Gas Hydrate-Associated Carbonates and Methane-Venting at Hydrate Ridge: Classification,...

Greinert et al.: Gas hydrate-associated carbonates and methane-venting at Hydrate Ridge 1

Gas Hydrate-Associated Carbonates and Methane-Venting at Hydrate Ridge: Classification, Distribution, and Origin of Authigenic Lithologies

Jens Greinert, Gerhard Bohrmann, and Erwin Suess

GEOMAR Research Center for Marine Geosciences, Kiel, Germany

Hydrate Ridge is part of the accretionary complex at the Cascadia margin and

is an area of widespread carbonate precipitation induced by the expulsion of methane-rich fluids. All carbonates on Hydrate Ridge are related to the methane-carbon pool either through anaerobic methanotrophy or through methanogenesis. Several petrographically distinct lithologies occur in boulder fields or in massive autochtonous chemoherm complexes which include methane-associated diagenetic mudstones and venting-induced breccias. The mudstones result from methane diagenesis in different sediment horizons and geochemical environ-ments related to very slow methane venting. Cemented bioturbation casts occur as fragments, complex framework or as clasts together with bivalve shells as part of intraformational breccias, which are restricted to chemoherm complexes. Here, fluids ascend from the sub-seafloor and support aragonite-dominated car-bonate precipitation near or at the sediment surface. Voids within mudclast brec-cias are either aragonite-rich indicating a formation near the surface at vent sites or are cemented by dolomite, which indicates formation in deeper parts of the sediment column. Brecciation is caused by tectonic or slump processes. In addi-tion, we recognized a direct relationship between gas hydrates and sediment frac-turing as well as the oxygen isotope composition of carbonate lithologies. Such gas hydrate-associated carbonates either show layered megapores and veins as relics of the original gas hydrate fabric or consist of aragonite-cemented intra-clast breccias formed by growing and decomposing gas hydrate near the sedi-ment surface. Both rock fabrics and the enrichment of 18O in high Mg-calcite demonstrate carbonate-forming mechanisms of gas hydrate.

1. INTRODUCTION

Authigenic carbonates are a common feature at cold vent sites where methane-enriched or bicar-bonate-oversaturated cold fluids escape from the seafloor. Cold vents and associated lithologies were discovered at the accretionary complex of the Cascadia margin in 1984 [Suess et al., 1985; Kulm et al. 1986; Ritger et al., 1987]. Due to the observation of active methane ebullition [Linke et al. 1994], the documentation of a well-developed bottom simulating reflector (BSR) [West-brook et al., 1994; Trehu et al., 1999], and the recovery of carbonates and gas hydrate at ODP-Site 892 [Sample and Kopf, 1995; Kastner et al., 1995], Hydrate Ridge (Fig. 1) became the focus of intense investigations of GEOMAR Research Center during RV SONNE cruises SO109 and SO110 in 1996. One goal of this international and interdisciplinary program was to investigate the carbonates near or at vent sites to elucidate their formation mechanism. The objectives in-cluded studies of the formation environment (1) at or near the sediment surface with carbonates showing the impact of slumping, bioturbation and the presumed influence of gas hydrates; (2) in deeper sediment horizons at or near fracture zones with carbonates exhibiting fissures and veins from post-consolidation stress conditions; (3) in different diagenetic zones and from various fluid sources. These geochemically distinct zones are (a) the oxic zone, (b) the sulfate reduction zone, (c) the upper part of the methanogenic zone, (d) the lower part of the methanogenic zone. These zones can be condensed and displaced towards the sediment surface due to the upward migration of flu-ids, but they still leave characteristic authigenic lithologies behind. Previous work on Hydrate Ridge has shown that the prominent features covering the northern ridge crest are extensive morphological complexes made up of a large variety of carbonate lithologies. Kulm and Suess [1990] have presented a classification of these authigenic carbonate structures that distinguishes between slabs, irregular edifices and four types of chimneys, which vary in shape from doughnuts to conical chimneys of more than 1 m in height. Earlier, Ritger et

mailto:[email protected]


al. [1987] classified block- and slab-like specimens as mudstones. In addition, Sample and Reid [1998] described mudstones and mudstone breccias from the northern summit of Hydrate Ridge. The observation of gas hydrate-associated carbonates [Bohrmann et al., 1998] was an important impetus for the attempt to classify the carbonate lithologies at Hydrate Ridge in greater detail and explain their origin.

2. GEOLOGICAL SETTING AND GEOCHEMICAL ENVIRONMENT OF CARBONATE FORMATION

Hydrate Ridge is part of the second ridge of the north-south striking accretionary ridge system of the continental margin off Oregon (Fig. 1). Tectonic uplift and thrusting driven by the plate con-vergence causes thrust faults, extensional fractures and breached folds to develop along the ridge crest [Carson et al., 1995]. The faults extend through the accreted Pliocene terrigenous silty-clayey sediments with interlayered sand to down below the BSR. Upward displacements in the BSR, which underlies the entire Hydrate Ridge [Trehu et al., 1999], indicates a focused upward flow of warm fluids along the fault [Westbrook et al., 1994] which serves as conduit [Kastner et al., 1995 b]. The ascending warmer fluids stimulate the decomposition of gas hydrate, which leads to further dewatering [Sample and Kopf, 1995; Suess et al., 1999] and generally drives the fluid venting phenomena and carbonate precipitation at cold vents. Important diagenetic processes for the pore water and fluid composition are the formation and dissolution of minerals (clay minerals, carbonates, zeolithes) and biogenic components (e.g. fo-raminifers) in the sediment which modify the concentration of major elements (e.g. Ca, Mg, Si, K) and isotopic compositions (e.g. oxygen, strontium) in the pore water. Gas hydrate decomposi-tion decreases the salt content as can be often recognized in lower chloride values [e.g. Kven-volden, 1998] and affects the oxygen isotope composition due to the release of 18O-enriched wa-ter (εgas hydrate/H2O ~ 3‰) [Davidson et al., 1983]. Methane from gas hydrate decomposition or from the methanogenic zone can be oxidized to HCO3

- via sulfate reduction, which increases to-tal alkalinity and partly controls whether aragonite, Mg-calcite or dolomite precipitates. Further, the formation of authigenic carbonates at cold vents is influenced by other reactions during the diagenetic turnover of organic matter, which also influences the total alkalinity and the isotopic composition of ΣCO2 [Claypool and Kaplan, 1974; Whiticar, 1996]. Methane generation via CO2 reduction lowers the δ13CCH4 to values of less than -80‰ PDB just below the sulfate reduc-tion zone [Whiticar, 1996; Elvert et al., 1999], whereas with increasing burial depth the residual CO2 pool of the methanogenic zone becomes isotopically heavier and methane generated from this CO2-pool becomes isotopically heavier, too. This 13C enrichment in the dissolved CO2 spe-cies is the reason for the extremely heavy δ13C values of up to +26‰ PDB observed in dolomite mudstones and reported by Sample and Reid [1998] from samples found at Hydrate Ridge. How-ever, typical δ13C values from carbonates formed at methane vents show light values between -70 and -35 ‰ PDB [Ritger et al., 1987; Matsumoto, 1990; Kulm and Suess, 1990; Bohrmann et al., 1998]. These values are generated by methane oxidation via sulfate reduction (CH4 + SO4

