q 2002 Geological Society of America. For permission to copy, contact Copyright Permissions, GSA, or editing@geosociety.org.Geology; July 2002; v. 30; no. 7; p. 631–634; 4 figures. 631

Sublacustrine mud volcanoes and methane seeps caused bydissociation of gas hydrates in Lake BaikalP. Van Rensbergen*M. De Batist

Renard Centre of Marine Geology, University of Ghent, Krijgslaan 281-S8, B-9000 Gent, Belgium

J. Klerkx Department of Geology and Mineralogy, Royal Museum of Central Africa, Tervuren, BelgiumR. HusJ. PoortM. Vanneste

Renard Centre of Marine Geology, University of Ghent, Krijgslaan 281-S8, B-9000 Gent, Belgium

N. GraninO. Khlystov

Limnological Institute, Siberian Branch of the Russian Academy of Sciences, Irkutsk, Russia

P. Krinitsky All Russian Institute for Geology and Mineral Resources of the World Ocean, Saint Petersburg, Russia

ABSTRACTFour lake-floor seeps have been studied in the gas-hydrate area in Lake Baikal’s South

Basin by using side-scan sonar, detailed bathymetry, measurements of near-bottom waterproperties, heat-flow measurements, and selected seismic profiles in relation to results fromgeochemical pore-water analysis. The seeps at the lake floor are identified as methaneseeps and occur in an area of high heat flow, where the base of the gas-hydrate layershallows rapidly toward the vent sites from ;400 m to ;150 m below the lake floor. Atthe site of the seep, a vertical fluid conduit disrupts the sedimentary stratification fromthe base of the hydrate layer to the lake floor. The seeps are interpreted to result fromlocal destabilization of gas-hydrate caused by a pulse of hydrothermal fluid flow along anactive fault segment. This is the first time that methane seeps and/or mud volcanoes as-sociated with gas-hydrate destabilization have been observed in a sublacustrine setting.The finding demonstrates the potential of tectonically controlled gas-hydrate destabiliza-tion to cause extreme pore-fluid overpressure and short-lived mud volcanism.

Keywords: gas hydrates, hydrothermal vents, Lake Baikal, methane, mud volcanoes, seepage.

INTRODUCTIONIn the marine environment, gas hydrate be-

low the seafloor is at places associated withmethane seeps at the seafloor, marked bypockmarks, fluid vents, authigenic carbonateprecipitation, and sometimes mud volcanism.The methane seeps are often interpreted to re-sult from gas-hydrate destabilization, but it isdifficult to differentiate the effect of gas-hydrate dissociation from other sources of flu-id and gas. Whether hydrate formed out ofupward methane-charged fluid flow or wheth-er methane expulsion or mud volcanism re-sults from decomposition of gas hydrates of-ten remains a ‘‘chicken or the egg’’ problem.

In modern rift lakes (e.g., Lake Baikal,Lake Tanganyika), fluid seeps are most com-monly hydrothermal in origin and occur alongpermeable, active faults in zones of elevatedheat flow (Crane et al., 1991; TanganydroGroup, 1992). Hydrothermal seeps are in gen-eral not accompanied by mobilization and ex-pulsion of subsurface sediment. More than 50hot springs associated with large heat-flowanomalies have been mapped in the Baikal riftzone, essentially in the North and Central Ba-sins (Golubev et al., 1993). At an offshore hy-drothermal seep in Frolikha Bay, Baikal North

*E-mail: pieterpvanrensbergen@yahoo.com.

Basin, heat flow reaches 8600 mW/m2 (Craneet al., 1991), and the flow of hydrothermal sa-line water from the seep can be traced withinthe freshwater basin (Kipfer et al., 1996). Inthe Baikal South Basin, heat flow varies be-tween 50 and 100 mW/m2 in a regular patternwithout such large isolated anomalies (Golu-bev et al., 1993; Poort et al., 1998).

Lake Baikal is one of the world’s largest riftlakes and the only freshwater basin with prov-en gas hydrates. Hydrates were found andsampled in a 225-m-long Baikal Drilling Pro-ject (BDP) core from the South Basin (Kuz-min et al., 1998). Gas from the hydrate sam-ples consisted mainly of methane (99% oftotal hydrocarbons) of biogenic origin (carbonisotope composition d13C between 258‰ and268‰; Kuzmin et al., 1998). In Lake Baikalthere are no known sources of thermogenicgas, and the gas of the gas hydrate is probablyderived from organic matter supplied by theSelenga River, the main source of terrigenousorganic supply to the lake. The hydrates occuraround the Selenga River delta (Fig. 1) at wa-ter depths of .580 m. The study area is lo-cated in a zone where multichannel seismicprofiles show the base of the hydrate stabilityzone (BHSZ) to be irregular, not at all mim-icking the lake floor as a bottom-simulatingreflector (Golmshtok et al., 2000).

