Deep-Sea Research I 74 (2013) 25–38

Contents lists available at SciVerse ScienceDirect

Deep-Sea Research I

0967-06

http://d

n Corr

E-m

journal homepage: www.elsevier.com/locate/dsri

Spatial distribution of seafloor bio-geological and geochemical processesas proxies of fluid flux regime and evolution of a carbonate/hydrates mound,northern Gulf of Mexico

Leonardo Macelloni a,n, Charlotte A. Brunner b, Simona Caruso c, Carol B. Lutken a,Marco D’Emidio a, Laura L. Lapham d

a Mississippi Mineral Resources Institute, Center for Marine Resources and Environmental Technology, Seabed Technology Research Center, University of Mississippi,

111 Brevard Hall, University, MS 38677, USAb Department of Marine Science, University of Southern Mississippi, Stennis Space Center, MS 39529, USAc Fugro Survey Limited, Survey House, Denmore Road, Bridge of Don, Aberdeen, Scotlandd Chesapeake Biological Lab, University of Maryland Center for Environmental Science, PO Box 38, 1 Williams St Solomons, MD 20688, USA

a r t i c l e i n f o

Article history:

Received 1 August 2012

Received in revised form

20 December 2012

Accepted 30 December 2012Available online 11 January 2013

Keywords:

Carbonate/hydrate mound

Hydrates

Gas hydrates

Hydrocarbon seeps

Chemosynthetic community

Fluid flux regime

Hydrates stability zone dynamics

Spatial distribution

37/$ - see front matter & 2013 Elsevier Ltd. A

x.doi.org/10.1016/j.dsr.2012.12.006

esponding author. Tel.: þ1 6629157320; fax:

ail address: lmacello@olemiss.edu (L. Macello

a b s t r a c t

Woolsey Mound, a carbonate/hydrate complex of cold seeps, vents, and seafloor pockmarks in

Mississippi Canyon Block 118, is the site of the Gulf of Mexico Hydrates Research Consortium’s

(GOMHRC) multi-sensor, multi-disciplinary, permanent seafloor observatory. In preparation for instal-

ling the observatory, the site has been studied through geophysical, biological, geological, and

geochemical surveys. By integrating high-resolution, swath bathymetry, acoustic imagery, seafloor

video, and shallow geological samples in a morpho-bio-geological model, we have identified a complex

mound structure consisting of three main crater complexes: southeast, northwest, and southwest. Each

crater complex is associated with a distinct fault. The crater complexes exhibit differences in

morphology, bathymetric relief, exposed hydrates, fluid venting, sediment accumulation rates, sedi-

ment diagenesis, and biological community patterns. Spatial distribution of these attributes suggests

that the complexes represent three different fluid flux regimes: the southeast complex seems to be an

extinct or quiescent vent; the northwest complex exhibits young, vigorous activity; and the southwest

complex is a mature, fully open vent. Geochemical evidence from pore-water gradients corroborates

this model suggesting that upward fluid flux waxes and wanes over time and that microbial activity is

sensitive to such change. Sulfate and methane concentrations show that microbial activity is patchy in

distribution and is typically higher within the northwest and southwest complexes, but is diminished

significantly over the southeast complex. Biological community composition corroborates the presence

of distinct conditions at the three crater complexes. The fact that three different fluid flux regimes

coexist within a single mound complex confirms the dynamic nature of the plumbing system that

discharges gases into bottom water. Furthermore, the spatial distribution of bio-geological processes

appears to be a valid indicator of multiple fluid flux regimes that coexist at the mound.

& 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Hydrate mounds are unique seafloor sites characterized by anextraordinary variability of bio-geological processes; the vent gasand gas hydrate are intimately associated with complex chemo-synthetic communities, whose initiation and stability dependupon hydrocarbon-driven microbial activity in sediments, includ-ing microbial hydrocarbon oxidation, reduction of CO2 via metha-nogenesis, sulfate reduction, and sulfide oxidation (Sassen et al.,

ll rights reserved.

þ1 6629155625.

ni).

1993, 1998, 2004). Microbial processes contribute to the devel-opment and stability of chemosynthetic communities by provid-ing required H2S. Anaerobic microbial processes lead todeposition of diagnostic authigenic minerals related to the carbon(carbonate minerals) and sulfur cycles (pyrite, elemental sulfur)that alter the structure of the seafloor. Seafloor cementationpromotes the formation of carbonate hard-grounds that providea hospitable environment for hard-bottom fauna, such as chemo-synthetic tubeworms and seep mussels (Sassen et al., 2004).Hydrate mounds, therefore, represent unique sites for under-standing the relationships between hydrate formation and dis-sociation, vent activity, sediment diagenesis, bacterial activity,and faunal distribution.

Fig. 1. Location of Mississippi Canyon block 118 (MC118).

L. Macelloni et al. / Deep-Sea Research I 74 (2013) 25–3826

The Woolsey Mound at Mississippi Canyon Block 118 (MC118)is an example of a complex carbonate/hydrate mound fed by anintricate system of faults overlying shallow salt and thereby withaccess to deeply-buried source rocks for gas and oil (Macelloniet al., 2012). Several submersible missions to the seafloor haveidentified large gas hydrate exposures, authigenic carbonatedeposits, several vents of methane and oil bubbles, and chemo-synthetic communities, including ice worms (Sassen and Roberts,2004; Woolsey et al., 2005). For these compelling reasons, MC118was chosen, in 2004, by the Gulf of Mexico Hydrates ResearchConsortium (GOMHRC) as the site for a long-term, gas hydratemonitoring station (McGee, 2006; Lutken et al., 2011a). In orderto characterize the site in preparation for installation of stationcomponents, and to build the baseline model to which to comparelong-term monitoring observations, MC118 has been investigatedextensively using a wide range of geophysical, geological, and bio-geochemical studies. Field investigations have targeted both themound’s subsurface – i.e., source of hydrocarbon gases (Laphamet al., 2008)), mechanism of fluid transport, and geometry of theshallow plumbing system – and processes ongoing at the mound’ssurface including hydrates formation and exposure, gas venting,sediment distribution, and faunal occurrence and distribution.Such observations have fostered the opportunity to assess marinegas hydrates formation/dissociation in a complex natural system.

The purposes of this paper are:

1)

To present the methodology we adopted to build a conceptualmodel of the spatial distribution of bio-geological processes onthe Mound;

2)

To investigate the relationship between the spatial distribu-tion of the bio-geological processes and the hydrocarbon fluidflux regime; and

3)

To derive the conceptual model that links the hydrocarbonfluid flux regime, fault activity, and hydrates formation/dis-sociation with mound formation/evolution.

2. Site location

MC118 is located approximately 170 km south of Pascagoula,Mississippi, and 100 km east of the Mississippi Canyon in�890 m of water (Fig. 1). The site is located on a gently seawarddipping (31 to 41) portion of the continental slope. A fault-controlled canyon, flanking the Whiting Dome and slump struc-ture to the east, is the only other prominent morphologicalfeature present in the block (Fig. 1).

Near MC118, on the upper slope, there is evidence, particularlyfrom bathymetry, of salt domes in the nearby shallow subsurface(Fig. 1). Exposures of gas hydrates, faulted carbonate hard-grounds and pockmark features consistent with gas and petro-leum seepage cover approximately 1 km2 of the seafloor at MC118(e.g., Sassen et al., 2006; Sleeper et al., 2006). The supply ofhydrocarbons (natural gas and petroleum) to the seafloor supportsan active biological seep community and microbial chemolitho-trophy in the immediate vicinity of active gas–fluid seepage.