2- → H2O + HS- + HCO3

-), which is thought to be the most important mechanism to increase alkalinity and trigger the authigenic precipitation of carbonates at cold vents [Ritger et al., 1987; Han and Suess, 1989; Kulm and Suess, 1990; Paull et al., 1992; Whiticar, 1996]. Apart from the availability of carbonate ions and the diagenesis of carbon, carbonate precipita-tion and dissolution is strongly influenced by the concentration of the cations Ca and Mg (Fe, Mn), complex forming anions like SO4

2- or PO43-and temperature [Burton, 1993; Fernández-Díaz

et al., 1996; Morse et al., 1997]. These factors determine the development of authigenic carbon-ates in general; hence, lithologies can be seen as a record of the geological and geochemical con-ditions under which they formed. With regard to cold vents, Burton [1993] and Savard et al. [1996] pointed out that aragonite seems to be favored in more oxic environments (high SO4

2-) with higher total alkalinity concentrations. The crystallization of Mg-calcite preferentially occurs under slightly more anoxic conditions with lower SO4

2- and total alkalinity concentrations. 3. CARBONATE COMPLEXES AND SAMPLING SITES

In 1996 we used a towed video sled (EXPLOS) to locate potential sample sites and map the dis-tribution of carbonate lithologies during SO109 and SO110. Based on this observations, we started sampling at the carbonate complex found by DSRV ALVIN in 1987 formerly called 'Bio-herm', 300 m west of ODP-Site 892 [Westbrook et al., 1994]. Samples were taken by a video-guided grab (television grab, TVG) at four locations near active cold vent fields (Fig. 2). Because of the large grab size carbonates of up to 70 cm in size and > 40 kg in weight were retrieved.


During SO109 a second complex was found approximately 800 m west of the northern summit of Hydrate Ridge. Again four samples were taken by TV-grab (Fig. 2). Despite the abundant clam and gastropod shells in the carbonates of these two complexes we use the term 'Chemoherm' for these both features [Bohrmann et al., 1998]. In the following, the former 'Bioherm' will be called 'ALVIN-Chemoherm' and the newly discovered complex west of the summit 'SONNE-Chemoherm' (Fig. 2). Samples from the summit itself were available from three TV-grabs. Be-cause of the massive occurrence of carbonate boulders which pave the seafloor in several layers and form a 20 m high and 400 m wide structure at the summit, we call this area 'Boulder-Complex'. In addition, samples from TVG 115 broadened the variety of carbonate lithologies from the northern summit of Hydrate Ridge. Along EXPLOS track 5 (Fig. 2), the seafloor around the Boulder-Complex and the SONNE-Chemoherm is characterized by soft sediment, sometimes with ripple marks and single or clusters of carbonate blocks and edifices. These carbonates vary in shape from well-rounded to subangu-lar disk-like slabs and boulders, doughnut-shaped chimneys, and cemented coiled bioturbation trails. In between the equivalent blocks and edifices at the Boulder-Complex (Plate I A and B), active venting is indicated by small fields of living chemoautotrophic clams (Calyptogena), whit-ish bacterial mats (Plate I B) and rising gas bubbles [Suess et al., 1999] as observed in the west-ern part of the Boulder-Complex. Based on the interpretation of seismic line OR 8 [Westbrook et al., 1994], the area of active fluid venting at the Boulder-Complex is probably fuelled by a sea-ward-dipping thrust fault which crops out at or near the western edge. At the SONNE-Chemoherm, very irregular and porous carbonate lithologies appear as a massive autochthonous carbonate complex incised by gullies several meters in depth and width. Only a thin sediment layer covers the rocks, but sediment accumulates inside the gullies, which allows clams to live. These gullies are probably induced tectonically by faults which cut the massive carbonate com-plex and also focus the fluid flow. The carbonates themselves contain clam and gastropod shells and typically show yellow-colored edges; the latter could be identified as pure aragonite precipi-tations. On the southern summit of Hydrate Ridge randomly distributed single carbonate blocks were found during the track of EXPLOS 17. (Fig. 1) Bacterial mats and irregular white gas hydrate patches of up to 10 m in diameter paving the seafloor together with shells and living clams of Ca-lyptogena indicating venting activity (Plate I C) could be observed. The sampling at Station TVG 18 (Fig. 1) yielded more than 50 kg of pure white gas hydrate intercalated with strongly reduced sediment and carbonate breccias [Bohrmann et al., 1998]. Additionally, angular edges of light yellow colored layers cropping out at the seafloor (Plate I C) were identified as aragonite crusts; they occurred in direct contact with and even inside gas hydrates.

4. ANALYTICAL METHODS

The carbonates were examined microscopically in thin sections and analyzed by x-ray diffraction (XRD). Subsamples were mechanically separated with a microdrill for mineral and isotope inves-tigations. Samples for XRD analyses were prepared by a standard procedure adding corundum as internal standard. XRD patterns were obtained from 20° to 60° 2θ (0.01° 2θ steps) to calculate semiquantitative abundances of the carbonate phases aragonite (arag), low Mg-calcite (lMc, <8 Mol% MgCO3), high Mg-calcite (hMc, 8 - 20 Mol% MgCO3), protodolomite (proto, 30 - 40 Mol% MgCO3) and dolomite (dolo, 40 - 55 Mol% MgCO3) using standard calibration curves of different mineral mixtures (accuracy ± 5 wt.%). Using the shift in d-spacing of the (104) reflec-tion, we estimated the MgCO3-content of the trigonal carbonate phases. Neither chemical analy-ses (ICP-OES; HCl and acetic acid leaching) nor cathodoluminescence microscopy of 10 samples yielded any evidence for the presence of Fe or Mn in the carbonate phases. The carbonate and organic carbon content was calculated from the carbon difference between bulk and HCl-treated sub-samples measured by a Carlo-Erba NA-1500 Element Analyzer. The CO2 extraction for δ13C and δ18O measurements was carried out with pure H3PO4 at 75°C with a Carbo-Kiel on-line device connected to a Finnigan MAT 252. Replicate analyses of a laboratory standard show a standard deviation better than 0.03‰ for δ18O and 0.02‰ for δ13C. Because of the unequal oxygen fractionation during the reaction to CO2, dolomite has an analyti-cal offset of +1.63 ‰ δ18O [Rosenbaum and Sheppard, 1986]. Our data are not corrected for this offset, which has to be taken into account in temperature calculations for dolomitic samples. For reasons outlined by Greinert [1999], we used the equations by Friedman and O'Neil [1977] for Mg-calcite, Hudson and Anderson [1989] for aragonite, Irwin [1968] for protodolomite and Clayton et al. [1968] for dolomite to calculate equilibrium formation temperatures or to deter-


mine a possible influence of "heavy water" from decomposing gas hydrate on the δ18O signal of the carbonate. Strontium isotope analyses were performed on acid leachates of carbonate precipitates with bulk or microdrilled samples. Following Sample and Kopf [1995], we used ca. 2 ml of 1.0 N acetic acid for 2 h to leach the samples after gently grinding and rinsing them with micropure water. Sr was separated on cation exchange columns with 2.5 N HCl and dried at 120 - 150°C. Analyses were carried out with a Finnigan MAT 262.

5. RESULTS AND DISCUSSION 5.1 Petrographic Description and Classification

Our carbonate samples (totaling 450 kg and > 130 individual hand specimens) were visually pre-classified into mudstones (M), mudclast breccias (McB), intraformational breccias (IB), ce-mented bioturbation trails (Bt) and gas hydrate-associated carbonates (GC). Shore-based analyses show that homogeneous, tectonized and bioturbated mudstones can be subdivided into different types (Plate I).

