolute chronological constraints for two generation events during the Late Hercynian (270 +/- 20 Ma) and Late Caledonian (459 +/- 81 Ma) orogeny, respectively. Isotope dating data provided crucial absolute age constraints for accurately restoring the complex hydrocarbon accumulation process. The method has great potential for application in the study of the accumulation process in similar complex basins worldwide. (c) 2026 China University of Geosciences (Beijing) and Peking University. Published by Elsevier B.V. on behalf of China University of Geosciences (Beijing). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

The Qiangtang Basin, situated in the northern part of the Qinghai-Xizang Plateau, is China's largest Meso-Cenozoic marine sedimentary basin. Jurassic dolomite of the Buqu Formation are regarded as the basin's most promising hydrocarbon-bearing interval, yet their palaeokarst reservoir architecture and petrophysical distribution have not been quantified systematically. In this study, building on 1∶250, 000 regional geological maps and new fieldwork, we measured three detailed stratigraphic sections-two N-S and one E-W-across the Long'eni, Angda'ercuo and Sairen blocks, collected 171 fresh hand specimens, and carried out optical microscopy, casting-thin-section analysis, SEM observation, helium porosimetry and steady-state permeametry. Integrating outcrop and subsurface data, we characterised carbonate facies associations, reservoir space types and the coupling between petrophysical properties, stratigraphic packages and tectono-stratigraphic blocks.Results indicate that: (1) The Buqu Formation comprises two primary lithologies:limestone and dolomite. Limestones are subdivided into grain-supported, mud-supported, and bioconstruction types, while dolomite include silty-crystalline, fine-crystalline, medium- to coarse-crystalline, and calcareous varieties. (2) Reservoir spaces are classified into two main categories-pores and fractures-encompassing intercrystalline pores, intercrystalline dissolution pores, intercrystalline dissolution pores, vugs, structual fractures, dissolution seams, and pressure-solution seams.(3) Porosity ranges from 0.41% to 20.15%, with an average of 2.526%;the majority of samples fall between 1% and 5%, indicating predominantly low to ultra-low porosity. Permeability varies from 0.0002 × 10−3 to 64.58 × 10−3 μm2, with an average of 4.434 × 10−3 μm2, and most values are below 1× 10−3 μm2, reflecting low to ultra-low permwability.Based on porosity criteria, the carbonate reservoirs of the Buqu Formation in the study area are classified as Class Ⅲ (41.18%), representing high-quality palekarst hydorcarbon reservoirs.We further propose a refined palaeokarst layering scheme, quantitatively constrain low-porosity/low-permeability cut-offs, and evaluate reservoir heterogeneity, thereby providing target optimisation and parameter calibration for future horizontal drilling and integrated acidising−volumetric fracturing. These findings not only advance the carbonate-reservoir assessment framework for the Qinghai-Xizang Plateau, but also offer a model for exploring complex karstification hydrocarbon systems in ultrahigh-elevation basins worldwide.

The shallow slip deficit (SSD) during strike-slip earthquakes raises a question of how the strain budget is accommodated over multiple cycles. However, the origin of variable SSD observed in different earthquakes is still under debate because each earthquake has its unique initial stress condition. Here, we derive the slip model of the 2021 MW7.4 Maduo earthquake in Qinghai, China, using multi-track radar images. Our results revealed that, in contrast to the large SSD on segments close to the epicenter, a much smaller SSD was observed at the west terminus of the rupture, where aftershock distribution indicates that the fault changes dip direction at 6 km depth. The 2021 Maduo earthquake thus represents an extraordinary case of significant along-strike SSD variation. After accounting for interseismic, postseismic, and diffuse off-fault deformation, we find that this variation is likely contributed by the along-dipping geometrical variation, implying that a multi-segment earthquake may leave heterogeneous stress condition on the fault with different amounts of SSD.