3. Materials and methods

3.1. Multibeam echo-sounder data

A complete high-resolution, high-precision, bathy-morphologicalsurvey was executed in 2005 by C&C Technologies (Lafayette, La.)using a HUGIN 3000 Autonomous Underwater Vehicle (AUV). TheAUV was equipped with a Simrad EM 2000 multibeam bathymetrysystem, a dual frequency (120 and 240 kHz) side-scan sonar, and a2–10 kHz chirp sub-bottom profiler. The Simrad EM 2000 is a high-frequency echo-sounder that simultaneously records high-resolution depth data and co-registered acoustic backscatter. Depthdata have been post-processed combining standard and non-standard processing techniques described by Bosman (2004) andBosman et al. (2009).

Fig. 2. (a) Combined acoustic backscatter and 3d bathymetry map of the mound

and relevant mound provinces, at MC118, Woolsey Mound; (b). Video and photo

surveys conducted over the Woolsey Mound plotted on plan-view of the back-

scatter imagery. (For interpretation of the references to color in this figure legend,

the reader is referred to the web version of this article.)

Fig. 3. Locations of the geological samples collected at the Woolsey Mound.

Samples were used both for lithological and geochemical studies. The figure

displays (a) core positions and results of lithological analysis (symbols indicate

lithofacies as reported in Table 1) and (b) microbial activity where symbols

correspond to core microbial activity that is low, moderate, and high, respectively.

The inset in (b) illustrates examples of high and low microbial activity.

L. Macelloni et al. / Deep-Sea Research I 74 (2013) 25–38 27

Acoustic backscatter has been processed in order to removethe sound artifacts due to seafloor topography and systemgeometry as reported by Beaudoin et al. (2002) and Ferrini andFlood (2006). A backscatter seafloor map with a pixel resolutionof 3 m has been created and combined with the computed digitalelevation 3D model (Fig. 2a).

3.2. Seafloor imagery

Video images and photographs have been collected since 2002;35 h of video tape and 215 high-resolution submarine images havebeen acquired. Fig. 2b illustrates the mound area, color-coded accord-ing to seafloor reflectivity (backscatter), where video/photo images areavailable, and the platforms used to collect them. The video data wereframe-separated, and then the single frames were georeferenced. Theimages were archived in ARCView 9.1 GIS visual software in order toposition them correctly over the bathymetric data.

3.3. Core sampling for lithological and geochemical studies

More than 40 gravity cores were collected from the WoolseyMound (Fig. 3a). The cores were collected from the R/V Pelican in

May 2005, October 2005, and April 2008. Although core sampleswere collected with different systems and for different purposes,results appropriate to this study have been incorporated andappropriate methods are described in Sections 3.3.1 and 3.3.2.Core-lengths ranged between 5 cm and 460 cm below the seafloor(b.s.f.). The cores were split lengthwise; one half was used forlithological studies and one half for geochemical analysis. Thecores were photographed using a digital camera and describedvisually. One push-core, PC4414, was collected in August 2002with the Johnson Sea-Link manned submersible.

3.3.1. Lithological studies

The character of the host sediment surrounding the WoolseyMound to a distance of at least 3 km consists of poorly sorted,fine-silt-rich sediment with minor amounts of clay and sand-sizeforaminifera (Brunner, 2007; Brunner and Ingram, 2008; Ingramet al., 2010). A homogeneous, heavily bioturbated, light olive mud,which contains as much as 10% planktonic, foraminiferal sand and40% carbonate, caps the section to a depth of about 1 m andis largely Holocene in age, based on the planktonic foraminifera(Z Zone of Kennett and Huddlestun (1972)). A dark olive mudwith large burrows filled by light olive mud underlies the

L. Macelloni et al. / Deep-Sea Research I 74 (2013) 25–3828

Holocene interval. The foraminiferal sand content in this unit andall those beneath is reduced to a fraction of a percent, and totalcarbonate is reduced to 10–20% by weight (Brunner, 2009,unpublished data). The mottles grade into thinly stratified, brownand olive mud layers below, each layer being 1–2 cm thick. Themottled unit corresponds to most of the deglacial period (Zone Y1of Kennett and Huddlestun (1972)).

A clay-rich horizon with a distinct reddish hue and filled withreworked, pre-Quaternary nannofossils (Marchitto and Wei,1995) lies at or just below the base of the mottled unit. Thereddish layer corresponds to deglacial meltwater event 1 A(Aharon, 2006; Montero-Serrano et al., 2009). Radiocarbon datingof the horizon (Brunner and Ingram, 2008; Ingram et al., 2010)produced an age of �14.2–14.7 ka. The reddish layer covers muchof the central and western slope of the Gulf of Mexico and is auseful datum level (Montero-Serrano et al., 2009).

The stratal, layered units below the mottled mud (to amaximum cored depth of 15 m) vary in color and sharpness ofstratal contacts, and some intervals contain frequent, well-sorted,silt layers, typically no thicker than 1 mm (Brunner andBrizzolara, 2011; Brizzolara and Brunner, 2012). Many intervalscontain Tasmanites, an algal cyst. The lithofacies is similar to thatreported in the Wisconsinan sections of DSDP Leg 96 (Boumaet al., 1986). The planktonic foraminifera are consistent with aWisconsinan age (Y Zone of Kennett and Huddlestun (1972)).

The sediment within the Woolsey mound, in general, has thesame lithostratigraphy as described above, but some cores appearoverprinted by diagenetic alteration. The sediment in these cores isa clear, light gray in contrast to the brown and olive hues describedabove. The gray intervals contain few or are barren of foraminifera,sometimes contain large, vesicomyid clam valves, sometimes con-tain the pin-prick marks of exsolving gas, and often containcarbonate nodules surrounded by light gray areole. Lapham et al.(2008) report that the geochemistry in such cores supports authi-genic carbonate formation related to methanogenesis, though other

Fig. 4. (a) Relevant morphological features and seafloor provinces at MC118. The sur

Magenta and Yellow, appear as solid lines that are colored as named; black arrows indi

are after Macelloni et al. (2012). (b) Enlargement of the Woolsey Mound with crestal fa

are located here and plotted in (c). (c) Circled topographic transects X, Y, and Z cross th

legend, the reader is referred to the web version of this article.)

diagenetic processes associated with fluid flow from deep-seatedfaults could be at work.

3.3.2. Geochemical studies

Data from geochemical analysis of pore fluids in shallow sedi-ments have been published by Lapham et al. (2008). From down-core measurements of dissolved methane and sulfate concentra-tions, as well as stable carbon isotopic signatures of methane anddissolved inorganic carbon, Lapham et al. (2008) categorized threemicrobial activity groups. Cores for which microbial activity wasdetermined are plotted in Fig. 3b. Low activity is defined as nodecrease in sulfate or increase of methane within the upper 3 m ofsediment. This case is represented by Core 23, whose downholemicrobial activity is plotted in the insert, Fig. 3b. Moderate activityis defined when sulfate concentrations decrease down-core at a rateof �0.07 mM/cm, indicating an active, albeit slow, population ofsulfate reducing bacteria. High microbial activity, represented byCore 9 in Fig. 3b, is defined when sulfate decreases at a rate of�1 mM/cm within the upper 3 m of sediment. High microbialactivity is usually associated with methane concentrations measur-ing near saturation at 1 atm (�2 mM). The microbial activity isused as an independent parameter to constrain the bio-geologicalzonation of the mound. Such microbial activity groupings can beextended to relative fluid expulsion rates, as done in Gay et al.(2006). The high microbial activity grouping also has the highestupward methane flux.