Homogeneous mudstones (M-h): Samples of this type are gray to light brown in color but often coated by dark brown Mn/Fe oxi-hydroxides (Plate I A). They are characterized by a dense homogene-ous matrix of terrigenous sediment (quartz, feldspar, amphibole, and clay minerals) and fo-raminifers. They are completely cemented by micrite to micro sparite (>67 wt% carbonate) con-sisting of pure dolomite or mixtures of hMc, proto and dolomite (Table 1). Dark specks of fram-boidal pyrite are common; enrichments of pyrite can also be found in dark layers sub-parallel to the rock surface (Plate I). Frequent light-colored rims of up to 2 cm in thickness indicate a secon-dary oxidation which is supported by oxidized pyrite and decreasing organic carbon contents. Some samples are enriched in dolomite (35-80 wt.%) in the outer rim suggesting dolomite recrys-tallization. Ritger et al. [1987] report a decrease in carbonate content for such rims, suggesting they are undergoing dissolution.

Tectonized mudstones (M-tc): These samples have a matrix similar to type M-h but are systematically enriched in dolomite (Table 1). The surface mirrors the internal texture of in situ brecciation with angular edges. Dark fissures and veins (<1 mm) represent clay mineral enrichment (Plate I). Other veins are cemented by dolomite-rich micrite to sparite formed by crystallization in an un-confined space. Occasionally these veins show different cement generations as described by Sample and Reid [1998]. Hand specimens of strongly tectonized mudstones show the transition to shear breccias.

Bioturbated mudstones (M-bt): Typical for this type are oval-shaped gray areas of bioturbation casts of 0.5 to 3 cm in diameter covering more than 80% of the rock cross-sections (Plate I). A wave-like surface is shaped as relief of the bioturbation trails. The terrigenous matrix is similar to that of other mudstones but is cemented by lMc, hMc and less than 35 wt.% by protodolomite (Table 1). No significant trends, neither in mineral composition nor in organic carbon content, were found between the bioturbation casts and the undisturbed beige-colored sediment matrix. Some samples showed sheared casts, indicating a secondary strain of the rigid carbonate block (Plate I).

Cemented bioturbation casts (Bt): These features frequently occur as elongated or bent single specimens on the seafloor (Plate I A) or build frameworks of crossed burrows (in particular at TVG 115). They are rich in glauconite (< 30%) and colored from light gray to dark green. Single casts usu-ally can be classified as matrix-supported calciarenitic breccias of sub-angular clay-clasts (hMc to dolomite cemented) within a silt- to sand-sized matrix cemented by aragonite. Others are com-posed of a silt to clay matrix with sand-sized quartz or feldspar grains (< 10%) and are cemented by aragonite, hMc or protodolomite.

Mudclast breccias (McB): These arenitic to ruditic breccias consist exclusively of intraclasts or have few extraclasts in addition (Plate I). A densely cemented type (McB-dc) can be differentiated from a porous, light and extremely aragonite-cemented type (McB-ap). Both types are grain- to matrix-supported. Hand specimens of type McB-ap are composed exclusively of internally uncemented angular mudclasts within a calcisiltite matrix which is cemented by sparitic aragonite (Table 1). In some samples aragonite cementation culminates in cm-wide white areas which are composed entirely of radially grown aragonite needles (10 to 40 µm across, up to 1 mm in length). Only a few samples are enriched in lMc and show transitions to the McB-dc type. This type consists of weakly cemented mudclasts in a dense matrix of terrigenous material with monomineralic or polyphased cements (Table 1). Dark veins and fractures provide evidence for post-cementation tectonization.


Intraformational breccias, chemoherm-typical (IBc): These polymict lithologies contain chaotically ori-entated ruditic components of matrix-like intraclasts (> 1 cm), dark extraclasts (< 5%), bioturba-tion casts, shells and poorly lithified mudclasts (Plate II). They are the dominant carbonates at the chemoherm complexes and identical to the irregular edifices reported by Kulm and Suess [1990], exhibiting extremely irregularly shaped and randomly interconnected plumbing network of mesopores. The IBc carbonates investigated can be divided into a porous type with a pore vol-ume of up to 20% (IBc-p) and a more densely cemented type (IBc-d; Plate II). Samples of the IBc-p type consist of more than 70% intraclasts, which are composed similarly to the surrounding matrix. This calciarenitic, clast-supported matrix of mudclasts (0.2 - 1 cm) and glauconite grains (ca. 5%; < 1 mm) in clay- to silt-sized terrigenous sediment is cemented by micritic and randomly orientated sparitic aragonite and little hMc (Table 1). The amount of bioclasts observed (Calyp-togena, Acharax and gastropod shells) was less than 15%; mesopores (0.1 to 3 mm) usually make up 5% to 10% of the volume. Within these mesopores, botryoidal, white aragonite cements occur (Plate II). Both clasts and matrix contain dispersely distributed pyrite (~1%) which also occurs inside foraminifers. Samples of type IBc-d show zonal rim cements of botryoidal aragonite (<7mm thickness) and an advanced, cloudy matrix cementation (slightly enriched in lMc, Table 1) which leads to a pseudobrecciated texture (Plate II). By simultaneously sampling massive gas hydrate layers and carbonates [Bohrmann et al., 1998], we could classify at least two gas hydrate-related rock structures (Plate II). Surface-near grow-ing gas hydrate at Hydrate Ridge builds massive, horizontal gas hydrate layers of sponge-like bubble fabric [Bohrmann et al., 1998]. This fabric is formed by the crystallization of gas hydrate at the rim of those gas bubbles which were trapped in the sediment [Greinert, 1999; Suess et al., 1999]. Both ascending bubbles and growing gas hydrate layers cause sediment brecciation. The resulting angular sediment clasts build an irregular, clast-supported fabric with high amounts of meso- and megapores. Gas hydrate layers or gas hydrate-cemented sediment parts in the host sediment prevent cementation by carbonate. In those cases the gas hydrate-free host sediment is cemented, the hydrate-bearing parts build pores after the gas hydrate has decomposed. Decompo-sition may occur in situ or during sample recovery. Lithologies with one of the above-mentioned rock structures were classified as gas hydrate carbonates (GC).

Aragonitic collapse breccia, formed by gas hydrate (GC-Ba): This type of breccia was recovered in di-rect contact to gas hydrate at Station TVG 18 (Fig. 1) [Bohrmann et al., 1998]. Samples show angular rudite- to arenite-sized clasts (Plate II). These monomict clasts are micrite-cemented, predominantly by aragonite, but consist also exclusively of hMc (Table 1). Due to gas hydrate decomposition the clasts build a grain-supported texture equivalent to collapse breccias with up to 35% pore space, which is cemented by botryoidal aragonite of several mm thickness (Plate II). Accessory, isolated layers of yellow aragonite - also observed at the seafloor surface during EXPLOS 17 (Plate I C) - were found in direct contact to gas hydrate and are interlayered with it (Plate II). Some layers show casts generated from the precipitation in the sponge-like bubble structure of hydrate layers [Suess et al., 1999]. Biomarker investigations of those layers show ex-tremely high amounts of components typical for methane-consuming and sulfate-reducing micro-organisms (Elvert et al., this volume). Those aragonite layers very likely represent a direct en-crustation of bio-films.