Ensuring secure long-term storage requires understanding how heterogeneous CO2 distributions affect rock properties and trapping mechanisms‐an issue that is particularly critical in tight sandstones with inherently low fluid mobility. To address this, we combined physical-simulation reaction data from natural and artificial tight sandstones to quantify the impact of CO2 gradients on core-scale mineralization. In numerical simulations, reaction formulations were reconstructed to emphasize the spatial distribution of CO2 and its influence on long-term mineralization heterogeneity. Under sufficient CO2‐H2O‐natural tight sandstone reaction, rock properties are significantly enhanced with porosity increasing by 7.38% and permeability by 67.81% after 15 days at 85 °C and 25 MPa. Mineralogical analyses indicate that feldspar dissolution dominates, releasing Ca²⁺ and Na⁺ and promoting secondary clay mineral precipitation. Experiments with artificial tight sandstones varying in CO2‐H2O ratios demonstrate that spatial CO2 gradients fundamentally partition the system into three reaction zones: a near-well dissolution zone, a transition buffering zone, and a far-well equilibrium zone. Numerical simulations show that strong acidification in the near-well zone is limited by low water availability; the transition zone progressively buffers acidity while dissolution continues; and the far-well zone sustains coupled dissolution‐precipitation. Over time, CO2 storage evolves from structural trapping (49.69% at 50 years) to a combination of dissolution-mineral trapping (50.32% at 1000 years), particularly within the far-well equilibrium zone, where CO2 is progressively immobilized as stable mineral phases, reinforcing long-term geochemical security. This work supplements the study of mineralization heterogeneity and evolution in space and can provide new insights into CO2 geosequestration in tight formations.

Continental saline lacustrine basins are characterized by complex mineral compositions and strong lithological heterogeneity. However, systematic studies on lithofacies depositional sequences, predictive frameworks, and spatiotemporal evolution remain limited, and investigations into controlling factors and depositional models are still insufficient. Taking the Qianjiang Formation in the Qianjiang Depression of the Jianghan Basin, China, as an example, this study identifies seven major lithofacies types and five lithofacies associations based on multi-scale petrographic observations of cores, thin sections, and SEM, combined with XRD and XRF analyses. The established lithofacies association framework refines the depositional sequence of rift-related saline lacustrine basins and highlights the control of paleoenvironmental conditions on lithofacies distribution. A lithofacies prediction method based on multiple well-logging parameters is further developed to characterize the spatial–temporal evolution of lithofacies. Vertically, the formation is dominated by a mudrock lithofacies–carbonate/mixed-rock/halite lithofacies–mudrock lithofacies succession, whereas laterally a provenance-controlled lithofacies belt develops from terrigenous clastic lithofacies through mixed-rock and carbonate lithofacies to sulfate and halite lithofacies toward the lake center. Climatic fluctuations related to the East Asian monsoon during the middle–late Eocene produced pronounced cyclic sedimentation, leading to two depositional models: steep-slope and gentle-slope types characterized by alternating freshening and salinization stages. These results improve the understanding of sedimentary responses of saline lacustrine systems to paleoclimate variations and provide implications for hydrocarbon exploration.

The Hangay Mountains in central Mongolia, with their distinctive dome-shaped topography, exert a key influence on the modern climate of Mongolia. However, the geodynamic mechanisms responsible for their uplift remain poorly constrained. We report newly identified Early Cretaceous volcanism around the Hangay Mountains, temporally correlated with a major uplift phase as evidenced by thermochronological data. Geochemical analyses indicate that these volcanic rocks were derived from the lithospheric mantle and display elstability in the source and thus melting depths exceeding similar to 80 km. Given that the present-day lithosphere is only similar to 70 km thick, this implies that lithospheric thinning or foundering must have occurred during or after Early Cretaceous magmatism. Integrating these findings with regional tectonic constraints, we propose that uplift of the Hangay Mountains was triggered by Early Cretaceous lithospheric foundering, following earlier orocline-driven lithospheric thickening. These insights provide a novel framework for understanding intracontinental magmatism and orogenesis.