4. Results: Data analysis

4.1. Bathymetric map and morphological units

The tridimensional, shaded-relief, bathymetric map of theWoolsey Mound at MC118 is presented in Fig. 4. The entire blockshows a generally smooth topography with a gentle 3% slope

face traces of normal faults, related to salt diapir activity and named Blue, Red,

cate seafloor pockmarks. Fault projections over the mound and fault nomenclature

ults appearing as dotted lines; W–E profiles through each of the crater complexes

e three vent complexes. (For interpretation of the references to color in this figure

Fig. 5. Characterization of seafloor imagery type(s). Seafloor imagery has been classified in terms of specific features at the seafloor. Type descriptions appear in the table, left.

L. Macelloni et al. / Deep-Sea Research I 74 (2013) 25–38 29

southward. The two obvious topographic anomalies are present inthe surface view: the Woolsey Mound in the south and a deep,narrow channel running northwest-southeast in the northeasterncorner (Fig. 4a). The mound has been subdivided into geomorphicsectors including three crater complexes: Southeast (SE), North-west (NW), and Southwest (SW) (Fig. 4b). Each complex consistsof several craters, each 5–20 m in diameter and low in relief(2–6 m) (Fig. 4c). The mound is further subdivided into threeother sectors. The Northern Sector is distinguished by a semicir-cular collapse structure just north of the NW crater complex. Thisstructure is characterized by rough seafloor morphology and isclearly circumscribed by a crown-shaped, slump scarp with apronounced edge (Fig. 4a and b). The Middle Sector is, in general,a depression with a bulge-and-hole morphology between thethree crater complexes. This depression shows a well-definedmargin in the east and north where the steep flanks step down tothe east, while to the south and west the perimeter is barelydiscernible (Fig. 4a and b). The last sector, the M Zone, consists ofa large, parabolic depression northeast of the Woolsey Mound.Several round pockmarks (indicated by black arrows in Fig. 4a)around the mound complete the morphological framework. Thefour solid lines in Fig. 4a (dashed in Fig. 4b) represent surfacetraces of normal faults. These faults, previously inferred fromseafloor morphology, have been confirmed by seismic imagingand are recognized as important tectonic structures produced byupward movement of a salt diapir beneath the mound (Macelloniet al., 2012). It is important to note that each of the three cratercomplexes lies on or near the intersection of a fault with the

seabed. Adopting the nomenclature of Macelloni et al. (2012), weobserve that the SE complex lies adjacent to the seafloor expres-sion of the Yellow fault, the NW complex lies at the intersection ofthe Red and Blue faults, and the SW complex occurs near theMagenta fault.

4.2. Seafloor image map and seabed types

Detailed mapping and semi-quantification of the areas coveredby the main bio-geological entities was accomplished via meti-culous analyses of video images and photographs. Emphasis hasbeen placed on recognition, description, and mapping of thefollowing features:

1)

Active gas vents and exposed hydrates; 2) Authigenic carbonate outcrops; 3) Chemosynthetic communities; 4) White bacterial mats (sulphide-oxidizing bacteria, predomi-

nantly Beggiatoa spp.); and

5) Areas covered by bioclastic materials such as shells and

shell hash.

The following seabed types have been thus recognized at theWoolsey Mound (Fig. 5):

�

Type A, soft-smooth: soft mud pierced by abundant, smallburrow openings, but with few epi-macroorganisms in view.

Fig. 6. Spatial distribution of seafloor features defined from imagery data. ‘‘Indeterminate video types’’ denotes less reliable results due to inadequate image quality.

L. Macelloni et al. / Deep-Sea Research I 74 (2013) 25–3830

The sediment is generally fine-grained with no more than 2%sand (Lutken et al., 2006).

�

Type B, soft-rough: mud similar to type A with some epi-macroorganisms present (i.e., bivalves, echinoids, crabs). Theorganisms never cover more than 50% of the field of view. � Type C, hard-smooth: seafloor where benthic organisms –

including whole valves and shell hash from dead clams –cover more than 50% of the sea floor.

� Type D, hard-rough: areas where hard carbonate deposits and

associated chemosynthetic assemblages are prevalent. Basedupon rock/outcrop type, three subclasses have been identifiedwithin type D:J D1: Carbonate rocks occur in small and scattered nodules

and blocks. The fauna is characterized by deep-watercorals, mollusks, crabs and worms;

J D2: Carbonate crusts or slabs of variable thickness (10–100 cm), often extensive, forming carbonate pavement onwhich a great variety of organisms thrives; and

J D3: Carbonate bridges comprised of massive carbonaterocks shaped as bridges, reaching several meters in length;further characterized by hydrate exposed at the surface.

�

Type E: Smooth mud: seabed area colonized by bacterial mats(Lloyd et al., 2010) that cover the sediment in irregular patchesthat often cover more than 60% of the field of view.

The distribution of seabed types is shown in Fig. 6 (note thatgeological samples have been used to ground truth sediments and

biological taxa).

4.3. Geo-acoustic facies map

Acoustic backscatter values have been used to identify sevendifferent geo-acoustic classes that have been plotted on an acousticseafloor map. Classes have been generated using the two endmember values of the backscatter intensity: class 1 is the lowestvalue and is associated with Gulf hemipelagic mud (�17 dB) andclass 7 is the highest intensity at the mound seafloor (�2 dB) andassociated with hard-grounds. Step-down intervals of 2.5 dB, assuggested by Johnson et al. (2003), have been used to define theintermediate classes. Once geo-acoustic classes were defined, theirspatial distribution at the seafloor was mapped (Fig. 7).

4.4. Lithology and relationship to seafloor morphology

Forty-one gravity cores from the Woolsey Mound (Table 1;Fig. 3a) were divided into four groupings based on their lithofacies:

(1)

Cores diagenetically altered throughout and characterized bycarbonate nodules and gray sediment bleached of the olivehue seen in unaltered sections throughout the region;

Fig. 7. The Woolsey Mound calibrated acoustic backscatter map. Intensity of acoustic backscatter anomalies was divided into seven intensity classes. Classes have been

mapped over the three complexes and surrounding areas.

Table 1Lithofacies classification of the 41 gravity cores from the Woolsey Mound.

MC118 core ID (A) Diagenesis throughout core

(nodules and bleached, gray sediment)

(B) Normal sediment

above diagenesis

(C) Hard-ground (D) Unaltered sediment

0505-2 X

0505-3 X

0505-4 X

0505-5 X

0505-6 X

0505-7 X

0505-8 X

0505-9 X

0505-10 X

0505-11 X

0505-12 X

1005-21 X

1005-22 X

1005-23 X

1005-24 X

1005-25 X X

1005-26 X

1005-27 X

1005-28 X

1005-29 X

1005-30 X

1005-31 X

1005-32 X

1005-33 X

1005-34 X

1005-35 X

1005-36 X

1005-37 X X

1005-38 X

1005-39 X

0408-01 X

0408-03 X

0408-06 X

0408-10 X

0408-11 X

0408-12 X

0408-13 X

0408-14 X

0408-15 X

0408-16 X

0408-17 X

L. Macelloni et al. / Deep-Sea Research I 74 (2013) 25–38 31

Fig. 8. A conceptual model of the spatial distribution of bio-geological process throughout the Woolsey Mound. Major morphological units, sediment types and biological

habitat are mapped, based upon video data, classes of acoustic backscatter anomalies and, where available, sediment core data. The model provides regionalization of the

bio-geological processes occurring at the mound and highlights significant differences within the three complexes.

L. Macelloni et al. / Deep-Sea Research I 74 (2013) 25–3832

(2)

Cores diagenetically altered at depth and overlain by unal-tered sediments above;

(3)

Cores with no recovery or 5 cm or less in length, suggestingimpact on carbonate hard-grounds; and

(4)

Cores containing a normal sequence of unaltered lithofaciesbut with shortened sections compared to the region outsidethe vent field.