Elongated pores, relics of gas hydrate cementation (GC-ep): Significant for this petrographically-unrestricted structural feature are elongated megapores of irregular shapes (Plate II) which look similar to gas hydrate layers observed in the sediment at site TVG 18 [Bohrmann et al., 1998; Suess et al., 1999]. We found such pores (up to 2 x 12 cm) in a mudclast breccia from SONNE-Chemoherm, but they can be expected in other lithologies (e.g. mudstones) as well, if environ-mental conditions favors surface-near gas hydrate formation.

5.2 Stable Isotope Signatures of C and O

Carbon and oxygen isotopes are widely used to distinguish between 'normal' marine carbonates and vent-related ones which are influenced by the sedimentary diagenetic carbon cycle. Whereas carbon isotopes reflect the carbon source, oxygen isotopes are effected by the formation tempera-ture, the isotopic composition of the source water and the respective fractionation of the carbon-ate phases [Friedman and O'Neil, 1977]. The widest range in carbon and oxygen isotope data was observed in samples from ALVIN-Chemoherm with strongly positive δ13C and δ18O values up to +26.2‰ and +14.8‰ PDB, re-spectively, and extremely negative values also for δ13C (down to -56‰ PDB, Fig. 3). Based on these different isotope values and the corresponding carbonate lithologies and mineral composi-


tions, we can discriminate six groups which represent different geochemical environments for carbonate formation. Specimens from ALVIN-Chemoherm exclusively belong to group A, but similar isotope data were reported by Sample and Reid [1998], Kopf et al. [1995], and Sample and Kopf [1995] from ODP Site 892. The mudstones and one clast of a mudclast breccia consist of pure dolomite or mixtures of dolo - proto - lMc (δ13C <10‰). The diagenetic mechanism generating such positive δ13C-values is most likely caused by CO2 reduction in methanogenic zones [Piscioto and Ma-honey, 1981; Mozley and Burns, 1993; Kopf et al., 1995; Whiticar, 1996]. Additionally, the tem-perature calculation for the formation of dolomite in water with 0‰ SMOW predicts crystalliza-tion at temperatures between 15.3 and 30.5 °C (corrected for the analytical offset of dolomite). Taking into account the temperature gradient of 51°C/km [Carson et al., 1995] and the decrease in δ18O in the pore water of -1‰ per 220m at ODP site 892 [Kastner et al., 1995], samples from group A theoretically should have formed either between 150 to 400 mbsf or between 0 and 220 mbsf depending on the mineral phase dolomite or protodolomite (Fig. 4). Samples which fall into group B are tectonized mudstones; they consist of dolomite (78 to 100 wt.%), hMc and protodolomite. Formation temperatures in equilibrium with water of 0‰ SMOW δ18O suggest a crystallization at 7°C for protodolomite or 17°C for dolomite. The conditions ob-served at ODP Site 892 predict a formation depth between 20 and 80 mbsf or 170 to 240 mbsf depending on the primary carbonate phase (Fig. 4). This indicates a formation of group B car-bonates at temperature conditions equivalent to greater sediment depth, too Samples with isotope data falling into group C (McB-dc, M-tc / h and Bt) were found at both chemoherms and the Boulder-Complex (Fig. 3). Caused by the dolomite-enriched mineralogy the oxygen isotope values are between 6.3 and 7.5 ‰ PDB. For dolomitic samples of TVG 45-2, formation in equilibrium with pore water of 0‰ SMOW requires 16.9°C or water of -3.5‰ SMOW at 4.55°C, which is the in situ bottom water temperature at the northern summit [Grein-ert, 1999]. Such isotopically negative pore water in surface-near sediments is unlikely, therefore higher temperatures must be assumed equivalent to a deeper origin (190 mbsf, Fig. 4). For the origin of primary protodolomite at 4.55°C the water would have to be -0.44‰ SMOW; such val-ues were measured at TVG 18 several cm below the surface [Suess et al., 1999] but only below 80 mbsf at ODP Site 892 [Kastner et al., 1995]. Based on the δ13C values from -34.6 to -46.7 ‰ PDB and an estimated amount of methane-derived carbon of more than 30%, using Rayleigh fractionation of a mixture of methane-derived carbon and CO2 of anaerobically degraded organic matter of -20‰ PDB (Greinert, 1999), formation of group C carbonates in the sulfate reduction zone is very likely. The samples of group D can be distinguished from group E by heavier δ13C values of up to -29.2 ‰ PDB and the absence of aragonite. The δ13C data likely indicate a mixture of CO2 generated by decomposition of organic matter via sulfate reduction (2CH2O + SO4

2- → 2HCO3- + H2S) and

methane-derived HCO3-, latter in smaller amounts than for group C. The lighter δ18O values,

compared to group C, results from the dominance of high Mg-calcite (12 mol% MgCO3). The es-timation of the δ18O-value of the source water in equilibrium with pure hMc carbonates (4.3 ‰ δ18O PDB) at a bottom water temperature of 4.55°C yielded 0.81‰ SMOW. However, waters below 510 m water depth show values between -0.15 and -0.24 ‰ SMOW, which indicates an apparently 18O-enriched source water or lower formation temperatures of around 1.1°C (water 0‰ SMOW) for the hMc phase of group D. We assume that hydrate water might be the reason for this effect. Group E contains most of our samples as was the case with the samples described by Ritger et al. [1987] and Kulm and Suess [1990] (Fig. 3). Typical carbonate lithologies are mudclast and intra-formational breccias, but bioturbation casts and some mudstones have similar isotopic values. The mineralogy is strongly dominated by aragonite and hMc with minor amounts of lMc and pro-todolomite. With δ13C data ranging from -38.2 to -55.2 ‰ PDB carbonates are typical for meth-ane-induced carbonate precipitations. With δ18O values of up to 6‰ PDB (Fig. 3), most samples of group E are too heavy to have formed from ambient seawater at 4.55°C at an assumed value of 0 ‰ SMOW. Linear correlations exist between the δ18O values and wt.% aragonite, which indi-cates an intercept of 3.34 ± 0.1 ‰ PDB δ18O for pure aragonite (Fig. 5; TVG 36-5, 11, 119). This supports the hypothesis that aragonite forms in equilibrium with ambient sea water tempera-tures and δ18O values between 0 and -0.27 ‰ SMOW at the northern summit of Hydrate Ridge. The corresponding high Mg-calcite phase (Fig. 5) shows δ18O values ranging from 4.86 to 7.3‰ PDB. This indicates formation in equilibrium with isotopically heavy water of around 3.7‰ SMOW at 4.55°C or of around 0‰ SMOW at -8.7°C at station TVG 36-5. For the hMc end


members at TVG 11 (5.6‰ δ18O PDB, 16 mol% MgCO3) and TVG 119 (6.5‰ δ18O PDB, 14 mol% MgCO3), formation temperatures of -2.7°C and -6.1°C (0‰ SMOW) or 1.9‰ and 2.9‰ SMOW at 4.55°C represent equilibrium conditions. The good correlations strongly support that the geochemical and isotopic conditions during the hMc precipitation are significantly different from those during aragonite formation. As temperatures below 0°C are truly unrealistic, heavier pore water has to be assumed during the formation of hMc. This is also shown by isotopic analy-ses of gas hydrate-related carbonates of type GC-Ba from station TVG 18 which fall exclusively into group E. As before, the δ18O values correlate with the aragonite content and 3.66‰ for pure aragonite and 4.86‰ for pure hMc (17 mol% MgCO3; Fig. 5). Calculations of the formation temperature in water with 0‰ SMOW resulted in 3.82°C for aragonite and 0.68°C for hMc. Us-ing the bottom water temperature of 4.12°C from the southern summit [Greinert, 1999], arago-nite should form from water with 0‰ SMOW δ18O, but the hMc-phase should form in "heavier" water of 0.86‰ SMOW. This further supports the assumption that gas hydrate influence hMc formation. Still enigmatic are the extremely high δ18O values of group F samples, whereas their carbon iso-topes are typical for a methane-generated carbon source. Taking into account the dolomite frac-tionation, even pure dolomite as found at station TVG 36-2 (14.9‰ δ18O PDB) would have to form at -8.3°C in water of 0‰ SMOW or at 4.55°C in water of 4.1‰ SMOW. Hydrate water as we know it from Hydrate Ridge [Suess et al., 1999] would not cause such a significant shift to-ward heavy oxygen isotopes.

