1

EAGE 65th Conference & Exhibition Stavanger, Norway, 2 - 5 June 2003

Z-99
Rock Physics of Gas Hydrate Reservoir
J. DVORKIN
1
AND A. NUR
2
Stanford University and Rock Solid Images, 2600 S.Gessner, Suite 650, Houston, TX 77063, USA
Geophysics Department, Stanford University


Summary. Enormous amounts of methane gas hydrate are present in sediments under the world's
oceans as well as in on-shore sediments in the Arctic. These hydrates are a potential future energy
resource. The most well-developed geophysical tool for exploring large volumes of the subsurface
where gas hydrate is found is seismic reflection profiling. To characterize a natural gas hydrate
reservoir with seismic data, we must be able to relate the elastic properties of the sediment to the
volume of gas hydrate present. One way of achieving this goal is through rock physics effective-
medium modeling. We offer a first-principle-based effective medium model for the elastic-wave
velocity in unconsolidated, high-porosity ocean-bottom sediments with gas hydrates. The elastic
constants of the dry-sediment frame depend on porosity, elastic moduli of the solid phase, and
effective pressure. To account for the effect of gas hydrate on sediment elastic moduli we assume that
hydrate becomes part of the solid phase thus modifying the porosity and elasticity of the frame. We
use this model to predict the position of sediments with gas hydrate and free gas in the elastic attribute
space.
Introduction. Gas hydrates are solids comprised of a hydrogen-bonded water lattice with entrapped
guest molecules of gas. Large amounts of methane gas hydrate are probably present in sediments
under the world's oceans as well as in on-shore sediments in the Arctic. Potentially, these hydrates
comprise the largest carbon as well as methane buffer on earth. As such, the hydrates located in the
subsurface may be the principal factor in the global climate balancing. Also, one may consider this
methane pool a potential future energy source. These considerations ignite the scientific communitys
interest in quantifying the amount of methane hydrate in the subsurface. In principle, remote detection
and quantification of gas hydrate reservoirs is no different from the traditional hydrocarbon
exploration. Similar and well-developed sensing techniques can be used, seismic reflection profiling
being the dominant among them.
Seismic response of the subsurface at a given time moment is determined by the spatial distribution of
the elastic properties. By mapping the elastic contrast, the geophysicist can illuminate tectonic
features and geobodies, including hydrocarbon reservoirs. To properly translate elastic-property
images into images of lithology, porosity, and pore-filling phases, quantitative knowledge is needed
that relates rocks elastic properties to its bulk properties and conditions. Specifically, to
quantitatively characterize a natural gas hydrate reservoir from seismic data, we must be able to relate
the elastic properties of the sediment to the porosity and volume of gas hydrate present. One way of
achieving this goal is through rock physics effective-medium modeling.
Rock Physics Model. The basis of our approach is the model of Dvorkin et al. (1999), which relates
the elastic moduli of high porosity ocean bottom sediments to porosity, pore fluid compressibility,
mineralogy, and effective pressure. At the critical porosity, we calculate the effective bulk and shear
moduli of the dry rock frame using the Hertz-Mindlin contact theory. For porosity below critical the
bulk and shear moduli of the dry frame are calculated via the modified lower Hashin-Shtrikman
bound. For porosity above critical these moduli are calculated via the modified upper Hashin-
2