Thiols play important roles in global carbon and sulfur cycling, hydrocarbon alteration and metal mineralization. However, their transformation pathways and thermal stability in hydrothermal systems remain poorly constrained. Here we use an in situ approach combining fused silica capillary capsules with Raman spectroscopy to investigate the fate of thiols during hydrous pyrolysis and thermochemical sulfate reduction (TSR). Our results show that thiols are sequentially transformed into alcohols, aldehydes/ketones, and carboxylic acids, which subsequently decarboxylate or oxidize to form light hydrocarbons and CO2. Complex organic sulfur compounds—including sulfides, disulfides, sulfones, and thiophenes—are also generated, and their formation can be explained by radical mechanisms. Thiol hydrous pyrolysis proceeds more slowly than anhydrous pyrolysis at high temperatures (e.g., >247.5 °C for 1‑propanethiol) but faster at lower temperatures. Notably, thiols significantly enhance the rate of sulfate reduction, providing direct kinetic evidence that they can promote TSR. Even in the absence of external sulfate, thiol formation and hydrous pyrolysis can still promote hydrocarbon decomposition while transferring H2S from formation fluid into solid bitumen, serving as a critical but previously overlooked reservoir desulfurization mechanism. Because TSR-derived H2S reacting with hydrocarbon is a major pathway for thiol formation, the overall impact of TSR on hydrocarbon decomposition may be substantially underestimated. These results establish a kinetic and mechanistic framework for thiol-mediated C-S cycling in hydrothermal systems, with implications for hydrocarbon decomposition and reservoir desulfurization.

The Central Asian Orogenic Belt is characterized by several large-scale oroclines, whose formation has been linked to along-strike variations in trench migration. However, establishing a genetic relationship between oroclinal bending and trench migration remains challenging due to strong overprinting of subduction-related records by subsequent collisional orogeny. In this study, we reconstruct trench migration patterns around the Kazakhstan Orocline using crustal thickness proxies to investigate how spatial and temporal variations in trench migration interacted with oroclinal bending. Our analysis of Devonian to Carboniferous detrital zircons from sedimentary rocks in the southern and northern limbs of the Kazakhstan Orocline (West Junggar and Chinese West Tianshan) reveals that these zircons were predominantly derived from proximal magmatic arcs associated with the subduction of the Junggar Ocean. Zircon-based crustal thickness proxies (LREE/HREE, Ce/Yb and Dy/Yb ratios) indicate a tectonic transition in the West Junggar from slab rollback (similar to 390-360 Ma) to trench advance (similar to 360-323 Ma). In contrast, the Chinese West Tianshan records an early phase of trench advance (similar to 390-355 Ma), followed by the trench retreat (similar to 355-323 Ma). Integrating these findings with the kinematic history of the Kazakhstan Orocline, we propose that the southern limb of the orocline was stabilized by an advancing double subduction system beneath the Yili Block, while simultaneous trench retreat in the West Junggar facilitated bending of the subduction system and clockwise rotation of the northern limb. Our study underscores how trench retreat governs the curvature of subduction systems, shedding new light on oroclinal bending processes in accretionary orogens.

