Enhanced lifetime of methane bubble streams within the deep ocean
Gregor Rehder,1Peter W. Brewer, Edward T. Peltzer, and Gernot Friederich
Monterey Bay Aquarium Research Institute, Moss Landing, California, USA
Received 23 August 2001; revised 16 July 2002; accepted 26 July 2002; published 7 August 2002.
rates of methane and argon bubbles experimentally released in the
ocean at depths from 440 to 830 m. The bubbles were injected
from the ROV Ventana into a box open at the top and the bottom,
and imaged by HDTV while in free motion. The vehicle was
piloted upwards at the rise rate of the bubbles. Methane and
argon show closely similar behavior at depths above the methane
hydrate stability field. Below that boundary (?520 m) markedly
enhanced methane bubble lifetimes are observed, and are
attribute to the formation of a hydrate skin. This effect greatly
increases the ease with which methane gas released at depth, either
by natural or industrial events, can penetrate the shallow ocean
INDEX TERMS: 4820 Oceanography: Biological and
Chemical: Gases; 4806 Oceanography: Biological and Chemical:
Carbon cycling; 3022 Marine Geology and Geophysics: Marine
sediments—processes and transport
We have made direct comparisons of the dissolution and rise
 Knowledge of the mechanism and rates of vertical oceanic
transport of methane is important for understanding the fate of
natural (seepage) and industrial (pipeline and well head) releases
today. It has long been recognized that at depths below about 500
m, methane forms a solid hydrate in the ocean [Sloan, 1997], and
both the size of this hydrate reservoir, and its stability are
important topics of debate. Hydrate layers in the sediment may
seal the seafloor, capturing large amounts of free gas below the
hydrate stability field. Slump slides associated with gas and
hydrate instability have been described, leading to the release
of large amounts of free gas from below the hydrate stability
field into the water column [Paull et al., 1991]. Release of free
gas into the water column well within the hydrate stability field
is known today from Hydrate Ridge, Cascadia margin [Suess et
al., 2001], from Blake Ridge at the Carolina continental rise
[Paull et al., 1995], from the Guayamas Basin [Merewether et
al., 1985], and from the Gulf of Mexico [MacDonald et al.,
 The deep ocean contains only trace amounts of methane
[Scranton and Brewer, 1978]. Thus a bubble plume from gas
seepage or release at depth creates an enormous driving force for
dissolution, and methane dissolved at depth will be microbially
transformed to CO2on a time scale shorter than the ventilation
time of deep water masses [Rehder et al., 1999; Warner et al.,
1996]. Only a fraction of CO2produced in the deep ocean would
enter the atmosphere due to the ocean’s storage capacity for CO2,
and the radiative forcing potential of CO2is smaller than that of
methane [Lelieveld et al., 1998]. As a consequence, the global
warming potential of a mole of methane oxidized to CO2in the
deep ocean is about a factor of 50 smaller than that of methane
released to the atmosphere without conversion.
 If however methane dissolution is inhibited during bubble
transport within the hydrate stability zone, then upward transport to
depths close to or even within the winter mixed layer, which are
ventilated on far shorter time scales, is possible. Here we report on
direct measurements of the rise rate and dissolution rate of bubbles
of both methane and (for comparison) argon in the deep sea. The
purpose of the experiments was to determine whether the dissolu-
tion rate of methane bubbles in free ascent is significantly
decreased within the hydrate stability field due to the formation
of a hydrate skin.