5.3 Sr-Isotopes and Seawater Influence

Sr-analyses of pure aragonite cements (McB-ap; IBc-p/d; GC-Ba), aragonite-rich clasts and ma-trixes (IBc-d/p; GC-Ba), and hMc, proto- and dolo-containing mudstones (M-h/tc) show values between 0.709191 and 0.709120 87Sr/86Sr. Hence, all these carbonates are formed from water with Sr-values similar to present day seawater (0.709175) [Paytan et al., 1993]. Only one hMc- and one dolomite-rich sample from the ALVIN-Chemoherm indicate a different source with 87Sr/86Sr values of 0.708320 and 0.708853. Such values are equivalent to pore water analyses at ODP-Site 892 at around 10 mbsf [Kastner et al., 1995 b].

5.4 Carbonates of Diagenetic Origin in Sediments of Very Slow Venting: Mudstones

Mudstones are formed as micrite-cemented concretions in the sediment pore space in carbonate-oversaturated sedimentary environments. They are generated in the sediment column and can precipitate whenever supersaturation is induced by the general diagenetic environment. It is still not clear if they form exclusively by diagenesis or require very slow ascending fluids. However, that the carbon mass balances cannot be maintained in a closed system suggest at least very slow venting. Typically, concretions seem to grow as horizontal slabs or round boulders, which can be broken by tectonic stress conditions by vertically oriented fault planes (which may serve as fluid pathways). Evidence for authigenic carbonate formation in deeper sediment horizons was found in samples of type M-tc/h (group A). This is shown by the dominance of dolomite (Table 1), which requires high Mg/Ca ratios, an extreme 13C-enrichment and a calculated formation tem-perature of up to 30°C. Appropriate environmental conditions occur at ODP-Site 892 between 10 to 20 mbsf [Kastner et al., 1995 b]. Methanogenesis via CO2-reduction and the fractionation in a closed system pushes the δ13C values up to +18‰ between 19 and 24 mbsf [Whiticar et al., 1995]. Thus, group A carbonates are supposed to represent the deepest genesis of any carbonates found at Hydrate Ridge (Fig. 6), which form in environmental conditions close to the primary precipitation of dolomite [Lippmann, 1973]. Specimens of group B are suggested to have formed near carbonates of group A because of the same dolomitic carbonate types, equivalent δ18O values and also higher formation temperatures (up to 17°C). The lighter carbon isotope values compared to group A can be explained by admix-tures of small amounts of methanotrophic CO2 with CO2 derived from methane oxidation. Thus, the geochemical environment appears to be the deepest part of the sulfate reduction zone (Fig. 6) where geochemical conditions (Mg/Ca ratio) strengthen the crystallization of Mg-enriched car-bonate phases [Kastner et al., 1995 b]. Pore water data from ODP-Site 892 show that the base of the sulfate reduction zone occurs at 20 mbsf. This depth of the methanogenesis sulfate reduction-boundary depends on the geochemical conditions which are themselves influenced by ascending fluids (Fig. 6) and can be pushed upward up to some cm below the seafloor surface [Boetius et al., 2000].


Group C carbonates are still dolomite-dominated and crystallized at elevated temperatures, which implies a deeper environment than the lMc- and hMc-dominated mudstones of group D and E. Because of the suggested higher methane-derived carbon content (δ13C <34‰ PDB) relative to group B, we assume a formation in the Mg-enriched sulfate reduction zone, which is not influ-enced by the methanogenic zone, for the mudstones of group C. Such geochemical conditions are probably present above the transition zone (Fig. 6) and thus above group B carbonates. Indicated by non-dolomite-dominated cements and even "too heavy" oxygen isotope values com-pared to the ambient bottom water temperature, we suggest a formation of group D mudstones in an geochemical environment with decreased Mg/Ca ratios relative to group C carbonates and thus at a shallower depth. A mixture of methane-derived carbon and carbon from degraded or-ganic matter can be assumed from the δ13C values (Fig. 3) which again indicate a formation in the sulfate reduction zone (Fig. 6). Significant negative δ13C values as in group E support the formation of M-h and M-bt lithologies from methane-derived carbon as main carbon source. Because of their dolomite-free, not-sheared occurrence and the high amount of bioturbation features in mudstones, they very likely represent a formation environment near the sediment surface (Fig. 6) but still buried in anaerobic condi-tions, which are characterized by methane oxidation.

5.5 Methane Venting-Induced Carbonates: Mudclast and Intraformational Breccias

The uncompacted and porous texture of the aragonite-rich mudclast (McB-ap) and intraforma-tional breccias (IBc-p/d; group E) exclude formation far below the sediment surface. Angularly shaped, uncemented mudclasts in McB-ap and -dc lithologies suggest in situ brecciation of semi-consolidated sediment. Brecciation may be triggered by sediment slumps or tectonic- and vent-ing-induced disturbances like the "sudden" release of trapped fluids or gas [Matsumoto, 1990]. In general, carbonates of the IBc and McB-ap lithologies represent par-autochthonous formations restricted to areas close to cold vent sites and elevated rates of channeled fluid flow. They consti-tute the typical lithologies at both chemoherms. The significant enrichment of aragonite in IBc, McB-ap and Bt type carbonates of group E supports the formation in a sulfate-bearing environ-ment of slightly reducing conditions, where aragonite precipitation is favored against high Mg-calcite crystallization [Burton, 1993; Savard et al., 1996]. Such conditions are realized very close to or directly at the seafloor where methane-derived HCO3-rich fluids mix with bottom or sur-face-near pore water (Fig. 6) as also shown by present day Sr-isotope values. A precipitation into the bottom water seems confirmed by pure sinter-like aragonite crusts found at both chemoherm complexes [Greinert, 1999].

