Bottom-Simulating Reflectors (BSRs) in Gas Hydrate Systems: A Comprehensive Review

Abstract

The bottom-simulating reflector (BSR) serves as an important seismic indicator for identifying gas hydrate-bearing sediments. This review synthesizes global BSR observations and demonstrates that spatial relationships among BSRs, free gas, and gas hydrates frequently deviate from one-to-one correspondence. Moreover, our analysis reveals that more than 35% of global BSRs occur shallower than the bases of gas hydrate stability zones, especially in deepwater regions, suggesting that the BSRs more accurately represent the interface between the gas hydrate occurrence zone and the underlying free gas zone. BSR morphology is influenced by geological settings, sediment properties, and seismic acquisition parameters. We find that ~70–80% of BSRs occur in fine-grained, grain-displacive sediments with hydrate lenses/nodules, while coarse-grained pore-filling sediments host <20%. BSR interpretation remains challenging due to limitations in traditional P-wave seismic profiles and conventional amplitude versus offset (AVO) analysis, which hinder accurate fluid identification. To address these gaps, future research should focus on frequency-dependent AVO inversion based on viscoelastic theory, multicomponent full-waveform inversion, improved anisotropy assessment, and quantitative links between rock microstructure and elastic properties. These innovations will shift BSR research from static feature mapping to dynamic process analysis, enhancing hydrate detection and our understanding of hydrate–environment interactions.

1. Introduction

The bottom-simulating reflector (BSR), as the most prominent seismic indicator of gas hydrate-bearing sediments (GHBS), serves three critical scientific and practical roles: ① providing the fundamental constraint (distribution and GHBS thickness) for estimating hydrate reservoir volumes in conjunction with well log and/or core data, ② delineating geohazards from dissociation-induced seafloor instability, and ③ resolving methane release scenarios for climate modeling. However, critical uncertainties remain regarding the formation mechanisms of the BSR, coupling with underlying free gas systems, and dynamic responses to environmental perturbations. Resolving these gaps through integrated geophysical and numerical approaches is essential for advancing hydrate system predictability.

BSR studies originated alongside early gas hydrate theories. Trofimuk et al. first proposed natural hydrate accumulation models [1], while seismic surveys (e.g., Blake Ridge) identified anomalous reflectors parallel to the seafloor but oblique to the strata [2,3]. However, due to the limited resolution of single-channel P-wave data and the lack of in situ drilling validation, the formation mechanism of BSR was not fully understood.

Multichannel seismic acquisition and advanced inversion methods (e.g., travel time [4,5] and full waveform inversion [6,7,8,9]) revealed a diagnostic velocity structure: a high-velocity hydrate-bearing layer overlying a low-velocity free gas zone [10]. AVO analysis revealed BSRs’ positive amplitude-versus-offset trends, though these responses were later shown to depend nonlinearly on hydrate/gas saturations. Critically, ODP Legs 164/204 drilling refuted the Hydrate Wedge Model [10], demonstrating that impedance contrast from high-saturation hydrates alone cannot generate strong reflections. Instead, BSR was confirmed to result from the presence of free gas beneath the base of the gas hydrate stability zone (BGHSZ), where even ~5% gas saturation reduces P-wave velocity sharply [11], and amplifies reflectivity [12]. These insights established the Free Gas Zone Model [13] as the dominant paradigm.

Since the early 21st century, BSR research has entered a more refined and multidisciplinary stage, characterized by three major trends:

First, the use of 3D seismic data significantly enhanced stratigraphic resolution [14]. When combined with multi-wave and multi-component techniques and AVO attribute analysis, researchers developed BSR identification frameworks based on elastic parameters such as P- to S-wave velocity ratios and Poisson’s ratio [15,16]. Further studies have shown that the microstructural occurrence of hydrates in sediments (e.g., cementation, pore-filling, or grain-contact modes) directly affects elastic responses: hydrates acting as cement or in grain contact enhance both P- and S-wave velocities, whereas pore-suspended hydrates predominantly increase only P-wave velocity [17,18]. This distinction provides a theoretical basis for inferring hydrate occurrence mechanisms.

Second, research has expanded from static property analysis to dynamic process monitoring. Methane plume imaging has confirmed that gas released from hydrate dissociation migrates vertically along faults, forming high-amplitude reflective zones [19]. The spatial distribution of these features is closely associated with the boundaries of the hydrate stability zone [20], offering indirect evidence for BSR identification.

Third, studies on viscoelastic absorption characteristics have revealed marked regional variability. In some areas, hydrate-bearing sediments exhibit strong seismic attenuation [21], while in others, weak attenuation is observed [22,23,24]. These contrasting results may stem from complex interactions among hydrate occurrence modes, free gas distribution, and local geological conditions.

Despite significant advances in BSR research, four critical knowledge gaps persist: ① the limited generalizability of localized case studies, ② persistent ambiguities in BSR identification accuracy, ③ poorly constrained spatial correlations between BSRs and BGHSZs, and ④ unresolved formation mechanisms for double or multiple BSRs.

To systematically address these challenges, this review undertakes a comprehensive approach that includes: ① synthesizing global BSR occurrences across active and passive margins, ② diagnosing methodological and interpretive anomalies in existing datasets, ③ evaluating the relative controls of thermal, compositional, and tectonic factors on BSR morphology, and ④ quantifying global BSR–BGHSZ relationships through integrated seismological and thermodynamic modeling. By bridging theoretical predictions with observational inconsistencies, these efforts aim to standardize BSR interpretation frameworks, thereby refining hydrate resource estimates and improving constraints on climate-sensitive methane release dynamics.

2. Seismic Identification and Global Distribution of BSRs

BSR is a distinctive seismic reflection with negative polarity, typically subparallel to the seafloor and often crosscutting underlying stratigraphy. It generally forms under conduction-dominated thermal regimes rather than convective environments [25]. The strength of the reflection can be quantified using the reflection coefficient (R) or amplitude (A):

$$R={\displaystyle \frac{Z_2-Z_1}{Z_2+Z_1}}={\displaystyle \frac{A_2}{A_1}}$$

(1)

$$Z=\rho \mathsf{\bullet }v$$

(2)

where Z represents acoustic impedance, calculated as the product of density (ρ) and P-wave velocity (v), and subscripts 1 and 2 correspond to the incident and reflected layers, respectively. The absolute value of R is positively correlated with reflection amplitude and typically falls within the range of −1 to 1 [26].

BSR arises from the acoustic impedance contrast between hydrate-bearing sediments and the underlying free gas zone. When gas hydrate saturation exceeds ~40%, it significantly increases the P-wave velocity in the overlying sediments (typically 1.8–2.5 km/s) due to grain-hydrate interaction [27]. Meanwhile, the presence of free gas in the underlying layer can significantly decrease the P-wave velocity (as low as 1.2–1.5 km/s) [11]. This sharp velocity contrast makes the BSR one of the most reliable seismic indicators for detecting GHBS [25]. A particular type of BSR, referred to as double or multiple BSRs, displays vertically stacked reflection patterns. These features are interpreted as the result of hydrate system re-stabilization following thermodynamic disequilibrium events and complex gas composition [28,29,30]. The global distribution of such multi-BSRs offers critical insights into the dynamic response of gas hydrate systems to climatic and geologic forcing over time.

2.1. Seismic Identification of BSR

Accurate seismic identification of BSR is critical for gas hydrate research, as misleading or pseudo-reflections may complicate BSR interpretation. Figure 1 presents representative seismic profiles from major global gas hydrate regions, emphasizing structural and stratigraphic features indicative of hydrate presence.

2.1.1. Seismic Characteristics of BSRs

BSR formation primarily depends on the regional thermal regime. In conduction-dominated systems (the most common scenario), the BSR parallels the seafloor or isotherms. In convection-dominated systems (e.g., fluid-venting zones), the BSR often deviates from seafloor geometry (see Section 4.1.3 described later). The BSR represents a phase boundary (gas hydrate stability zone, GHSZ) rather than a lithological contact, allowing it to cross-cut dipping strata. In terms of polarity, the BSR is characterized by a high-impedance hydrate-bearing layer overlaying a low-impedance free gas zone, resulting in a negative polarity reflection—an inverse polarity signature that contrasts with diagenetic BSRs, which typically exhibit positive polarity [31,32].

2.1.2. Seismic Anomalies Associated with BSR

The identification of the BSR on seismic profiles should rely on a combination of multiple seismic anomalies rather than a single seismic attribute. Beyond the characteristic negative-polarity BSR, several associated reflections commonly occur: ① a strong positive-polarity reflector beneath the BSR, marking the impedance contrast at the base of the free gas zone [33]; ② a positive-polarity reflector above the BSR in some cases, resulting from the hydrate-bearing layer’s upper boundary; and ③ acoustic blanking zones—both below the BSR (due to gas attenuation in chimneys or diapirs) and above it (caused by amplitude suppression in homogeneous hydrate-bearing sediments) (Figure 1). These reflection events typically parallel the seafloor [34].

2.1.3. Seismic Attribute Analysis of BSR

Conventional seismic profiles derived from raw data can effectively illustrate amplitude variations and stratigraphic geometry. However, they provide limited information on physical properties, lithology, and fluid content. As a result, distinguishing fluid differences in similar lithologies becomes difficult, which may lead to misinterpretation of BSRs in some cases. In contrast, seismic attribute analysis enables more detailed and accurate characterization of subsurface features by extracting amplitude, frequency, phase, and other derived parameters [35].

Seismic attributes can be derived from either post-stack or pre-stack seismic data. Post-stack attributes primarily include instantaneous amplitude, instantaneous phase, and instantaneous frequency [36], while pre-stack attributes are mainly extracted from Amplitude Versus Offset (AVO) analysis.

Figure 1. Seismic profiles of representative global gas hydrate regions: (a) Blake Ridge, modified from Hornbach et al. [37]; (b) Nyegga, Norway, modified from Hjelstuen et al. [38]; (c) Barents Sea, modified from Vadakkepuliyambatta et al. [39]; (d) Shenhu Area, South China Sea, modified from Jin et al. [40]; (e) Gulf of Mexico, modified from Portnov et al. [41]; and (f) Makran Accretionary, modified from Liu et al. [42].

Figure 1. Seismic profiles of representative global gas hydrate regions: (a) Blake Ridge, modified from Hornbach et al. [37]; (b) Nyegga, Norway, modified from Hjelstuen et al. [38]; (c) Barents Sea, modified from Vadakkepuliyambatta et al. [39]; (d) Shenhu Area, South China Sea, modified from Jin et al. [40]; (e) Gulf of Mexico, modified from Portnov et al. [41]; and (f) Makran Accretionary, modified from Liu et al. [42].

Instantaneous amplitude, or reflection strength, represents the energy of a seismic signal at a given time and is always positive and phase-independent. The BSR at the interface between hydrate-bearing layers and underlying free gas zones typically exhibits strong amplitudes, while hydrate-bearing layers themselves often produce weak reflections due to their acoustic homogeneity. Compared with conventional profiles, amplitude attributes more clearly highlight high-contrast reflections linked to the BSR and hydrate accumulations [43]. The instantaneous phase reflects time-dependent wave phase changes and helps track reflection continuity. Phase shifts occur as seismic waves pass through lithological boundaries, making the BSR more distinguishable in phase sections. Since phase attributes are independent of amplitude, they are useful for identifying weak or discontinuous BSR, although their visibility decreases when the BSR is nearly parallel to bedding planes [44,45]. Instantaneous frequency, the time derivative of the instantaneous phase, detects low-frequency shadows beneath gas-saturated sediments, complementing amplitude data for gas/hydrate differentiation [36].