Eight altered cores lie within the NW and SW complexes(1005-24, -25, -26, and -37 and 0505-2, -3, -7, and -8). Fouraltered cores lie on the scarp of the northern sector (0408-1, -3,and -6 and 0505-9) and one lies on the scarp surrounding the SEcrater (0408-12). Two lie in the Middle sector (0408-11 and 1005-31), and one lies in a pockmark west of the Northwest crater(0408-10). Four of the five cores with altered sediment overlainby normal sediment (1005-21, -22, -28, and -29) lie in anortheast-trending swath between the NW and SE crater com-plexes extending into the Middle Sector, and one core (1005-30)lies in the SE crater. Four coring attempts in the SW crater eitherrecovered no material or 5 cm or less of sediment (1005-25 and -37, and 0408-16 and -17).

Eighteen cores from the vent field recovered unaltered sedi-ment with a succession of lithofacies comparable to that recov-ered from the surrounding continental slope, within 3 km of theWoolsey Mound (0505-4, -5, -6, -10, -11, and -12; 1005-23, -27, -32, -33, -34, -35, -36, -38, and -39; 0408-13, -14 and -15).However, the depth to lithostratigraphic datum levels, like thereddish layer dated at 14.2 to 14.7 kiloannum (ka), is significantlyless in the mound than outside the mound. The median depth to

the reddish layer in 11 cores from the Woolsey Mound is 100 cmwhere as the median depth to the reddish layer in fourteen coresoutside the Woolsey Mound is 190 cm (Brunner, 2007; Ingramet al., 2010).

4.5. Integrating morpho-acoustic imagery and geological data.

The multibeam-derived seafloor backscatter map (Fig. 7)shows a seafloor that is extremely heterogeneous in acousticresponse; this heterogeneity is related to the high variability ofseabed types recognized on the video data and reported in Fig. 6.Seafloor backscatter covers the entire mound; however, video anddigital images have been collected only along profiles in targetedareas. Integration of the two datasets allows:

1)

Ground truthing of the backscatter values and 2) correlation of backscatter values with specific seafloor

sediment types.

The interpretation of seafloor type was extended throughoutthe mound based on acoustic class (Fig. 8). The distribution ofbackscatter values (Fig. 7) corresponds very well with the sedi-ment types classified by analysis of video and images (Fig. 5),when the subtypes D1, D2 and D3 are grouped into the moregeneral type D for carbonate outcrops. The seabed patterns are asfollows:

�

(A) hemipelagic mud with infaunal, biogenic structures; � (B) mud with some epibenthic organisms;

L. Macelloni et al. / Deep-Sea Research I 74 (2013) 25–38 33

�

(C) bioclastic deposits (bioclastic fraction more than 50%); � (D) carbonate outcrops; and � (E) mud covered by patches of bacterial mat.

The integrated imagery and acoustic seafloor classification areused to derive the distribution of bio-geological processes at themound surface (Fig. 8).

5. Results: The spatial distribution of bio-geological andgeochemical processes at the Woolsey Mound

5.1. SE crater complex

The SE crater complex presents a generally smooth seafloormorphology in which fine-grained sediments dominate as baremud with bioturbational features (Fig. 5, type A). The complex isfurther characterized by localized fields of chemoautotrophicclam valves and shell hash (Fig. 5, type C)—no live clams wereobserved. Gas bubble streams have not been observed at thiscomplex. Four gravity cores were collected from the complexperiphery (1005-30, 36, 0408-12, -13; Fig. 3a). One core from thesoutheast rim of the structure (0408-12) contains sedimentapparently altered by methanogenesis throughout its length,one core (1005-30) contains altered sediments in the subsurfacecovered by unaltered, hemipelagic sediments (Holocene age?),and the other two cores, collected on the north rim of thestructure (0408-13) and outside the structure on its southwestflank (1005-36), contain an unaltered succession of facies,although the units are thinner than those outside the mound onthe adjacent continental slope. Lack of evidence for recent ventingin the SE crater complex suggests that the conduits that oncesupplied sufficient methane to support an abundant population ofclams have become blocked. Geochemical evidence suggests thisarea is older and experienced venting activity in the past. Forexample, dissolved methane in pore water extracted from sedi-ments recovered from the SE complex has been more oxidizedthan that at either of the other two crater complexes as shownwith stable carbon isotope ratios (Lapham et al., 2008).

5.2. NW crater complex and the northern sector

A combination of carbonate nodules, hydrates chunks andmud irregularly pave the crater of the NW complex (Fig. 5, typeD2). The seafloor is characterized by hemipelagic mud cover andby patches of white bacterial mats, which Lloyd et al. (2010) haveclassified as Beggiatoa spp. (Fig. 5, type E). Eight gravity coreswithin the NW crater and along the scarp of the Northern sector(0408-1, -3, and -6; 0505-2, -3, -7, -8 and -9; Fig. 3a) recovereddiagenetically altered sediment with subsurface carbonatenodules. A small bubble stream was observed with a drift cameraat a spot on the scarp that bounds the Northern Sector, and mass-spectrometer, in-situ measurements nearby recorded high con-centrations of methane in the water column (Camilli et al., 2009).

5.3. SW crater complex

The SW crater complex comprises multiple nested vents(Fig. 4a and b). It is divided into western and eastern parts, eachcomprised of several intersecting, smaller craters. The two partsare separated by a ridge of fine-grained material that overlies awell-defined horizon of authigenic carbonate that consists of sub-horizontal tabular blocks about 1-m thick and several metersacross (Fig. 5, type D2, D3). In the western portion of the SWcomplex, carbonate blocks pave the floors of craters, and in someplaces, streams of gas bubbles rise from fractures alongside the

carbonate rocks (Fig. 5, type D3). A set of intersecting cratersapproximately 60 m in diameter and 6 m deep characterizes theeastern part of this complex. Craters in this area are about 5 mdeeper than those in the western complex and appear to experi-ence more active venting of hydrocarbon fluids, including oil, invarious stages of microbial degradation. An exposure of authi-genic carbonate rock in various stages of submarine erosionoutcrops on the northern and eastern flanks of the intersectingeastern craters at the same elevation as the paving blocks in thewestern part. It is overlain by about 2 m of sediment, the uppermeter of which appears to be a berm of expelled material. Lookingdown into the eastern crater from a point on that berm, a largecarbonate/hydrate outcrop (6 m long by 2 m wide by 1.5 m thick)protrudes from the crater’s eastern flank (Fig. 5, D3) forming aprominent seafloor feature named the ‘‘Sleeping Dragon.‘‘ Fourattempts were made to gravity-core the craters and either failedto recover sediment or recovered short section o5 cm (respec-tively, 0408-16 and -17 and 1005-25 and -37; Fig. 3a), presum-ably because they struck carbonate hard-grounds. Three gravitycores recovered diagenetically altered sediment supporting thesurface observations of methanogenic diagenesis (1005-24, -26,37; Fig. 3a).

The SW crater complex is the most diverse of the mound’scraters in terms of seabed types (i.e., visible seeping, abundance ofhydrates exposures, extent of authigenic carbonate formation).Furthermore, the SW crater complex hosts the most diversefaunal assemblage. Photos and videos document taxa that includeclams, ice worms, tube worms, crabs and spectacular, deep-watercorals. The benthic fauna is not randomly distributed, but appearsto follow a predictable pattern within the complex. If sediment/benthic fauna typology/zonation are plotted versus distance fromthe seep (Fig. 9), an ordered zonation emerges between sedimenttype/faunal position and distance from the gas emission point:With increasing distance from the emission point, benthic fauna,bacterial consortia, and authigenic carbonate sediment decreaseprogressively in number and type, and over a distance of about250 m, the seabed returns to an unaltered state. Geochemical datafor the SW complex (Fig. 3b) show very high microbial activity atthe crater, and decreasing activity moving away from it.