 Measurements were made using MBARIs ROV Ventana in
Monterey Bay, California, where the upper boundary of the
methane hydrate stability field is close to 520 m depth [Peltzer
and Brewer, 2000] (see auxiliary material, GRL online1). Single
gas bubbles were released from compressed gas cylinders through
a 1/4-inch stainless steel nozzle at various depths between 820 and
430 m into an imaging box attached to the ROV (Figure 1). The
ROV was piloted upwards at a speed adjusted to match the rise
velocity of the bubbles, and the ascent of individual bubbles within
the imaging box was recorded using an HDTV camera system (see
suppl. material, GRL online). The bubbles initially had an ellip-
soidal shape with an initial equivalent diameter between 5 and 9
mm. The equivalent diameter (i.e. the diameter of a sphere with the
same volume) was calculated using the relation de= (a2b)1/3, where
a and b are the long and the short axis of the ellipsoid, respectively
[Sam et al., 1996]. The imaging box (89 cm long, 25 cm deep, 30
cm wide) restrained lateral motions, while permitting almost
undisturbed upward motion of the gas bubble. The backwall
consisted of two opaque PVC plates spaced 8 cm apart, reducing
visual clutter from the ubiquitous background of marine snow and
acting as diffusers for four light sources mounted behind the
 Argon served as a reference gas in the experiments. A wide
range of parameters, including coverage with surfactants, different
shapes depending on bubble size, temperature, and hydrostatic
pressure can influence the dissolution kinetics of a single bubble
[Clift et al., 1978; Leifer et al., 2000]. These parameters are quite
variable in the ocean and some of them are not well established.
The only gas-specific parameter used in theoretical descriptions of
gas dissolution kinetics is the quotient of the solubility and the
square root of the Schmidt number (b/Sc0.5) [Clift et al., 1978]. In
the temperature range in which the experiments were carried out
(4–7.5?C), b/Sc0.5is almost identical for methane and argon
[Wanninkhof, 1992] (see suppl. material, GRL online), but argon
does not form hydrates under these P,T-conditions [Marshall et al.,
1964]. By comparing the behavior of both gases in experiments
performed under identical oceanic conditions, the uncertainties
1Supporting material is available via Web browser or via Anonymous
FTP from ftp://ftp.agu.org, directory ‘‘apend’’ (Username = ‘‘anonymous’’,
Password = ‘‘guest’’); subdirectories in the ftp site are arranged by paper
number. Information on searching and submitting electronic supplements is
found at http://www.agu.org/pubs/esupp_about.html.
GEOPHYSICAL RESEARCH LETTERS, VOL. 29, NO. 15, 10.1029/2001GL013966, 2002
1Now at GEOMAR Research Center, Wischhofstr, 1-3, D-24148 Kiel,
Copyright 2002 by the American Geophysical Union.
21 - 1
involved in the parameterizations needed to derive the absolute
dissolution rate are avoided.
Bubble Rise Rates
 The rise rate of the gas bubbles was determined in two
ways. The pressure log of the ROV provides data on the change of
depth of the bubbles with time, and the (dynamic) rise velocity can
be derived from the slope of this function. For comparison with the
‘‘vehicle in motion’’ experiments, the rise rate was also determined
with the ROV stationary on the seafloor (847 dbar; 4.5?C). HDTV-
frames were taken at least 15 cm above the nozzle, and the distance
traveled was typically 60 cm. The mean rise velocity of argon (25.1
cm/sec ± 0.21 cm/sec; N = 11) was about 5% smaller than for
methane (26.3 cm/sec ± 0.4 cm/sec; N = 9). This reflects the
buoyant forces driving the ascent due to the different densities of
methane and argon at depth (see suppl. material, GRL online).
 The mean dynamic rise rate was 28.3 ± 1.7 cm/s and 29.2
± 1.8 cm/s for argon and methane, respectively (Figure 2a). This
is about 12% larger than the result obtained from the static
experiment, which implies that the drag produced by the walls of
the imaging box had only small impact on the rise velocity.
While the bubbles shrank significantly during ascent, the rise rate
stayed almost constant (Figure 2). The rise velocity of bubbles is
a function of size [Clift et al., 1978], but the bubbles monitored
fall in the range where this effect is small. Apparently, the effects
of changing diameter, increasing density difference between gas
and ambient seawater, and changing shape of the bubble all
combined to yield a constant rise rate within the error of our
3.2. Bubble Dissolution
 The equivalent diameter of argon bubbles at all depths
(between 830 and 360 m) decreased linearly with time (Figure
2b). The shrinking rate was very consistent at all depths, with a
mean rate of 18.9 ± 2.2 mm/sec. Methane above the hydrate
stability field (Figure 2b) showed the same linear trend and
consistent behavior as argon, although the bubbles dissolved
somewhat more slowly (15.5 ± 3.2 mm/sec). Within the hydrate
stability field, the behavior was completely different (Figure 2c).