METHODS AND DATAThe data discussed are located in a study

area of ;15 km 3 16 km in the Baikal SouthBasin in a water depth of 1320–1440 m. Inthis study area, a side-scan sonar mosaic (Fig.2) covers an area of ;90 km2. The 30 kHzSONIC sonar imaged 0.8 km at each side ofthe track line, and the across-track footprint ofthe acoustic beam ranged from 0.75 m to 3.8m. In addition, 9 single-channel airgun seis-mic reflection profiles (total length ;180 km)were obtained using a 3 L Impulse-1 airgun(frequency range of 45–330 Hz) with a pen-etration as deep as 700 ms two-way traveltime(;560 m) and a vertical resolution of ;3 m.Thermal-probe measurements (n ; 30) weretaken along two seismic profiles to calculatethe lateral variation in heat flow (assumingpurely conductive heat flow). The seeps werestudied in detail by using conductivity-temperature-depth (CTD) casts (and measure-ments of light transmission and oxygen con-centration) and a 12 kHz echo sounder. In theMalenki seep area, samples of hydrates in di-atom-rich silts and silty clays were retrievedfrom shallow hydrate accumulations (;20 cmbelow the lake floor).

INTERPRETATION AND DISCUSSIONOF THE DATAMorphological Expression of Seeps at theLake Floor

The four documented seeps are Bolshoy(large), Stari (old), Malyutka (very small), andMalenki (small). Malenki and Malyutka arelow-relief craters, and Bolshoy and Stari aremud cones on the lake floor. The seeps occursouth of a small fault antithetic to the Posol-sky fault (downthrown side to the north).

Malenki is the largest of the low-relief cra-ters, and occupies an irregular low area dis-sected by fault escarpments. The seep activityis visible on acoustic data as a reflectiveplume in the lake water. The seep area has amaximum depth of 8 m and a diameter of;500 m. Malyutka is a low-relief seep areaof ;200 m diameter with small parallel es-

632 GEOLOGY, July 2002

Figure 1. Setting of study area and bathymetry within complex rift structure of LakeBaikal. Gas-hydrate accumulations (shaded regions) occur almost symmetricallyaround Selenga delta. Inset shows Posolsky fault and locations of Figures 2 and 3.

Figure 2. Side-scan sonar mosaic showing four methane seeps in their structural setting.

carpments within the low area. The seep ac-tivity is seen on bathymetry data as well ason multichannel seismic data. The poor sur-face expression, lacking extrusive mudflowsor well-developed craters, suggests that theMalenki and Malyutka seeps are youngfeatures.

Bolshoy is the largest of the four vents. Itappears as an irregular cone 24 m high and800 m in diameter (Fig. 3). Acoustic anoma-lies are present at the top of the cone. On side-scan sonar images the Bolshoy cone appearsto be built from several smaller cones, givinga rough appearance to the slopes. Lake-floor

sediments accumulate against the northwestflank of the cone, locally smoothing the irreg-ular topography and causing a difference inlevel of 10 m between the opposing sides ofthe mud cone. The seismic data show possiblelenses of extrusive mud just below the lakefloor (maximum 40 m deep and as thick as 20m) in the vicinity of the Bolshoy cone, butthere is no seismic line through the center ofthe seep. The side-scan sonar data show notraces of mudflows at the lake floor. Stari isthe farthest from the small antithetic fault. Itis an ellipsoidal mound ;500 m long and 10m high with an irregular surface.

Relationship with Gas Hydrate and HeatFlow

The seismic reflection profiles show that theBHSZ in Lake Baikal is in general marked bya continuous high-amplitude bottom-simulat-ing reflection (BSR) not affected by intrabasinfaults or dipping stratigraphic reflections. Inthe study area, however, the BHSZ is irregularand strongly disrupted in the vicinity of theantithetic fault (Fig. 4A). Bright reflections,interpreted as free-gas pockets below theBHSZ (Vanneste et al., 2001), occur in patch-es that appear to be displaced along faults, al-though their apparent vertical displacement isat places much larger than the actual fault dis-placement. The distribution of bright reflec-tions indicates that at places the BHSZ shal-lows to a subbottom depth of 150 m in thefootwall block of the fault (as opposed to;400 m in the surrounding areas). The faultsegment along which this occurs is ;10 kmlong; adjacent segments of the same faultshow no anomalies in the BHSZ morphology.

