14 Gas Hydrates in Marine Sediments
496
of the surrounding water. The change in interstitial ion concentration resulting from this “ion
exclusion” mechanism is proportional to the
amount of gas hydrate that is formed. Ussler and
Paul (2001) use a simple cartoon to represent the
effect on pore water salinity at the foci of gas
hydrate formation (Fig. 14.13). They further
modelled the diffusive attenuation of the chloride
anomaly over time, and showed that a positive
anomaly of 56 mM (created by formation of
hydrate that occupies ~9% of the pore space) will
not be detected with current analytical methods
after about 40,000 years (Fig. 14.13).
Various numerical models have shown that
generating gas hydrate to a concentration of
~10% of the pore space, in both passive and
active settings (e.g. Blake Ridge, Hydrate Ridge;
Nimblett and Ruppel 2003) probably required
formation times of at least 10
3
, and perhaps as
much 10
6
years. Therefore, if chloride behaved
conservatively, the pore water in contact with
these deposits should have a chloride concentration similar to seawater. This is, however, not
commonly the case. Indeed, fluids with chloride
concentration significantly lower than seawater
have been sampled from most convergent margins
and such “freshening” has been attributed to gas
hydrate dissociation and dehydration of hydrous
minerals at depth (e.g. Gieskes et al. 1990;
Kastner et al. 1991). The issue of background
chloride concentration, and an example of a
chloride ano-maly created by natural gas hydrate
dissociation is described in the following
sections.
In situ chloride concentration in pore fluids of
hydrate-bearing sediments also show enrichments relative to seawater in some natural systems. These occur when the geological setting
supports formation of brines, or when gas hydrate forms so rapidly that the resulting excess
ions do not have sufficient time to diffuse away.
These scenarios are also discussed below.
Estimating Gas Hydrate Abundance Using
Dissolved Chloride Data
Because gas hydrate is not stable at the temperature
and pressure conditions that exist at the sea surface,
most estimates of the in situ distribution and concentration of gas hydrate rely on a variety of proxies.
Perhaps the most widely used of these proxies is based
on the accurate measurement of dissolved chloride in
the pore fluids. During core recovery, gas hydrate
dissociates, resulting in dilution of the chloride
concentration by addition of water sequestered in the
gas hydrate lattice prior to core recovery. The negative
chloride anomalies relative to in situ chloride concentrations are proportional to the amount of gas hydrate
in a sediment sample. Uncertainties in the estimates of
gas hydrate abundance using the dissolved chloride
proxy arise from a paucity of information on (1) the in
situ dissolved chloride values, (2) the chloride content
potentially trapped within the pores of the gas
hydrates, and (3) the spatial sampling resolution.
There is to date no reliable data on the amount
of Cl
–
sequestered by the hydrate cage because
the physical separation of the water released by
natural hydrate dissociation from pore water contamination can be very difficult. Suess et al.
(2001) suggest that there may be residual chloride
trapped within the hydrate pore space. Nevertheless, since this number is small and very
poorly defined, most estimates of hydrate abundance in marine sediments assume that hydrate
formation excludes all dissolved ions.
If the amount of chloride ions trapped in the hydrate
structure is assumed to be negligible, the measured
chloride concentration after hydrate dissociation can
be related to the hydrate abundance by the following
equation (see Ussler and Paull 2001 for derivation).
Cl
-
s
/Cl
-
o
= 1-[V h /(w-V h (w-1))]
(1)
where Cl
-
s
is the chloride concentration in the
sample (i.e. after hydrate decomposition), Cl
-
o
is
the pore water concentration in situ (prior to
decomposition), V h is the volume fraction of
hydrate filling pore space, and w represents the
occupancy-density characteristics of the gas
hydrate formed, as calculated from:
w = ρ w M h /(ρ h M w m w )
(2)
Here, ρ w and ρ h are the densities of fresh water and
gas hydrate respectively, and m w is the number of
moles of fresh water contained in 1 mole of gas
hydrate. M w and M h represent the molecular weights
of water and gas hydrate, respectively. The value of
M h depends on the degree of occupancy of the
hydrate structure. When the structure is fully
occupied, 1 mole of gas hydrate contains 5.9 moles
of water, its density is 910 kg m
-3
and its molecular
weight is 122.2 g mol
-1
(Ussler and Paull 2001).
