Dissolved and particulate organic carbon in hydrothermal plumes from the East Pacific Rise, 9°50′N

Deep-Sea Research I 58 (2011) 922–931

Contents lists available at ScienceDirect

Deep-Sea Research I journal homepage: www.elsevier.com/locate/dsri

Dissolved and particulate organic carbon in hydrothermal plumes from the East Pacific Rise, 91500 N Sarah A. Bennett a,n, Peter J. Statham a, Darryl R.H. Green b, Nadine Le Bris c, Jill M. McDermott d,1, Florencia Prado d, Olivier J. Rouxel e,f, Karen Von Damm d,2, Christopher R. German e a

School of Ocean and Earth Science, National Oceanography Centre, Southampton SO14 3ZH, UK National Environment Research Council, National Oceanography Centre, Southampton SO14 3ZH, UK c Universite´ Pierre et Marie Curie—Paris 6, CNRS UPMC FRE3350 LECOB, 66650 Banyuls-sur-mer, France d University of New Hampshire, Durham, NH 03824, USA e Woods Hole Oceanographic Institution, Woods Hole, MA 02543, USA f Universite´ Europe´enne de Bretagne, European Institute for Marine Studies IUEM, Technopˆ ole Brest-Iroise, 29280 Plouzane´, France b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 19 November 2010 Received in revised form 23 June 2011 Accepted 27 June 2011 Available online 3 July 2011

Chemoautotrophic production in seafloor hydrothermal systems has the potential to provide an important source of organic carbon that is exported to the surrounding deep-ocean. While hydrothermal plumes may export carbon, entrained from chimney walls and biologically rich diffuse flow areas, away from sites of venting they also have the potential to provide an environment for in-situ carbon fixation. In this study, we have followed the fate of dissolved and particulate organic carbon (DOC and POC) as it is dispersed through and settles beneath a hydrothermal plume system at 91500 N on the East Pacific Rise. Concentrations of both DOC and POC are elevated in buoyant plume samples that were collected directly above sites of active venting using both DSV Alvin and a CTD-rosette. Similar levels of POC enrichment are also observed in the dispersing non-buoyant plume,  500 m downstream from the vent-site. Further, sediment-trap samples collected beneath the same dispersing plume system, show evidence for a close coupling between organic carbon and Fe oxyhydroxide fluxes. We propose, therefore, a process that concentrates POC into hydrothermal plumes as they disperse through the deep-ocean. This is most probably the result of some combination of preferential adsorption of organic carbon onto Fe-oxyhydroxides and/or microbial activity that preferentially concentrates organic carbon in association with Fe-oxyhydroxides (e.g. through the microbial oxidation of Fe(II) and Fe sulfides). This potential for biological production and consumption within hydrothermal plumes highlights the importance of a multidisciplinary approach to understanding the role of the carbon cycle in deep-sea hydrothermal systems as well as the role that hydrothermal systems may play in regulating global deep-ocean carbon budgets. & 2011 Elsevier Ltd. All rights reserved.

Keywords: Hydrothermal Dissolved organic carbon Particulate organic carbon East Pacific Rise

1. Introduction Hydrothermal circulation is an important source and sink of elements to the ocean (Elderfield and Schultz, 1996; German and Von Damm, 2004). In particular, hydrothermally sourced iron (Fe), stabilised by organic complexes within dispersing plumes may be important to global ocean budgets (Bennett et al., 2008; Toner et al., 2009; Tagliabue et al., 2010; Wu et al., 2011). In a typical basalt hosted hydrothermal field, end-member fluids

n Corresponding author. Now at NASA Jet Propulsion Laboratory, Caltech, 4800 Oak Grove Drive, Pasadena, CA 91109, USA. Tel.: þ1 6266765372. E-mail addresses: [email protected]. [email protected] (S.A. Bennett). 1 Now at Woods Hole Oceanographic Institute, Woods Hole, MA 02543, USA. 2 Deceased.

0967-0637/$ - see front matter & 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.dsr.2011.06.010

are enriched in Fe(II) but depleted in organic carbon (Von Damm, 1995a; Lang et al., 2006). Therefore, any elevated concentrations of organic carbon present within a hydrothermal plume must be entrained from adjacent areas of diffuse flow and chimney walls (Cowen et al., 1986; Winn et al., 1986; Lang et al., 2006) and/or produced in-situ within the plumes themselves (Roth and Dymond, 1989; McCollom, 2000; Lam et al., 2004). Hydrothermal plumes are dynamic, 3-dimensional biologically active zones in the deep-sea and biological communities within the hydrothermal system rely on chemoautotrophic primary production fueled by hydrothermal fluids (McCollom, 2000). As hot, chemically reduced fluids mix with oxygenated seawater, complex redox disequilibria are established that provide chemical energy for microbial metabolism. Areas of diffuse flow and chimney walls are well known for their rich biomass and symbiotic bacteria, with elevated concentrations of dissolved and particulate organic carbon