AVO analysis identifies lithology and potential hydrocarbon zones by investigating changes in reflection amplitude with offset (or incidence angle). In GHBS, BSRs frequently exhibit positive AVO anomalies, where reflection amplitude strengthens with increasing offset. However, this response depends strongly on hydrate/gas saturation and thickness, making AVO responses alone insufficient for accurately describing elastic properties [13,46,47,48,49]. To address this limitation, AVO attribute analysis focuses on hydrocarbon indicators such as the intercept and fluid factor. The intercept, representing the zero-offset P-wave reflection coefficient, is useful for identifying BSRs and inferring their geological nature. The fluid factor, highly sensitive to free gas content, plays a key role in distinguishing between hydrate-bearing sediments and underlying gas zones [50].

In practice, reliable BSR interpretation demands an integrated approach, combining multiple seismic attributes with drilling and well-logging data (see

Table 1. Seismic identification of BSR in gas hydrate systems.

Category	Key Features/Parameters	Description	Strengths	Limitations	Refs.
Conventional seismic-section analysis	Morphological Features	- Thermal conduction control: BSR is sub-parallel to the seafloor/isotherms. - Thermal convection control: BSR appears as an upward-convex structure and often develops in fluid migration zones. - Usually a cross-cutting, high-amplitude reflector (>30% of seafloor-reflection amplitude).	Visually straightforward; enables rapid delineation of hydrate-bearing provinces over large areas.	- Highly dependent on data quality; low signal-to-noise ratios or low dominant frequencies can render the BSR hard to recognize - In structurally complex settings the BSR may be masked or confused with other stratigraphic boundaries. - Where free gas is limited, the reflector weakens or disappears, leading to missed detections.	[25,51]
	Polarity Features	A hydrate-related BSR typically shows negative polarity (central negative lobe flanked by positive side lobes), distinguishing it from diagenetic BSRs with positive polarity.	Directly reflects the impedance reversal at the hydrate/gas interface; a useful auxiliary discriminator.	- May be confused with negative-polarity reflections generated solely by underlying high-saturation gas layers or overlaying carbonates.
Associated reflection phenomena	Positive-polarity reflection below the BSR	Resulting from significant impedance contrasts between free gas zones and underlying sediments.	To enhance the reliability of hydrate-free gas system identification by incorporating multiple lines of evidence.	- Characterized by weak amplitudes and simple geometries, these features are observable only in high-resolution, low-noise profiles. - They can be easily mistaken for other bright spots or stratigraphic boundaries; therefore, meticulous validation is imperative during interpretation.	[33,34]
	Positive-polarity reflection at the hydrate top boundary	Associated with impedance contrasts between hydrate-bearing sediments and overlying normal sediments.
	Acoustic blanking zone below the BSR	Attributed to seismic attenuation linked to high gas concentrations within gas chimneys or mud diapirs.
	Acoustic blanking zone within the hydrate-bearing layer	Due to the homogeneous nature of hydrate-bearing sediments, leading to reduced seismic reflection amplitudes.
Seismic-attribute analysis	Instantaneous Amplitude	- Highlights prominent BSR reflections and amplitude anomalies within hydrate-bearing sediments. - Independent of seismic phase and thus polarity-insensitive.	To overcome the shortcomings of traditional methods and facilitate the detection of the weak or discontinuous BSR.	- High data quality requirements; noise or multiples can introduce artifacts. - Subjectivity in the selection of attribute parameters. - AVO analysis requires well-sampled multi-offset data and is unsuitable for datasets with limited offset coverage or conventional single-offset data.	[43]
	Instantaneous Phase	- Improves BSR visibility when cross-cutting strata. - Limited sensitivity to parallel bedding.			[44,45]
	Instantaneous Frequency	- Free gas absorption leads to low-frequency shadowing, aiding in free gas identification.			[36]
	AVO Attributes	- Intercept parameter facilitates BSR identification. - Fluid factor is highly sensitive to free gas detection.			[50]

for a comparative summary).

2.2. Global Distribution of BSR

Based on a systematic statistical analysis, this study delineates the global distribution characteristics of BSR associated with GHBS. Figure 2 provides an intuitive visualization of the spatial locations, distribution patterns, and morphological features of all identified BSR sites. To support this analysis, we systematically compile key parameters—including Tectonic Geological Features (TGF), pore habit, sediment features, seismic morphology types of BSRs (SMT), geological setting types of BSRs (GST), BSR depth (Z_BSR, mbsf), water depth (WD, m), seafloor temperature (T_sf, °C), and geothermal gradient (GG, °C/km)—organized by region in Table 2, Table 3, Table 4, Table 5, Table 6, Table 7 and Table 8 (Atlantic: Table 2; Pacific: Table 3; Arctic: Table 4; Antarctic: Table 5; Indian Ocean: Table 6; inland seas/lakes: Table 7; continental permafrost: Table 8). Table 2, Table 3, Table 4, Table 5, Table 6, Table 7 and Table 8 serve as the data backbone of this research, with their core value reflected in two key aspects: ① providing a comprehensive data foundation for the analytical calculations and discussions in subsequent sections and ② establishing a standardized reference database to facilitate further research within the academic community. A one-to-one correspondence between the codes in Figure 2 and those in Table 2, Table 3, Table 4, Table 5, Table 6, Table 7 and Table 8 confers explicit spatial traceability on the dataset, markedly increasing its practical value and multidimensional analytical potential of the dataset. For the sediment types associated with BSR, it is assumed that the representative lithology at the BSR location corresponds to the stratigraphy within 50 m above it. In the absence of specific sedimentological data for certain locations, lithological information from nearby sites can be obtained from reports of the Ocean Drilling Program (ODP), Deep Sea Drilling Project (DSDP), or Integrated Ocean Drilling Program (IODP). These reports are accessible through the International Ocean Drilling Data Archive. While the lack of co-located data introduces some uncertainty, this approach provides the best available estimates for sedimentological characteristics. Through this statistical evaluation, several noteworthy phenomena have been identified, warranting further discussion. These observations can be categorized into four main types, as outlined in the following subsections.

2.2.1. Depth Distribution of BSR

The depth distribution of BSR is distinctly right-skewed (skewness = +1.49), with approximately 68% of observations occurring between 200 and 500 mbsf, and a mean depth of ~365 mbsf. Notably, BSRs deeper than 600 mbsf are predominantly found in rapidly subsiding basins with low geothermal gradients (typically <30 °C/km), suggesting that the BSR burial depth is primarily controlled by the interplay between geothermal gradient and water depth. However, the uppermost sediments (<100 mbsf) often lack the conditions necessary for BSR formation due to the significant consumption of methane by anaerobic oxidation of methane (AOM) processes mediated by microbial communities [52]. Nonetheless, field observations identify two exceptional cases where shallow BSRs develop: ① seepage-induced BSR shoaling and ② landward pinch-out of shallow BSRs.

The first type, BSR shoaling in seepage environments, is characterized by significant upward displacement of the BSR beneath mud volcanoes or pockmarks, occasionally intersecting the seafloor. This phenomenon has been reported in regions such as the El Arraiche Mud Volcano Field [53], Congo-Angola Margin [54], Husmus hydrocarbon field in Mid-Norway [55], and Rio Grande Cone on the Brazilian continental margin [56]. The primary driver of this process is the sustained migration of deep-sourced thermogenic gas, facilitated by mud diapirism and gas chimney systems, which transport methane to shallower depths. Additionally, localized high-intensity hydrothermal activity alters pressure–temperature (P-T) conditions, further contributing to BSR shoaling.

The second type, landward pinching-out of the shallow BSR, is primarily observed at the landward edges of hydrate-bearing regions, such as the Hydrate Ridge [57] and offshore Mauritania [58]. This phenomenon is controlled by multiple geological factors. First, a decrease in water depth toward the continental shelf modifies local P-T conditions, influencing hydrate stability. Second, high methane flux combined with highly permeable turbidite layers facilitates methane accumulation in shallow sediments, thereby elevating the sulfate–methane transition zone (SMTZ). Third, localized thermal perturbations induced by the high thermal conductivity of salt diapirs increase the geothermal gradient, further affecting BSR stability. A particularly important concern associated with BSR outcrops, where the BSR intersects the seafloor, is their positioning at the feather edge of the hydrate stability zone. This makes them highly susceptible to seafloor temperature fluctuations, increasing the risk of warming-induced hydrate dissociation. As a result, such settings are more prone to methane release, posing significant geohazard risks and environmental implications.

2.2.2. Temperature Distribution of BSR

The temperature at the BSR ranges from 10.7 °C to 27.8 °C, with a mean value of 17.2 °C. Fifty percent of the data (interquartile range) fall between 14.8 and 19.7 °C, and the distribution shows a moderate positive skewness (skewness = +0.66). Notably, several high-temperature outliers (>25 °C) are identified, primarily concentrated in tectonically active regions such as the Lima Basin [59] and the offshore area of the western Indian Ocean near Tanzania [60]. What is more, in regions such as the Western Marmara Sea, Western Indian Ocean offshore Tanzania, and Site 860 in Chile, BSRs have been observed at depths significantly deeper than the predicted BGHSZ for pure methane systems. Thermodynamic analyses suggest these anomalies correspond to structure II (sII) gas hydrates, characterized by enhanced thermal stability due to their distinct clathrate cage architecture, enabling persistence under elevated temperature regimes.

2.2.3. Geothermal Gradient Distribution for BSR Occurrence

Statistical analysis indicates that the geothermal gradient in global BSR-bearing regions ranges from 19 to 130 °C/km, with a mean value of 46.8 °C/km and a moderately right-skewed distribution (skewness = +1.86). The interquartile range is 32–52 °C/km, consistent with the global average thermal conduction regime of marine sediments. Notably, abnormally high geothermal gradients (>80 °C/km) are concentrated in areas such as the El Arraiche Mud Volcano Field [61] and the Scotia Sea [62], which are associated with intense heat flow upwelling within volcanic rift systems or nascent faulted margins. In these environments, elevated heat flux significantly raises the BGHSZ, resulting in a shallower BSR depth (mbsf). This effect underscores the dominant influence of tectonic activity on geothermal regimes.

Statistical comparisons further demonstrate that TGF is a primary control on the spatial variability of geothermal gradients: active continental margins—typically associated with subduction, rifting, or volcanism—exhibit an average geothermal gradient of approximately 55 °C/km, while passive continental margins—characterized by tectonic stability—have significantly lower values, averaging around 40 °C/km. Particularly low gradients (<32 °C/km) are observed in thickly sedimented basins such as the Blake Ridge, where massive sediment accumulations act as a thermal blanket, effectively impeding the upward transfer of deep heat [63,64,65,66,67,68].

The strong coupling between the geothermal gradient and TGF is further reflected in its negative correlation with BSR depth (Pearson’s r ≈ –0.48), indicating that areas with high heat flow tend to have shallow BSRs. These observations reveal a global bimodal distribution pattern of BSRs:

• “Shallow BSRs with high geothermal gradients”, typically occurring in venting settings (e.g., gas chimneys or mud diapirs), where rapid fluid migration thins the gas hydrate stability zone and increases subsurface temperatures.
• “Deep BSRs with low geothermal gradients”, commonly found in thick sedimentary basins, where a stable thermal regime allows the BGHSZ to extend below 300–600 mbsf.

