With the rapid advancement of urbanization and industrialization, global energy consumption has entered a phase of accelerated depletion (Khan et al., 2022; Wang et al., 2025). As onshore energy resources approach their development limits, submarine energy reserves have increasingly drawn attention (Cai et al., 2024). China, with its vast marine areas, harbors abundant coastal resources, particularly offshore oil, natural gas, and gas hydrates (Wang et al., 2022; Su et al., 2022). However, the exploitation of subsea resources inevitably induces shallow gas leakage, as these resources are typically located at significant depths (Zhang and Zhai, 2015; Vielstaedte et al., 2017). Submarine shallow gas release, a pervasive geological process, has garnered growing attention due to its dual role in shaping seafloor topography and disrupting marine ecosystems (Kunath et al., 2022). When shallow gas breaches sediment and interacts with ocean currents, it forms plumes that entrain surrounding sediments, leading to the development of pockmarks. This not only compromises the bearing capacity of offshore infrastructure but also exacerbates seafloor instability under seismic or wave loads, potentially triggering submarine landslides, foundation settlement, and platform overturning (Li et al., 2023; Singhroha et al., 2023). Moreover, it alters sediment stability, pore water chemistry, and nutrient cycling, influencing benthic habitat and migration behaviors (Wallace et al., 2017; Ashford et al., 2021). Additionally, CH₄ and H₂S, which are abundant in shallow gas, may leak into the atmosphere and potentially cause fireball and vapor cloud explosions (Dando and Hovland, 1992; Ruppel and Kessler, 2017; Leifer and Judd, 2015; Dissanayake et al., 2023), directly threatening avian species. Thus, investigating pockmark development and plume morphology provides intuitive insights into the impact scope of shallow gas leakage.

Shallow gas, rich in CH₄, H₂S, and CO₂, forms plumes through entrainment effects during ascent due to density differences. Sufficient gas source pressure drives gas upward to breach the water surface, manifesting as fountains. Plume morphology is linked to gas release intensity and diffusion rates in water. To understand plume formation and evaluation methods, researchers have conducted experimental and simulation studies. Experimentally, Geng et al. (2021) investigated underwater gas diffusion behaviors, discussing the effects of leakage pressure, water depth, and burial depth on plume dynamics, providing references for gas release research and consequence assessment. Wang et al. (2022) reviewed progress in oil-gas leakage plume studies, highlighting droplet observation and plume dynamics as key to marine oil spill risk assessment. Ren et al. (2022) proposed a multi-beam water column point cloud model for plume extraction, validating its effectiveness in gas hydrate-rich areas of the Gulf of Mexico via density clustering (DBSCAN) and feature matching (FPFH). Wang et al. (2024) further revealed the positive correlation between fountain effects and leakage pressure/aperture size, analyzing plume dynamic equilibrium based on Froude (Fr) and Reynolds (Re) numbers to offer new methods for marine environmental impact assessment. Numerically, Sun et al. (2020) simulated using a 3D Eulerian-Lagrangian coupled model, analyzing the effects of leakage rate, flow velocity, and water depth on plume formation. Li et al. (2020) developed an Eulerian-Lagrangian model to capture jet-to-plume transition and entrainment. For complex scenarios like submarine cold seep plumes, Duan et al. (2020) designed seismic scattering models, finding strong scattering energy and directional seismic wavefields in cold seep plume areas. Although some studies focus on other fluids (e.g., liquid hydrogen), their methods are transferable—for example, Pu et al. (2019) revealed three-stage diffusion characteristics (heavy gas, ascent, passive phases) in liquid hydrogen leakage, providing generalizable insights into buoyancy-driven diffusion.

Plumes entrain sediments during formation, and continuous gas eruption leads to pockmarks centered at the gas breakthrough site. To understand pockmark formation and assess sediment strength changes, studies based on integral models, simulations, and experiments have been conducted. Hovland and Judd (1988) through seabed sediment acoustic data and ROV observations in the North Sea, revealed the genetic link between pockmark development and natural gas leakage, and proposed a “fluid escape-sediment erosion” dynamic formation model. Cathles et al. (2010) established a “piston-water drive” model to reveal the process where rapid gas release following capillary seal failure forms gas chimneys, which in turn triggers sediment liquefaction leading to pockmark formation, emphasizing that complete seal failure is the key mechanism. Mazzini et al. (2017) linked the formation of Norwegian Sea pockmark clusters to gas hydrate dissociation 10 ka ago, indicating alternating cementation-disintegration in pit sediments during depressurization. Sun et al. (2019) analyzed mechanical properties around two large pockmarks near the Zhongjie Mountain Islands in the East China Sea, clarifying causes of sediment strength variations. Jordan et al. (2019) combined vibrocore analysis and geochemical testing in Bantry Bay, Ireland, finding that microbial anaerobic methane oxidation (AOM) reduces pore water sulfate, weakening soil cementation and decreasing compressive strength by ~28%. Pape et al. (2020) quantified the impact of gas hydrate dissociation on pockmark dynamics through pressure coring and geochemical analysis in Nigerian deepwater pockmark A. Wu et al. (2023) proposed a “gas explosion pockmarking” model, explaining similar depth-radius ratios between pockmarks and volcanic craters worldwide. Vaknin et al. (2024) analyzed shear strength in sand-clay bilayer media experiments, identifying dome rupture, brittle faulting, and bubble ascent as gas breakthrough mechanisms in sealing layers.