S.A. Bennett et al. / Deep-Sea Research I 58 (2011) 922–931

in the surrounding waters (Rau and Hedges, 1979; Karl et al., 1980; Comita et al., 1984; Lang et al., 2006). Free living microbes within hydrothermal plumes are also an important, but often overlooked, primary producer (Cowen et al., 1999; Lam et al., 2008; Dick and Bradley, 2010; German et al., 2010; Sylvan et al., in review). Both symbiotic and non-symbiotic microbes play an important role in chemical cycling in these environments. Early plume studies detected elevated microbial biomass at plume depths both near- and far-field, entrained from bottom waters near the vents (Cowen et al., 1986; Winn et al., 1986; Straube et al., 1990). This was followed by the demonstration of chemoautotrophic carbon fixation and biological removal of organic matter in the hydrothermal plume (Roth and Dymond, 1989). Bacteria and viruses have been detected within hydrothermal plumes (Juniper et al., 1998; Ortmann and Suttle, 2005; German et al., 2010), along with zooplankton directly above the plume, apparently grazing off the hydrothermal constituents (Burd and Thomson, 1994; Vereshchaka and Vinogradov, 1999; Cowen et al., 2001). The microbial community within the hydrothermal plume has been inferred to include methane (Cowen et al., 2002), ammonia (Lam et al., 2004; Lam et al., 2008), hydrogen and sulfide oxidizers (Jannasch and Mottl, 1985; Sunamura et al., 2004), as well as microbes that utilize Mn and Fe (Cowen et al., 1998; Dick et al., 2009). Sulfide and hydrogen oxidation is considered to be important in the buoyant or early non-buoyant plume, whereas methanotrophy and ammonia oxidation are more likely to be important in non-buoyant plumes (Lam et al., 2008). The plume environment is biologically and chemically rich and is able to host both auto- and hetero-trophic communities, yet bulk dissolved and

Table 1 Details of cruise number and dates during which the samples for this study were collected. Cruise number

Dates

Sample collection

AT15-6 AT15-12 AT15-13

June 2006–July 2006 October 2006–November 2006 November 2006–December 2006 December 2006–January 2007

Sediment trap deployment Sediment trap recovery Vent fluid collection

AT15-14

Plume samples

923

particulate organic carbon (POC and DOC) analyses are lacking for this setting. Initial vent end-member carbon measurements made by Lang et al. (2006), suggested that the flux of carbon from hydrothermal vents to the open ocean is minor compared to most oceanic source and sink terms. However, the dynamic plume environment has been suggested to be a source of ‘new’ organic carbon to the deep-ocean (Karl et al., 1984). This study focuses on bulk organic carbon variations, both particulate and dissolved, directly above a hydrothermal source, its evolution through a dispersing plume and its presence within sinking particles settling beneath the buoyant plume.

2. Sample collection, processing and analysis All samples for this study were collected during 2006 from the well-characterized basalt hosted vent system at 91500 N, East Pacific Rise (EPR), which is one of the best characterized vent systems worldwide. Samples were collected from four separate research cruises of the area (Table 1) during the months immediately following a volcanic eruption event that occurred in winter 2005–2006 (Tolstoy et al., 2006; Cowen et al., 2007). Within the axial summit collapse trough (ASCT), five sites of high-temperature hydrothermal activity remained actively venting after the eruption: Biovent, Bio9, P vent, Ty and Io (Fig. 1). There were also microbial mats covering basaltic substrates and areas of diffuse flow, occupied by patches of Tevnia and Alvinella. The majority of the Riftia and vent mussels present prior to the eruption had been destroyed. It is unlikely that the volcanic eruption affected our samples to a major extent, even though increased microbial activity and generation of biomass within diffuse flow areas have been reported in previous eruptions at mid-ocean ridges (Haymon et al., 1993; Juniper et al., 1995; Juniper et al., 1998; Huber et al., 2003). The samples collected in this study were obtained at least 6 months following the eruption event, and even though microbial mats were still largely visible in the surrounding diffuse vents, chemosynthetic megafauna already appeared to dominate the biomass of vent biological assemblages, with Tevnia dominating. This stage in the colonization of vent systems was reported by Shank et al. (1998), 10 months post the 1991 eruption at 91500 N EPR.

Fig. 1. Locations of CTD stations 83 and 91 relative to the positions of the Ty, Io, Bio90 , P vent and Biovent vent sites at 91500 N EPR.

924

S.A. Bennett et al. / Deep-Sea Research I 58 (2011) 922–931

2.1. Vent fluid samples End-member vent fluid samples were collected with titanium ‘majors’ bottles using DSV Alvin as reported in Von Damm (2000). High-temperature vent fluid samples were collected from four smoker orifices; Bio90 , P vent, Biovent and Ty, with titanium major sampler pairs using the DSV Alvin. Bottles were triggered when the Alvin high-temperature probe or inductively coupled link temperature probes, both calibrated to a NIST traceable standard, stabilised at a maximum temperature. Shipboard sample treatments were as reported in Von Damm et al. (1995b, 2000). On the day of sampling, high-temperature vent fluid samples were processed shipboard for H2S (5% precision, standardised to Dilut-its standard solutions) by iodimetric starch titration. The remaining vent fluid samples were acidified with 0.5–1.0 ml of concentrated 3x quartz distilled HCl, and stored in high-density polyethylene (HDPE) bottles. Any remaining precipitates were rinsed out of the bottles during the sample draw and saved in HDPE bottles. At the University of New Hampshire (UNH), USA, high-temperature vent fluid samples were filtered through 0.45 mm nuclepore filters in a laminar flow bench. The acidified, filtered fluid fraction samples were stored at room temperature in HDPE bottles, while the filters and filtered particles were saved in separate HDPE bottles. Fluid fractions were analyzed for Fe and Mn (both 1% precision, standardised to Ricca Chemical Company AA standards) by flame atomic absorption spectroscopy (FAAS). Fe and Mn end-members were calculated for a high-temperature vent by performing a least-squares regression of an individual chemical species versus Mg and assuming a linear relationship passing through the ambient bottom seawater composition, and extrapolated to 0 mmol/kg Mg. Reported Fe and Mn end-members were calculated using only data derived from filtered samples (dissolved fraction). As such, this does not take into account the precipitated or suspended fractions (but does include colloidal material), and so will under-estimate the true endmember values. Detailed studies on the vent fluid evolution and chemical evolution following the winter 2005–2006 eruptions are beyond the scope of this paper and will be reported in detail elsewhere (Von Damm et al., unpublished data). In this study, we will use end-member vent-fluid Fe/Mn ratios to help calculate the relative dilution factors that can be ascribed to buoyant plume samples collected directly above each vent-site and H2S concentrations to determine the potential for microbial sulfide oxidation. 2.2. Plume samples Near-field buoyant plume samples were collected from above four high-temperature vent sites; Bio90 , P vent, Biovent and Ty, using 1 L externally sprung Niskin bottles placed in the basket attached to DSV Alvin. A total of eight samples was collected, with two samples from each site obtained sequentially as DSV Alvin flew a few meters above the active chimneys. The distance of DSV Alvin above the seafloor was recorded with a 200 kHz Benthos (ex-Datasonics) Model PSA-900D altimeter. The samples were collected based on increases in water temperature (up to 15 1C) recorded by the temperature probe on Alvin and on direct observations of black particle-laden plumes through the submersible view-ports. Duplicate Niskin samples were collected at each sampling location along with in-situ measurements of pH, free sulfide and temperature. The comprised pH and free sulfide concentrations within the plume were estimated from an autonomous electrochemical probe. The probe was made of pH and sulfide electrodes connected to autonomous potentiometric data loggers, as used in Le Bris et al. (2003, 2006) for pH and Laurent