5.6 Carbonate Lithologies Generated by Gas Hydrate

The influence of gas hydrate on carbonates on the one hand is illustrated by the irregular and flat-tened 'gas hydrate pores' in a mudclast breccia (GC-ep; Plate II) from the northern summit. On the other hand, gas hydrate growth and decomposition produce the aragonitic collapse breccias characteristic for TVG 18 (GC-Ba, Plate II). As shown in the linear correlation between the ara-gonite content and δ18O values (Fig. 5) and in agreement with observations by Sample and Reid [1998], we propose that the oxygen isotope data of hMc phases reflect the influence of decom-posed gas hydrates on the formation of carbonates at both Hydrate Ridge summits. The domi-nance of aragonite in the matrix, the occurrence of pure aragonite crusts (GC-Ba) on the seafloor (Plate I C) or in gas hydrate fabric [Bohrmann et al., 1998] indicates a surface-near origin (Fig. 6). Here, the conditions typical for carbonates of group E in general stimulate the aragonite pre-cipitation with their high alkalinity and higher sulfate concentrations [Burton, 1993; Savard et al., 1996]. In contrast, Mg-calcite formation is favored at a relatively lower alkalinity and low sulfate concentrations. For hMc of group E, this has to be conform to a sediment environment with 18O-enriched pore water (Fig. 5). Such conditions are suggested to be present after the de-composition of gas hydrate when sulfate has been consumed for the oxidation of hydrate-methane via sulfate reduction. The released hydrate water further changes the isotopic composition of the pore water and decreases the alkalinity by dilution.

6. CONCLUSIONS

Petrographic investigations and detailed isotope analyses allow to characterize the formation en-vironment of varied venting-related carbonate lithologies. Excluding dolomitic mudstones (δ13C


+10 to +26 ‰PDB), which represent the deepest formation environment within the methanogenic zone, all analyzed carbonates from Hydrate Ridge contain methane-derived carbon. Suggested by the mineralogical and isotopic composition, other mudstones are formed within the sulfate reduc-tion zone from mixed carbon sources (residual CO2 from the methanogenic zone, oxidized meth-ane, decomposed organic matter) at different depths. Surface-near growing carbonates show in-creased lMc, hMc and dominantly aragonite contents, whereas carbon mainly derived from meth-ane (δ13C below -38‰ PDB). Oxygen isotope data indicate low formation temperatures and in-fluences of decomposed gas hydrate during the hMc crystallization. Intraformational breccias di-rectly formed at vent sites near the sediment surface are the typical carbonates building massive carbonate complexes, "Chemoherms". Structural signs of carbonate-forming mechanisms by growing and decomposing gas hydrate are given by aragonite-cemented collapse breccias and elongated megapores which can occur in different lithologies. Again, oxygen isotope data of the hMc phase suggest an influence of decomposed gas hydrate during the formation of these "gas hydrate carbonates".

Acknowledgement. This contribution is a part of the Ph.D. thesis of J.G., he wishes to thank all persons who helped from start to finish. We all acknowledge in particular the excellent technical assistance by B. Domeyer and J. Heinze (GEOMAR) as well as M. Joachimski (University Erlangen) and H. Strauß (Uni-versity Bochum) for isotope analyses. Further thanks to E. Hütten and T. Nähr for helpful discussions as well as C. Paull and an anonymous reviewer for useful comments. We thank the Federal Ministry of Educa-tion, Research and Technology, Bonn for supporting these work (Grants 03G0110A/B and 03G0109A/C).

REFERENCES

Boetius, A., Ravenschlag, K., Schubert, C.J., Rickert, D.,

Widdel, F., Gieskes, A., Amann, R., Jørgensen, B.B., Witte, U. and Pfannkuche, O. (2000): A microbial consortium apparently mediating anaerobic oxidation of methane. Nature, 407, 623-626.

Bohrmann, G., Greinert, J., Suess, E., and Torres, M. (1998): Authigenic carbonates from the Cascadia subduction zone and their relation to gas hydrate sta-bility. Geology, 26, 647-650.

Burton, E.A. (1993): Controls on marine carbonate cement mineralogy: review and reassessment. Chem. Geol., 105, 163-179.

Carson, B., Westbrook, G.K., Musgrave, R.J., and Suess, E. (eds.) (1995): Proc. ODP, Sci. Results, 146 (Pt. 1): College Station, TX (Ocean Drilling Program).

Claypool, G.E. and Kaplan, I.R. (1974): The Origin and Distribution of Methane in Marine Sediments. In: Kaplan, I.R. (ed.): Natural Gases in Marine Sedi-ments. Plenum Press, New York and London. 99-139.

Clayton, R.N., Jones, B.F., and Berner, R.A. (1968): Iso-tope studies of dolomite formation under sedimentary conditions. Geochim. Cosmochim. Acta, 32, 415-432.

Davidson, D.W., Leaist, D.G., and Hesse, R. (1983): Oxy-gen-18 enrichment in the water of a clathrate hydrate. Geochim. Cosmochim. Acta, 47, 2293-2295.

Elvert, M, Suess, E., and Whiticar, M. J. (1999): Anaerobic methane oxidation associated with marine gas hy-drates: superlight C-isotopes from saturated and un-saturated C20 and C25 irregular isoprenoids. Naturwis-senschaften, 86(6), 295-300.

Fernández-Díaz, L., Putnis, A., Prieto, M., and Putnis, C.V. (1996): The role of magnesium in the crystallization of calcite and aragonite in a porous medium. J. Sed. Res., 66, 482-491.

Friedman, I. and O'Neil, J.R. (1977): Compilation of stable isotope fractionation factors of geochemical interest. In: M. Fleischer (ed.): Data of Geochemistry, 6th ed., U.S.G.S Professional Paper, 440-KK

Greinert, J. (1999): Rezente submarine Mineralbildungen: Ab-bild geochemischer Prozesse an aktiven Flui-daustrittsstellen im Aleuten- und Cascadia-Akkretionskomplex. Dissertation, Christian-Albrechts-University, Kiel; GEOMAR Report 87, 217 pp.

Han, M.W. and Suess, E. (1989): Subduction-induced pore fluid venting and the formation of authigenic carbonates along the Cascadia continental margin: Implications for the global Ca-cycle. Palaeogeogr., Palaeoclimatol., Pa-laeoecol., 71, 97-118.

Hudson, J.C. and Anderson, T.F. (1989): Ocean temperatures and isotopic compositions through time. In: Clarkson, E.N.K., Curry, G.B., and Rolfe, W.D.I. (eds.): Environ-ments and physiology of fossil organisms, Trans. R. Soc. Edinburgh, 80, 183-192.

Irwin, H. (1980): Early diagenetic precipitation and pore fluid migration in the Kimmeridge Clay of Dorset, England. Sedimentology, 27, 577-591.

Kastner, M., Kvenvolden, K.A., Whiticar, M.J., Camerlenghi, A., Lorenson, T.D. (1995): Relation between pore fluid chemistry and gas hydrate associated with bottom-simulating reflectors at the Cascadia margin, Sites 889 and 892. In: Carson et al. (eds.) (1995): Proc. ODP, Sci. Results, 146 (Pt. 1): College Station, TX (Ocean Drilling Program), 175-187.

Kastner, M., Sample, J.C., Whiticar, M.J., Hovland, M., Cragg, B.A., Parkes, J.R. (1995 b): Geochemical evi-dence for fluid flow and diagenesis at the Cascadia con-vergent margin. In: Carson, B. et al. (eds.) (1995), Proc. ODP, Sci. Results, 146 (Pt. 1): College Station, TX (Ocean Drilling Program), 375-384.

Kopf, A., Sample, J.C., Bauer, P., Behrmann, J.H., and Er-lenkheuser, H. (1995): Diagenetic carbonates from Cas-cadia Margin: Textures, chemical composition, and oxy-gen and carbon stabile isotope signatures. In: Carson, B. et al. (eds.) (1995), Proc. ODP, Sci. Results, 146 (Pt. 1): College Station, TX (Ocean Drilling Program), 117-136.