6. Discussion

The integration of high resolution bathymetry, acoustic back-scatter, video images, lithological studies and feedback fromgeochemical pore fluid analysis has allowed the creation of aseafloor model that provides insight into the morphology and bio-geological processes at MC118, the Woolsey Mound. Moundmorphology has been shown to consist of three main complexesthat overly active crestal faults formed by salt tectonics(Macelloni, et al., 2012). All three crater complexes have formedwhere deep faults intersect the seafloor, but each complexcurrently presents a different complement of sediment types,fluid emissions, and benthic communities. Roberts and Carney(1997), Roberts (2001), Roberts et al. (2006), and Lapham et al.(2008) have shown that these seafloor attributes and theirassociation/distribution can be used as a proxy to evaluaterelative upward fluid flux.

6.1. Multiple fluid flux regimes coexisting at the Woolsey Mound

The SE crater complex exhibits lower relief than the other twocomplexes, no observed authigenic carbonate – except subsurfacecarbonate nodules – large fields of shells that indicate that thearea was formerly inhabited by a thriving community of macro-organisms, and hemipelagic mud. The proportion of seabed

Fig. 9. Fluid flux regime at cold seeps can be derived, qualitatively, by observing localized seafloor response in terms of sediment type, chemosynthetic community

complexity, and presence/absence of diagnostic morphologies. The cartoon, adapted from Boetius et al. (2000), and Knittel et al. (2003), illustrates a typical cold seep with

a moderate and persistent fluid flux regime and the zonation (according to distance from the seeps) of the different bio-geological entities. Note that the origin of the

distance axis lies on the left at the gas emission point. The distribution of bio-geological processes at the SW crater complex is consistent with the proposed model, i.e.,

numbers of specimens and diversity of taxa diminish with increasing distance from the seep. This relationship supports the existence of a moderate and persistent fluid

flux regime at the SW complex. The NW and SE complexes diverge from the model, leading to the conclusion that fluid flux can be highly variable even within a

single mound.

L. Macelloni et al. / Deep-Sea Research I 74 (2013) 25–3834

covered by shell fragments decreases almost linearly from thecenter to the periphery of the complex. Studies have demon-strated that a similarly large chemoautotrophic clam populationcan exist today only in association with abundant discharge ofhydrocarbon fluids (Roberts et al., 2006); therefore it is reason-able to conclude that venting conditions have changed signifi-cantly since the shelled organisms thrived at the SE complex. Itappears that a formerly open venting system has been sealed sothat hydrocarbon fluids are no longer delivered to the seafloor tosustain the fauna that once lived there in abundance.

The western area of the mound, only 500 m distant, shows atotally different scenario; the SW and NW complexes exhibitgreater bathymetric relief, are morphologically younger and hosta greater variety of sediments and fauna. Solid hydrates arepresent at the seafloor and bubble streams have been observed.

The SW complex is populated by a diverse assemblage ofbenthic organisms, presents the whole suite of sediment types,including thick carbonate rocks, outcropping hydrates, and gasbubble streams. Most importantly, the seabed types, and theirspatial distribution (Fig. 9), portray a predictable relationship orgradient between benthic species/sediment type and distancefrom the active vent. Boetius et al. (2000) and Knittel et al. (2003)

have presented studies that illustrate bacterial consortia, benthicfauna and authigenic sedimentary products that can be present atand near a cold seep. These studies show that a moderate (tohigh) but steady gas flux at the seafloor favors conditions todevelop complex, chemosynthetic communities, thick authigeniccarbonate rocks and the concomitant occurrence of outcroppinghydrate and gas bubble streams. Similar conclusions have beenderived in other parts of the world where seafloor features wereconnected to fluid flux (i.e., Olu et al., 1996; Sibuet and Olu, 1998;Sibuet and Olu-Le Roy, 2002). These attributes will also follow aroughly regular distribution as a function of the distance from theseep: diversity and numbers of individuals decrease with increas-ing distance from the seep. The SW complex closely approximatesthe Boetius–Knitell mature, steady flux regime seep. In addition,considering the thickness and abundance of the carbonate rocksand the presence of a well-developed, deep coral community, thisflux regime must have persisted for hundreds of years (Lutkenet al., 2011b).

The NW complex presents evidence of ongoing fluid flux –authigenic carbonates, bacterial mats, and a chemosyntheticcommunity – but the spatial zonation does not reflect thepredictable pattern and biological complexity of the SW complex.

L. Macelloni et al. / Deep-Sea Research I 74 (2013) 25–38 35

Mats are abundant and are hypothesized (Hovland, 2002) andshown (Joye et al., 2004; Lloyd et al., 2010) to be indicators ofupward methane flux. Previous studies have also seen elevatedmethane concentrations in bottom water (i.e., Ondreas et al., 2005)as was measured at this location by Camilli et al. (2009). Likewise,their presence on the seafloor might also indicate where theformation/precipitation of authigenic carbonates is occurring(Sassen et al., 1998). The NW complex also hosts two large seafloorpockmarks, manifestations of seafloor instability. The evidenceexamined leads to the hypothesis that the NW complex has veryhigh flux, and its venting is more episodic than that of the SWcomplex, with possible violent emissions of gas. These violentepisodes, responsible for the formation of the large pockmarks,can maintain the well-established bacterial mat, but will notsupport the formation of a complex chemosynthetic communitywhose establishment and survival require more steady conditions.

Geochemical data (Lapham, et al., 2008) further confirm thedifference in fluid flux among the three complexes. High micro-bial activity in the very shallow sediment (up to 5 m b.s.f.), thatindicates intense anaerobic oxidation of methane, is found in theSW and NW complexes, while the SE complex has moderate tolow activity (Fig. 3b).

Regarding the lithological studies from the gravity cores, it isinteresting to note that apparently methanogenetically alteredsediments occur in several of the cores outside the crater com-plexes. Diagenetically altered sediment overlain by Holoceneaged, unaltered sediments occur in five cores that lie in anortheast-trending swath between the Northwest and Southwestcraters extending into the Middle sector (1005-21, -22, and -28, -29, -30; Fig. 3a). Either methanogenesis is presently insufficient toaffect the surface sediment or the diagenesis extended to the

Fig. 10. Conceptual model of the Woolsey Mound HSZ and fluid flux regime. Geophysica

the observations at the seafloor to understand the mechanism of hydrocarbon gas migra

different fluid flux regimes coexist. In fact, each of the crater complexes is associated

hydrocarbon gases transit. In the shallow subsurface, the gas appears to be trapped in

shallower stratigraphic position, seems to mark the base of a segmented stability zone a

for each complex. Three types have been defined: hydrates dominated, gas dominat

Differences in the distribution of bio-geological processes reflect how the seafloor respo

times (TWTT) is represented accurately to about 1 s, the vertical scale is afterwards a

migration in a single figure). (For interpretation of the references to color in this figur

surface in the early Holocene but has ceased since, allowingunaltered sediment to accumulate.

6.2. Hydrocarbon gas migration and dynamics of the hydrates

stability zone

We suggest that the diverse surface morphologies and relatedbio-geological and geochemical processes are driven by seafloorresponse to venting and the seepage of hydrocarbons. Macelloniet al., (2012) have illustrated the complexity of the hydrocarbonplumbing system and the rapid transience of the hydrate stabilityzone (HSZ). Subsurface data show that the MC118 HSZ issegmented along the faults, making it laterally variable. Suchvariability produces different mechanisms of migration of thethermogenic gas through the HSZ and the venting modalities atthe seafloor. Combining the evidence from subsurface studies andour observations at the seafloor, we attempt to outline WoolseyMound dynamics in the conceptual model, Fig. 10, which depictsa multi-attribute (geology, biology, tectonics) transect passingacross the three complexes.