et al. (2009) for both pH and sulfide. The tips of the electrodes were combined with a thermistance sensor in order to define temperature at measurement point (NKE, France). The combined probes were placed on the Alvin basket, about 1 m from the Niskin bottles. Electrode signals were calibrated in the laboratory before the dive, and later converted into pH and concentration values using the sulfide acidity constant given in Rickard and Luther (2007). The calculated temperature, pH and free sulfide content were averaged over 15 s around the time the maximum temperature was reached. A further suite of water column samples was collected by CTD rosette, from both buoyant and non-buoyant plumes overlying the EPR 91500 N vents, using 10 L Ocean Test Equipment (OTE) bottles internally sprung with silicone tubing. These bottles were mounted on a 24-position rosette equipped with a SeaBird SBE 9/11þCTD system and a SeaPoint STM in-situ turbidity sensor. Initially, the rosette was deployed directly above the 91500 N vent sites in order to collect buoyant plume samples, as detected by the presence of particle, density and temperature anomalies (CTD 83). This was followed by a second CTD cast occupied  0.5 km southwest of the vent sites, selected to be down-plume along the prevailing current direction, which was detected with a loweredacoustic Doppler current profiler (L-ADCP) attached to the CTD rosette. Here, a non-buoyant plume was intersected  150 m above the seafloor and detected by the presence of particle anomalies (CTD 91, Fig. 1) (Baker, 1998). On recovery of the CTD rosette or DSV Alvin aboard ship, water samples were collected from the OTE/Niskin bottle, after shaking, directly into Teflon bottles (with 3-fold rinsing). Water collected directly above the vent orifice was sub-sampled into acid-cleaned LDPE bottles and acidified to pH 1.6 using HNO3 (Fisher Scientific, OPTIMA grade). In a dedicated clean area of the shipboard laboratory, 1 L of sample was filtered through pre-combusted (400 1C, 44 h) GF/F filters (0.7 mm, 25 mm diameter, Whatman) into 40 ml I-CHEM, certified low-level total organic carbon (TOC) vials. These were acidified with 40 ml trace metal grade HCl (Fisher Scientific, stored in glass) to pH 2 and stored at 4 1C for analysis in our shore laboratories. All GF/F filters were folded in combusted foil and frozen, with volumes of seawater filtered recorded for subsequent calculations of particulate material concentrations per litre of ocean/plume water. Analyses of total dissolvable Fe and Mn (TdFe and TdMn) in the near-field plume samples collected directly above the vent orifices were carried out at the National Oceanography Center, Southampton (NOCS), UK. A 4% solution of each sample in 0.4 MdHNO3 was analyzed using a Perkin Elmer Optima 4300DV ICP-OES that was calibrated against matrix-matched standard solutions that bracketed the concentration ranges of the sample solutions. Standard solutions were prepared using single element 1000 mg L  1 standards (Specpure, Spex) of Fe and Mn. A 4% IAPSO seawater standard was also run, even though it contained very low Fe and Mn concentrations, to check for contamination. The limit of detection of the technique was 6.5 nM for Fe and 1.3 nM for Mn, determined by repeated analysis of the seawater matrix blank (3s, n¼10). While these detection limit values are high compared to open-ocean trace metal concentrations, they are much lower than the concentrations reported for the near-vent hydrothermal plume samples investigated here (mM concentrations). Determination of DOC was carried out on a coupled hightemperature combustion total organic carbon-nitrogen chemiluminescence detection (HTC TOC–NCD) system at NOCS. The measurements were performed using a Shimadzu TOC 5000A total carbon analyzer coupled with a Sievers NCD 255 nitrogen chemiluminescence detector (Pan et al., 2005). A procedural blank of MQ water, gave an analytical precision of 70.8 mM (1s, n ¼3) and the accuracy of the technique was determined from the