The rheology of Earth's lithosphere fundamentally governs tectonic processes and landscape evolution. Postseismic deformation following large earthquakes has been widely used to constrain rheological structures globally. However, regional-scale rheological variations remain poorly understood due to the infrequency of large earthquakes. The Bayan Har Block in the central-eastern Tibetan Plateau is a unique tectonic unit to investigate rheology regional-variations by pronounced high seismicity, heat flow anomalies, and localized lithospheric thinning. Here, we image postseismic deformation following the 2021 Maduo earthquake using InSAR and GNSS observations to probe the rheology over the block's north-east margin. We integrate long-term and short-term interferograms to separate deformation from atmospheric signals. After accounting for interseismic velocity, InSAR measurements extends out to 200 km from the rupture and its tips. Compared with prior studies, higher-quality far-field measurements allow the separation of viscoelastic relaxation from near-field afterslip effects. Our preferred model indicates a Burgers (bi-viscous) lower-crust rheology with transient and steady-state viscosities of 2.5 (+2.5/-1.3) x 10(1)(8) Pas and 2.5 (+0.7/-0.5) x 10(1)(9) Pas, respectively, beneath a 20-km-thick elastic layer. By testing laterally-variable and depth-dependent rheological structures, we identify a smaller intraplate-to-interplate viscosity contrast (factor of similar to 2) than that observed after the 2008 Wenchuan (factor of similar to 10) and 2001 Kokoxili (factor of similar to 5) earthquakes. This result suggests a positive correlation between viscosity contrasts, fault slip rates, and topographic gradients at a regional scale. These findings highlight that the rheological contrasts within the Bayan Har block exert a fundamental control on the long-term topographic evolution of the central Tibetan Plateau.

This study systematically reveals the fundamental mechanisms controlling redox-induced phase transformations occurring in basalt melting processes via integrated high-temperature redox experiments combined with thermodynamic simulations. Our findings demonstrate that oxidizing conditions drive clinopyroxene dissolution and concurrent crystallization of refractory phases—hematite [(Fe,Ti,Al)2O3] and magnesioferrite [(Mg,Fe)(Fe,Al)2O4]—where distinct crystallization pathways govern magnesioferrite morphology evolution. Conversely, reducing environments suppress oxide mineral formation while promoting phase transformation from high-melting-point plagioclase to low-melting-point clinopyroxene solid solutions, thus lowering the system’s liquidus temperature to achieve full melting. This provides a theoretical basis for optimizing energy consumption in basalt fiber production and offers new insights into the effects of material melting temperature.

Earthquakes with large magnitude induce massive post-seismic deformation lasting for months to years. Modeling the post-seismic deformation gains invaluable insights to understanding the physics of fault zone and the lower crustal rheology. However, the observed post-seismic deformation is originated from sources with variant mechanisms, including afterslip, poroelastic rebound, and viscoelastic relaxation, which occur at different spatial and temporal scales. Decomposing and interpreting deformation resulted from deep afterslip and viscoelastic relaxation especially remains challenging. The 2021 Mw 7.4 Maduo earthquake, which occurred on a secondary fault ~80 km south of the previously identified major block boundaries, east Kunlun fault, has generated clear afterslip signal reported by several studies. However, the interpretations regarding viscoelastic models remained debated in two aspects: 1) How can we quantify the contribution from deep afterslip and viscoelastic relaxation during the early post-seismic phase? 2) Does the lower crust exhibit the same rheological property across the ruptured Jiangcuo fault and east Kunlun fault? In this context, acquiring high-resolution and extensive coverage of post-seismic deformation data becomes critically important.Here, we derived a high-resolution post-seismic deofrmation extending over ~1000 kilometers for 2.5 years, using 6 tracks of Sentinel-1 SAR images and 32 continuous GNSS stations. Far-field deformations showed a smooth decay, ranging from 2 cm/year at the fault to 200 kilometers away on both sides of the fault rupture, extending over 500 kilometers along the strike. Notably, no discontinuity was observed along the east Kunlun fault, indicating that the boundary fault kept silent following the Maduo earthquake. We constrained the spatial pattern of post-seismic deformation with high-resolution InSAR observations, offering significant constrains into the depth and viscoelastic structure. Additionally, we utilized GPS time-series data to accurately ascertain the viscosity magnitude. By extracting the contribution of shallow afterslip from the initial observations, we explored the trade-off between deep afterslip and viscoelastic relaxation.We firstly used a three-layer Maxwell and Burgers model for far-field deformation (100-200 km) and then incorporated deep afterslip and viscoelastic relaxation for mid-field observations (10-100 km). Our best-fit results reveal that deep afterslip dominates in mid-field areas, while viscoelastic relaxation significantly impacts far-field deformation. The optimal model presents an upper crust depth of 20 km, with transient and steady-state viscosities in the lower crust at 10^18 and 4*10^19 Pa·s, respectively, and a steady-state upper mantle viscosity of 10^20 Pa·s. As with the preliminary results, the model did not require a strong variant viscosity to explain the data. Disregarding deep afterslip could lead to overestimating viscosity by 1-1.5 orders of magnitude. Our results imply that the ruptured secondary fault can continue to ~20 km and kept slip after earthquakes. However, for the deeper lower crust and upper mantle, the material keeps the same strength across the northeastern boundary of Bayankara block.