2.2.4. Seismic Morphologies of BSR

A global statistical analysis of BSR seismic morphologies reveals four dominant types with distinct distribution patterns:

Continuous BSRs account for 37.50% of the total in the statistics and are mainly distributed along passive continental margins. These BSRs typically appear as continuous reflections in seismic profiles, parallel to the seafloor, and intersect with the stratigraphic reflections. A typical example is found in the Blake Ridge [25,51,69,70].

Discontinuous BSRs are the most widespread globally, comprising 54.69% of the total. They are primarily located along active continental margins and manifest as segmented, discontinuous lateral reflections, following the topography of the seafloor [51,69].

Clustered BSRs are rare, making up only 6.25% of the total. They are primarily localized at the crests of domes or anticlines. In such structural settings, the lateral migration of free gas is impeded, promoting its vertical accumulation beneath the structural crest. This process results in the entrapment of gas within discrete stratigraphic layers, producing a sequence of vertically stacked, high-amplitude reflections beneath a BSR that is roughly parallel to the seafloor. They often occur in areas of folding or salt-related domes, indicating a high concentration of natural gas hydrates and free gas beneath [41,71,72,73,74,75,76].

Pluming BSRs are also infrequent, accounting for just 1.56% of the total, such as in the Gulf of Mexico GC-955 [77]. These BSRs represent a unique type of continuous seismic reflection, typically found in localized areas associated with surface mud volcanoes or salt domes. They are smaller in scale and generally occur at relatively shallow depths, typically less than 1200 mbsf [51]. In seismic profiles, their distribution is not parallel to the seafloor, and their shape appears semicircular. They are often accompanied by acoustic blank zones [77].

Table 2. Global statistical analysis of known BSR data (Atlantic).

Code	Location	TGF	Sediment Features	Pore Habit	SMT	GST	ZBSR (mbsf)	WD (m)	Tsf (°C)	GG (°C/km)	ZBGHSZ (mbsf)	ZBGHSZ* (mbsf)
1/ 2	Continental margin of Brazil (Foz do Amazonas Basin/Rio Grande Cone)	PCM	(Channel-levee turbidite systems and MTDs; 85% silt, 13% clay, 0.15% sand [78])/(fine-grained siliciclastic material [56])	NA	Disc./ Cont. [56]	Stru./ Vent. [56]	450 [56]/ (200–300) [56])	(600–2800 [56])/ (500–3500 [56])	NA	30–34 [79]	NA	NA
3/ 4/ 5/ 6	Gulf of Mexico (Green Canyon/Perdido Canyon/Terrebonne/Orca Basin)	ACM	20–50% silt, 50–80% clay, <1% sand [80]	Nodules [81]	Clus./ Clus./ Disc./ Plum. [51,77]	Stru./ Stru./ Stru./ Vent. [51,77]	450 [77]/ NA/ NA/ NA	2000 [77]/ NA/ NA/ NA	NA	NA	NA	NA
7/ 8/ 9	Blake Ridge (Site 533/Site 995/Site 997)	PCM	(11–28% silt, 70–86% clay, 0.3–3% sand)/(Nannofossil-rich clay and claystone, with sediments exhibiting moderate to intense bioturbation)/(Homogeneous dark green clay and claystone, rich in diatoms) [63,64,65,66,67,68]	Nodules [66]	Cont. [64]/ (Cont.-Plum.) [68]/ Cont. [67]	Stra. [64]/ Stra. [68]/ Stra. [67]	600 [64]/ 450 [68]/ 464 [67]	3184 [64]/ 2779.5 [68]/ 2770.1 [67]	2.15 [64]/ 3.6 [68]/ 3.39 [67]	36 [64]/ 33.5 [68]/ 36.8 [67]	584.34/ 553.22/ 505.41	627.46/ 600/ 547.7
10	Carolina Trough (Site 991)	PCM	15% silt, 85% clay, 0% sand [82]	Nodules [83]	Disc. [84]	Stru. [84]	300 [84]	1500 [84]	3.1 [82]	NA	NA	NA
11	Continental Rise (Site 107)	PCM	25% silt, 75% clay, 0% sand [85]	NA	Cont. [86]	Stra. [86]	400 [86]	2700–3400 [86]	NA	36–51 [86]	NA	NA
12	Namibe Basin (Site 1080)	PCM	Diatom bearing 5–10% silt, 30–60% clay, 0% sand [87]	Nodules [88]	Diag. [87]	NA	250 [87]	1000 [87]	4.4–4.6 [88]	NA	NA	NA
13	Punta del Este Basin	PCM	NA	NA	Cont. [89]	Vent. [89]	400 [89]	350–2200 [89]	NA	NA	NA	NA
14/ 15	Norwegian Continental Slope (Husmus/Nyegga)	PCM	(Soft clay) [55]/ (45% silt, 54% clay, 1% sand) [90]	Nodules [91]	Disc./ Cont. [55]	Vent./Vent. [55]	5/ 280 [55]	330/ 730 [55]	(6–7)/ (−0.7) [55]	NA/ 50 [92]	NA	NA
16/ 17/ 18/ 19/ 20	Newfoundland (Makkovik Slope/Hamilton Spur/Haddock Channel/Mohican Channel/Barringto)	PCM	Closely related to the drift deposition area [93]	NA	Disc./ Cont./ Cont./ Cont./ Disc. [93]	Stru./ Stra./ Stra./ Stra./ Stra. [93]	443/ 335/ 375/ 290/ 378 [93]	(620–2555)/ (1075–2100)/ (1700–2150)/ (1324–2875)/ (2220–2850) [93]	(2.1–4.15) [93]	32 [93]	NA	NA
21	Newfoundland Sackville Spur	PCM	Closely related to the drift deposition area [93]	NA	Disc. & (Doub/Mult.) [94]	Stra. [94]	350 [94]	1125 [94]	3.9 [95]	35.8 [95]	314.94	360.82
22	Offshore Mauritania	PCM	Fine-grained turbidites and nannofossil-rich muds [58]	NA	Disc. [58]	Stru. [58]	83 [58]	711 [58]	8 [96]	32 [96]	39.96	105.52
23	Colombia Basin (Site 502)	ACM	Mostly nannofossils and clay; 25–75% clay [97,98]	NA	Cont. [99]	Stru. [99]	485 [99]	2640 [99]	6 [76]	20 [76]	823.06	905.67
24	Panama	ACM	28% silt, 70% clay, 2% sand [100]	NA	Cont. [101]	Stru. [101]	180–300 [101]	1800–2800 [101]	NA	59 [101]	NA	NA
25	El Arraiche Mud Volcano Field	PCM	Mud breccia & clay [61]	Nodules [102]	Cont. [53]	Vent. [53]	0 [53]	380 [53]	10 [53]	110 [53]	NA	NA
26	Congo Deep Sea Fan (Site 1076)	PCM	25–75% silt, 25–75% clay, 0% sand [54,103]	Nodules [54]	Disc. [104]	Stru. [104]	220 [104,105]	2980 [104,105]	2.5 [105]	80 [105]	241.87	260.57
27	Niger Delta Front	PCM	80% fines, 20% sand [106]	Nodules/ lenses [107]	Disc. [108]	Stra. [108]	300–380 [108]	2400–2900 [107,108]	4.45–4.53 [107]	72 [107]	NA	NA
28	Offshore UK (Faroe-Sheland Basin)	PCM	Coarse-grained sediments, linked to paleochannels	NA	Cont. [109]	Stru. [109]	300–350 [109]	>750 [109]	NA	36.5 [109]	390 [109]	NA

Note: NA = Not Available; PCM = Passive Continental Margin; TGF = Tectonic Geological Features; ACM = Active Continental Margin; MCM = Mixed Continental Margin; TRL = Tectonic Rift Lake; TPB = Transition Plate Boundary; ISB = Inland Sedimentary Basin; CC = Continental Collision; SMT = Seismic Morphology Types of BSRs; Cont. = Continuous BSR; Disc. = Discontinuous BSR; Clus. = Clustered BSR; Plum. = Pluming BSR; Diag. = Diagenetic BSR; GST= Geological Setting Types of BSRs; Stru. = Structural Setting; Stra. = Stratigraphic Setting; Vent. = Venting Setting; Z_BSR = Depth of BSR; WD = Water Depth; T_sf = Seafloor Temperature; GG = Geothermal Gradient; Z_BGHSZ = Depth of BGHSZ in Salt Water (3.5 wt.% NaCl); Z_BGHSZ* = Depth of BGHSZ in Pure Water. The notes for Table 3, Table 4, Table 5, Table 6, Table 7 and Table 8 are identical to those for Table 2.

Table 2. Global statistical analysis of known BSR data (Atlantic).