S.A. Bennett et al. / Deep-Sea Research I 58 (2011) 922–931

925

analysis of a certified reference material. The standard deviation of each sample was obtained from 3 to 5 injections of the same sample into the instrument. Determination of POC was carried out at the Plymouth Marine Laboratory, UK. The GF/F filters were unfolded, removed from the foil and placed in small polystyrene containers. The filters were then treated with sulfurous acid under vacuum for 24 h to remove the inorganic carbonates (Verardo et al., 1990). The filters were dried at 60 1C for 24 h, quartered and packaged in pre-combusted aluminum foil (Hilton et al., 1986). The samples were then analyzed on a Thermo Finnigan Flash EA1112 Elemental analyzer using acetanilide as a calibration standard. Filter blanks were used for blank correcting all samples. The analytical precision, determined from analysis of the filter blanks, was 0.55 mg (1s, n¼6).

were sieved to retain coarse fragments and the remaining o1 mm fraction passed through a 10-port rotating wet sediment splitter. Total dry mass for mass flux calculations was determined from three of the sample splits. Dried subsamples (5–10 mg) were analyzed for particulate inorganic carbon and POC using a CHN analyzer, and Al, Fe, Mn and V, after acid digestion, using ICP–OES and ICP–MS. A number of geo-reference standards (BHVO-1 and IFG) were analyzed along with the samples to confirm analytical accuracy, which was better than 5% for all elements reported.

2.3. Sediment trap samples

The end-member temperature, Fe, Mn and H2S concentrations for each high-temperature vent are listed in Table 2. The endmember concentrations for vent-fluids were sampled over several days. End-member regressions were calculated from major ion analyses for 2–6 discrete samples.

3. Results 3.1. Vent fluid samples

Sediment trap samples analyzed for this study were collected using a McLane 21-position time-series trap deployed at the Bio9 vent-site at 9150.290 N, 140117.490 W and at a depth of 2505 m, 3 m above the seafloor. This ‘R1’ trap was deployed during AT15-6 from the RV Atlantis and repositioned during Alvin submersible dive 4206 on 30th June, 2006. The sediment trap was positioned 25 m laterally away from the Bio9 vent-site with the intention that the trap would then be sufficiently close to underlie (hence, collect particles settling from) both the buoyant and non-buoyant portions of the hydrothermal plume dispersing away from this vent-site. The R1 trap then collected samples continuously, with each sample representing a 14.8-day cycle (to coincide with spring/neap tidal cycles in the overlying water column) until its recovery during Alvin dive 4262 (3rd November, 2006, RV Atlantis AT15-12). Further information on this deployment will be reported in German et al. (manuscript in preparation, 2011). Prior to deployment, each 250 mL polyethylene sample cup was filled with dimethyl sulfoxide (DMSO) and buffered to pH 9.0 70.5. This preservative prevents any biological activity continuing within the sample-cups, post-collection, while retaining sample integrity for mineralogical, bulk geochemical, and molecular microbiological investigations (Comtet et al., 2000). Upon trap recovery, each sample cup was capped and refrigerated at 4 1C for return to the laboratory and shore-based analysis. All subsequent sample processing was conducted at Woods Hole Oceanographic Institution using standard methods established for deep-ocean sediment trap analyses (Honjo et al., 1995). Samples

3.2. Plume samples The POC and DOC concentrations in the near-field vent samples together with their respective TdFe and TdMn concentrations, Fe/Mn ratios, maximum temperature and calculated percent endmember fluid in these samples (see later) are listed in Table 2. The POC concentrations ranged from 0.87 to 3.81 mM and the DOC concentrations ranged from 37.9 to 47.1 mM. The averaged temperature, pH and free sulfide recorded by the in-situ probes for these samples and height of DSV Alvin above the seafloor during each sampling event are shown in Table 3. The CTD depth profiles of POC concentrations, DOC concentrations, potential temperature anomalies and particle anomalies are shown in Fig. 2 and the depth vs density plots for these two CTD casts are shown in Fig. 3. Potential temperature and particle anomalies were determined as described in Baker (1998). The POC concentrations in the plume ranged from 0.07 to 0.46 mM compared to an average ‘background’ value of 0.1670.05 mM measured above and below the plume. DOC concentrations were approximately one order of magnitude greater than POC concentrations and ranged from 35.3 to 43.2 mM in these samples compared to ‘background’ values of 38.270.9 mM above and below plume-height. The upper and lower background limits are shown on Fig. 2.

Table 2 Chemical characteristics of vent fluids and near-field buoyant plume samples collected at 91500 N EPR. Vent fluids

Near-field buoyant plume

Temperature (1C)

[Fe] (lM)

[Mn] (lM)

[H2S] (mM)

Max temp (1C)

[Fe]Td (lM)

[Mn]Td (lM)

Fe/Mn

% end-member fluid

[POC] (lM)

[DOC] (lM)

Biovent

321

854

523

12.3

6

3870 4080 1080 1300 3780

149 156 92.2 97.1 582

11.4 8.6 12.8 10.6 28.7

13 11

Ty

382 335 392 386 359

Background

–

–

–

–

2

2.00 1.70 0.30 0.27 0.53 0.77 0.60a 0.59 –

1.8 1.7 23.8 22.6 8.9 8.8 2.9a 4.8 –

0.38 0.33 0.20 0.18 0.56 0.81 0.10a 0.10 –

0.87

Bio90

3.50 2.83 7.10 6.18 4.74 6.80 1.75a 2.87 –

40.0 37.9 46.1 43.9 41.5 47.1 44.0a 39.2 38.2 7 0.9

P vent

7

1.48 3.74 3.81a 0.167 0.05

[Fe] and [Mn] are the concentrations of Fe and Mn with subscript ‘Td’ indicating ‘total dissolvable’. Reported vent fluid concentrations have been corrected to allow for seawater entrainment (Section 2.1), [POC] is the concentration of particulate organic carbon and [DOC] is the concentration of dissolved organic carbon. The Fe/Mn ratio was calculated using the concentrations of total dissolvable Fe and Mn measured in the near-field buoyant plume samples. The % end-member fluid represents the dilution of the vent fluids with the surrounding seawater and was calculated using Mn as a conservative tracer (see Section 4.1 for further details). a

One of the Niskin bottles fired at Ty, was not shaken prior to sample collection.

