Previous   |   Next

doi:10.2204/iodp.proc.311.213.2010

Constraints on the vertical extent of the gas hydrate stability zone and the occurrence of gas hydrate

A primary objective of Expedition 311 was to define the vertical extent of the gas hydrate stability zone (GHSZ) and the occurrence of gas hydrate. The depth of the base of gas hydrate stability zone (BGHSZ) was estimated before drilling by assuming that the seismically observed BSR represents the BGHSZ. Among the proxies used to determine the depth of the BGHSZ and also to define the occurrence of gas hydrate and to compare these occurrences to the predicted depth of the BSR are

• Downhole temperature measurements,

• Well-log measurements of P-wave velocity and electrical resistivity,

• Degassing experiments on pressure cores,

• Pore water chlorinity,

• IR imaging, and

• Hydrocarbon gas ratios (e.g., C₁/C₂ and i-C₄/n-C₄) of the void gas.

There are notable differences between the individual techniques because each is highly dependent on how the measurement is conducted and the resolution or sensitivity of the particular measurement and because the results are biased by core recovery (IR imaging, pore water chemistry, and gas chemistry) and sampling density (e.g., the frequency of temperature tool deployments and linear regression analysis). The results of the various techniques used to estimate the BGHSZ and predict the occurrence of gas hydrate at all sites are shown in Table T1.

The expedition included 36 temperature tool deployments to characterize the thermal regime of all five drilled sites. Three standard IODP temperature tools were deployed, including the Advanced Piston Corer Temperature Tool (APCT, as well as the newer version [APCT-3; Heesemann et al.], 20 times), the Davis-Villinger Temperature Probe (DVTP, 11 times), and the Davis-Villinger Temperature-Pressure Probe (DVTPP, 5 times). A compilation of the in situ temperature estimates from Expedition 311 is compared to previous results from Site 889 (Fig. F3). The temperature data acquired were used to estimate the depth of the BGHSZ at each site. A pure methane gas chemistry was assumed for the in situ hydrate (confirmed by shipboard analysis), and interstitial water salinities from the onboard analysis of core samples were used in the calculations. Furthermore, the pore pressure gradient was assumed to be hydrostatic (9.795 kPa/m) and the geothermal gradients linear. Results of all regression analyses at all sites are summarized in Table T1.

One of the more important methods used to identify and quantify the occurrence and extent of gas hydrate is pressure core degassing. Pressure cores retrieved at in situ pressure conditions were used to determine gas hydrate quantity (see also below) using mass balance calculations (e.g., Dickens et al., 1997; Milkov et al., 2004). Pressure cores were also used to investigate gas hydrate distribution using nondestructive physical property measurements of the cores at in situ pressures. Pressure coring is crucial for understanding the concentrations of gas hydrate and free methane gas in marine sediments, their nature and distribution, and their effect on the intrinsic properties of the sediment. Pressure cores were collected using the IODP Pressure Core Sampler (PCS), the Fugro Percussion Corer (FPC), and the HYACE Rotary Corer (HRC). Combined results from all degassing experiments at the five sites visited are shown in Figure F4.

Top of gas hydrate occurrence

Gas hydrate occurrence, as inferred from downhole logs, IR core images, pore water chlorinity freshening, and physical recovery in cores, was much shallower than expected based on previous studies at this margin. At Site U1326, gas hydrate was first recovered at ~47 meters below seafloor (mbsf), and, at Site U1325, gas hydrate was inferred from the electrical resistivity logs to occur as shallow as 73 mbsf. In contrast, at Site U1327, gas hydrate was first recovered at 111 mbsf. Site U1329 did not show any significant evidence of gas hydrate content; however, a small pore water freshening trend at ~123 mbsf may indicate low concentrations of gas hydrate just above the BSR. The apparent progressive decrease in the top depth of gas hydrate occurrence along the transect was investigated by Malinverno et al. (2008) and Torres and Kastner, who used one-dimensional geochemical diffusion modeling to explain the observed landward deepening of the top of gas hydrate occurrence on a site-by-site basis. The model computes methane concentration in the pore fluid for a given in situ bacterial methane production rate, sedimentation rate, and fluid advection velocity. Modeling shows that lower rates of sedimentation or fluid advection result in lower methane concentrations with depth and thus a deepening of the first occurrence of gas hydrate. Sedimentation rates decrease landward along the coring transect (Akiba et al.), corresponding to an increase in depth to the top of the first gas hydrate occurrence. Fluid advection rates, due to dewatering of the accretionary wedge, are expected to first increase with distance from the deformation front, to a maximum of ~15 km into the accretionary prism (Hyndman and Davis, 1992), and then progressively decrease farther landward. A combination of these two mechanisms can explain the deepening of the top of the gas hydrate occurrence along the transect. However, considerable variability along the margin is induced by locally variable sedimentation rates and erosion along uplifted ridges. Also, because gas hydrate forms predominantly in sandy turbidite sediments (see next section), the presence of an appropriate host strata further affects gas hydrate occurrence.