Hydrocarbon fluids (red arrows) migrate from the deep oilreservoir to the shallow subsurface via salt tectonic faults andfractures. They then accumulate in stratigraphic intervals of highporosity and permeability sediments. In the shallow subsurface,within the HSZ, the thermogenic gases are alternately bound inthe sediment or are being expelled into the water-column. Thisprocess is mainly controlled by the dynamic of two particular gas-bearing reservoirs identified by Macelloni et al. (2012), and by thehydraulic connection between them. The authors named thesetwo gas reservoirs BS-1 and BS-2 (see Fig. 10 as reference) andsuggest that the BS-2, which occupies the shallower stratigraphic

l data of the mound’s subsurface (Macelloni et al., 2012) have been combined with

tion/venting. The Woolsey Mound appears to be a complex system in which three

with a specific salt tectonic fault or faults, along which deep-sourced, dissolved

two particular gas horizons: BS-1 and BS-2. The BS-2 horizon, which occupies the

nd to be directly related to the seafloor venting processes, which occur differently

ed, and steady systems acting, respectively at the SE, NW, and SW complexes.

nds to these three different fluid flux regimes. (Note that although two-way travel

djusted to allow the portrayal of the deeper salt body and associated faults and

e legend, the reader is referred to the web version of this article.)

L. Macelloni et al. / Deep-Sea Research I 74 (2013) 25–3836

position, seems to mark the base of a very particular ‘‘segmented’’hydrates stability zone, directly related to the venting processesat the seafloor. Seafloor observations suggest that gas is migrat-ing/venting as three distinct fluid flux regimes: hydrates domi-nated, gas dominated and steady system.

Hydrate-dominated flux occurs when conditions favor theformation of hydrate. Solid hydrate then acts as a temporary sealof the plumbing system, stopping completely the advectingthermogenic gas and sealing the conduits. In this case, thermo-genic gas is trapped in the stability field as solid hydrate, orbeneath the stability field as free gas. This equilibrium persistsuntil the thermogenic gas pressure exceeds the lithostatic pres-sure, or stress along the fault, depressurizing the system andallowing gas to escape at the seafloor. We recognize this mechan-ism acting along the Yellow fault at the SE complex. The seafloorresponds to this flux regime in different ways, depending onwhether it is in a quiescent period (hydrates sealing the conduits)or active period (gas venting at the sea floor). When conduits areopen, chemosynthetic organisms begin to populate the seafloor,but once the conduits are sealed again the communities die,leaving a dead or relict crater.

The gas-dominated flux is present in the NW Complex, wherethe hydrocarbon reservoirs BS-1 and BS-2 have the most efficienthydraulic connection due to the intersection of the red and bluefaults and where the BS-2 is at its thickest. Here, large volumes ofgas can migrate through the stability zone and escape into thewater-column at the seafloor. The process can be violent, withvast amounts of gas passing, largely unaltered, through thestability zone and creating pockmarks at the seafloor. Alterna-tively, gas may migrate upward, at constant but still high flux,resulting in the formation of hydrates only in or near faultswhere fine-grained sediments are fractured. In this case, largecolonies of bacterial mats form, as well as specific chemosyntheticbenthic fauna that tolerate high concentrations of hydrocarbonsand the episodic expulsions. Here, small nodules of authigeniccarbonates form, by-products of the metabolism of biologicalcommunities.

The flux acting at the SW complex, has been classified as asteady system because it is similar to the steady, focused flowhypothesized by Liu and Flaming (2006) for a seep at HydratesRidge, offshore of Oregon. Liu and Flaming (2006) have evidencesuggesting that it is necessary to have focused, moderate butsteady flow, in order for active venting and hydrate formation tooccur simultaneously at the seafloor. The existence of this situa-tion implies a permanent – or at least long-term – open conduitbetween the gas reservoir and the seafloor (marked by thecharacteristic wipe-out zone in our seismic data) to support theformation of these features. They also suggest that this processoccurs because the conduits where the gas transits support theremarkable scenario of a three-phase equilibrium among free gas,gas hydrates, and gas dissolved in pore water. Gas, transported byits buoyancy, can partially bypass the HSZ. Areas where this fluxregime acts present large hydrate outcrops at and near theseafloor, often imbedded in thick authigenic carbonates deposits.If steady conditions last for a long time, the site will establish awell-developed and diverse, cold seep, benthic fauna (Fig. 9).Wedo not know if the three regimes portray different stages of seeplife: namely generation, maturation, and sealing. Several authors,such as Hovland et al. (2002) and Naudts et al. (2008), haveinvoked formation of authigenic carbonate as the primary causeof seep self-sealing. Our data, however, show the presence ofbubble streams and venting activity in the same craters wherecarbonate crusts are generally present. We do see relocation ofthe bubble streams around the cemented, impermeable areacovered by the authigenic carbonate, but we do not observe an‘‘extinction’’ like that of the SE complex. Surficial authigenic

carbonate crusts and slabs can cause the temporary, small-scalediversion of gas migration while our integrated data, whichembrace both surface and subsurface evidence of deeper, largerscale hydrates and related seeps, provide a more comprehensiveview of a self-sealing mechanism that includes the formation anddissociation of hydrates along the master faults.

7. Conclusions

The Woolsey Mound at MC118 has been investigated exten-sively by geological, geophysical, and geochemical means. Usinghigh-resolution bathymetry, seafloor acoustic backscatter, seabedvideo surveys, and geological/geochemical data, a bio-geological,conceptual model of the mound has been generated. The modelillustrates the morphological complexity of the mound and the highvariability of the sediments and benthic fauna associated with it.

Three main crater complexes have been recognized, each oneowing its genesis to thermogenic, hydrocarbon migration alongsalt-tectonic-induced, normal faulting. Despite their commonorigin, the three complexes show significant differences in thespatial distribution of sediments and habitats. Evidence of directrelationships between hydrocarbon fluid flux and bio-geologicalprocesses for each complex and relevant results have beenderived by analyzing the spatial distribution of the associatedsediments and biogeochemical communities versus seabedacoustic reflectivity. Data suggest that the three complexes arein different stages of activity/fluid flux and that their correspond-ing seafloor expressions provide compelling evidence that can beused to define those stages:

1)

A relict complex with low relief and no observable ventingactivity is associated with barren sediments devoid of livingmacrofauna and only the remains of once-living vent organisms;

2)

A complex with moderate-to-low relief, a moderate-to-lowlevel of venting and an abundant epifauna is transitioningfrom inactivity to activity;

3)

A mature, venting complex that exhibits moderate-to-highrelief, is associated with a thriving chemosynthetic community.

Geochemical studies corroborate these generalizations andindicate a strong connection between sedimentary microbialactivity and fluid flux regimes. In fact, activity waxes and waneswith changes in upward fluid flux that can be followed bymonitoring biogeochemical cycling of sulphur and carbon. Micro-bial ‘‘hot spots’’ are thus a proxy for the presence of upwardlyadvecting fluids. The spatial distribution of seafloor bio-geologicalprocesses seems to be a similar proxy at a larger scale. Thecombination of the mound surface observations with results fromsubsurface study, show that different fluid flux regimes are adirect consequence of complexity in the shallow plumbingsystem, lateral variability within the HSZ, and the dynamics ofspecific gas-bearing intervals that act as temporary reservoirs forthermogenic hydrocarbons. Three general mechanisms of ther-mogenic fluids migration have been recognized acting along themain faults. However, we still know little about why the threemechanisms coexist at the same site and at the same time.