Kulm, L.D., Suess, E., Moore, J.C., Carson, B., Lewis, B.T., Ritger, S.D., Kadko, D.C., Thornburg, T.M., Embley,


R.W., Rugh, W.D., Massoth, G.J., Langseth, M.G., Cochrane, G.R., and Scamman, R.L. (1986): Oregon subduction zone: Venting, fauna and carbonates. Sci-ence, 231, 561-566.

Kulm, L.D. and Suess, E. (1990): Relationship between carbonate deposits and fluid venting: Oregon accre-tionary prism. J. Geophys. Res., 95, B6, 8899-8915.

Kvenvolden, K.A. (1998): A primer on the geological oc-currence of gas hydrates. In: Henriet, J.-P. and Mienert, J. (eds.): Gas hydrates, Relevance to world margin stability and climatic change. Geological Soc. Spec. Pub., 137, 9-30.

Linke, P., Suess, E., Torres, M., Martens, V., Rugh, W.D., Ziebis, W., and Kulm, L.D. (1994): In-situ measure-ments of fluid flow from cold seeps at the active con-tinental margins. Deep Sea Res., 41, 721-739.

Lippmann, F. (1973): Sedimentary carbonate minerals. Berlin, Heidelberg: Springer, 228 pp.

Matsumoto, R. (1990): Vuggy carbonate crust formed by hydrocarbon seepage on the continental shelf of Baf-fin Island, northeast Canada. Geochemical Journal, 24, 143-158.

Morse, J.W., Wang, Q., and Tsio, M.Y. (1997): Influence of temperature and Mg:Ca ratio on CaCO3 precipi-tates from seawater. Geology, 25;, 85-87.

Mozley, P.S. and Burns, S.J.(1993): Oxygen and Carbon isotopic composition of marine carbonate concretions: An Overview. J. Sed. Pet., 63, 73-83.

Paull, C.K., Chanton, J.P., Neumann, A.C., Coston, J.A., and Martens, C.S. (1992): Indicators of methane-derived carbonates and chemosynthetic organic car-bon deposits: Examples from the Florida Escarpment. Palaios, 7, 361-375.

Paytan, A., Kastner, M., Martin, E.E., MacDougall, J.D., and Herbert, T. (1993): Marine barite as a monitor of seawater strontium isotope composition. Nature, 366, 445-449.

Pisciotto, K.A. and Mahoney, J.J (1981): Isotopic survey of diagenetic carbonates, Deep Sea Drilling Project Leg 63. In: Orlofsky, S. (ed.), Initial Reports of Deep Sea Drilling Project 63, Texas A. & M. University, Ocean Drilling Program College Station, Texas, United States, 595-609.

Ritger, S., Carson, B., and Suess, E. (1987): Methane-derived authigenic carbonates formed by subduction-induced pore-water expulsion along the Ore-gon/Washington margin. Geo. Soc. Am. Bull., 98, 147-156.

Rosenbaum, J. and Sheppard, S.M.F. (1986): An isotopic study of siderites, dolomites and ankerites at high temperatures. Geochim. Cosmochim. Acta, 50, 1147-1150.

Sample, J.C. and Kopf, A., 1995, Isotope geochemistry of syntectonic carbonate cements and veins from the Oregon margin: Implications from the hydrogeologic evolution of the accretionary wedge. In: Carson, B. et al. (eds.) (1995): Proc. ODP, Sci. Results, 146 (Pt. 1): College Station, TX (Ocean Drilling Program), p. 137-147.

Sample, J.C. and Reid, M.R. (1998): Contrasting hydro-geologic regimes along strike-slip and thrust faults in the Oregon convergent margin: Evidence from the chemistry of syntectonic carbonate cements and veins. Geol. Soc. Am. Bull., 110, 48-59.

Savard, M.M., Beauchamp, B., and Veizer, J. (1996): Signifi-

cance of Aragonite Cements around Cretaceous Marine Methane Seeps. J. Sed. Res., 66, 430-438.

Suess, E., Carson, B., Ritger, S.D., Moore, J.C., Jones, M.L., Kulm, L.D., and Cochrane, G.R., (1985): Biological communities at vent sites along the subduction zone off Oregon. In: Jones, M.L. (ed.): The hydrothermal vents of the eastern Pacific: An Overview. Biol. Soc. Wash. Bull., v. 6, p. 475-484.

Suess, E., Torres, M., Bohrmann, G., Collier, R.W., Greinert, J., Linke, P., Rehder, G., Trehu, A., Wallmann, K., Winckler, G., and Zuleger, E. (1999): Gas hydrate desta-bilization: enhanced dewatering, benthic material turn-over and large methane plumes at the Cascadia conver-gent margin. Earth Planet. Sc. Lett., 170, 1-15.

Trehu, A.N., Torres, M.E., Moor., G.F., Suess, E. and Bohr-mann, G. (1999): Temporal and spatial evolution of a gas hydrate-bearing accretionary ridge on the Oregon conti-nental margin. Geology, 27, 939-942.

Westbrook, G.K., Carson, B., Musgrave, R.J., et al (1994): Proc. ODP, Init. Repts., 146 (Pt.1): College Station, TX (Ocean Drilling Program), 55-125.

Whiticar, M.J., Hovland, M., Kastner, M., Sample, J.C. (1995): Organic geochemistry of gases, fluids, and hy-drates at the Cascadia accretionary margin. In: Carson, B. et al. (eds.) (1995): Proc. ODP, Sci. Results, 146 (Pt. 1): College Station, TX (Ocean Drilling Program), 385-397.

Whiticar, M.J. (1996): Isotope tracking of microbial methane formation and oxidation. Mitt. der intern. Verein. für theoretische und angewandte Limnol., 25, 39-54.

G. Bohrmann, J. Greinert, and E. Suess, GEOMAR Research

Center for Marine Geosciences, Wischhofstrasse 1-3, 24148 Kiel, Germany. Contact: [email protected].


Table 1. Mineral composition and isotope data of carbonates from Hydrate Ridge.

Lithology araga wt.% lMcb wt.% hMgc wt.% protod wt.% doloe wt.% δ13C ‰ PDB δ18O ‰ PDB

M-h n=31

- 10f (8-70)g n=10

33 (28-93) n=13

13 (7-73) n=17

44 (5-100) n=20

-37.0f

-54.4 to +26.3g 6.1

4.6 to 13.1

M- tc n=15

- 7 (36-66) n=2

20 (14-100) n=5

5 (19-35) n=3

68 (34-100) n=12

-37.9 -50.0 to -25.5

6.2 4.5 to 7.3

M-bt n=9

- 51 (38-79) n=7

25 (25-87) n=4

29 (13-34) n=9

- -47.4 -50.9 to -38.1

5.0 4.7 to 5.3

McB-dc n=28

3 (93) n=1

33 (13-100) n=11

20 (6-87) n=10

2 (1-16) n=4

42 (51-100) n=13

-32.3 -51.8 to +26.2

6.3 3.3 to 14.9

McB-ap n=13

94 (61-99) n=13

1 (4) n=1

4 (1-40) n=6

1 (8) n=1

- -52.6 -55.2 to -39.1

4.2 3.4 to 4.8

IBc-p n=23

65 (61-99) n=17

1 (1-11) n=10

27 (2-85) n=21

5 (1-21) n=12

1 (1-8) n=9

-49.4 -54.8 to -38.0

4.0 3.0 to 5.1

IBc-d n=31

59 (9-100) n=23

7 (1-100) n=12

20 (1-82) n=23

7 (1-60) n=16

7 (2-100) n=7

-48.9 -55.1 to -40.1

5.2 3.5 to 12.2

Bt n=9

41 (23-94) n=5

4 (1-16) n=5

36 (1-78) n=9

11 (3-30) n=6

8 (3-38) n=5

-45.9 -49.8 to -39.5

4.8 3.3 to 6.3

GC-ep n=6

57 (0-97) n=4

- 38 (3-91) n=6

5 (2-15) n=4

- -53.8 -54.4 to -53.0

3.8 3.2 to 4.4

GC-Ba n=54

77 (0-100) n=46

- 23 (0-100) n=40

- - -48.5 -54.2 to -40.6

3.9 3.2 to 4.9

(-) Not detected by XRD; n = number of samples; a aragonite; b low Mg-calcite; c high Mg-calcite; d protodolomite; e dolomite; f mean; g range.