Code	Location	TGF	Sediment Features	Pore Habit	SMT	GST	ZBSR (mbsf)	WD (m)	Tsf (°C)	GG (°C/km)	ZBGHSZ (mbsf)	ZBGHSZ* (mbsf)
1/ 2	Continental margin of Brazil (Foz do Amazonas Basin/Rio Grande Cone)	PCM	(Channel-levee turbidite systems and MTDs; 85% silt, 13% clay, 0.15% sand [78])/(fine-grained siliciclastic material [56])	NA	Disc./ Cont. [56]	Stru./ Vent. [56]	450 [56]/ (200–300) [56])	(600–2800 [56])/ (500–3500 [56])	NA	30–34 [79]	NA	NA
3/ 4/ 5/ 6	Gulf of Mexico (Green Canyon/Perdido Canyon/Terrebonne/Orca Basin)	ACM	20–50% silt, 50–80% clay, <1% sand [80]	Nodules [81]	Clus./ Clus./ Disc./ Plum. [51,77]	Stru./ Stru./ Stru./ Vent. [51,77]	450 [77]/ NA/ NA/ NA	2000 [77]/ NA/ NA/ NA	NA	NA	NA	NA
7/ 8/ 9	Blake Ridge (Site 533/Site 995/Site 997)	PCM	(11–28% silt, 70–86% clay, 0.3–3% sand)/(Nannofossil-rich clay and claystone, with sediments exhibiting moderate to intense bioturbation)/(Homogeneous dark green clay and claystone, rich in diatoms) [63,64,65,66,67,68]	Nodules [66]	Cont. [64]/ (Cont.-Plum.) [68]/ Cont. [67]	Stra. [64]/ Stra. [68]/ Stra. [67]	600 [64]/ 450 [68]/ 464 [67]	3184 [64]/ 2779.5 [68]/ 2770.1 [67]	2.15 [64]/ 3.6 [68]/ 3.39 [67]	36 [64]/ 33.5 [68]/ 36.8 [67]	584.34/ 553.22/ 505.41	627.46/ 600/ 547.7
10	Carolina Trough (Site 991)	PCM	15% silt, 85% clay, 0% sand [82]	Nodules [83]	Disc. [84]	Stru. [84]	300 [84]	1500 [84]	3.1 [82]	NA	NA	NA
11	Continental Rise (Site 107)	PCM	25% silt, 75% clay, 0% sand [85]	NA	Cont. [86]	Stra. [86]	400 [86]	2700–3400 [86]	NA	36–51 [86]	NA	NA
12	Namibe Basin (Site 1080)	PCM	Diatom bearing 5–10% silt, 30–60% clay, 0% sand [87]	Nodules [88]	Diag. [87]	NA	250 [87]	1000 [87]	4.4–4.6 [88]	NA	NA	NA
13	Punta del Este Basin	PCM	NA	NA	Cont. [89]	Vent. [89]	400 [89]	350–2200 [89]	NA	NA	NA	NA
14/ 15	Norwegian Continental Slope (Husmus/Nyegga)	PCM	(Soft clay) [55]/ (45% silt, 54% clay, 1% sand) [90]	Nodules [91]	Disc./ Cont. [55]	Vent./Vent. [55]	5/ 280 [55]	330/ 730 [55]	(6–7)/ (−0.7) [55]	NA/ 50 [92]	NA	NA
16/ 17/ 18/ 19/ 20	Newfoundland (Makkovik Slope/Hamilton Spur/Haddock Channel/Mohican Channel/Barringto)	PCM	Closely related to the drift deposition area [93]	NA	Disc./ Cont./ Cont./ Cont./ Disc. [93]	Stru./ Stra./ Stra./ Stra./ Stra. [93]	443/ 335/ 375/ 290/ 378 [93]	(620–2555)/ (1075–2100)/ (1700–2150)/ (1324–2875)/ (2220–2850) [93]	(2.1–4.15) [93]	32 [93]	NA	NA
21	Newfoundland Sackville Spur	PCM	Closely related to the drift deposition area [93]	NA	Disc. & (Doub/Mult.) [94]	Stra. [94]	350 [94]	1125 [94]	3.9 [95]	35.8 [95]	314.94	360.82
22	Offshore Mauritania	PCM	Fine-grained turbidites and nannofossil-rich muds [58]	NA	Disc. [58]	Stru. [58]	83 [58]	711 [58]	8 [96]	32 [96]	39.96	105.52
23	Colombia Basin (Site 502)	ACM	Mostly nannofossils and clay; 25–75% clay [97,98]	NA	Cont. [99]	Stru. [99]	485 [99]	2640 [99]	6 [76]	20 [76]	823.06	905.67
24	Panama	ACM	28% silt, 70% clay, 2% sand [100]	NA	Cont. [101]	Stru. [101]	180–300 [101]	1800–2800 [101]	NA	59 [101]	NA	NA
25	El Arraiche Mud Volcano Field	PCM	Mud breccia & clay [61]	Nodules [102]	Cont. [53]	Vent. [53]	0 [53]	380 [53]	10 [53]	110 [53]	NA	NA
26	Congo Deep Sea Fan (Site 1076)	PCM	25–75% silt, 25–75% clay, 0% sand [54,103]	Nodules [54]	Disc. [104]	Stru. [104]	220 [104,105]	2980 [104,105]	2.5 [105]	80 [105]	241.87	260.57
27	Niger Delta Front	PCM	80% fines, 20% sand [106]	Nodules/ lenses [107]	Disc. [108]	Stra. [108]	300–380 [108]	2400–2900 [107,108]	4.45–4.53 [107]	72 [107]	NA	NA
28	Offshore UK (Faroe-Sheland Basin)	PCM	Coarse-grained sediments, linked to paleochannels	NA	Cont. [109]	Stru. [109]	300–350 [109]	>750 [109]	NA	36.5 [109]	390 [109]	NA

Note: NA = Not Available; PCM = Passive Continental Margin; TGF = Tectonic Geological Features; ACM = Active Continental Margin; MCM = Mixed Continental Margin; TRL = Tectonic Rift Lake; TPB = Transition Plate Boundary; ISB = Inland Sedimentary Basin; CC = Continental Collision; SMT = Seismic Morphology Types of BSRs; Cont. = Continuous BSR; Disc. = Discontinuous BSR; Clus. = Clustered BSR; Plum. = Pluming BSR; Diag. = Diagenetic BSR; GST= Geological Setting Types of BSRs; Stru. = Structural Setting; Stra. = Stratigraphic Setting; Vent. = Venting Setting; Z_BSR = Depth of BSR; WD = Water Depth; T_sf = Seafloor Temperature; GG = Geothermal Gradient; Z_BGHSZ = Depth of BGHSZ in Salt Water (3.5 wt.% NaCl); Z_BGHSZ* = Depth of BGHSZ in Pure Water. The notes for Table 3, Table 4, Table 5, Table 6, Table 7 and Table 8 are identical to those for Table 2.

Table 3. Global statistical analysis of known BSR data (Pacific).

Code	Location	TGF	Sediment Features	Pore Habit	SMT	GST	ZBSR (mbsf)	WD (m)	Tsf (°C)	GG (°C/km)	ZBGHSZ (mbsf)	ZBGHSZ* (mbsf)
29	Acapulco (Site 491)	ACM	50% silt, 30% clay, 20% sand [110]	Nodules [110]	Cont. [110]	Stru. [110]	380 [101]	2000–3800 [101]	NA	21 [101]	NA	NA
30	Hydrate Ridge (Site 1245)	ACM	50% silt, 50% clay [111]	Layered [111]	Cont. [57]	Stru. [57]	134 [57]	870 [57]	4 [57]	54 [112]	150.91	180.32
31	Cascadia Margin (Site 889)	ACM	30–50% silt, 40–70% clay, 0–4% sand [113,114]	Nodules [115]	Disc. [114]	Stru. [114]	224 [114]	1313 [114]	2.7 [114]	54 [114]	243.22	271.66
32	Lima Basin (Site 685)	ACM	Mainly composed of diatom mud and calcareous mudstone; 45% silt, 55% clay [59,116]	Nodules [59]	Disc. [59]	Stru. [59]	612 [59]	5070 [59]	1.5 [59]	43 [59]	590.16	625.49
33	Nankai Trough	ACM	26–36% silt, 1–3% clay, 61–72% sand [117,118]	Pore-filling [119]	Disc. [118]	Stru. [118]	204 [119]	945 [119]	3 [119]	46 [120]	225.55	260.52
34	Sea of Okhotsk (Site 796)	MCM	73–79% silt, 20.4–26.3% clay, 0–5% sand [121]	Nodules [122] /lenses [123]	Disc. [124]	Stru. [124]	450 [125]	1300 [125,126]	2.5 [126]	35 [126]	402.49	448.72
35	Ulleung Basin	ACM	100% <75 μm, median grain size: 2.3–3.0 μm [127]	Nodules/ Lenses [128]	Disc. [128]	Stru.-Stra. [128]	150 [128]	2100 [127,128]	NA	NA	NA	NA
36/ 37/ 38/ 39	Shenhu Area (SH1/SH2/SH3/SH7)	PCM	(Clay & Silt)/ (70% silt, 25% clay, 5% sand)/ (Clay & Silt)/ (Silt & Sand) [129,130]	Pore-filling [130]; Nodules [131]	Disc./ Cont./ Cont./ Cont. [130]	Stra./ Vent./ Vent./ Vent. [130]	219/ 221/ 204/ 181 [130]	1500 [130]	5.2/ 4.84/ 5.53/ 6.44 [130]	47.53/ 46.95/ 49.34/ 43.65 [130]	222/ 229/ 206/ 184 [130]	276.4/ 288.72/ 257.78/ 272.06
40	Hikurangi Trough (Site U1517)	ACM	40% silt, 58% clay, 2% sand [132]	Lenses [133]	(Cont.) & (Doub./Mult.) [134]	Stra. [134]	165 [134]	725 [134]	5.32 [135]	39.8 [135]	130.58	175.57
41/ 42/ 43	Chile (Site 859/Site 860/Site 861)	ACM	(Silty clay & Clayey silt)/ (44% silt, 49% clay, 7% sand)/ (Coarse-grained clastic sediment) [136,137]	NA	Disc./ Cont./ Disc. [136]	Stru./ Stru./ Stru. [136]	100/ 200/ 250 [138]	2741.2/ 2145.9/ 1652 [136]	NA/ 2.5 [137]/ NA	NA/ 103.5 [137]/ NA	NA/ 160.2/ NA	NA/ 174.46/ NA

Table 3. Global statistical analysis of known BSR data (Pacific).

Code	Location	TGF	Sediment Features	Pore Habit	SMT	GST	ZBSR (mbsf)	WD (m)	Tsf (°C)	GG (°C/km)	ZBGHSZ (mbsf)	ZBGHSZ* (mbsf)
29	Acapulco (Site 491)	ACM	50% silt, 30% clay, 20% sand [110]	Nodules [110]	Cont. [110]	Stru. [110]	380 [101]	2000–3800 [101]	NA	21 [101]	NA	NA
30	Hydrate Ridge (Site 1245)	ACM	50% silt, 50% clay [111]	Layered [111]	Cont. [57]	Stru. [57]	134 [57]	870 [57]	4 [57]	54 [112]	150.91	180.32
31	Cascadia Margin (Site 889)	ACM	30–50% silt, 40–70% clay, 0–4% sand [113,114]	Nodules [115]	Disc. [114]	Stru. [114]	224 [114]	1313 [114]	2.7 [114]	54 [114]	243.22	271.66
32	Lima Basin (Site 685)	ACM	Mainly composed of diatom mud and calcareous mudstone; 45% silt, 55% clay [59,116]	Nodules [59]	Disc. [59]	Stru. [59]	612 [59]	5070 [59]	1.5 [59]	43 [59]	590.16	625.49
33	Nankai Trough	ACM	26–36% silt, 1–3% clay, 61–72% sand [117,118]	Pore-filling [119]	Disc. [118]	Stru. [118]	204 [119]	945 [119]	3 [119]	46 [120]	225.55	260.52
34	Sea of Okhotsk (Site 796)	MCM	73–79% silt, 20.4–26.3% clay, 0–5% sand [121]	Nodules [122] /lenses [123]	Disc. [124]	Stru. [124]	450 [125]	1300 [125,126]	2.5 [126]	35 [126]	402.49	448.72
35	Ulleung Basin	ACM	100% <75 μm, median grain size: 2.3–3.0 μm [127]	Nodules/ Lenses [128]	Disc. [128]	Stru.-Stra. [128]	150 [128]	2100 [127,128]	NA	NA	NA	NA
36/ 37/ 38/ 39	Shenhu Area (SH1/SH2/SH3/SH7)	PCM	(Clay & Silt)/ (70% silt, 25% clay, 5% sand)/ (Clay & Silt)/ (Silt & Sand) [129,130]	Pore-filling [130]; Nodules [131]	Disc./ Cont./ Cont./ Cont. [130]	Stra./ Vent./ Vent./ Vent. [130]	219/ 221/ 204/ 181 [130]	1500 [130]	5.2/ 4.84/ 5.53/ 6.44 [130]	47.53/ 46.95/ 49.34/ 43.65 [130]	222/ 229/ 206/ 184 [130]	276.4/ 288.72/ 257.78/ 272.06
40	Hikurangi Trough (Site U1517)	ACM	40% silt, 58% clay, 2% sand [132]	Lenses [133]	(Cont.) & (Doub./Mult.) [134]	Stra. [134]	165 [134]	725 [134]	5.32 [135]	39.8 [135]	130.58	175.57
41/ 42/ 43	Chile (Site 859/Site 860/Site 861)	ACM	(Silty clay & Clayey silt)/ (44% silt, 49% clay, 7% sand)/ (Coarse-grained clastic sediment) [136,137]	NA	Disc./ Cont./ Disc. [136]	Stru./ Stru./ Stru. [136]	100/ 200/ 250 [138]	2741.2/ 2145.9/ 1652 [136]	NA/ 2.5 [137]/ NA	NA/ 103.5 [137]/ NA	NA/ 160.2/ NA	NA/ 174.46/ NA

Table 4. Global statistical analysis of known BSR data (Arctic).