This work confirms, unequivocally, that hydrate mounds areproducts of large-scale geological process such as salt tectonics,but also, spatially and temporally, of small-scale bio-geochemicalsystems. It also confirms that only multidisciplinary study,embracing geophysics geochemistry, biogeology, and oceanogra-phy will be able to resolve, fully, these exceptional marineenvironments.

L. Macelloni et al. / Deep-Sea Research I 74 (2013) 25–38 37

Acknowledgments

This paper is dedicated to the memory of Dr. J. Robert (Bob)Woolsey, Jr., long-time Director of The Mississippi Mineral ResourcesInstitute, and founding Director of its two Centers, the Center forMarine Resources and Environmental Technology and the SeabedTechnology Research Center (MMRI-CMRET-STRC) at The Universityof Mississippi and founder/Director of the GOMHRC. Bob was anoutstanding and innovative scientist, an incredible man, mentor, andcolleague, and an incomparable friend.

The authors are grateful to MMRI-CMRET-STRC ResearchSystems Specialists Brian Noakes, Andy Gossett and Matt Loweand Electronics Technician Larry Overstreet, to the LouisianaUniversities Marine Consortium (LUMCON) and the Captain andCrew of the R/V Pelican for at-sea support, to C&C Technologies(Lafayette, La.) for data acquisition, expertise and guidance, and toHarbor Branch Oceanographic Institution and the Captain andcrew of the R/V Seward Johnson/Johnson SeaLink for at-seasupport, and to Paul Mitchell for graphics support. The com-ments/suggestions of the three anonymous reviewers expandedthe scope and greatly enhanced the final version of this paper.

This project has been supported by Gulf of Mexico HydratesResearch Consortium through funds provided by DOI BOEM (formerMMS), DOE National Energy Technology Laboratory, and DOC-NOAA’s National Institute for Undersea Science and Technology.

References

Aharon, P., 2006. Entrainment of meltwaters in hyperpycnal flows duringdeglaciation superfloods in the Gulf of Mexico. Earth Planet. Sci. Lett. 241,260–270.

Beaudoin, J.D., J.E. Hughes Clarke, E.J. Van Den Ameele, J.V. Gardner (2002),Geometric and radiometric correction of multibeam backscatter derived fromReson 8101 systems. In: Proceedings from the Canadian Hydrographic Con-ference, Toronto, Canada, May 28–31, pp. 24.

Boetius, A., Ravenschlag, K., Schubert, C., Rickert, D., Widdel, F., Gieseke, A., Amann,R., Jørgensen, B.B., Witte, U., Pfannkuche, O., 2000. A marine microbialconsortium apparently mediating anaerobic oxidation of methane. Nature407, 623–626.

Bosman, A., (2004), High-Resolution Side Scan Sonar and Multibeam Data Proces-sing and Merging to Study Submarine Instability Phenomena on VolcanicIsland (Pontine, Campanian and Aeolian Archipelagos). Ph.D. Dissertation. In‘‘Geoinformazione’’ XVI Cycle. Rome University ‘‘Tor Vergata’’.

Bosman, A., Chiocci, F.L., Romagnoli, C., 2009. Morpho-structural setting ofStromboli volcano revealed by high-resolution bathymetry and backscatterdata of its submarine portions. Bull. Volcanol. 71, 1007–1019.

Bouma, A.H., Coleman, J.M., Meyer, A.W. (Eds.), 1986. U.S. Government PrintingOffice, Washington D.C.

Brizzolara, J.L., Brunner, C.A., 2012. Lithostratigraphy of jumbo piston cores fromthe Woolsey Mound, the northern Gulf of Mexico. J. Mississippi Acad. Sci. 57(1), 121.

Brunner, C.A. Brizzolara, J.L., 2011, Lithostratigraphy and Physical Properties ofJumbo Piston Cores from the Woolsey Mound. Oral Presentation at the MMRIMethane Consortium Annual Meeting, Jackson, MS., 8 November 2011.

Brunner, C.A., (2007), Stratigraphy and paleoenvironment of shallow sedimentsfrom MC118. In: Proceedings of Gulf of Mexico Hydrates Research ConsortiumAnnual Meeting, October 10, 2007, Oxford, MS.

Brunner, C.A., W. Ingram (2008), Stratigraphy and paleoenvironment of shallowsediments from MC118. In: Proceedings of Gulf of Mexico Hydrates ResearchConsortium Annual Meeting, October 14, 2008, Oxford, MS.

Camilli, R., L. Macelloni, V. Asper, M. Woolsey, J. William, A. Dierks, K. Sleeper andC. Lutken, 2009, Discovery and Characterization of Cold Seep Vents Using aMass Spectrometer Operating aboard an Autonomous Underwater Vehicle,Transaction of American Geophysical Union, 2009 Fall Meeting, San Francisco,14–18 December, 2009.

Ferrini, V.L., Flood, R.D., 2006. The effects of fine-scale surface roughness and grainsize on 300 kHz multibeam backscatter intensity in sandy marine sedimentaryenvironments. Mar. Geol. 228, 153–172.

Gay, A., Lopez, M., Ondreas, H., Charlou, J.L., Sermondadaz, G., Cochonat, P., 2006.Seafloor facies related to upward methane flux within a giant pockmark of theLower Congo Basin. Mar. Geol. 226, 81–95.

Hovland, M., 2002. On the self-sealing nature of marine seeps. Cont. Shelf Res. 22(2002), 2387–2394.

Hovland, M., Gardner, J.V., Judd, A.G., 2002. The significance of pockmarks tounderstanding fluid flow processes and geohazards. Geofluids 2 (2), 127–136.

Ingram, W.C., Meyers, S.R., Brunner, C.A., Martens, C.S., 2010. Evaluation of latePleistocene–Holocene sedimentation surrounding an active seafloor gas

hydrate and cold seep field on the northern Gulf of Mexico slope. Mar. Geol.278, 43–53.

Johnson, J.E., Goldfinger, C., Suess, E., 2003. Geophysical constraints on the surfacedistribution of authigenic carbonates across the hydrate ridge region, Cascadiamargin. Mar. Geol. 202, 79–120.

Joye, S.B., Boetius, A., Orcutt, B.N., Montoya, J.P., Schulz, H.N., Erickson, M.J.,Lugo, S.K., 2004. The anaerobic oxidation of methane and sulfatereduction in sediments from Gulf of Mexico cold seeps. Chem. Geol. 205,219–238.

Kennett, J.P., Huddlestun, P., 1972. Late Pleistocene paleoclimatology, foraminif-eral biostratigraphy and tephrochronology, western Gulf of Mexico. Quat. Res.2, 38–69.

Knittel, K., Boetius, A., Lemke, A., Eilers, H., Lochte, K., Pfannkuche, O., Linke, P.,Amann, R., 2003. Activity, distribution, and diversity of sulfate reducersand other bacteria above gas hydrate (Cascadia margin, OR.). GeomJ 20,269–294.

Lapham, L.L., Chanton, J.P., Martens, C.S., Woolsey, R.J., 2008. Microbial activity insurficial sediments overlying acoustic wipe-out zones at a Gulf of Mexico coldseep. Geochem. Geophys. Geosyst. 9, Q06001, http://dx.doi.org/10.1029/2008GC001944.

Liu, X., Flaming, P.B., 2006. Passing gas through the hydrate stability zoneat southern hydrate ridge, offshore Oregon. Earth Planet. Sci. Lett. 241,211–226.