After an initial stage of fast dissolution, with a rate comparable to
that of methane bubbles released above the hydrate stability field
(mean 12.8 mm/sec), the shrinking rate markedly slowed by a factor
of 3.5 to 6 (mean 3.0 mm/s). The transition appeared at de= 3.5 to 4
mm, with one exception at de= 2.3 mm. Gas bubbles with a
reduced shrinking rate also had a more spherical appearance
(Figure 3a). The difference in dissolution behavior within and
out of the hydrate stability field for methane was not found for
argon, and can thus be attributed to a gas specific process, most
likely the nucleation of a hydrate skin.
 For methane bubbles released within the stability field and
in the state of slow dissolution, shape and path oscillations were
suppressed. The low shrinking rate and non-oscillatory motion
remained after the bubbles left the hydrate stability field, suggest-
ing strong residual order in the boundary layer.
3.3.Observations in a Natural System
 Field observations made at Hydrate Ridge, Oregon, suggest
that the observed effect exists strongly in natural systems. We
monitored the rise of natural gas bubbles, emerging from the
seafloor at the southern summit of Hydrate Ridge, Oregon, at a
water depth of 780 m. The gas consists of >98% methane
[Peltzer et al., 2000], and the upper boundary of the methane
mounted in front of the HDTV camera. An array of hydraulic
valves was used for the release of different gases (left side). (b)
Modified imaging box using a set of two opaque back walls and
illumination from the back.
(a) ROV Ventana before the dive with the imaging box
mean rise rate for methane (filled symbols) was slightly higher than for argon (open symbols). For methane, the rise rate for bubbles
released within the hydrate stability field (points and diamonds) and above the hydrate stability field (triangles) was undistinguishable.
Also given are the results of the (static) rise rate measurements with the ROV stationary at the seafloor at 847 dbar. (b) Decrease of the
equivalent diameter (de) of single bubbles during the ascent versus time. Argon bubbles (black symbols) shrink with a consistent rate of
18.9 ± 2.2 mm/sec in the entire depths range covered (830–440 m release depth). Methane bubbles above the hydrate stability (blue
symbols) dissolve at a similar rate (15.5 ± 3.3 mm/sec). (c) Same as Figure 2b but for methane bubbles released within the hydrate stability
field. The data follow a bilinear trend, suggesting a drastic change in behavior after transition from a state without hydrate effect to a stage
with the influence of a hydrate skin. The slope of the first part (mean 12.8 mm/sec) is close to the average shrinking rate observed on CH4
bubbles out of the hydrate stability field (blue hatched line). The slopes after the transition average 3.0 mm/sec.
(a) Rise distance of single bubbles after release. The slope of this function represents the (dynamic) bubble rise velocity. The
21 - 2 REHDER ET AL.: ENHANCED LIFETIME OF METHANE BUBBLES
hydrate stability field was at ?500 m. Bubbles were about 6–7
mm in diameter at the source, but appeared to have an almost
spherical shape immediately on emerging from the sediment
(Figure 3b). This unusual shape for a bubble of this size indicates
the presence of a more rigid surface, which suggests that a hydrate
skin has already been formed when the bubble leaves the seabed.
 The growth of hydrate on the gas-water interface requires
two steps. The initial formation of a hydrate nucleus, and growth of
hydrate after the nucleation site has formed. The onset of hydrate
formation on a plane water/gas interface has been shown to be
almost immediate in water which has been in contact with hydrates
before, while it could take times long compared to a bubble
lifetime in distilled water or fresh tap water [Vysniauskas and
Bishnoi, 1983]. This is explained by the preservation of water
microstructures favoring hydrate nucleation and growth [Sloan,
1997]. For a certain water environment, time of nucleation in
multiple experiments is randomly distributed. In laboratory experi-
ments on methane and natural gas bubbles, it has been shown that
nucleation was delayed in newly replaced water and that pressures
considerably in excess of the equilibrium pressure were required
for this case [Maini and Bishnoi, 1981; Topham, 1984]. After
hydrate had once been formed in the system, the nucleation process
was accelerated dramatically in following nucleation experiments.