Code	Location	TGF	Sediment Features	Pore Habit	SMT	GST	ZBSR (mbsf)	WD (m)	Tsf (°C)	GG (°C/km)	ZBGHSZ (mbsf)	ZBGHSZ* (mbsf)
44	Sverdrup Basin	PCM	Sandstone [139,140]	NA	Disc. [141]	Stru. [141]	900 [141]	NA	NA	20–40 [139]	NA	NA
45	Fram Strait (Site 986)	PCM	30% silt, 65% clay, 5% sand [142]	NA	Disc. [143]	Stru. [143]	200 [143]	2500 [143]	0 [143]	75 [143]	274.71	294.68
46	Aleutian Trench (Site 186)	ACM	25% silt, 45% clay, 5% spicules, 25% diatoms [144]	NA	Disc. [145]	Stru. [145]	926 [145]	4500 [145]	1 [145]	20 [145]	1306.05	1385.6
47	Shirshov Ridge/Koryak	ACM	11.3–25.8% silt, 5.6–12.1% clay, 62.1–83.1% sand [146]	NA	Disc. [147]	Stru. [147]	200–500 [147]	1500–3000 [147]	NA	58 [147]	NA	NA
48	Barents Sea (Site 7316/03-U-01)	PCM	20% silt, 60% clay, 20% sand [148]	NA	Clus. [39,149]	Stru. [39,149]	180 [150]	345 [150]	2 [149]	NA	NA	NA
49	Offshore NW Greenland (Baffin Bay)	PCM	NA	NA	Disc. [151]	Stru. [151]	200 [151]	625–720 [151]	10–11.6 [151]	40–59 [151]	NA	NA

Table 4. Global statistical analysis of known BSR data (Arctic).

Code	Location	TGF	Sediment Features	Pore Habit	SMT	GST	ZBSR (mbsf)	WD (m)	Tsf (°C)	GG (°C/km)	ZBGHSZ (mbsf)	ZBGHSZ* (mbsf)
44	Sverdrup Basin	PCM	Sandstone [139,140]	NA	Disc. [141]	Stru. [141]	900 [141]	NA	NA	20–40 [139]	NA	NA
45	Fram Strait (Site 986)	PCM	30% silt, 65% clay, 5% sand [142]	NA	Disc. [143]	Stru. [143]	200 [143]	2500 [143]	0 [143]	75 [143]	274.71	294.68
46	Aleutian Trench (Site 186)	ACM	25% silt, 45% clay, 5% spicules, 25% diatoms [144]	NA	Disc. [145]	Stru. [145]	926 [145]	4500 [145]	1 [145]	20 [145]	1306.05	1385.6
47	Shirshov Ridge/Koryak	ACM	11.3–25.8% silt, 5.6–12.1% clay, 62.1–83.1% sand [146]	NA	Disc. [147]	Stru. [147]	200–500 [147]	1500–3000 [147]	NA	58 [147]	NA	NA
48	Barents Sea (Site 7316/03-U-01)	PCM	20% silt, 60% clay, 20% sand [148]	NA	Clus. [39,149]	Stru. [39,149]	180 [150]	345 [150]	2 [149]	NA	NA	NA
49	Offshore NW Greenland (Baffin Bay)	PCM	NA	NA	Disc. [151]	Stru. [151]	200 [151]	625–720 [151]	10–11.6 [151]	40–59 [151]	NA	NA

Table 5. Global statistical analysis of known BSR data (Antarctic).

Code	Location	TGF	Sediment Features	Pore Habit	SMT	GST	ZBSR (mbsf)	WD (m)	Tsf (°C)	GG (°C/km)	ZBGHSZ (mbsf)	ZBGHSZ* (mbsf)
50	Ross Sea (Site 273)	PCM	40% silt, 59% clay, 1% sand [152]	NA	Disc.& (Doub./Mult.: BSR1-SI hydrate; BSR2-sII hydrate) [153]	Stru. [153]	500 [153]	880 [153]	−1.5 [153]	36 [153]	442.96	489.1
51	Scan Basin (BSR1/BSR2/BSR3)	ACM	NA	NA	((Cont.)/(Diag.: Opal-A/CT)/(Diag.: Opal-CT/Quartz)) & (Mult.) [62]	Vent. [62]/ NA/ NA	145/ 429/ 1360 [62]	2100 [62]	−0.4 [62]/ NA/ NA	130/ (60–70)/ NA [62]	148.25	159.51
52	Weddell Sea (Site 695)	PCM	NA	NA	Diag. [154]	NA	690 [154]	1300 [154]	NA	52 [154]	NA	NA

Table 5. Global statistical analysis of known BSR data (Antarctic).

Code	Location	TGF	Sediment Features	Pore Habit	SMT	GST	ZBSR (mbsf)	WD (m)	Tsf (°C)	GG (°C/km)	ZBGHSZ (mbsf)	ZBGHSZ* (mbsf)
50	Ross Sea (Site 273)	PCM	40% silt, 59% clay, 1% sand [152]	NA	Disc.& (Doub./Mult.: BSR1-SI hydrate; BSR2-sII hydrate) [153]	Stru. [153]	500 [153]	880 [153]	−1.5 [153]	36 [153]	442.96	489.1
51	Scan Basin (BSR1/BSR2/BSR3)	ACM	NA	NA	((Cont.)/(Diag.: Opal-A/CT)/(Diag.: Opal-CT/Quartz)) & (Mult.) [62]	Vent. [62]/ NA/ NA	145/ 429/ 1360 [62]	2100 [62]	−0.4 [62]/ NA/ NA	130/ (60–70)/ NA [62]	148.25	159.51
52	Weddell Sea (Site 695)	PCM	NA	NA	Diag. [154]	NA	690 [154]	1300 [154]	NA	52 [154]	NA	NA

Table 6. Global statistical analysis of known BSR data (Indian Ocean).

Code	Location	TGF	Sediment Features	Pore Habit	SMT	GST	ZBSR (mbsf)	WD (m)	Tsf (°C)	GG (°C/km)	ZBGHSZ (mbsf)	ZBGHSZ* (mbsf)
53	Gulf of Oman (Site 222)	MCM	64–76% clay, 24–36% silt, 0% sand [155]	NA	Disc. [156]	Stru. [156]	680 [156]	3000 [156]	2 [156]	30 [156]	701.02	753.51
54/ 55/ 56/ 57/ 58/ 59	Krishna-Godavari Basin (NGHP-01-01/NGHP-01-07/NGHP-01-09/NGHP-01-17/NGHP-01-21)	PCM	(Carbonate oozes)/ (Clay with silt or sand beds)/ (Clay or silt)/ (Clay or silt)/ (Clay or silt with volcanic ash beds)/ (50–70% clay, 30–50% silt) [157,158]	Nodules/ Lenses [159]	Diag./ Disc./ Disc./ Clus./ Clus./ Disc. [158]	NA/ Stra./ Stru./ Stru./ Stra./ Stra. [158]	220/ 188/ 290/ 160/ 608/ 160 [158]	2663/ 1285/ 1935/ 1038/ 1344/ 1049 [158]	2.4/ 5.21/ 5/ 6.5/ 5.6/ 6 [158]	52/ 51/ 51/ 45/ 19/ 45 [158]	NA/ 200.6/ 268.81/ 156.82/ 645.98/ 172.2	NA/ 230.98/ 298.85/ 192.61/ 740.87/ 207.9
60	Sunda Arc	ACM	Interbedded sandstone, argillaceous siltstone, dolomitic limestone, and volcanic deposits form alternating high-low permeability sequences [160]	NA	Disc. [160]	Stru. [160]	150 [160]	1500–2200 [160]	NA	NA	NA	NA
61/ 62	Western Indian Ocean Offshore Tanzania (BSR1 (Biogenic)/BSR2 (thermogenic))	ACM	NA/ Hemipelagic units, slope channel deposits & turbidite sands [60]	NA	Disc./ Disc. [60]	Stru./ Stru. [60]	350/ 450 [60]	1800/ 2362.5 [60]	9/ 0.7 [60]	28.2/ 54 [60]	335.09/ 366	393.24/ 294.2
63	North Carnarvon Bsin	PCM	Carbonate ooze [161]	NA	Disc. [161]	Stru. [161]	NA	> 1000 [161]	NA	54–63 [161]	258 [161]	NA

Table 6. Global statistical analysis of known BSR data (Indian Ocean).

Code	Location	TGF	Sediment Features	Pore Habit	SMT	GST	ZBSR (mbsf)	WD (m)	Tsf (°C)	GG (°C/km)	ZBGHSZ (mbsf)	ZBGHSZ* (mbsf)
53	Gulf of Oman (Site 222)	MCM	64–76% clay, 24–36% silt, 0% sand [155]	NA	Disc. [156]	Stru. [156]	680 [156]	3000 [156]	2 [156]	30 [156]	701.02	753.51
54/ 55/ 56/ 57/ 58/ 59	Krishna-Godavari Basin (NGHP-01-01/NGHP-01-07/NGHP-01-09/NGHP-01-17/NGHP-01-21)	PCM	(Carbonate oozes)/ (Clay with silt or sand beds)/ (Clay or silt)/ (Clay or silt)/ (Clay or silt with volcanic ash beds)/ (50–70% clay, 30–50% silt) [157,158]	Nodules/ Lenses [159]	Diag./ Disc./ Disc./ Clus./ Clus./ Disc. [158]	NA/ Stra./ Stru./ Stru./ Stra./ Stra. [158]	220/ 188/ 290/ 160/ 608/ 160 [158]	2663/ 1285/ 1935/ 1038/ 1344/ 1049 [158]	2.4/ 5.21/ 5/ 6.5/ 5.6/ 6 [158]	52/ 51/ 51/ 45/ 19/ 45 [158]	NA/ 200.6/ 268.81/ 156.82/ 645.98/ 172.2	NA/ 230.98/ 298.85/ 192.61/ 740.87/ 207.9
60	Sunda Arc	ACM	Interbedded sandstone, argillaceous siltstone, dolomitic limestone, and volcanic deposits form alternating high-low permeability sequences [160]	NA	Disc. [160]	Stru. [160]	150 [160]	1500–2200 [160]	NA	NA	NA	NA
61/ 62	Western Indian Ocean Offshore Tanzania (BSR1 (Biogenic)/BSR2 (thermogenic))	ACM	NA/ Hemipelagic units, slope channel deposits & turbidite sands [60]	NA	Disc./ Disc. [60]	Stru./ Stru. [60]	350/ 450 [60]	1800/ 2362.5 [60]	9/ 0.7 [60]	28.2/ 54 [60]	335.09/ 366	393.24/ 294.2
63	North Carnarvon Bsin	PCM	Carbonate ooze [161]	NA	Disc. [161]	Stru. [161]	NA	> 1000 [161]	NA	54–63 [161]	258 [161]	NA

Table 7. Global statistical analysis of known BSR data (Inland seas and lakes).