926

S.A. Bennett et al. / Deep-Sea Research I 58 (2011) 922–931

Table 3 Plume characteristics as measured by the in-situ probe on board DSV Alvin. Site

Dive number

Time

Height above seafloor (m)

Average temp (1C)

pH

Free H2S (lM)

% end-member fluid

BioVent Bio90 P Vent Ty

4295 4292 4292 4294

20:52:25–20:52:40 20:04:16–20:04:34 20:47:34–20:47:49 21:07:51–21:08:11

7.5 7.0 4.1 4.2

5.0 7.7 7.2 4.7

6.9 7.0 7.0 7.6

22 26 26 9

0.9 1.5 1.3 0.8

Data were averaged over 15 s at the maximum temperature anomaly. The height of DSV Alvin above the seafloor was recorded with an altimeter attached to the bow of the submersible. The % end-member fluid was calculated using the average temperature measured in the plume and the maximum temperature measured in each end-member (Table 2), subtracting the background temperature from each value.

POC (μM)

POC (μM) 0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7 0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

2200 CTD 91

CTD 83

Depth (m)

2250 2300 2350 2400 Temperature Turbidity POC

2450 2500 0.00 0.06 0.02 0.04 Potential temperature anomalies, Δθ (°C)

0.00 0.06 0.02 0.04 Potential temperature anomalies, Δθ (°C)

-0.2

-0.2

-0.1 0.0 0.1 Particle anomalies, Δc (NTU)

-0.1 0.0 0.1 Particle anomalies, Δc (NTU) DOC (μM)

DOC (μM) 30

40

50

60

70

30

40

50

60

70

2200 CTD 83

CTD 91

Depth (m)

2250 2300 2350 2400 Temperature Turbidity DOC

2450 2500 0.00 0.06 0.02 0.04 Potential temperature anomalies, Δθ (°C) -0.2

-0.1 0.0 0.1 Particle anomalies, Δc (NTU)

0.00 0.06 0.02 0.04 Potential temperature anomalies, Δθ (°C) -0.2

-0.1 0.0 0.1 Particle anomalies, Δc (NTU)

Fig. 2. Depth profiles of POC, DOC, potential temperature anomalies and turbidity anomalies for CTD stations that intercepted a buoyant and non-buoyant plume at 91500 N EPR. The gray line indicates temperature, the solid line indicates turbidity, the solid circles indicate POC and the open circles indicate DOC. The vertical dashed lines on the plots indicate the upper and lower limits of the background concentration.

3.3. Sediment trap samples The sediment trap flux data for elements pertinent to this study are shown in Table 4 and Fig. 4. A more complete discussion of post-eruption EPR sediment trap fluxes will be presented in German et al. (manuscript in preparation, 2011). For this study,

we have calculated the relative abundance of hydrothermally sourced Fe-oxyhydroxide material, [Fe-oxy]hyd, in each of the sediment trap samples, using a protocol previously described for hydrothermal sediments, sediment trap samples and hydrothermal plume particles (German et al., 1997; German et al., 2002; Bennett et al., 2009). This method exploits the scavenging

S.A. Bennett et al. / Deep-Sea Research I 58 (2011) 922–931

2200

In order to correct for these continental eolian inputs, we have used sediment data reported in Kyte et al. (1993) to provide us with detrital vanadium:aluminum ratios and assumed that total Al content within the sediment traps is a result of such inputs (German et al., 1997). From this, we can calculate the predicted detrital concentration of vanadium in each sample using ½Vdet ¼ ½Altrap  ð½V=½AlÞeolian

where [V]det is the predicted detrital concentration of vanadium in a sediment trap sample, [Al]trap is the measured aluminum concentration in a sample and [V]/[Al]eolian is the average V:Al ratio in the upper 0–1 m of sediment as reported by Kyte et al. (1993) and shown in Table 4. The excess of vanadium in each sample is assumed to be hydrothermal plume fall out, [V]hyd,

CTD 83 CTD 91

2250

ð1Þ

100

Mass flux Organic carbon FeOOH

80

3.0

10

2.5

Depth (m)

2300

2350

2400

60 40

6

20 4

0 -20

2

-40

2450

-60 0 2500 27.71

27.72

27.72

27.72

27.73

27.73

Potential density (Kg/m3) Fig. 3. Depth vs density plot demonstrating the buoyancy of the plume at CTD 83.

20

40

60 80 Day

100

120

0 140

Corg (mg/m2/day)

Mass flux (mg/m2/day)

8

2.0 1.5 1.0

FeOOH (mg/m2/day)

properties of Fe oxyhydroxides for oxyanions (e.g. vanadium (V)) within the water column, resulting in V:Fe ratios which vary systematically with the composition of the ambient seawater (Feely et al., 1998). By knowing the concentration of vanadium that is derived from hydrothermal Fe-oxyhydroxide scavenging, we can calculate the concentration of [Fe-oxy]hyd. The chemical constituents within the trap will be sourced from both eolian inputs sinking through the water column and hydrothermal inputs dispersed laterally within the hydrothermal plume. Therefore, we must first determine how much of the vanadium in the trap samples was sourced from the hydrothermal system as opposed to eolian inputs.