Lloyd, K.G., Albert, D.B., Biddle, J.F., Chanton, J.P., Teske, A., 2010. Spatial structureand activity of sedimentary microbial communities underlying a Beggiatoaspp. Mat in a Gulf of Mexico hydrocarbon seep. PLoS One 5 (1), e8738.

Lutken, C., T.M. McGee, A. Lowrie, C. Brunner, R. Rogers, L. Macelloni, A. Bosman, K.Sleeper, J. Dearman, J. Robert Woolsey, L. Lynch (2006), Comparison of TwoGas Hydrates Sites for Sea Floor Monitoring. Transactions, Gulf Coast Associa-tion of Geological Societies, 56th Annual Meeting, Lafayette, LA.

Lutken, B., L. Macelloni, K. Sleeper, M. D’Emidio, T.M. McGee, A. Simonetti, J.H.Knapp, C.C. Knapp, S. Caruso, J.P. Chanton, L.L. Lapham, M. Lodi, M. Ingrassia, P.Higley, C.A. Brunner, R. Camilli, B.M. Battista, T. Short, R. Bell, P. Fietzek(2011a), New discoveries at Woolsey Mound, MC118, northern Gulf of Mexico.In: Proceedings of the Seventh International Conference on Gas Hydrates(ICGH 2011), Edinburgh, Scotland, United Kingdom, July 17–21, 2011.

Lutken, B., Antonello Simonetti, Michela Ingrassia, Leonardo Macelloni, H. Knapp,Charles Fisher, Simona Caruso, M. D’Emidio, 2011b. Biogeophysical classifica-tion of seafloor seeps at a carbonate–hydrate mound, northern Gulf of Mexico.In: AAPG International Conference and Exhibition, Milan, Italy, October 23–27,2011.

Macelloni L., A. Simonetti, J.H. Knapp, C.C. Knapp, C.B. Lutken, L.L. Lapham (2012),Multiple resolution seismic imaging of a shallow hydrocarbon plumbingsystem, Woolsey Mound, northern Gulf of Mexico. Mar. Petrol. Geol. 38 (1),128–142.

Marchitto, T.M., Wei, K.-Y., 1995. History of the Laurentide meltwater flow to theGulf of Mexico during the last deglaciation, as revealed by reworked calcar-eous nannofossils. Geology 23, 779–782.

McGee, T.M., 2006. A seafloor observatory to monitor gas hydrates in the Gulf ofMexico. Leading Edge 25 (5), 644–647.

Montero-Serrano, J.,.C., Bout-Roumazeilles, V., Tribovillard, N., Sionneau, T.,Riboulleau, A., Bory, A., Flower, B., 2009. Sedimentary evidence of deglacialmegafloods in the northern Gulf of Mexico (Pigmy Basin). Quat. Sci. Rev. 28,3333–3347.

Naudts, L., Greinert, J., Artemov Y., Beaubien, S.E., Borowski C., De Batist M. (2008)Anomalous sea-floor backscatter patterns in methane venting areas, Dneprpaleo-delta, NW Black Sea. Mar. Geol. 25, 253–267.

Olu, K., Sibuet, M., Harmegnies, F., Foucher, J.-P., Fiala-Medioni, A., 1996. Spatialdistribution of diverse cold seep communities living on various diapiricstructures of the southern Barbados prism. Prog. Oceanogr. 38, 347–376.

Ondreas, H., Olu, K., Fouquet, Y., Charlou, J.L., Gay, A., Dennielou, B., Donval, J.P.,Fifis, A., Nadalig, T., Cochonat, P., Cauquil, E., Bourillet, J.-F., Le Moigne, M.,Sibuet, M., 2005. Integrated ‘‘in situ’’ study of a deep giant pockmark on theCongo-Angola margin. Geo-Mar. Lett. 25, 281–292.

Roberts, H.H., Carney, R.S., 1997. Evidence of episodic fluid, gas, and sedimentventing on the northern Gulf of Mexico continental slope. Econ. Geol. Bull. Soc.92, 863–879.

Roberts, H.H., 2001. Fluid and gas expulsion on the Northern Gulf of Mexicocontinental slope: mud-prone to mineral-prone responses. In: Paull, C.K.,Dillon, W.P. (Eds.), Natural Gas Hydrates: Occurrence, Distribution, andDetection. Geophysical Monograph Series, 124. AGU, Washington, D. C.,pp. 315, http://dx.doi.org/10.1029/GM124.

Roberts, H.H., Hardage, B.A., Shedd, W.W., Hunt, J.L., 2006. Seafloor reflectivity—animportant seismic property for interpreting fluids/gas expulsion geology andthe presence of gas hydrates. Leading Edge 25, 620–628.

Sassen, R., Roberts, H.H., Aharon, P., Larkin, J., Chinn, E.W., Carney, R., 1993.Chemosynthetic bacterial mats at cold hydrocarbon seeps, Gulf of Mexicocontinental slope. Org. Geochem. 20, 77–89.

Sassen, R., MacDonald, I.R., Guinasso Jr., N.L., Joye, S.A., Requejo, A.G., Sweet, S.T.,Herrera, -J., DeFreitas, D.A., Schink, D., 1998. Bacterial methane oxidation insea-floor gas hydrate: significance to life in extreme environments. Geology26, 851–854.

Sassen, R., H.H. Roberts (2004), Site Selection and Characterization of Vent Gas,Gas Hydrate, and Associated Sediments. DOE Technical Report, 28-227.

Sassen, R., Roberts, H.H., Carney, R., Milkov, A.V., DeFreitas, D.A., Lanoil, B., Zhang,C., 2004. Free hydrocarbon gas, gas hydrate, and authigenic minerals in

L. Macelloni et al. / Deep-Sea Research I 74 (2013) 25–3838

chemosynthetic communities of the northern Gulf of Mexico continentalslope: relation to microbial processes. Chem. Geol. 205, 195–217.

Sassen, R., H.H. Roberts, W. Jung, C.B. Lutken, D.A. DeFreitas, S.T. Sweet, N.L.GuinassoJr. (2006), The Mississippi Canyon 118 gas hydrate site: a complexnatural system. In: OTC Paper no. 18132; Offshore Technology Conference,Houston, TX.

Sibuet, M., Olu, K., 1998. Biogeography, biodiversity and fluid dependence of deep-sea cold-seep communities at active and passive margins. Deep-Sea Res. II 45(1998), 517–567.

Sibuet, M., Olu-Le Roy, K., 2002. Cold seep communities on continental margins:structure and quantitative distribution relative to geological and fluid ventingpatterns. In: Wefer, G., Billet, D., Hebbeln, D., Jorgensen, B.B., Schluter, M.,

Weering, T.C.E.V. (Eds.), Ocean Margin Settings, 2002. Springer-Verlag, Berlin-Heidelberg, pp. 235–251.

Sleeper, K.G., A. Lowrie, A. Bosman, L. Macelloni, C.T. Swann (2006), Bathymetric

mapping and high resolution seismic profiling by AUV in MC 118 (Gulf ofMexico). In: OTC Paper no. 18133, Offshore Technology Conference, Houston, TX.

Woolsey, J.R., P. Higley, L.L. Lapham, J.P. Chanton, C.B. Lutken, K.G. Sleeper, R. Culp,

S. Sharpe, D. Ross (2005), Operations Report of Cruise GOM2-05-MC118:Deployment of the Initial Components of the sea Floor Monitoring Station –The Pore-Fluid Array and the Geophysical Line Array –Via the Sea-Floor Probe

System and Collection of Core Samples, Mississippi Canyon 118. Gulf ofMexico Hydrates Research Consortium Cruise Report.