Code	Location	TGF	Sediment Features	Pore Habit	SMT	GST	ZBSR (mbsf)	WD (m)	Tsf (°C)	GG (°C/km)	ZBGHSZ (mbsf)	ZBGHSZ* (mbsf)
64	Black Sea	MCM	Particle size distribution: >0.1 mm (not detected), 0.01–0.1 mm (10–30%), <0.01 mm (70–90%) [162]	Nodules and lenses [163]	(Disc.) & (Doub./Mult.) [164]	Stru. [164]	320 [164]	1330 [164]	9 [165]	24.5 [165]	293.45	364.95
65	Baikal Lake	TRL	70% clay, 5% silt, 25% sand [166]	Nodules and lenses [163]	Disc. [163]	Stru. [163]	160 [163]	1600 [163]	3.5 [163]	NA	NA	NA
66	Anaximander Mud Volcanoes	ACM	56–67% clay, 19–30% silt, 14% sand [167]	NA	Diag. [167]	NA	40–200 [167]	2025 [167]	13.75 [167]	26–38 [167]	NA	NA
67	Nile Deep-Sea Fan	PCM	Fine grained turbidite sediments and MTDs [168]	NA	Disc. [168]	Stru. [168]	190 [168]	2443 [168]	12.5 [168]	40 [168]	191.9	230.81
68	Western Marmara Sea	TPB	50–70% clay, 30–40% silt, 0–1% sand [169]	Nodules [170]	Cont. [171]	Vent. [171]	25 [170,171]	680 [170,171]	14.5 [170,171]	39.3 [170]	−324.95	−232.16

Table 7. Global statistical analysis of known BSR data (Inland seas and lakes).

Code	Location	TGF	Sediment Features	Pore Habit	SMT	GST	ZBSR (mbsf)	WD (m)	Tsf (°C)	GG (°C/km)	ZBGHSZ (mbsf)	ZBGHSZ* (mbsf)
64	Black Sea	MCM	Particle size distribution: >0.1 mm (not detected), 0.01–0.1 mm (10–30%), <0.01 mm (70–90%) [162]	Nodules and lenses [163]	(Disc.) & (Doub./Mult.) [164]	Stru. [164]	320 [164]	1330 [164]	9 [165]	24.5 [165]	293.45	364.95
65	Baikal Lake	TRL	70% clay, 5% silt, 25% sand [166]	Nodules and lenses [163]	Disc. [163]	Stru. [163]	160 [163]	1600 [163]	3.5 [163]	NA	NA	NA
66	Anaximander Mud Volcanoes	ACM	56–67% clay, 19–30% silt, 14% sand [167]	NA	Diag. [167]	NA	40–200 [167]	2025 [167]	13.75 [167]	26–38 [167]	NA	NA
67	Nile Deep-Sea Fan	PCM	Fine grained turbidite sediments and MTDs [168]	NA	Disc. [168]	Stru. [168]	190 [168]	2443 [168]	12.5 [168]	40 [168]	191.9	230.81
68	Western Marmara Sea	TPB	50–70% clay, 30–40% silt, 0–1% sand [169]	Nodules [170]	Cont. [171]	Vent. [171]	25 [170,171]	680 [170,171]	14.5 [170,171]	39.3 [170]	−324.95	−232.16

Table 8. Global statistical analysis of known BSR data (Continental permafrost).

Code	Location	TGF	Sediment Features	Pore Habit	SMT	ZBSR (mbsf)	WD (m)	Tsf (°C)	GG (°C/km)	ZBGHSZ (mbsf)	ZBGHSZ* (mbsf)
69	Qilian Mountains Tibet	CC	Sandstones (medium to fine-grained), siltstones, mudstones, and oil-bearing shales [172]	Fracture filling in rock [173,174]	NA	133–396 [173]	0	NA	NA	NA	NA
70	West Siberia Basin	ISB	Particle size distribution: 8% at 0.2 mm, 4% at 0.5 mm, 4% at 0.8 mm, and 84% exceeding 1 mm. [175]	NA	NA	800 [175]	0	NA	NA	NA	NA
71	Mallik	ISB	Fine/medium sand, mean grain size: 149.9–502.5 μm [117,176]	Pore-filling [176]	NA	1000 [176]	0	NA	NA	NA	NA
72	Lena-Tunguska Basin	ISB	NA	NA	NA	800–2000 [139]	0	NA	NA	NA	NA
73	Mt Elbert	CC	56–61% <75 μm, mean grain size: 0.07–0.074 mm [177,178]	Pore-filling [117,176]	NA	850 [177]	0	NA	NA	NA	NA

Table 8. Global statistical analysis of known BSR data (Continental permafrost).

Code	Location	TGF	Sediment Features	Pore Habit	SMT	ZBSR (mbsf)	WD (m)	Tsf (°C)	GG (°C/km)	ZBGHSZ (mbsf)	ZBGHSZ* (mbsf)
69	Qilian Mountains Tibet	CC	Sandstones (medium to fine-grained), siltstones, mudstones, and oil-bearing shales [172]	Fracture filling in rock [173,174]	NA	133–396 [173]	0	NA	NA	NA	NA
70	West Siberia Basin	ISB	Particle size distribution: 8% at 0.2 mm, 4% at 0.5 mm, 4% at 0.8 mm, and 84% exceeding 1 mm. [175]	NA	NA	800 [175]	0	NA	NA	NA	NA
71	Mallik	ISB	Fine/medium sand, mean grain size: 149.9–502.5 μm [117,176]	Pore-filling [176]	NA	1000 [176]	0	NA	NA	NA	NA
72	Lena-Tunguska Basin	ISB	NA	NA	NA	800–2000 [139]	0	NA	NA	NA	NA
73	Mt Elbert	CC	56–61% <75 μm, mean grain size: 0.07–0.074 mm [177,178]	Pore-filling [117,176]	NA	850 [177]	0	NA	NA	NA	NA

3. BSR as Indicators for the Presence of BGHSZ and Gas Hydrate/Free Gas

The GHSZ is primarily controlled by pressure (P) and temperature (T), with temperature variations exerting a significantly greater impact on hydrate stability than pressure fluctuations [179]. BSR has long been recognized as the indicator for identifying gas hydrate occurrences; however, the precise correspondence between BSR depth and the BGHSZ remains debated. Some studies suggest that BSR depth accurately represents the BGHSZ [26,180,181,182,183,184], whereas others argue that BSR depth typically occurs shallower than the BGHSZ, indicating a systematic offset [185].

3.1. Comparison of BSR Depth and Predicted Depth of BGHSZ

To assess this relationship, we used the Colorado School of Mines Hydrate (CSMHYD) program to compute methane hydrate phase equilibrium curves under both 3.5 wt.% salinity and pure water conditions [186]. By integrating measured data from 34 global hydrate-bearing sites with observed BSR, we derived the sub-seafloor depths of the BGHSZ (in meters below seafloor, mbsf) for both scenarios: Z_BGHSZ, 3.5 wt.% salinity (seawater); and Z_BGHSZ*, pure water (derived from the datasets in Table 2, Table 3, Table 4, Table 5, Table 6, Table 7 and Table 8 and visualized in Figure 3c,d) [187]. We then conducted linear correlation analyses comparing these predicted depths with observed BSR depths ($Z_{\mathrm{B}\mathrm{S}\mathrm{R}}$). Notably, we introduced ΔZ as a quantitative measure of the depth deviation between BGHSZ and BSR (Equation (3)). Considering the vertical resolution of seismic data (10–20 m), a 20 m threshold criterion was established: BSR was considered coincident with BGHSZ when ΔZ ≤ 20 m, while BSR was determined to be shallower than BGHSZ when this threshold was exceeded.

$$\mathsf{\Delta }Z=\mathrm{M}\mathrm{i}\mathrm{n}\mathrm{i}\mathrm{m}\mathrm{u}\mathrm{m}(\left|Z_{\mathrm{B}\mathrm{G}\mathrm{H}\mathrm{S}\mathrm{Z}}-Z_{\mathrm{B}\mathrm{S}\mathrm{R}}\right|, \left|Z_{\mathrm{B}\mathrm{G}\mathrm{H}\mathrm{S}\mathrm{Z}}\mathrm{*}-Z_{\mathrm{B}\mathrm{S}\mathrm{R}}\right|)$$

(3)

In contrast to theoretical expectations, our analysis identified three sites where BSR depths exceeded the predicted BGHSZ, all corresponding to Type II gas hydrate systems (confirmed from the literature review, as described in Figure 3a). These anomalies were consequently excluded from further analysis. As illustrated in Figure 3b, the surveyed depths of BSRs coincide with the theoretical base of the BGHSZ at 64.5% of the surveyed sites. However, 35.48% exhibited BSRs shallower than the BGHSZs, indicating that BSR should not be treated as a definitive proxy for the BGHSZ. Instead, it more reliably marks either the base of the hydrate occurrence zone (BGHOZ) or the top of the free gas zone (TFGZ).

This discrepancy arises from the potential existence of free gas between BGHOZ and BGHSZ under two specific conditions:

• Capillary pressure effects in fine-grained sediments can inhibit hydrate formation, allowing methane to persist in gaseous form [188].
• Local variations in P-T conditions triggered by seasonal fluctuations, elevated bottom-water temperatures, or tectonic uplift can induce partial dissociation of gas hydrates, temporarily releasing free gas and creating conditions of disequilibrium [28,29].

Theoretical models demonstrated that BGHSZ, BGHOZ, and TFGZ coincide spatially only when methane concentration exceeds solubility thresholds and flux surpasses critical rates [185]. Moreover, we found that when the BSR depth is less than approximately 2700 mbsl, its deviation from the BGHSZ is relatively small (average ΔZ = 11.5 m); whereas when the BSR depth exceeds ~2700 mbsl, it is generally much shallower than the BGHSZ (average ΔZ = 112.2 m). It is important to note that the interpretation of this depth-dependent deviation must be based on reliable data quality and accurate BSR identification. Excluding potential errors from literature data or misinterpretation of BSR, the substantial ΔZ observed in deeper waters may be attributed to reduced methane flux caused by a sharp decline in organic matter input, preventing the BSR from fully developing down to the theoretical base of the stability zone. Accordingly, BSRs located above 2700 mbsl can be considered approximate indicators of the BGHSZ, whereas, in deeper regions, systematic deviations between the BGHSZ and BGHOZ must be accounted for.

These findings improve the accuracy of GHSZ delineation for resource evaluation while enhancing methane leakage monitoring capabilities. Rigorous validation of BSR interpretations remains essential, especially in deep-water environments where data uncertainties amplify.

3.2. BSR as an Indicator of Gas Hydrate and Free Gas Presence

The relationship between BSRs, free gas, and gas hydrates exhibits complex variability, with three primary scenarios requiring differentiation.

3.2.1. Gas Hydrate Presence Without Observable BSR

Drilling results from the Gulf of Mexico and Blake Ridge demonstrate that gas hydrates can exist without seismic BSR expression when free gas is absent below the hydrate layer [181,189]. This eliminates the acoustic impedance contrast needed for BSR generation, as hydrate-bearing sediments and water-saturated strata exhibit similar seismic properties.

When the thickness of the free gas zone causes destructive interference between seismic reflections from its upper and lower boundaries, the BSR signal may weaken or disappear. Specifically, the seismic waves traveling between these interfaces generate phase differences. If this travel path difference is an odd multiple of the wavelength, the reflected waves interfere destructively, eliminating the observable BSR signal [190].