927

0.5 0.0

Fig. 4. Time profile of the mass flux, organic carbon (Corg) and hydrothermally sourced Fe oxyhydroxides (FeOOH) for sediment trap (R1) deployed 25 m from the Bio9 vent site. The error for organic carbon analysis is too small to show on the plot (RSD ¼ 0.92%).

Table 4 Sediment trap data. Days elapsed

Mass mg/m2/day

Fe mg/m2/day

Corg mg/m2/day

Al mg/m2/day

V lg/m2/day

[Fe-oxy]hyd mg/m2/day

RSD %

6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108 114 120 N. Pac. clay Average 0 1 m

81.0 79.6 16.9 20.0 26.3 12.5 22.0 25.8 46.1 26.8 26.9 18.5 33.8 17.0 18.2 19.0 23.7 52.3 60.0 57.2

14.4 16.2 2.6 3.0 3.8 2.5 5.6 5.0 9.1 3.7 1.8 2.0 5.8 1.7 3.0 1.7 3.2 7.9 9.1 9.2

1.8 1.6 2.3 1.0 2.1 0.5 0.8 0.9 1.7 3.8 3.6 1.1 1.7 1.1 1.5 1.3 2.2 4.7 4.4 2.8

1.09 1.44 0.30 0.29 0.50 0.09 0.17 0.14 0.81 0.40 0.42 0.22 0.39 0.12 0.21 0.19 0.17 0.57 0.68 0.43 98.5 mg/g

5.07 6.29 1.28 1.44 2.02 0.67 1.10 0.80 3.84 1.96 1.54 1.03 2.04 0.76 1.22 1.05 1.25 3.79 4.95 3.64 152 mg/g

1.27 1.49 0.30 0.36 0.46 0.19 0.30 0.21 0.94 0.49 0.34 0.26 0.53 0.21 0.33 0.28 0.36 1.08 1.43 1.08

12.3 13.1 13.2 12.1 13.7 10.5 11.0 11.3 12.5 12.1 14.3 12.2 11.7 10.8 11.1 11.4 10.3 10.5 10.3 10.0

from LL44-GPC3 (Kyte et al., 1993)

Fe, organic carbon (Corg), Al and V flux data are shown for bulk compositions within the trap samples. These data were used to calculate the Fe-oxyhydroxide hydrothermal component ([Fe-oxy]hyd), as described in the text (Section 3.3), along with a propagated relative standard deviation (RSD).

928

S.A. Bennett et al. / Deep-Sea Research I 58 (2011) 922–931

such that ½Vhyd ¼ ½Vtrap 2½Vdet

ð2Þ

where [V]trap is the measured vanadium concentration in the sediment trap sample. The uptake of V by Fe oxyhydroxides along the EPR has a V/Fe ratio of 0.0028 (Feely et al., 1998), therefore, the hydrothermal Fe oxyhydroxide component ([Fe–ox]hyd) of the sediment trap samples can be calculated using ½Fe-oxhyd ¼ 1=0:0028  ½Vhyd

ð3Þ

The results of these calculations are shown in Table 4.

4. Discussion 4.1. Dilution of near-field plume samples Within the short time frame of buoyant plume rise (  1 h), from emission at the vent orifice to emplacement at non-buoyant plume height (Baker et al., 1995), dissolved Mn can be treated as a conservative tracer. By contrast, Fe does not show conservative behavior over this time frame, due to the formation of polymetallic sulfide precipitates within the first few seconds of venting and the rapid oxidation of Fe(II) to form Fe oxyhydroxide precipitates. Both of these particulate Fe species can fall out of a plume during buoyant plume rise. By contrast, Mn oxidation is much slower and dissolved reduced Mn species can be considered to be quantitatively transported to and emplaced within nonbuoyant hydrothermal plumes. Using the vent fluid data collected in November 2006 (Table 2), we can investigate the evolution of near-field plumes and calculate the percent dilution for each near-field buoyant plume sample using % endmember fluid ¼ ½MnBP =½MnVF  100

ð4Þ

where [Mn]BP is the Mn concentration measured in the near-field buoyant plume samples and [Mn]VF is the corresponding vent fluid end-member Mn concentration. Dilution factors calculated by this method fall in the range of 125- to 1000-fold dilution with ambient seawater (i.e. the samples contain just 0.1–0.8% endmember vent fluid, Table 2), demonstrating the dilution that is associated with turbulent mixing within the first few meters of buoyant plume rise. Except for Ty, duplicate near-field plume samples taken successively at each site exhibit very similar Fe/Mn ratios, demonstrating that each plume was rather homogeneous. Dilution factors calculated from in-situ temperature measurements ranged from 0.8% to 1.5%. These values were calculated as for Mn but using the average temperature measured within the plume and the maximum temperature measured within the vent fluid at each site (and subtracting background temperature of 2 1C from each value) (Table 3). These lower dilution ratios reflect minimum dilution conditions experienced in the core of the plume, as illustrated by steep peaks simultaneously recorded for all three parameters (temperature, pH and sulfide) while DSV Alvin flew a few meters above the high-temperature chimneys. In comparison, Niskin sample contents have integrated several mixing fractions from the turbulent plume, with larger seawater dilution. 4.2. Buoyant plume rise and dispersal As buoyant plumes rise up into the water column, further dilution continues. At CTD 83, directly above the EPR 91500 N vents, strong particle anomalies coupled with positive temperature anomalies were detected from just above the seafloor up to 2460 m. Importantly, the density profile for this station (Fig. 3)