While single-orifice leakage models have been well-studied, research on two-orifice models remains insufficient. It remains unclear whether plume and pockmark formation processes interact in two-orifice leakage or if diffusion characteristics converge with single-orifice models. To deepen understanding of gas leakage mechanisms and accurately evaluate diffusion characteristics during shallow gas release, this study used customized experimental systems to conduct single and two orifice gas leakage tests. Variations in plume and pockmark parameters under different gas source pressures and burial depths were investigated, and correlations between sediment strength, pressure, and distance from pockmark centers were discussed. The findings provide experimental insights into how gas release parameters influence submarine geomorphology and sedimentary environments, contributing to risk assessment of marine ecological disturbances associated with shallow gas activities.

An experimental physical model was established to simulate shallow gas release and its ecological impacts, modified based on sediment-biology interactions. The experiment was conducted in a flume measuring 50 m in length and 1 m in width ( Figure 1 ). To minimize soil usage, a transparent acrylic box measuring 40 cm by 30 cm by 30 cm was placed within the flume, filled with a sand-clay mixture at a 9:1 ratio and with a dry density of 1.65 to 1.80 g/cm³, simulating typical marine sediments. A compressed air system (0–700 kPa capacity) was used to simulate gas release, with pressures set at 100, 200, 300, 400, and 500 kPa. Release configurations included a single orifice (5 mm diameter) and two orifices (5 cm spacing), while burial depths were 10, 15, and 20 cm. CO₂ gas was used to represent shallow gas in the test, as it is easy to prepare and has high safety.

Hydrodynamic processes were recorded using a high-speed camera (200 fps) to measure fountain height, plume diameter, and cone angle. Post-experimentally, pockmark depth and diameter were measured with a stainless steel ruler, and volume was calculated via integration. Sediment strength was assessed using a WXGR-3.0 micro penetrometer (17–130 kPa range) in four zones: pockmark center, waistline, edge, and external area (2 cm from the edge), targeting sediment stability indices critical for benthic colonization. Coefficients of variance were calculated to evaluate parameter sensitivity, linking physical disturbances to potential ecological impacts. Each test was repeated three times for reproducibility, and analysis of variance was used to determine the significant effects of pressure, depth, and orifice number on the plume-pockmark-sediment system.

Soil entrained by gas and plume entrainment alters the shear strength of sediments inside and around pockmarks (Zhang et al., 2022). To analyze the disturbance caused by uncontrolled shallow gas release, soil regions have been classified by previous studies into intense disturbance zones, severe disturbance zones, slight disturbance zones, and gas-bearing compaction zones ( Figure 9a ). Based on the compaction principle, compaction occurs when sediments are subjected to external forces with frequencies close to their natural frequencies. This resonance expels pore water between particles, reducing voids and increasing sediment density. To assess sediment compactness, pockmarks were divided into center, waistline, edge, and external regions according to their morphology ( Figure 9b ).

To investigate vibration compaction and variations in soil strength across different pockmark regions, soil strength was measured at distances of R/2, 5R/6, and R+2 from the pockmark center at a burial depth of 10 cm. The influence of pressure on soil strength at measurement points is shown in Figure 10 . For both single and two orifices, soil strength in the pockmark center and external areas increased relative to the initial strength, whereas strength in the pockmark waistline and edge regions decreased. The pockmark center was compacted due to increased fluid pressure and sediment settlement. The waistline, located in the maximum shear stress zone, exhibited pressure gradient reversal according to the Darcy-Forchheimer equation. Meanwhile, the external region showed enhanced soil strength due to the formation of a dense “edge skirt”.

Where, ▽P denotes the pressure gradient; β is the inertial resistance coefficient;k represents the permeability

As shown in ( Figure 10a ), at the same pressure, soil strength first decreased then increased with distance from the pockmark center, consistent with previous regional classifications by disturbance intensity. Increasing pressure homogenized the disturbed area, shifting the position of uniform soil strength backward. Unlike the high concentration of gas in single-orifice release, the strong pressure and plume merging entrainment in double-orifice release distributed sediment displacement more widely. Significantly affected by pressure, the rapid development of plumes and pockmarks in double-orifice release increased soil strength compared to single-orifice release ( Figure 10b ), reducing sediment settlement heterogeneity and enhancing the symmetry of soil strength distributions under different pressures.

Shallow gas leakage poses significant risks to burrowing organisms, aquatic life, and atmospheric biota. Pockmark formation and changes in soil strength disrupt habitats of soil-dwelling organisms and affect soil oxygen content. Plumes influence aquatic migration behaviors, while fountains and escaped gases directly threaten avian species under specific conditions. Among the chain reactions of shallow gas leakage, pockmark impacts are most readily assessable. Exploiting the soil strength modification by gas leakage, some offshore platforms have enhanced surrounding sediment strength. However, practical applications must balance these benefits against potential ecological impacts.

An experimental platform for gas leakage was independently developed to investigate the effects of gas source pressure, burial depth, and number of release orifices on plume morphology and pockmark development. Correlations between sediment strength, pressure, and distance from pockmark centers were discussed, leading to the following conclusions:

Traditional single-orifice leakage studies provide preliminary local hazard assessments, but real-world marine conditions often involve chain reactions of gas leakage. Notable differences exist between single and two orifice leakages in terms of plume characteristics, pockmark development, and sediment strength. Although this study compared single- and two-orifice scenarios, using CO₂ to simulate CH₄;-dominated shallow gas may introduce discrepancies in plume dynamics and sediment interactions due to differences in physical properties (e.g., density, solubility). Additionally, the influence of orifice spacing and water flow on plume merging characteristics remains to be investigated, which will be a focus of future research.