3.2.2. False “BSR” Without Gas Hydrates Presence

The BSR is caused by contrasts in acoustic impedance; however, similar strong reflections can also be generated by other factors even in the absence of gas hydrates. High-saturation free gas zones can produce negative-polarity reflections mimicking BSRs through abrupt P-wave velocity reduction, even in hydrate-free systems [191]. Diagenesis forms pseudo-BSR features through two distinct mechanisms. Silicate diagenesis (e.g., opal-A→CT→quartz and smectite→illite transitions) generates acoustic impedance contrasts via pore compression and dehydration, forming sub-seafloor-parallel interfaces controlled by geothermal gradients [31]. These interfaces exhibit positive polarity and occur at depths significantly below the hydrate stability zone, distinguishing them from the hydrate-related BSR [62,154,192]. Carbonate diagenesis, observed in the Xisha Trough, Namibe Basin Site 1080, and Krishna-Godavari Basin NGHP-01-01, produces high-amplitude continuous reflectors with negative polarity and geometry akin to the hydrate-related BSR [32,87,158]. Liang et al. developed a Carbonate-Hydrate Identification Template (CHIT) using a hybrid matrix velocity model, confirming that these carbonate-derived reflectors are independent of thermobaric conditions, fundamentally differing from hydrate phase equilibria [32].

Mass-transport deposits (MTDs), formed by gravity-driven submarine slope failures, often display complex internal architectures. Variations in density and seismic velocity between displaced sediment blocks can generate strong reflections and polarity reversals. While these reflections may exhibit polarity characteristics similar to the BSR, their irregular geometries and discontinuous reflection patterns reflect sediment dynamics governed by slope morphology and are not related to the hydrate stability zone [168,193].

4. Controls on BSR Morphology in Gas Hydrate Systems

The morphology of BSR is influenced by three main factors: geological settings, sediment types, and seismic data acquisition methods. The first two reflect natural geological conditions, whereas the third arises from instrumental and methodological constraints.

4.1. Influence of Geological Settings

The geological settings in which BSR occurs can be broadly categorized into structural settings, stratigraphic settings, and venting settings. Each setting is governed by distinct geological characteristics and formation mechanisms [184], as illustrated in Figure 4.

4.1.1. BSRs in Structural Settings

Structural settings account for 58.46% of all documented BSR occurrences, making them the predominant category (as detailed in Table 2, Table 3, Table 4, Table 5, Table 6, Table 7 and Table 8). However, structural deformation—including faulting, mud diapirism, and salt tectonics—typically limits their lateral continuity, resulting in smaller-scale BSRs. The gas responsible for forming the BSR in these environments often originates from deep, focused hydrothermal fluids. Based on its distribution and the gas hydrate accumulation process, the BSR in tectonic settings can be further classified into three types, as illustrated in Figure 5 [195].

The ridge-type BSR develops at structural highs and appears as a convex-shaped reflection (Figure 5a). In this setting, free gas ascends preferentially along the ridge axis and accumulates beneath the GHSZ, generating a well-defined BSR. Prominent examples include gas hydrate systems in the Sea of Okhotsk [124,125].

The buried anticline-type BSR forms within anticlines that are overlain by younger sediments (Figure 5b). This BSR typically exhibits a flat-lying seismic reflector, parallel to the seafloor, where upward-migrating gas accumulates beneath the overlying seal. Such configurations have been documented in several sedimentary basins, including the southwestern margin of the Kumano Basin in the Nankai Trough [195], the eastern Beaufort outer slope [196], and the South Makassar Basin in Indonesia [197].

The accretionary prism-type BSR is mainly found within accretionary wedges (Figure 5c). Its distribution is less influenced by local geological structures or seafloor topography. Instead, it is primarily controlled by intense and diffuse upward fluid flow driven by tectonic thickening and sediment loading. Representative examples have been reported in the Gulf of Mexico and Acapulco [110], as well as offshore Mauritania [58].

Clustered and discontinuous forms of the BSR are more commonly observed in structural settings (Figure 4(a1–a2,b1–b2)). Among the worldwide BSR observations, clustered BSRs account for 10.53% and discontinuous BSRs for 73.68% of all BSR occurrences identified in structural settings, as summarized in Table 2, Table 3, Table 4, Table 5, Table 6, Table 7 and Table 8. Notably, all instances of the clustered BSR are found exclusively in such settings.

This prevalence can be attributed to the role of faults and fracture systems, which serve as vertical pathways for deep-sourced fluids. These conduits facilitate the focused migration and accumulation of gas, resulting in zones of high hydrate saturation near the BGHSZ, and ultimately lead to the development of the localized, clustered BSR [198,199].

In addition, tectonic activity often generates numerous faults that enable the upward movement of hot fluids from depth, which can inhibit hydrate formation and cause interruptions in the continuity of the BSR [200,201]. Furthermore, processes such as salt tectonics and folding can lead to tilting of the underlying strata. This structural deformation alters porosity and permeability at different positions along the BGHSZ/BGHOZ, resulting in uneven distributions of gas and hydrates and ultimately giving rise to the discontinuous BSR [184].

4.1.2. BSRs in Stratigraphic Settings

BSRs in stratigraphic settings account for 24.62% of all identified BSRs, as indicated in Table 2, Table 3, Table 4, Table 5, Table 6, Table 7 and Table 8. These BSRs lack obvious fluid venting structures or seismically detectable faults. The methane responsible for hydrate formation in these settings is more likely derived from in situ microbial degradation of organic matter [202] or from deep thermogenic gas migrating through high-permeability layers or undetectable fractures [203]. In these settings, continuous BSRs (56.25%) dominate over discontinuous BSRs (43.75%) with their distribution strongly influenced by the stratigraphic geometry relative to the BGHSZ (Figure 3c,d).

Uniform porosity/permeability along the BGHSZ/BGHOZ promotes homogeneous fluid distribution, forming continuous BSRs (e.g., Puke Ridge in the Hikurangi Trough [204], Site 997 at Blake Ridge [67], and Site NGHP-01-17 in the Krishna-Godavari Basin [158]). Heterogeneous sediment permeability leads to compartmentalized gas/hydrate accumulations, generating discontinuous BSRs (e.g., Sackville Spur offshore Newfoundland [94]).

4.1.3. BSRs in Venting Settings

Venting-related BSRs represent 16.92% of observed occurrences and are linked to focused fluid expulsion through mud volcanoes, gas chimneys, or hydrothermal vents. In venting environments, continuous BSRs and pluming BSRs are the most commonly observed types, comprising 81.82% and 9.09% of venting-related BSRs in the statistical region, respectively (Figure 4e1–e2). On seismic profiles, they typically exhibit circular or semi-circular structures, often accompanied by acoustic blanking zones in the surrounding sediments [51,205].

4.2. Influence of Sediment Types

Triangular grain-size distribution analysis (Figure 6) indicates that BSR preferentially occurs in fine-grained sediments with high clay (40–80%) and silt (20–60%) content, particularly in sediment types such as clay, silty clay, clayey silt, and sand–silt–clay. In contrast, BSR is rare or nearly absent in coarser sediments where sand content exceeds 50%, such as sand and silty sand.

Continuous BSRs predominantly develop in fine-grained sediments (clay, sandy clay, and silty clay), characterized by small grain sizes and high clay content (40–80%). This preference may arise from the higher sealing capacity of fine-grained materials, which facilitates stable and continuous reflections. In contrast, discontinuous BSRs exhibit broader sedimentological affinity, occurring in mixed grain-size deposits (e.g., sandy clay, sand–silt–clay, silty clay) with medium-to-fine textures. Their irregular distribution likely results from heterogeneous geological controls, including permeability variations and localized fluid migration pathways. Clustered BSR is relatively rare and is exclusively found within structural settings. Previous studies suggest that its occurrence is often associated with coarse-grained sediments such as channel sands [76].

Sediment type not only influences the seismic morphology of the BSR but also determines the pore habit of gas hydrates. In fine-grained sediments, high capillary pressure promotes hydrate growth within small pores, replacing sediment grains and forming discrete hydrate lenses. This pore habit is referred to as “grain-displacive” hydrate formation. In contrast, in clean, coarse-grained sediments, hydrates experience lower capillary pressure within larger pores, allowing them to fill or invade pores without altering the sediment framework. This behavior is known as “pore-invasive” or “pore-filling” hydrate formation [203,206]. Pore-invasive hydrates are primarily observed in the Nankai region, Shenhu area, and Arctic permafrost zone, characterized by extensive distribution but lower saturation, making them less economically viable for extraction.

4.3. Influence of Seismic Data Acquisition Methods

The BSR morphology is shaped not only by geological settings and sediment types but also by seismic data acquisition parameters, including sampling intervals, frequency, and survey line orientation. Smaller sampling intervals and higher frequencies enhance seismic resolution, which comprises vertical and horizontal components. The vertical resolution (h) approximates one-quarter of the seismic wavelength, while the horizontal resolution (r) can be expressed by the radius of the Fresnel zone (Equations (4) and (5)). The continuity of BSR is predominantly influenced by horizontal resolution; higher seismic frequencies result in shallower propagation depths and smaller Fresnel zone radii, thus improving horizontal resolution.

$$h={\displaystyle \frac{v}{4f}}$$

(4)

$$r=\sqrt{{\displaystyle \frac{v·z}{2f}}}$$

(5)

where v is the seismic wave velocity, f is the dominant seismic frequency, and z is the target depth.

Studies by Hillman and Crutchley demonstrated that low-frequency seismic data can better capture extensive features, such as gas–water interfaces, thereby presenting BSR as a continuous reflection. However, these data often lack sufficient detail to elucidate complex gas migration pathways. Conversely, high-frequency data provide more detailed images of the specific distribution of hydrates and free gas near the GHSZ base, yet these data generally exhibit lower amplitudes due to greater attenuation [69,207]. Consequently, a BSR appearing continuous in low-frequency data might appear discontinuous in high-frequency data from a lateral perspective. Vertically, this also implies that a BSR represents a transition zone of a certain thickness, within which hydrate progressively transitions into free gas, rather than a simple binary interface between a hydrate-bearing layer and a free gas-bearing layer as traditionally hypothesized.

Furthermore, the orientation of seismic survey lines significantly affects the continuity of BSR, particularly when sediment layers are inclined. If the seismic line is oriented parallel to the direction of sediment tilt, it traverses multiple inclined layers intersected by the BSR. Variations in porosity and permeability along the same isotherm, resulting from sediment tilt, typically cause BSR to appear discontinuous. Conversely, if the seismic line is perpendicular to the direction of sediment tilt, the distribution of porosity and permeability within the same stratigraphic layer is relatively uniform, leading to continuous BSR reflections. However, in this case, the absence of information regarding sediment inclination may result in a BSR that does not accurately reflect the true distribution of hydrates and free gas [69].

As shown in Section 4.1 and Section 4.2, disregarding the influence of seismic acquisition methods, both geological settings and sediment types jointly affect the morphology of the BSR. Continuous BSRs prevail in stable seepage environments with homogeneous lithology (e.g., fine-grained seal layers overlying permeable pathways). These settings facilitate fluid/gas migration and accumulation, yielding consistent gas hydrate formation. Discontinuous BSRs dominate tectonically active regions with heterogeneous sediments (e.g., sandy clay and sand–silt–clay mixtures). Here, structural deformation (e.g., faulting) disrupts gas distribution, while lithological variability further amplifies BSR segmentation. This multiscale analysis clarifies the interplay between sedimentation and tectonics in BSR formation, providing critical insights for gas hydrate prospecting and reservoir characterization.