Determining the maximum temperature or burial depth where liquid hydrocarbon is preserved (i.e., oil preservation window) is a critical scientific problem. However, large discrepancy remains in the documented oil preservation window (e.g., 150-175 degrees C versus >240 degrees C), significantly influencing our understanding of the distribution of oil in deep earth. In this study, complete diagenesis and fluid inclusion records of oil charging and cracking were discovered in a Permian carbonate reservoir from the eastern Sichuan Basin (South China). Reconstructions of the diagenesis and petroleum evolution history reveal dynamic temperature-pressure controls on oil stability. Specifically, oil can be cracked to form methane-dominated gas after heating at <= 192 degrees C (sigma = 4) and in a hydrostatic pressure regime for similar to 20 m.y. Extreme overpressure formed postdating oil cracking due to intense calcite cementation, which resulted in isolated system and favored generation and accumulation of abnormally high pressure (up to 199 MPa) hydrocarbon fluids. The new oil preservation window can be applied to normal pressure or weak overpressure reservoirs, predicting the distribution of liquid hydrocarbon in deep earth.

Lacustrine gravity-flow deposits have enormous hydrocarbon potential.However,their deposi-tional architecture has long been considered very complicated and is controlled by various factors,making petroleum exploration and development highly challenging.The influence of palaeogeomorphology on the extent and formation mechanisms of lacustrine gravity-flow deposits remains poorly understood.This study uses 3D seismic data,well-log data,and core data to characterize the palaeogeomorphology and depositional architecture of the lacustrine gravity-flow deposit of the Yanchang Formation in the Ordos Basin,while analyzing the influence of palaeogeomorphology on the architecture,distribution,and evolution of lacustrine gravity-flow deposit systems.The results showed that three palaeogeomorphologic units—shelf,slope,and basin floor—developed in the basin.Gravity-flow deposits can be further divided into channel and lobe sys-tems.Gravity-flow channel systems have developed in the upper slope area,which include three types of architectural elements:confined channels,unconfined channels,and levee-overbank.As the transport dis-tance of gravity flow increases,the erosion ability of gravity flow weakens,resulting in the terrain formed by gravity flow erosion gradually weakening the constraint on sediment transport.Consequently,gravity-driven flow pathways shift from linear,confined channels to curved,unconfined channels,finally leading to the development of numerous distributary channels.In the lower slope area,gravity flow sediments distribute rapidly and accumulate to form lobes after encountering slope breaks,due to the decrease of dynamics,including three types of architectural elements:distributary channels,lobe axis and lobe fringe.These lobes are tongue-shaped and fan-shaped on the plane.A series of lobes are superimposed and form a lobe complex set.These lobe complex sets are distributed in a branch-like manner.There are many branch-shaped low-bend amplitude channels inside the lobe complex set.Palaeogeomorphology plays an important role in controlling the spatial distribution of depositional architecture.As the slope gradient increases,the influence of slope break on sediment gravity flow weakens,leading to a delay of the sediment gravity flow slumping to the basin bottom.Additionally,the extension length of the channel system on the slope increases.Furthermore,the extension length of the lobes toward the center of the basin also increases.This work improves the funda-mental understanding of the depositional architecture of lacustrine gravity-flow deposits and may help enhance sand prediction for the same type deposit reservoirs.

