The study of deepwater channels is important for the understanding of the sedimentary evolution mechanism and the sedimentary process of the marginal sea. In 2019, thick pore-filling gas hydrate with high saturation was firstly discovered in the Quaternary sands of the Qiongdongnan Basin (QDNB), which expanded the reservoir types of gas hydrates in the South China Sea. However, the distribution of sand-related channels is not well characterized, which limits the ability to predict sand reservoirs with gas hydrate. Using integrated 2D/3D seismic, multi-beam, well logging, and coring data, the current study documents the distribution characteristics of channel systems in the Quaternary strata and discusses their controlling factors. The integrated analysis shows that the channel-related sedimentary facies include channel-filling facies, levee facies, crevasse splay facies, and lobes facies. A total of six periods of channel systems is identified in the Quaternary strata. There are obvious distribution differences between the Channel 1 and Channel 3 systems when comparing the western, middle, and eastern sections: the channels in the western and eastern sections are mainly dominated by near straight V-shaped channels, while the middle section mainly consists of large braided channels, where channel-levee sedimentary facies developed. Compared with the distribution of the Central Canyon that developed in the Miocene, the Channel 1 and Channel 3 systems in the western section show southward migration since the Miocene. The distribution and evolution of Quaternary channels were likely collectively controlled by seafloor morphology, tectonic movement, sea-level fluctuations, and provenance supply. Tectonic movement controls seafloor morphology, which directly controls the flow of channels and their distribution characteristics; provenance supply determines the scale and sedimentary characteristics of each channel. The periodic changes in sea-level determine the evolution of multi-stage channel systems. This study has implications for the prediction of gas hydrate–bearing sands in the Quaternary QDNB and deepens our understanding of the Quaternary tectonic and sedimentary evolution in the QDNB.

Various factors controlling the accumulation of natural gas hydrates (NGHs) form various enrichment and accumulation modes through organic combination. This study mainly analyzes the geological and geophysical characteristics of the NGHs occurrence in the uplifts and their slope zones within the deep-water area in the Qiongdongnan (QDN) Basin (also referred to as the study area). Furthermore, it investigates the dominant governing factors and models of NGHs migration and accumulation in the study area. The results are as follows. (1) The uplifts and their slope zones in the study area lie in the dominant pressure-relief direction of fluids in central hydrocarbon-rich sags in the area, which provide sufficient gas sources for the NGHs accumulation and enrichment through pathways such as gas chimneys and faults. (2) The top and flanks of gas chimneys below the bottom simulating reflectors (BSRs) show high-amplitude seismic reflections and pronounced transverse charging of free gas, indicating the occurrence of a large amount of gas accumulation at the heights of the uplifts. (3) Chimneys, faults, and high-porosity and high-permeability strata, which connect the gas hydrate temperature-pressure stability zones (GHSZs) with thermogenic gas and biogenic gas, form the main hydrate migration system. (4) The reservoir system in the study area comprises sedimentary interlayers consisting of mass transport deposits (MTDs) and turbidites. In addition, the reservoir system has developed fissure- and pore-filling types of hydrates in the pathways. The above well-matched controlling factors of hydrate accumulation enable the uplifts and their slope zones in the study area to become the favorable targets of NGHs exploration.

2D and 3D seismic data and basin simulation were used to investigate the gas hydrate distribution and natural gas migration pathways in the Qiongdongnan Basin (QDNB). Hydrate-related amplitude anomalies and extensive bottom simulating reflectors (BSRs) were mapped within the uppermost part. Based on their seismic reflection characteristics, the three main types of natural gas migration pathways and their distributions in the QDNB were identified through high-resolution seismic data. Basin modeling was carried out to document the migration efficiency of different migration pathways and their effects on hydrate enrichment. The basin modeling results show the following: (1) Diapirs, fault structures, and fractures constitute the three main types of natural gas migration pathways that transport the thermogenic gas from the deep to shallow layers in the QDNB. (2) The three migration pathways impact hydrate enrichment in different ways. Diapirs and faults contribute significantly to hydrate enrichment due to their higher migration efficiency. In comparison, the migration efficiency of the fracture systems is lower, with minimal benefit to hydrate enrichment. (3) The natural gas hydrate in the QDNB is mainly distributed along the diapirs and deep faults and generally scattered around the fracture system. These conclusions indicate that the migration pathways in the QDNB are regionally distributed and are closely related to hydrate accumulation.

《地球科学》(ISSN 1000-2383;CN42-1874/P) (原刊名《地球科学——中国地质大学学报》ISSN 1000-2383;CN42-1233/P)，创刊于1957年，是由教育部主管、中国地质大学主办的综合性地学学术理论期刊，月刊，2015年改名为《地球科学》,她以反映中国地质大学教学与科研最新的、高水平的基础地质、应用地质、资源与环境地质及地学工程技术研究成果为主要任务。

The origin of deep-water mounds has been a topic of debate in recent years. In this study, newly collected seismic data were employed to characterize the mounds within the Meishan Formation in the Qiongdongnan Basin and a novel model was proposed. The result showed that pervasive mounds and ‘V’-shaped troughs were alternately distributed at the top of the Meishan Formation. They appeared as elongated ridges flanked by similarly elongated gullies, with the trending parallel with the strike of the basinward slope. The mounded features were considered to be formed in response to the tectonically induced seabed deformation. The differential subsidence steepened the slope that was equivalent to the top of the Meishan Formation (ca. 10.5 Ma), which offered sufficient driving forces triggering the slope’s instability. Correspondingly, the uppermost deposits glided along a bedding-parallel detachment surface, creating a number of listric detachment faults that ceased downward to this surface. The uppermost layer was cut into a range of tilted fault blocks with tops constituting a seemingly mounded topography. Some of the downfaulted troughs between mounds steered the gravity flows and were filled by sand-rich lithologies. The differential subsidence played a decisive role in the formation of a mounded stratigraphy, which in turn acted as clues to the important tectonic phase since the late Miocene.

Conventional high-resolution marine seismic surveys often face challenges in detecting small features in the shallow subsurface because acquisition parameters involve trade-offs among spatial resolution, penetration, signal-to-noise ratio (SNR), and operational efficiency. This study presents an intelligent marine seismic acquisition framework that integrates selectable acquisition hardware with modified ML Real domain adaptation, deep learning reconstruction, quantitative quality-control (QC) evaluation, and a human-supervised adaptive agent. Rather than directly predicting acquisition parameters with the neural network, the deep learning model reconstructs high-resolution seismic images from degraded acquisitions, which are subsequently evaluated using QC metrics including reflector continuity, spectral bandwidth, SNR, and depth penetration. The adaptive agent uses these metrics to rank discrete, hardware-feasible configurations of source type, source energy, dominant frequency, and shot interval and recommends the configuration for the subsequent acquisition cycle. Synthetic benchmark tests demonstrate improved imaging of thin layers and localized subsurface features under the tested conditions. A two-cycle nearshore field experiment further demonstrates the operational feasibility of the workflow, in which QC evaluation of the initial acquisition guided the subsequent acquisition configuration and resulted in improved reflector continuity, penetration, and SNR. The results show that integrating AI-assisted numerical modeling and evaluation can guide acquisition-parameter refinement and reduce reliance on repeated offshore trial-and-error surveys. The proposed framework provides a proof-of-concept for integrating configurable high-resolution seismic hardware with AI-assisted, QC-guided acquisition optimization for marine shallow-subsurface imaging.