Fluid advection rates required for modeling the thickness of the gas hydrate occurrence zone may be constrained by values of sulfate (and thus methane) flux obtained from reactions of sulfate and methane at the sulfate–methane transition zone (e.g., Borowski et al., 1996). However, there is a complex, uncertain correlation between the depth of the sulfate–methane transition zone and the thickness of the gas hydrate occurrence when data from various sites of gas hydrate are compared globally (Kastner et al., 2008; Dickens and Snyder, 2009; Torres and Kastner).

The coupled microbial reactions of anaerobic oxidation of methane (AOM) are also a sink for sulfate, and the presence of AOM can best be determined from δ¹³C isotopic data. Evidence for AOM is present at all sites studied during Expedition 311, and results from the northern Cascadia margin compare well with observations made at SHR during Leg 204 (Claypool et al., 2006).

The base of gas hydrate stability zone and the nature of the bottom-simulating reflector

A regional BSR was observed along the entire northern Cascadia margin. This BSR is believed to represent the base of gas hydrate stability and thus to mark the transition from gas hydrate–bearing sediments and sediments containing some free gas. Associated changes in the physical properties of the sediment at the BSR from (possibly) higher to lower P-wave velocity are the cause of this prominent seismic reflection. Combined geophysical data analyses have shown that the amount of free gas required to yield the BSR reflection strength observed on the northern Cascadia margin is <1% of the pore space (e.g., Hyndman et al., 2001; Yuan et al., 1996, 1999) and that the free gas zone is relatively thin (<10 m; Chapman et al., 2002).

In order to trap some free gas at the BGHSZ, a temporal permeability barrier of some sort most likely exists. At all sites investigated during Expedition 311, the gas hydrate concentration is <5% (of the pore space) on average; additionally, no evidence was found for a systematic and regionally distributed increase in gas hydrate concentration just above the BSR, as was previously suggested (e.g., Hyndman et al., 2001).

It is also important to note that the upward movement of gas bubbles in porous sediments can be restricted by capillary forces. For free gas to pass through the pore throat of sediment, the gas pressure inside the bubble must overcome the capillary pressure. Vertical migration of bubbles requires an interconnected gas phase and a gas column thick enough that the pressure difference between gas bubbles and pore water (due to different densities) can overcome the capillary pressure in the pore throats (Schowalter, 1979). Gas bubbles beneath the BGHSZ can also be stuck or trapped in the pore space of a fine-grained sediment with a gas saturation of only a few percent (as estimated at the northern Cascadia margin).

Thus, as little as a few percent gas hydrate in a clay-rich, low-permeability sediment could already be an effective barrier or at least a boundary that impedes gas and water from flowing across this horizon and results in a concentration gradient across the boundary with free gas being trapped.

As described by Haacke et al. (2007), gas hydrate recycling at the BGHSZ is a common phenomenon at convergent margins. The continued growth of the accretionary prism (tectonic uplift) combined with sedimentation processes (erosional and depositional) results in continuous changes in the BGHSZ and thus the dissociation of gas hydrate and the release of free gas and water. The resulting free gas zone below the BGHSZ is also much thinner compared to passive margins where the free gas zone is generally thicker (Haacke et al., 2007). It is possible that the free gas migrates back into the GHSZ (if buoyant enough, or, for example, by processes such as the self-generated permeability described by Flemings et al., 2003); however, in most cases, a residual amount (concentrations of a few percent or less) of free gas is trapped by capillary forces in the pore space.

Top of page   |   Previous   |   Next