Figure 1 The map shows the location of the second accretionary ridge at the Cascadia margin called Hydrate Ridge. The northern summit is enlarged in Fig. 2 (square). Line EXPLOS 17 and the open circle TVG 18 at the southern summit indicate an observa-tion track and sampling site of video-guided systems. JdF-P = Juan de Fuca Plate; NA-P = North American Plate.


Figure 2 Distribution of the carbonate lithologies based on weight percent of the samples from the northern summit of Hydrate Ridge. Dashed lines mark the estimated outline of the carbonate complexes. Numbers represent the station number of SONNE Cruise 109 (TVG 36-1, 36-2, 36-4, 36-5, 43-1, 43-2, 45-2, 110, 115, 119) and SONNE Cruise 110 (TVG 9, 11).


Figure 3 Carbon and oxygen isotope values of carbonates from the northern summit of Hydrate Ridge and two dolomitic mudstones from station TVG 18 (southern summit). Isotope data from GC-Ba type carbonates of TVG 18 fall exclusively into group E. Six groups (A to F) were distin-guished which differ in their dominating carbonate cement phases as well as in their carbon sources and conditions of formation. Also shown are fields which mark analytical results taken from literature.


Figure 4 Calculated δ18O values for different carbonate phases in equilibrium with the tem-perature gradient and oxygen isotope values found at ODP Site 892 (solid lines). Dashed lines for protodolomite and dolomite represent the analytical offset of 1.63‰ for dolomitic samples. They were used for the depth estima-tion of isotope investigations reported here.


Figure 5 Group E samples (Fig. 4) reveal linear correlation between the arag/hMc ratio and the δ18O values (black sym-bols) with an 18O-enriched hMc phase. Pure aragonite yields 3.5 ± 0.1 ‰ δ18O PDB, whereas 18O is strongly enriched in the high Mg-calcite phase (hMc). There is no clear correlation between the mineralogy and δ13C values (open symbols). TVG 36-5 from ALVIN-Chemoherm; TVG 11 and 119 from SONNE-Chemoherm; TVG 18 from the southern summit of Hydrate Ridge.


Figure 6 Sketch of the origin of petrographically and isotopically different lithologies relative to the controlling diagenetic environ-ment. Group A is restricted to the deep methanogenic zone resulting in the formation of dolomite in M-h and M-tc carbonates. Group B, also dolomite-dominated, is representative of the transition between sulfate reduction and methanogenesis. Group C lithologies, with typical methane-derived δ13C values but protodolomite and dolomite as carbonate phases, have formed in the sulfate reduction zone. Group D contains HCO3

- from a mixture of degraded organic matter and methane-derived carbon; carbonates are generated close to the surface which can be deduced from the absence of dolomite and the calculated formation temperatures. The aragonite-rich car-bonates of group E represent a formation near to or at the surface, influenced by methane oxidation via sulfate reduction; typical are irregular brecciated fabrics of chemoherm carbonates as well as the gas hydrate-related carbonates GC-Ba and GC-ep.

Plate I

shearedbioturbation trace

5 cm

M-bt

pyrite-richdark layers

M-h

5 cm

M-tc

area of radial grownaragonite needles

McB-ap

subsamples

uncemented mudclasts

5 cm

McB-dc

uncementedmudclast

A A

mudstone withMn/Fe coating mudstone withMn/Fe coating mudstone withMn/Fe coating

Bt Bt

slab slab

Bt

slab

A B B B

white bacterialmat

white bacterialmat

white bacterialmat

gas hydrate gas hydrate gas hydrate

aragonite crust aragonite crust aragonite crust

C C C

burrow

5 cm 5 cm

dark veins

extraclast

GREINERT ET AL: 103

Plate I Seafloor images (locations see Fig. 3) taken during EXPLOS 5 (A, B) and EXPLOS 17 (C); scale: horizontal length = 2m, length of bottom-weight = 20 cm. A: block- and slab-like mudstones (M) with brown Fe/Mn coatings and bioturbation casts (Bt); B: irregular chemoherm block (center, IBc), white bacterial overgrowth on a mudstone surrounded by white clams (left). C: white patches of outcropping gas hydrate, bacterial mats, and aragonite crusts in direct contact with gas hydrate, dead and living clams on the sediment. Mudclast breccias McB-dc TVG 36-2, McB-ap TVG 36-4; mudstones M-bt TVG 9, M-tc TVG 36-1, M-h TVG 36-1. The McB-ap sample is fixed in resin, which darkens the rims of the mudclasts.

Greinert, J., Bohrmann, G., and Suess, E. (2001): Gas Hydrate-Associated Carbonates and Methane-Venting at Hydrate Ridge: Classification, Distribution, and Origin of Authigenic Lithologies. In: Paull, C.K. and Dillon, P.W. (Eds): Natural Gas Hydrates: Occurrence, Distribution, and Detection. Geophysical Monograph 124, 99-113.

106 GASHYDRATE-ASSOCIATED CARBONATES AT HYDRATE RIDGE

Greinert, J., Bohrmann, G., and Suess, E. (2001): Gas Hydrate-Associated Carbonates and Methane-Venting at Hydrate Ridge: Classification, Distribution, and Origin of Authigenic Lithologies. In: Paull, C.K. and Dillon, P.W. (Eds): Natural Gas Hydrates: Occurrence, Distribution, and Detection. Geophysical Monograph 124, 99-113.

layered pore space;primarily filled with gas hydrate

5 cm GC-ep

subsample

pore space primarilyfilled with gas hydrate GC-Ba

Plate II

2 cm

gas hydrate (white)filling pores in a pure

aragonite crust (yellow)

5 cm

IBc-d

5 cm

aragonitecement

bioturbation trace

IBc-p

5 cm

brecciated area

aragoniterim cements

aragonite coating

5 cm

1

2

3

4

denselycemented area

Plate II Chemoherm-type intraformational breccias IBc-d TVG 119; IBc-p TVG 36-5; carbonates influenced by and formed in contact with gas hydrate GC-ep TVG 11, GC-Ba TVG 18. The images 1 to 4 show the process of gas hydrate dissolution in the pores of an aragonite crust.

Date post:	30-Nov-2023
Category:	Documents
Upload:	independent
View:	0 times
Download:	0 times

Gas Hydrate-Associated Carbonates and Methane-Venting at Hydrate Ridge: Classification,...

Documents