5. Formation Mechanisms of Double and Multiple BSRs

Although BSR is widely distributed along continental margins, double or multiple BSRs are relatively rare. Global observational data indicate that their presence primarily occurs along the continental margin of western Norway [208,209], the Nankai Trough offshore Japan [195], the Manila subduction zone [210], Hydrate Ridge offshore Oregon [30], the Ross Sea in Antarctica [153], the Danube deep-sea fan in the Black Sea [165], the Tumbes Basin offshore Peru [211], the Hikurangi margin offshore New Zealand [212], the South China Sea [213,214,215], Sackville Spur offshore Newfoundland [94], the Mauritanian continental margin in West Africa [216], and the Makran accretionary wedge along the northwestern coast of the Indian Ocean [217].

Double BSRs typically reflect disequilibrium states in gas hydrate systems, with the primary reflector (BSR1) corresponding to equilibrium conditions under current or recent P-T conditions, whereas the secondary reflector (BSR2) represents a relict paleo-BSR formed due to changes in P-T conditions leading to partial dissipation [218]. In specific instances, differences in hydrate type may also result in the formation of double BSRs [219]. Typically situated near ridge crests, double BSRs appear as two approximately parallel reflectors that mimic the seafloor topography and display polarity reversals. Notably, BSR2 often exhibits weaker amplitude and poorer continuity than BSR1 [211,220]. High acoustic impedance anomalies above and adjacent to both BSR1 and BSR2 indicate accumulations of natural gas hydrates, whereas low acoustic impedance anomalies directly below these reflectors suggest the presence of free gas accumulations [221]. Fundamentally, double/multiple BSRs reflect the dynamic response of gas hydrate systems to variations in P-T conditions. The mechanisms of their formation can be classified into several types, as described in Section 5.1, Section 5.2, Section 5.3, Section 5.4, Section 5.5 and Section 5.6.

5.1. Rapid Sedimentation

As depicted in Figure 7 and Figure 8a, continuous burial (with the seafloor shifting from Seafloor2 to Seafloor1) facilitates the transfer of deep heat to shallower, newly deposited strata. As the geothermal gradient is re-established, the GHSZ migrates upward from BGHSZ2 to BGHSZ1, leading to the formation of a new BSR in the shallower strata. Multiple rapid sedimentation events can induce repeated adjustments, resulting in the identification of double or multiple BSRs. Upward adjustments of the BGHSZ due to this mechanism have been observed in various marine regions, including the Black Sea, the Hikurangi subduction zone off New Zealand, the Nankai Trough in Japan, and the Manila subduction zone, as shown in Table 3, Table 5 and Table 7. Through numerical simulations, Gupta et al. (2024) identified the primary factors influencing the formation of multiple BSRs: continuous sedimentation layers several tens of meters thick, low-permeability sediment layers (which facilitate the accumulation of free gas), and appropriate kinetics of hydrate formation and dissociation—specifically, slower dissociation rates (10⁻¹⁶ mol/m² Pa·s to 10⁻¹⁴ mol/m² Pa·s) coupled with higher formation rates (10⁻¹⁴ mol/m² Pa·s to 10⁻¹² mol/m² Pa·s) [222].

5.2. Tectonic Uplift and Subsidence

Figure 8b demonstrates that during tectonic uplift, a reduction in pressure drives the upward migration of the BGHSZ. As gas hydrates dissociate between the former BGHSZ and the newly established BGHSZ, two nearly parallel BSRs may temporarily coexist during this transition. If hydrates are present above the former BGHSZ, their dissociation during uplift can be significantly slowed due to the latent heat-buffering effect, thereby preserving BSR2 for a prolonged period [225]. It is important to note that the inferred uplift associated with double BSRs may result from multiple uplift events. A new BSR can only form if the quiescent period between successive uplift events is sufficiently long, allowing the dissociation of hydrates in the sediments above the former BGHSZ, the upward migration of free gas, and its subsequent accumulation beneath the newly established BGHSZ. Meanwhile, during tectonic subsidence, increased pressure drives the downward migration of the BGHSZ, in contrast to the process observed during tectonic uplift.

5.3. Upwelling of Thermal Fluids

According to Figure 8c, in regions of localized basement uplift, deep fluids migrate upward along faults, leading to an increased geothermal gradient in the shallow sediments. This results in the upward adjustment of the BGHSZ and the formation of anomalously shallow, arching BSRs [40]. This mechanism is similar to the BGHSZ migration induced by bottom water temperature (BWT) increases during interglacial periods [30,211,226]. However, unlike BWT-induced changes, the influence of thermal fluids on the original hydrate layer is typically more persistent and direct. As a result, this process does not necessarily lead to the formation of double or multiple BSRs.

5.4. Fluid Overpressure Induced by Slope Failure

As presented in Figure 8d, observations of microseismic activity and seismic attenuation indicate that highly overpressured fluids, generated at compressional margins, are expelled through the fracture network in the overlying strata following slope failure [227,228]. Shortly after slope failure, the upward migration of fluids increases pore pressure beneath the thrust ridge, causing the BGHSZ to shift downward [212]. In addition to slope failure, fluid overpressure can also result from various factors, including changes in fluid volume (e.g., temperature-induced expansion, fluid hydrocarbons generated from kerogen maturation, or H₂O release during smectite-to-illite transformation), fluid movement (e.g., buoyancy-driven flow or permeability effects), and compaction (e.g., reservoir compaction induced by tectonic stress or rapid sedimentation) [229].

5.5. Canyon Erosion and Subsequent Sedimentation

As shown by Figure 8e, an increase in seafloor depth on the canyon’s eroded side leads to a downward shift of the BGHSZ. Conversely, on the channel deposition side, sediment accumulation tends to cause an upward migration of the BGHSZ. Between BGHSZ1 and BGHSZ2, residual hydrates and/or free gas may persist. Under these contrasting P-T conditions, a distinct intersecting double BSR can ultimately form [215,230].

5.6. Different Types of Gas Hydrate Structure

In the scenario illustrated by Figure 8f, the mechanism is relatively unique, as both BSR1 and BSR2 remain in equilibrium under current P-T conditions. Their formation is closely linked to the presence of different gas hydrate structures [30,208,209,224]. Heavy hydrocarbons preferentially form sII hydrates, which exhibit stability under higher P-T conditions compared to structure I (sI) hydrates [211]. Therefore, BSR2 indicates the lower boundary of the Type sII GHSZ, typically situated beneath BSR1, which marks the base of the Type SI GHSZ [219]. In the case study from the Norwegian continental margin, the occurrence of a double BSR is more reasonably explained by this theoretical model than by invoking the existence of a paleo-BSR [209].

In addition, another category of double/multiple BSRs is unrelated to gas hydrates but instead results from variations in sediment lithology or mineralogy [220,231]. However, since this category is beyond the scope of this study, it will not be further discussed here.

The interpretations of double/multiple BSRs described above are reasonable within specific regions but cannot be broadly generalized due to the complexity and variability of geological controlling factors worldwide. Consequently, research on double/multiple BSR phenomena remains exploratory. Studying double BSRs contributes to a better understanding of the evolutionary mechanisms, timescales, and distribution patterns of gas hydrate systems.

6. Challenges

As a key seismic indicator for identifying natural gas hydrates, the accurate interpretation of the BSR faces multiple technical challenges. Conventional P-wave seismic profiles often exhibit non-uniqueness in BSR morphology, making it difficult to reliably distinguish real BSRs from pseudo ones. While AVO analysis provides hydrocarbon indicators, its dependence on empirical relationships and indirect S-wave estimation introduces uncertainties in fluid characterization. Although multi-wave, multi-component techniques based on OBS enable in-situ S-wave measurements, their widespread application is constrained by high acquisition costs and complex data processing and interpretation. Therefore, economically acquiring reliable P- and S-wave data and establishing efficient interpretation workflows have become critical bottlenecks in seismic exploration of gas hydrates.

The fundamental challenge in BSR research lies in characterizing the coupled seismic responses between the overlying hydrate-bearing layer and the underlying free gas zone. While conventional P-wave seismic interpretation provides qualitative geometrical analysis of BSRs, it proves insufficient for comprehensive reservoir assessment. To address this limitation, the research community must develop integrated quantitative methodologies that incorporate ① joint inversion of elastic properties characterizing the gas hydrate-free gas system, ② mechanistic studies of seismic attenuation and dispersion, and ③ physics-based correlations between BSR attributes (both dynamic and static) and hydrate saturation through rigorous rock physics modeling. These advances will ultimately transform BSR analysis from simple phase-boundary identification to the quantitative inversion of reservoir physical parameters.

7. Conclusions

This study synthesizes global BSR observations to systematically investigate the origins of pseudo-reflections and double/multiple BSRs, the relationship between BSR and the base of the BGHSZ, and the primary controls on BSR seismic morphology. The following conclusions can be drawn:

• BSRs, free gas zones, and GHBS exhibit complex, non-unique correlations. Reliable BSR identification requires integrating multiple seismic attributes rather than relying on a single diagnostic feature.
• Over 35% of global BSRs occur significantly shallower than theoretical BGHSZ predictions, particularly in deepwater basins. This confirms that the BSR primarily marks the base of the gas hydrate occurrence zone (BGHOZ) and the top of the free gas zone (TFGZ), not necessarily the thermodynamic GHSZ boundary.
• BSR morphology exhibits systematic variability controlled by three interlinked factors: ① geological setting, ② sediment properties, and ③ seismic acquisition constraints. Continuous BSRs typically develop in venting settings and in stratigraphic contexts where homogeneous strata are parallel to the BGHSZ. Discontinuous BSRs are often found in structural settings or where heterogeneous strata dip relative to the BGHSZ. Clustered BSR is predominantly associated with coarse-grained sediments in structural settings while pluming BSR is confined to venting settings linked to surface mud volcanoes or salt diapirs. The majority (~70–80%) of BSRs occur in fine-grained, grain-displacive sediments, where gas hydrates typically form as isolated lenses or nodules. In contrast, coarse-grained, pore-filling sediments account for a smaller proportion of BSR occurrences, typically less than 20%. Additionally, seismic data quality parameters—notably lateral resolution and survey-line orientation relative to geological structures—fundamentally govern the apparent continuity and sharpness of BSR reflections in seismic imagery.
• Double and multiple BSRs result from dynamic adjustments in gas hydrate systems, driven by sedimentation, tectonics, fluid migration, or differing hydrate phase stability. While these mechanisms provide regional explanations, their complexity demands further interdisciplinary research to predict hydrate system evolution.
• By integrating these four frontiers—frequency-dependent AVO, viscoelastic full waveform inversion (FWI), multicomponent anisotropy, and microstructural modeling—the next generation of BSR research will achieve quantitative, physics-constrained reservoir evaluation.
• The integration of high-resolution exploration technologies and intelligent algorithms is anticipated to shift BSR research from “static feature identification” to “dynamic process analysis”, potentially optimizing gas hydrate exploration efficiency while providing critical scientific insights into the interaction mechanisms between hydrate systems and marine environments.