Lacustrine basins are rich in unconventional oil and gas resources and the heterogeneity of matrix reservoir quality seriously affects its development. Previous studies have recognized that diagenesis is the key factor affecting the heterogeneity of matrix reservoir quality, but diagenesis controlled by lithofacies assemblages is poorly studied. In this study, tight sandstones in the Triassic (Ladinian-Carnian) Chang 7 Member of Ordos Basin were taken as examples. Lithofacies, lithofacies assemblages, diagenesis evolution and distribution were investigated by casting thin section analysis, scanning electron microscopy (SEM), X-ray diffraction (XRD), fluid inclusion and carbon-oxygen isotope. The results indicate that eight lithofacies assemblages in Chang 7 exhibit variations in lithofacies, sand-mud ratio and grain size rhythm, controlling the distribution and intensity of diagenesis. Variations in average proportion of lithofacies assemblages are linked to lake level cycles. Compared to cementation, the mechanical compaction has a higher adverse effect on primary porosity of tight sandstones in Chang 7. Siltstone lithofacies show stronger compaction and cementation (mainly clay cementation), whereas fine sandstone lithofacies exhibit stronger dissolution, especially FSpc (fine sandstone with parallel or cross-bedding). Modest proportions of illite (<4%) and chlorite (<1.5%) in fine sandstones tend to form clay coatings, resisting compaction and preserving more porosity. Conversely, abundant pore-filling illite (>4%) and chlorite (>1.5%) result in a noticeable decline in porosity. Particle size and sorting affect reservoir quality by controlling compaction strength. The origin of carbonate cement in sandstones is closely related to mud-rich lithofacies within lithofacies assemblages. Carbonate cementation preferentially occurs in fine sandstones within 0.4 m from the sand-mud interface and the thickness of adjacent mud-rich lithofacies is greater than 0.1 m. Interfaces between fine sandstones serve as fluid migration pathways and bidirectional erosion, facilitated by high porosity and feldspar content. In contrast, the clay-rich matrix in siltstones acts as a barrier hindering fluid migration, creating unidirectional erosion in fine sandstones near the interfaces between siltstones and fine sandstones. Ultimately, five evolution processes linking diagenesis with lithofacies assemblages are summarized, highlighting the differential reservoir quality. This study contributes to understanding the mechanism behind diagenetic variations in tight sandstones and the reservoir evaluation. Consequently, it offers geological reference for the development of unconventional oil and gas resources in comparable lacustrine basins.

Figure S1: Plot of amphibole and plagioclase compositions. Figure S2: Cathodoluminescence images of the dated zircon grains. Figure S3: Chondrite normalized REE patterns of dated zircon grains. Table S1: Data of standard materials for bulk-rock chemical analyses. Table S2: Representative mineral compositions of the studied samples. Table S3: SHRIMP zircon U-Pb dating results. Table S4: Trace element concentrations of zircon grains of the selected samples. Table S5: SIMS monazite U-Pb dating results for sample 21DB89. Table S6: Matrix-type biotite compositions and temperature results from Ti-in-biotite thermometry. Table S7: Ti contents in zircon and temperatures from Ti-in-zircon thermometry.

The Tabei uplift in the Tarim Basin is one of the deepest and most important petroleum-producing areas in China, with more than 3 billion t (21 billion bbl) of oil equivalent discovered in the Paleozoic carbonate reservoirs. Further petroleum exploration and development in the Tabei and neighboring areas will greatly bene fit from an in-depth understanding of the hydrocarbon charge and accumulation history of these deeply buried carbonate reservoirs. The molecular correlation of reservoir oils indicates that oils from major accumulations in the area share similar geochemical characteristics and were presumably derived from the same source rocks deposited in a marine environment. The Shunbei reservoir oil has the highest thermal maturity, followed by the Yuecan reservoir oil, whereas the Tahe reservoir oil has the lowest thermal maturity. Six generations of calcite cementation spanning over 130 m.y. have been delineated in calcite veins, with U-Pb ages ranging from ca. 446 Ma to ca. 316 Ma. The second and fifth generations of calcite cementation were accompanied by oil charge events, as indicated by the occurrence of bitumen and primary oil inclusions. Fluid inclusion analysis coupled with basin modeling results reveal that the Tabei area experienced two major oil charges, with the first charge occurring during the late Caledonian Orogeny, at 426 to 415 Ma, and the second charge during the middle-late Hercynian Orogeny, at 339 to 278 Ma. The Shunbei and Yuecan reservoirs contain well-preserved oils accumulated during the two charge events, whereas the Tahe reservoir oil has been partially biodegraded.