The use of the chloride proxy is predicated on the
assumption that the background chloride concentration is known and that the rate of hydrate formation
496
of the surrounding water. The change in interstitial ion concentration resulting from this “ion
exclusion” mechanism is proportional to the
amount of gas hydrate that is formed. Ussler and
Paul (2001) use a simple cartoon to represent the
effect on pore water salinity at the foci of gas
hydrate formation (Fig. 14.13). They further
modelled the diffusive attenuation of the chloride
anomaly over time, and showed that a positive
anomaly of 56 mM (created by formation of
hydrate that occupies ~9% of the pore space) will
not be detected with current analytical methods
after about 40,000 years (Fig. 14.13).
Various numerical models have shown that
generating gas hydrate to a concentration of
~10% of the pore space, in both passive and
active settings (e.g. Blake Ridge, Hydrate Ridge;
Nimblett and Ruppel 2003) probably required
formation times of at least 10
3
, and perhaps as
much 10
6
years. Therefore, if chloride behaved
conservatively, the pore water in contact with
these deposits should have a chloride concentration similar to seawater. This is, however, not
commonly the case. Indeed, fluids with chloride
concentration significantly lower than seawater
have been sampled from most convergent margins
and such “freshening” has been attributed to gas
hydrate dissociation and dehydration of hydrous
minerals at depth (e.g. Gieskes et al. 1990;
Kastner et al. 1991). The issue of background
chloride concentration, and an example of a
chloride ano-maly created by natural gas hydrate
dissociation is described in the following
sections.
In situ chloride concentration in pore fluids of
hydrate-bearing sediments also show enrichments relative to seawater in some natural systems. These occur when the geological setting
supports formation of brines, or when gas hydrate forms so rapidly that the resulting excess
ions do not have sufficient time to diffuse away.
These scenarios are also discussed below.
Estimating Gas Hydrate Abundance Using
Dissolved Chloride Data
Because gas hydrate is not stable at the temperature
and pressure conditions that exist at the sea surface,
most estimates of the in situ distribution and concentration of gas hydrate rely on a variety of proxies.
Perhaps the most widely used of these proxies is based
on the accurate measurement of dissolved chloride in
the pore fluids. During core recovery, gas hydrate
dissociates, resulting in dilution of the chloride
concentration by addition of water sequestered in the
gas hydrate lattice prior to core recovery. The negative
chloride anomalies relative to in situ chloride concentrations are proportional to the amount of gas hydrate
in a sediment sample. Uncertainties in the estimates of
gas hydrate abundance using the dissolved chloride
proxy arise from a paucity of information on (1) the in
situ dissolved chloride values, (2) the chloride content
potentially trapped within the pores of the gas
hydrates, and (3) the spatial sampling resolution.
There is to date no reliable data on the amount
of Cl
–
sequestered by the hydrate cage because
the physical separation of the water released by
natural hydrate dissociation from pore water contamination can be very difficult. Suess et al.
(2001) suggest that there may be residual chloride
trapped within the hydrate pore space. Nevertheless, since this number is small and very
poorly defined, most estimates of hydrate abundance in marine sediments assume that hydrate
formation excludes all dissolved ions.
If the amount of chloride ions trapped in the hydrate
structure is assumed to be negligible, the measured
chloride concentration after hydrate dissociation can
be related to the hydrate abundance by the following
equation (see Ussler and Paull 2001 for derivation).
Cl
-
s
/Cl
-
o
= 1-[V h /(w-V h (w-1))]
(1)
where Cl
-
s
is the chloride concentration in the
sample (i.e. after hydrate decomposition), Cl
-
o
is
the pore water concentration in situ (prior to
decomposition), V h is the volume fraction of
hydrate filling pore space, and w represents the
occupancy-density characteristics of the gas
hydrate formed, as calculated from:
w = ρ w M h /(ρ h M w m w )
(2)
Here, ρ w and ρ h are the densities of fresh water and
gas hydrate respectively, and m w is the number of
moles of fresh water contained in 1 mole of gas
hydrate. M w and M h represent the molecular weights
of water and gas hydrate, respectively. The value of
M h depends on the degree of occupancy of the
hydrate structure. When the structure is fully
occupied, 1 mole of gas hydrate contains 5.9 moles
of water, its density is 910 kg m
-3
and its molecular
weight is 122.2 g mol
-1
(Ussler and Paull 2001).
The use of the chloride proxy is predicated on the
assumption that the background chloride concentration is known and that the rate of hydrate formation