also reveals an instability in the water column coincident with the positive temperature anomalies from the seafloor up to 2460 m, confirming that we had intercepted a buoyant (rising) hydrothermal plume at these depths (Lupton, 1995). During this same cast, a lowered-acoustic Doppler current profiler (L-ADCP) attached to the CTD rosette, measured a current velocity at 2400 m depth of 8.6 cm s  1 along heading 1991. Because the depth of the ASCT is only  20 m, this rising plume would not have been constrained by the local topography but, rather, should have been dispersed away from the ridge axis along the prevailing current direction. At CTD 91, occupied  500 m southwest of the 91500 N vents (i.e. in the projected down-plume direction), particle anomalies were observed at a typical non-buoyant plume height of between 2350 and 2275 m, 150–225 m above the seafloor. At these depths, no density anomalies were observed indicating that the plume was no longer buoyant (Fig. 3). Further, the L-ADCP measured a current velocity at plume-height of 5.8 cm s  1 on a heading of 2331, consistent with the 91500 N high-temperature vents being the source for this plume. 4.3. DOC and POC in the near-field plume POC concentrations in our near-field plume samples exhibit elevated concentrations over background, ranging from 0.87 to 3.81 mM with highest concentrations at the Ty and P vents (Table 2). The DOC concentrations in these same near-field plume samples ranged from 37.9 to 47.1 mM, a range that spans the local value for background seawater (38.2 70.9 mM), but also reflects enrichments of up to 20% at three of the sites investigated. The analysis of POC and DOC is operationally defined as the separation of particles through a combusted GF/F glass fiber filter (0.7 mm nominal pore-size). A direct consequence is that the majority of any POC fraction measured is often dominated by microorganisms, including microbial biomass and larvae (Mullineaux et al., 2010), whereas DOC sources commonly include extracellular release from bacteria, grazer mediated release and excretion, release via cell lysis (from viruses and bacteria), solubilization of particles and bacterial transformation and release (Carlson, 2002). DOC sinks include biotic consumption (e.g. by heterotrophic bacteria) and abiotic sorption onto particles. Winn et al. (1986), estimated living carbon concentrations within a hydrothermal plume by using ATP measurements as an indicator of microbial biomass, and reported particulate living carbon concentrations of 1.2 mM at 20 m above a vent-chimney and 0.3 mM at 50 m off-bottom at hydrothermal sites along the Juan de Fuca Ridge. These concentrations are similar to the total POC measured in our near field plume samples collected from DSV Alvin as well as in the buoyant and non-buoyant plume samples collected during CTD casts 83 and 91 (POCr0.46 mM; Fig. 2). We do not believe that EPR 91500 N vent fluids can be the source for the POC and DOC enrichments in our near-field plume samples, because the required end-member DOC and POC concentrations would be unreasonably high. Typically, thermal degradation of organic matter is to be expected at the hightemperatures associated with typical ‘black smoker’ venting. For example, at the Main Endeavor field, and at Axial Volcano, on the Juan de Fuca Ridge, average DOC concentrations in end-member fluids are 15 and 17 mM, respectively (Lang et al., 2006). By contrast, if our Alvin-collected Niskin samples (average DOC concentration of 42.5 mM) have experienced a 125- to 1000-fold dilution with seawater (Table 2), the required end-member ventfluid DOC concentration would fall in the range 5.3–42.5 mM, some three orders of magnitude higher than previously measured at the Juan de Fuca Ridge vents.

S.A. Bennett et al. / Deep-Sea Research I 58 (2011) 922–931

A more plausible source for the DOC and POC in our near field plume samples is from entrainment of diffuse flow and chemosynthetic life on the chimney walls into the base of the buoyant plume (Cowen et al., 1986; Karl et al., 1988; Bailly-Bechet et al. 2008). At the time of sampling, the high-temperature chimney walls were covered with Alvinella colonies and it is expected that hydrodynamic constraints will favor the entrainment of organic material from the chimney walls or alvinellid colonies (Bailly-Bechet et al., 2008). In addition, diffuse systems typically host the most abundant chemosynthetic communities associated with seafloor hydrothermal venting and during our dive program, areas of high biomass were observed in the areas surrounding the vents at 91500 N EPR. Within these environmental niches, DOC concentrations have been observed to reach values (39–69 mM) that are nearly double background deep-ocean values (Lang et al., 2006) and are closer to those measured in the most productive waters associated with the equatorial Pacific upwelling region, where DOC concentrations reach 67 mM (Carlson and Ducklow, 1995). Elevated POC concentrations (of up to 18.3 mM) have also been observed in warm (20 1C) diffuse-flow fluids at 211N, EPR (Comita et al., 1984). In that study, DOC concentrations were 53–71 mM, similar to the range reported by Lang et al. (2006), who concluded that the high DOC concentrations present in diffuse flow ventfluids were a result of high microbial activity sustained by energy produced from the oxidation of sulfur species and H2 (McCollom and Shock, 1997; Sarradin et al., 1999). Thus, while entrainment of pre-formed DOC and POC into our plume samples could explain the data reported here, we cannot preclude the possibility that on-going microbial productivity within the buoyant and nonbuoyant plume might also play an important role in producing biomass 7organic ligands7exopolymers. Certainly, in-situ production of microbial biomass has the potential, energetically, to occur within plumes themselves, due to the redox disequilibria set up between the chemically reduced ventfluids and oxygenated seawater (McCollom, 2000; Lam et al., 2008; Dick et al., 2009). Of course, any in-situ production within the buoyant plume samples obtained using Alvin, prior to their collection, is likely to be minimal because the samples were taken from turbulent rising plumes just a few meters above each vent orifice. In-situ production could have occurred subsequently, however, within the closed Niskin bottles, during the ascent of Alvin but prior to being filtered aboard the ship. This would be expected to result in the production of biomass (increasing POC) as well as the production and consumption of DOC. Within the early stages of buoyant plume rise, microbial activity is most likely the result of hydrogen and sulfide oxidation (Lam et al., 2008), with the oxidation of elemental sulfur and sulfide precipitates providing the biggest biomass potential (McCollom, 2000). Hydrogen sulfide in the vent fluids ranged from 8.6 to 28.7 mM and the free sulfide concentrations measured in-situ as DSV Alvin flew through the plume were high compared to background seawater (Table 3). It is also important to remember that free sulfide represents only part of the total sulfide transported by hydrothermal plumes. Since aqueous and polymeric iron sulfides in equilibrium with H2S and HS- may contribute more than 60% of the total ‘dissolved’ sulfide pool (Luther et al., 2001; Le Bris et al., 2003) the total amount of reduced sulfur that was trapped within each Niskin bottle, and hence available for microbial oxidation following each sampling event, was most probably still larger than the values reported in Table 3. Previously, McCollom (2000) carried out a theoretical study to quantify the component of biomass production within hydrothermal plumes that can be directly attributed to inorganic chemical reactions. He determined that approximately 50 mg of biomass could potentially be produced per kg of vent fluid (from EPR 211N OBS vent) as the fluid dilutes 1000-fold with the surrounding seawater in the rising plume. This is equivalent to