Shtrikman bound. In a common case of mixed mineralogy, the elastic constants of the solid phase are
calculated from those of the individual mineral constituents using Hill's average formula. This
approach is also used to account for the presence of gas hydrate. Specifically, gas hydrate in the pore
space is treated as part of the load-bearing frame. Its presence acts to reduce the porosity space
available for water and, at the same time, alter the elastic properties of the composite solid matrix
phase. The effect of the pore fluid on the effective elastic moduli is calculated from Gassmanns
equation.
This gas hydrate rock physics model has been successfully applied to well log data (Helgerud et al.,
1999) and seismic data (Ecker et al., 2000) at the Outer Blake Ridge in the Atlantic which is marked
by a prominent Bottom Simulating Reflector (BSR) that supposedly manifests the seismic reflection at
the interface between the sediment with methane hydrate and the underlying sediment with free gas.
The hydrate-filled sediment possibly serves as the caprock to the underlying gas reservoir. The
sediment at the location is mostly comprised of high-porosity unconsolidated clay. A good match
between the hydrate volume obtained from sonic via rock physics modeling and hydrate volume
obtained from resistivity (Helgerud et al., 1999) indicates that the model is an accurate tool for seismic
gas hydrate quantification. The same rock-physics-based approach has been used by Ecker et al.
(2000) to estimate gas hydrate volume from seismic data. The interval velocity computed from
seismic traces has been used as the input into the model. The resulting hydrate and free gas volume
estimates lie within the expected bounds.
Numerical Examples. We use this model to calculate the elastic properties of sand with porosity
ranging from 30 to 40% filled with solid methane hydrate with the hydrate concentration in the pore
space ranging from zero to 70%. The rest of the pore space is filled with brine. The assumed
mineralogy is 90% quartz and 10% clay. The results shown in Figure 1 indicate that the larger gas
hydrate concentration at a fixed porosity value the larger the impedance and the smaller the Poissons
ratio. These two plots can be used to extract both the total porosity and gas hydrate concentration
from elastic well log or seismic inversion data. The result (the product of porosity and hydrate
concentration) will give the bulk amount of gas hydrate in the sediment per unit volume. For example,
if the product of porosity and hydrate concentration is 0.1, the volume of gas hydrate in 1 m
3
of the
formation is 0.1 m
3
. This volumetric hydrate concentration is plotted versus the P-wave impedance
and Poissons ratio in Figure 2. It is common to determine the latter two elastic parameters from
seismic. Figure 2 tells us which domain in the Poissons ratio versus P-wave impedance cross-plot
corresponds to best hydrate reserves. A similar model-based cross-plot can be made for other
commonly used seismic elastic parameters, such as the acoustic and elastic impedance and AVO
intercept and gradient.
The same rock physics model allows us to estimate the influence of free gas (that is likely to be
trapped beneath the gas hydrate layer) on the elastic rock properties. Here we use the same model
sand as in the above example. We assume that there is no gas hydrate present in the pore space and
that water saturation pattern is either uniform or patchy (Dvorkin and Nur, 1998).

3
EAGE 65th Conference & Exhibition Stavanger, Norway, 2 - 5 June 2003
Figure 1. Left -- The P-wave impedance versus porosity and methane hydrate concentration in the
pore space for model sand as described in the text above. Darker shade means smaller value.
Right -- Same for Poissons ratio.

Figure 2. Volumetric hydrate concentration versus P-wave impedance and Poissons ratio for
model sand as described in the text above. Darker shade means smaller value. Encircled is the
domain where the amount of gas hydrate is likely to be maximum. It is the target of seismic gas
hydrate exploration.

Figure 3. Free gas volume versus P-wave impedance and Poissons ratio for model sand as
described in the text above. Darker shade means smaller value. Left -- Uniform saturation. Right
-- Patchy saturation.
In Figure 3 we lot the gas saturation of the pore space times the total porosity versus the P-wave
impedance and Poissons ratio. This product gives the free gas volume per unit volume of the
sediment. It is evident from the display that the interpreted free gas saturation depends on what
saturation pattern is assumed to be present in situ. The two results, for uniform and patchy patterns,
can be treated as the lower and upper bounds for the free gas volume, respectively.
Position in Attribute Space. Figure 4 summarizes the position of sediment with water, gas, and gas
hydrate in the Poissons ratio/Acoustic Impedance attribute space. The areas are color-coded
according to the volume concentration of gas hydrate and free gas (for uniform saturation). The
domain occupied by water-saturated sediment is color-coded by the clay content. This map indicates
that it is possible to discriminate layers filled with gas hydrate from water-saturated sand and shale and
gas-saturated sand, and quantify gas hydrate concentration in the sediment.
This map can be used as a guide during seismic gas hydrate interpretation and for forward seismic
modeling. It is synthetic modeling that will ultimately help identify the seismic visibility of thin layers
with gas hydrate.
4

Hydrate
Free
Gas
Water
V
o
l
u
m
e

C
o
n
c
e
n
t
r
a
t
i
o
n

Figure 4. Sand with hydrate and free gas and water-saturated sand/shale in elastic attribute space.
References
Dvorkin, J., and Nur, A., 1998, Acoustic Signatures of Patchy Saturation, Int. J. Sol. Struct., 35, 4803-
4810.
Dvorkin, J., Prasad, M., Sakai, A., and Lavoie, D., 1999, Elasticity of marine sediments, GRL, 26,
1781-1784.
Ecker, C., Dvorkin, J., and Nur, A., 2000, Estimating the amount of gas hydrate and free gas from
marine seismic data, Geophysics, 65, 565-573.
Helgerud, M., Dvorkin, J., Nur, A., Sakai, A., and Collett, T., 1999, Elastic-wave velocity in marine
sediments with gas hydrates: Effective medium modeling, GRL, 26, 2021-2024.