The actual seeps occur on top of narrowvertical zones of chaotic reflections—seismicchimneys—that disrupt continuous subsurfacereflections on the seismic profiles (Fig. 4B).From the crest of the dome-shaped BHSZ tothe lake floor, these seismic chimneys can beas high as 200 m. Such a seismic chimney ismost commonly interpreted as a zone of ver-tical fluid conduction caused by hydraulicfracturing of the overburden by overpressured,often gas-charged, fluids (Van Rensbergen etal., 1999).

In the study area, heat-flow values from insitu thermal-probe measurements range be-tween 55 and 110 mW/m2, slightly higherthan average heat-flow values for the BaikalSouth Basin (50–70 mW/m2). The variationsof the measured values correlate well with theobserved changes in depth of the BHSZ. Thisfinding indicates that the irregular morphologyof the BHSZ is predominantly temperaturecontrolled and directly related to the existenceof a local anomaly of higher heat flow. At theMalenki seep, heat flow rises to 160 mW/m2,much higher then the heat flow calculatedfrom the depth of the BHSZ (Fig. 4A), indi-cating that hydrate is probably not stable atthe observed depth.

At the Malenki seep, the volume of meth-ane released by decomposition of hydrate isestimated as 3.9 3 109 m3 (on the basis of theBHSZ morphology and according to Athy’slaw for porosity decrease with depth and a hy-drate saturation of 10%, which was observedat the hydrate level in the BDP core). As thisfree gas rises, a large part of it will be fixedagain in the form of gas hydrate. One exampleis a shallow hydrate accumulation encounteredin the Malenki seep area.

GEOLOGY, July 2002 633

Figure 3. Detail of Bolshoy seep on side-scan sonar. Bolshoy seep is characterized by mudcone with irregular surface; its eastern flank is partly smoothed by lake-floor sediment. Low-backscatter anomaly in center represents gas cloud at apex of Bolshoy cone. A: Uninterpreteddetail of side-scan sonar image. B: Side-scan sonar with superimposed bathymetry from echo-sounder survey. C: Interpretation highlighting irregular morphology of Bolshoy cone.

Geochemical Characterization of the SeepsNine vertical CTD casts are located at and

around the Malenki crater. Figure 4C showsthe results of a series of four CTD casts mea-sured within a few hours time. The data showa small average positive temperature anomalyof 1.5 6 0.7 mK and an average negativeanomaly in oxygen concentration of ;0.0876 0.024 mgO2/kg at station 04, but no sig-nificant variations in light transmissivity orconductivity. The temperature and oxygenanomalies occur in an interval of 60 m abovethe northwestern part of the crater (stations 09and 07). To the northwest of the crater, thisanomaly occurs in an interval between 70 mand 125 m above the lake floor (station 04).The anomalies of temperature and oxygen canbe attributed to the oxidation of rising meth-ane bubbles (R. Kipfer, 2001, personal com-mun.): the measured average oxygen deficien-cy corresponds to an oxidation of 1.36 3 1026

mol/L of CH4 in the free gas, which results inan estimated temperature increase of ;0.3 60.1 mK. The estimated temperature increase isin agreement within a factor of two with themeasured average temperature increase. Thesetemperature and oxygen anomalies are transi-tory phenomena; CTD casts at almost thesame locations measured a week earlier didnot show any significant anomaly. These mea-surements confirm that gas is intermittentlyescaping at the Malenki crater. If fluids arealso expelled at Malenki, they cannot be muchdifferent from the lake water in terms of tem-perature and salinity.