Our observations of natural and artificial bubbles are in agreement
with these findings. In case of the artificial release in Monterey
Bay, the gas was surrounded by water that had not had any contact
with hydrates before, and it took in general a long — and
unpredictable — time before the onset of nucleation. Lateral
spreading of hydrate along a gas/water interphase after nucleation
is a very rapid process, which explains the sharp transfer to the
skin-covered condition (Figure 2c). In contrast, gas hydrates and
free gas have been found in the upper sediment layer of the
southern summit of Hydrate Ridge [Suess et al., 2001]. Thus, it
is more than likely that the porewaters in contact with these
hydrates should contain the water precursors required for hydrate
nucleation, which should yield to the immediate presence of a
hydrate skin on the gas bubbles emerging the seafloor at this site,
in accordance with our observations.
 Our experiments and observations show that hydrate
nucleation on a methane bubble can occur even in a highly
undersaturated water column. The methane concentration in the
water column below 500 m in Monterey Bay is ?1 nmol/l, which
is typical for the oceanic methane background [Rehder et al., 1999;
Scranton and Brewer, 1978]. Seawater CH4concentrations in the
vicinity of Hydrate Ridge are enhanced due to several processes,
but measurements on the northern summit in July, 2000, were
always below 100 nmol/l and hence, negligible compared to the in
situ saturation value (data not shown). The reduced shrinking rate
has considerable impact on the upward transport of methane
through the water column. Using the mean rise rate of 28.7 cm/
sec and a shrinking rate of 12.8 mm/s and 3 mm/s for a non-hydrate-
coated and a hydrate-coated bubble, respectively, we calculate the
mass loss of a bubble of an initial size of 7 mm, released at 800 m
water depth. The bubble without a hydrate skin would loose 50%
of its mass after 106 sec, or 30 m. In contrast, a bubble with a
hydrate film, assumed to shrink with a rate of 3 mm/sec, would
travel 110 m before half of its mass would be dissolved (see suppl.
material, GRL online).
 The slowing of the dissolution rate of methane bubbles
within the hydrate stability field provides a mechanism by which
methane gas released from the seafloor can be efficiently trans-
ported above the hydrate stability boundary. Methane is oxidized in
the open ocean on a time scale of decades or longer [Rehder et al.,
1999; Scranton and Brewer, 1978], which is short compared to the
ventilation time of the deep oceans (?500 yrs Pacific; ?250 yrs
Atlantic), but long compared to the annual ventilation by winter
mixing of the upper few hundred meters of the water column
[Warner et al., 1996]. In some areas of the ocean, the upper
boundary of the methane hydrate stability field is even shallower
than the maximum depth of the mixed layer in winter [Harvey,
1982]. Thus, this transport mechanism increases the possibility for
a direct interaction of methane released from deep water gas
sources with the atmosphere. Our findings increase understanding
of the methane distribution above modern deep gas vents and gas
leakage from under-sea pipelines, and will help to refine scenarios
of ancient, present, and future climate feedback associated with
R/V Western Flyer and the ROV technicians and pilots of ROV Ventana
and Tiburon for their skillful support during offshore operation. We are
indebted to Chris Rodgers-Walz and Kyra Schlining for invaluable and
friendly help in the digital and video labs. Comments from Ira Leifer,
University of California Santa Barbara, considerably helped to improve this
manuscript. This work was supported by a grant to MBARI from the David
and Lucile Packard foundation.
We thank the crews of R/V Point Lobos and
Clift, R., J. R. Grace, and M. E. Weber, Bubbles, Drops, and Particles, 380,
Academic Press, New York, 1978.
Harvey, J., Theta-S relationships and water masses in the eastern North
Atlantic, Deep-Sea Res., 29A, 1021–1033, 1982.
Leifer, I., R. K. Patro, and P. Bowyer, A study on the temperature variation
of rise velocity for large clean bubbles, J. Atmospheric and Oceanic
Technology, 17, 1392–1402, 2000.
Lelieveld, J., P. J. Crutzen, and F. J. Dentener, Changing concentration,
lifetime, and climate forcing of atmospheric methane, Tellus, 50, 128–
MacDonald, I. R., N. L. Guinasso Jr., R. Sassen, J. M. Brooks, L. Lee, and
K. T. Scott, Gas hydrate that breaches the sea floor on the continental
slope of the Gulf of Mexico, Geology, 22, 699–702, 1994.