Lacustrine gravity-flow deposits are common in sedimentary basins, and sand bodies formed in the basin center mingled with organic rich mudstone may form excellent tight reservoirs of hydrocarbons. However, the complex architecture and intensive heterogeneities of lacustrine gravity-flow deposits commonly pose great risks for hydrocarbon exploration and development. To address this problem, a new workflow for predicting tight sandstone is proposed in this work. The workflow combines well log analysis, intelligent seismic inversion and 3D geological modelling to improve the prediction effectiveness of tight sandstone and sedimentary facies. The original 3D seismic data was firstly decomposed into three components with low, middle and high frequencies to improve the resolution of seismic inversion, and a nonlinear relationship between well logs and the decomposed 3D seismic data was then established by using an algorithm of support vector machine. A 3D lithofacies model was subsequently established integrating well logs and the seismic inversion result. The distribution of sand thickness was imaged relying on the lithofacies model. Correlations between the actual and predicted sand thickness based on the seismic attributes and the seismic inversion result was improved from 0.66 to 0.79. The prediction of sand bodies was further improved through the 3D lithofacies modelling, especially for these thin sandstone layers. Sedimentary facies of the Chang_7 member in the Qingcheng Oilfield, Ordos Basin, were characterized based on the distribution of sand bodies. Fan-shaped sand bodies are mapped in each zone of the Chang_7 member, revealing that the entire Chang_7 successions are of sub-lacustrine fan in origin. The Chang_7 member mainly contains six sedimentary architectural elements: main channel, branch channel, lobe, lobe edge, slump body and inter lobe/inter channel. The distribution of sedimentary facies is different in different zones, and appears to be largely related to the change of base-level. A horizontal well was designed according to the prediction of sand bodies, which show a higher probability of drilling into sandstone by approximately 10% compared with that of other horizontal wells.

Semantic segmentation using cross-modal data is a hot topic in the field of Earth observation. Compared with single-modal strategies, cross-modal networks fuse multiaspect information and yield higher segmentation accuracy, which is widely used in urban planning, environmental monitoring and so on. In this study, an end-to-end adaptive cross-modal fusion network (ACFNet) is proposed for semantic segmentation task using high resolution and light detection and ranging images, because of the difference of sensor resolution, different modal data have different abilities of ground object expression. Therefore, multimodal data fusion should consider the features with different spatial scales, while most existing methods simply use the same spatial scale features for fusion. In this work, we first design an adaptive scale fusion module that can automatically choose the features with optimal spatial scales, making full use of the representation properties of ground object details. Second, the important feature guidance module is designed, which can evaluate the influence weights of deep semantic features and shallow spatial detailed features, achieving adaptive deep and shallow feature fusion, and reducing the semantic-spatial information dilution caused by layer-by-layer up and down sampling. Finally, we introduce a divide Fourier context learning (DFCL) module to transform the feature maps from spatial domain to frequency domain. Compared to the limited perception of current spatial convolution kernels, the DFCL module can easily model the contextual dependencies of cross-modal features, which will improve the segmentaion accuracy for complex ground objects of cities, especially for occlusion. To demonstrate the generalisation performance of our module, we conduct extensive experiments and ablation studies on three datasets: Potsdam, Vaihingen, and IEEE GRSS DFC 2018. Results show that the proposed ACFNet is effective in semantic segmentation.