However, geochemical analysis of gas-hydrate–associated waters and subsurface porewaters in the Malenki seep area by Granina etal. (2000) showed that chloride concentrationsin gas-hydrate–associated water (19.8 mg/L)and sediment pore water (11.8 mg/L) in theMalenki seep area are much higher than av-erage concentrations in pore waters in LakeBaikal’s South Basin (0.8 mg/L). An increaseof chloride concentration in pore waters at theMalenki seep area is associated with an in-crease in sulfate concentration. The alkalinityof Baikal water is low (average con-2HCO3

centration 66 mg/L; Falkner et al., 1991) andis even lower in the pore water sampled at theMalenki seep area (minimum concen-2HCO3

tration 12 mg/L; Granina et al., 2000), pre-venting carbonate precipitation. Lake Baikal isa freshwater basin with very low mineraliza-tion; hence the enrichment in chlorides andsulfates of the hydrate-associated waters andpore waters in the Malenki seep seems to sug-gest a deeper subsurface or onshore source.This conclusion is in agreement with thechemical and isotopic analysis of interstitialwaters at the Malenki seep area by Matveevaet al. (2001). They also found an increase inmineral content in the gas-hydrate–associatedwaters interpreted as indications of subsurfaceinjection of exotic fluids from deeper subsur-face levels or from onshore sources along ac-tive rift faults.

Shanks and Callender (1992) explainedsimilar geochemical signatures from hydro-thermal vents in Frolikha Bay, northern Lake

Baikal, as meteoric water intrusion from on-shore mountain ranges along active rift faultsand leaching of Cambrian evaporite rocks inthe vicinity of Lake Baikal. The geochemicaldata at the Malenki seep may be interpretedsimilarly. Only in this case, venting of hydro-thermal fluids at the lake floor is hindered bya low-permeability hydrate layer, and heat isadvected from the warm hydrothermal fluidsto the base of the hydrate layer.

CONCLUSIONSThe lake floor seeps in Lake Baikal’s South

Basin are methane seeps that occur where thebase of the hydrate stability zone shallows inresponse to elevated heat flow. The heat-flowanomalies are most likely caused by hydro-thermal activity along active fault segments ina way similar to that at hydrothermal ventsites in the Baikal North Basin. This studyprovides evidence that pore-fluid volume in-crease by localized dissociation of gas hydratecan cause an increase in pore pressure and hy-drofracturing of the sedimentary overburdenfollowed by vigorous expulsion of methaneand water and extrusion of entrained mud.

The role of gas hydrates in the global car-bon budget is still not fully understood. Sincethe discovery of submarine gas hydrates(Markl et al., 1970), the possibility of suddendestabilization and gas blowouts has been sug-gested with dramatic consequences for cli-mate, sediment stability, and buoyancy ofships. These consequences have been down-played (e.g., Kvenvolden, 2002). In fact, evi-dence is building that gas hydrates in seep ar-eas accumulate as a consequence of apreexisting upward fluid flow (e.g., Makranaccretionary wedge; Von Rad et al., 2000),rather than acting as a trigger for the flow.This is not the case in Lake Baikal, wheremethane is almost entirely of biogenic originand where no deep sources of gas are known.Methane seeps and mud volcanoes in LakeBaikal are interpreted to be examples of vig-orous gas and fluid expulsion caused by tec-tonically controlled gas-hydrate dissociation byan upward flow of fluids advecting heat to theBHSZ. The process is short-lived, being limitedby the amount of hydrate, by the duration ofthe hydrothermal pulse, or by the time neededfor a new equilibrium to be achieved.

ACKNOWLEDGMENTSThe research project was funded by the Belgian

Federal Office for Scientific, Technical and CulturalAffairs (OSTC) and the INTAS (International Co-operation with the New Independent States of theformer Soviet Union) project 1915. Van Rensbergenis supported by the Fund for Scientific Research–Flanders (FWO); Hus and Vanneste have a grantfrom the Flemish Institute for Scientific TechnologicResearch (IWT). We thank the captain and crew ofR/V Vereshchagin and the Director of the Limno-logical Institute (Irkutsk, Russia) for their support,E. Chapron and R. Kipfer for their expert advice,

634 GEOLOGY, July 2002

Figure 4. A: Seismic cross section showing localization of Malenki seep in its structuralsetting. Measured and calculated heat flow values are plotted along same profile. B: Detailof seismic section showing seismic chimney from crest of elevated base of gas-hydratelayer to lake floor at Malenki seep. Also shown is acoustic anomaly in near-bottom lakewater. C: Near-bottom measurements of temperature and oxygen concentration indicatebuoyant methane plume northwest of Malenki seep.

and Peter Vogt, John Peck, John W. King, and anunknown referee for their positive reviews.

REFERENCES CITEDCrane, K., Hecker, B., and Golubev, V., 1991, Hy-

drothermal vents in Lake Baikal: Nature,v. 350, p. 281.