Maini, B. M., and P. R. Bishnoi, Experimental investigation of hydrate
formation behaviour of a natural gas bubble in a simulated deep sea
environment, Chem. Engineering Science, 36, 183–189, 1981.
Marshall, D. R., S. Saito, and R. Kobayashi, Hydrates at high pressures:
and a CH4bubble (right) between 622 and 544 m water depth. The
length of the white scale bar is 1 cm. (b) Bubbles of natural gas at
the seafloor at 780 m depth, showing a spherical appearance
immediately after contact with the seawater.
(a) Development of the size of two argon bubbles (left)
REHDER ET AL.: ENHANCED LIFETIME OF METHANE BUBBLES
21 - 3
Part I. Methane-water, argon-water, and nitrogen-water systems, AIChE Download full-text
Journal, 10, 202–205, 1964.
Merewether, R., M. S. Olsson, and P. Lonsdale, Acoustically detected
hydrocarbon plumes rising from 2-km depths in Guayamas Basin, Gulf
of California, J. Geophys. Res., 90, 3075–3085, 1985.
Paull, C. K., W. Ussler III, and W. P. Dillon, Is the extend of glaciation
limited by marine gas-hydrates, Geophys. Res. Lett., 18, 432–434,
Paull, C. K., W. Ussler III, W. S. Borowski, and F. N. Spiess, Methane-rich
plumes on the Carolina continental rise: Associations with gas hydrates,
Geology, 23, 89–92, 1995.
Peltzer, E. T., and P. G. Brewer, Practical physical chemistry and empirical
predictions of methane hydrate stability, in Natural Gas Hydrate in Ocea-
nic and Permafrost Environments, edited by M. D. Max, 17–28, Kluwer
Academic Publishers, Dordrecht, 2000.
Peltzer, E., D. Donohoue, C. K. Paull, G. Rehder, G. Friederich, W. Ussler,
and P. G. Brewer, Novel observations of vent gas chemistry, and evidence
for the presence of structure H hydrate (Abstract), Eos Trans., AGU 8,
(Fall Meeting Supplement), F637, 2000.
Rehder, G., R. S. Keir, E. Suess, and M. Rhein, Methane in the northern
Atlantic controlled by oxidation and atmospheric history, Geophys. Res.
Lett., 26, 587–590, 1999.
Sam, A., C. O. Gomez, and J. A. Finch, Axial velocity profiles of single
bubbles in water/frother solutions, Int. J. Miner. Process., 47, 177–196,
Scranton, M. I., and P. G. Brewer, Consumption of dissolved methane in the
deep ocean, Limnol. Oceanogr., 23, 1207–1213, 1978.
Sloan, E. D., Clathrate hydrates of natural gases, Marcel Dekker, Inc.,
New York, 1997.
Suess, E., et al., Sea floor methane hydrates at Hydrate Ridge, Oregon, in
Natural Gas Hydrates: Occurance, Distribution, and Detection, edited
by C. K. Paull and W. P. Dillon, pp. 87–98, American Geophysical
Union, Washington, DC, 2001.
Topham, D. R., The formation of gas hydrates on bubbles of hydrocarbon
gases rising in seawater, Chemical Engineering Science, 39, 821–828,
Vysniauskas, A., and P. R. Bishnoi, A kinetic study of methane hydrate
formation, Chemical Engineering Science, 38, 1061–1072, 1983.
Wanninkhof, R., Relationship between wind speed and gas exchange over
the ocean, J. Geophys. Res, 97, 7373–7382, 1992.
Warner, M. J., J. L. Bullister, D. P. Wisegarver, R. H. Gammon, and R. F.
Weiss, Basin-wide distributions of chlorofluorocarbons CFC-11 and CFC-
12 in the North Pacific, J. Geophys. Res, 101, 20,525–20,542, 1996.
? ? ? ? ? ? ? ? ? ? ?
G. Rehder, P. G. Brewer, E. T. Peltzer, and G. Friederich, Monterey
Bay Aquarium Research Institute, 7700 Sandholdt Road, Moss Landing,
CA 95039, USA. (firstname.lastname@example.org; email@example.com; firstname.lastname@example.org;
21 - 4 REHDER ET AL.: ENHANCED LIFETIME OF METHANE BUBBLES