Falkner, K.K., Measures, C.I., Herbelin, S.E., andEdmond, J.M., 1991, The major and minor el-ement geochemistry of Lake Baikal: Limnol-ogy and Oceanography, v. 36, p. 413–423.

Golmshtok, A.Y., Duchkov, A.D., Hutchinson,D.R., and Khanukaev, S.B., 2000, Heat flowand gas hydrates of the Baikal rift zone: In-

ternational Journal of Earth Sciences, v. 89,p. 193–211.

Golubev, V.A., Klerkx, J., and Kipfer, R., 1993,Heat flow, hydrothermal vents and static sta-bility of discharging thermal water in LakeBaikal (south-eastern Siberia): Bulletin duCentre de Recherches Elf Exploration Produc-tion, v. 17, p. 54–65.

Granina, L.Z., Callender, E., Lomonosov, I.S., Mats,V.D., and Golobokova, L.P., 2000, Anomaliesin the composition of Baikal pore waters:Geologija i Geofizika, v. 42, p. 362–373 (inRussian).

Kipfer, R., Aeschbach-Hertig, W., Hofer, M., Hoh-mann, R., Imboden, D.M., Baur, H., Golubev,V., and Klerkx, J., 1996, Bottomwater forma-tion due to hydrothermal activity in FrolikhaBay, Lake Baikal, eastern Siberia: Geochimicaet Cosmochimica Acta, v. 6, p. 961–971.

Kvenvolden, K.A., 2002, Methane hydrate in theglobal organic carbon cycle: Terra Nova (inpress).

Kuzmin, M.I., Kalmychkov, G.V., Geletij, V.F., Gni-lusha, V.A., Goreglyad, A.V., Khakhaev, B.N.,Pevzner, L.A., Kavai, T., Ioshida, N., Duch-kov, A.D., Ponomarchuk, V.A., Kontorovich,A.E., Bazhin, N.M., Mahov, G.A., Dyadin,Y.A., Kuznetsov, F.A., Larionov, E.G., Mana-kov, A.Y., Smolyakov, B.S., Mandelbaum,M.M., and Zheleznyakov, N.K., 1998, Firstfind of gas hydrates in sediments of Lake Bai-kal: Doklady Akademii Nauk, v. 362,p. 541–543 (in Russian).

Markl, R.G., Bryan, G.M., and Ewing, J.I., 1970,Structure of the Blake-Bahama Outer Ridge:Journal of Geophysical Research, v. 75,p. 4539–4555.

Matveeva, T.V., Mazurenko, L.L., Soloviev, V.A.,Klerkx, J., Kaulio, V.V., and Khlystov, O.,2001, Gas hydrate accumulation associatedwith fluid discharge structure in the Lake Bai-kal: European Union of Geosciences Journalof Conference Abstracts, v. 6, p. 154.

Poort, J., van der beek, P., and ter Voorde, M., 1998,An integrated modelling study of the centraland northern Baikal rift: Evidence for non-uniform lithospheric thinning: Tectonophysics,v. 291, p. 101–122.

Shanks, W.C., and Callender, E., 1992, Thermalsprings in Lake Baikal: Geology, v. 20,p. 495–497.

Tanganydro Group, 1992, Sublacustrine hydrother-mal seeps in northern Lake Tanganyika, EastAfrican Rift: 1991 Tanganydro Expedition:Bulletin du Centre de Recherches Elf Explo-ration Production, v. 16, p. 55–81.

Vanneste, M., De Batist, M., Golmshtok, A., Krem-lev, A., and Versteeg, W., 2001, Multi-frequency seismic study of gas hydrate-bear-ing sediments in Lake Baikal, Siberia: MarineGeology, v. 172, p. 1–21.

Van Rensbergen, P., Morley, C.K., Ang, D.W.,Hoan, T.Q., and Lam, N.T., 1999, Structuralevolution of shale diapirs from reactive rise tomud volcanism: 3D seismic data from the Bar-am Delta, offshore Brunei Darussalam: Geo-logical Society of London Journal, v. 156,p. 633–650.

Von Rad, U., Berner, U., Delisle, G., Doose-Rolin-ski, H., Fechner, N., Linke, P., Luckge, A.,Roeser, H.A., Schmaljohann, R., and Wiedi-cke, M., 2000, Gas and fluid venting at theMakran accretionary wedge off Pakistan: Geo-Marine Letters, v. 20, p. 10–19.

Manuscript received November 8, 2001Revised manuscript received March 25, 2002Manuscript accepted April 9, 2002

Printed in USA