929

23 mg of C per kg of fluid (when 2.2 g biomass contains 1 g of C (Battley, 1998)) or  1.9 mM C. In a diluted buoyant plume, this would only result in a 1.9 mM increase in POC concentrations over background which, for EPR 91500 N (background¼0.1670.05 mM), would imply maximum predicted POC values within our Alvincollected near-field samples of  2.1 mM. Intriguingly, that value compares extremely well with the POC concentrations that we did measure in our Niskin samples, processed aboard ship, which fall in the range 0.87–3.81 mM (Table 2). 4.4. POC and DOC in the dispersing plume As hydrothermal plumes rise and become progressively more dilute, they can be tracked readily using in situ optical sensors that respond to the high concentrations of suspended particulate matter within those plumes (Baker et al., 1995). In this study, such a phenomenon is most readily apparent in cast CTD 91 where a particle-rich lens of water at 2275–2350 m depth provides clear evidence for a dispersing non-buoyant hydrothermal plume (Fig. 2). What is particularly notable at this station is that the profile of measured POC concentrations directly mimics this turbidity profile, indicating that highest values of POC occur toward the core of the dispersing non-buoyant plume (Fig. 2). Further, the maximum POC concentrations observed at CTD 91 are directly comparable to the highest buoyant plume concentrations observed at CTD 83, situated immediately above the seafloor. Given that we should expect plume samples at CTD 91 to have undergone at least one order of magnitude further dilution, relative to samples from the deep water column at CTD 83 (Lupton, 1995), this indicates that there must have been some further process resulting in the addition of POC to the dispersing non-buoyant plume. The close correlation observed between particle anomalies and POC data, suggests that an additional source of POC to the plume may result from passive scavenging of dissolved organic carbon onto plume particles and/or from a microbial community active within the dispersing plume. Adsorption of organic carbon onto plume particles should result in a shift in the size fractionation of the total organic carbon present, increasing the POC fraction and decreasing the DOC fraction. Likewise, a hetero- or mixotrophic microbial community would consume DOC (decreasing that concentration) while increasing its biomass (leading to increased POC concentrations). In either case, one would expect any increase in POC to coincide with a complementary decrease in DOC concentrations. In our plume samples— and within the precision of our measurements—none of the DOC concentrations measured at plume height differ significantly from background. Nevertheless, across the depth-range for the nonbuoyant plume at CTD 91, as defined from in-situ turbidity signals, there does appear to be a concave curvature in the profile of DOC that mirrors the concave curvatures observed in both the POC profile and the in-situ turbidity sensor profile. This could be consistent with scavenging and/or biological consumption of DOC from the dispersing non-buoyant plume (Fig. 2). At the core of the plume, however, maximum POC concentrations are  0.4 mM representing an increase of just 0.2 mM over background. By contrast, DOC concentrations appear to decrease from  39 to 37 mM over the same depth range—a deficit of  2 mM. If this much larger decrease were due to uptake of DOC onto/into particulate matter at plume-height, the apparent net deficit would need to be explained by some additional removal process—e.g. in the form of settling plume particulates. Evidence that such a mechanism is plausible can be gained from a consideration of fluxes to sediment traps underlying the EPR 91500 N plume (Fig. 4). In those sediment traps, particulate organic carbon fluxes during the period immediately prior to our Alvin and CTD-rosette based sampling appear to relate more closely to the

930

S.A. Bennett et al. / Deep-Sea Research I 58 (2011) 922–931

calculated Fe oxyhydroxide (FeOOH) fluxes than to total mass-flux or total Fe flux (Fig. 4, Table 4). Since this Fe oxyhydroxide fraction represents just 1–2% of the total mass flux, this would appear to be consistent with a more specific Corg-transporting process than just surface-independent particle scavenging or adsorption. Such removal of organic carbon in association with Fe-oxyhydroxide species could occur as a direct result of the preferential adsorption of organic carbon onto Fe oxyhydroxide mineral surfaces (Balistrieri and Murray, 1987; Gschwend and Schwarzenbach, 1992), and/or through microbial oxidation of Fe(II) and Fe sulfides that preferentially concentrates organic carbon with Fe oxyhydroxides. While detailed microscopic, biogeochemical and mineralogical analyses on similar trap samples have recently demonstrated that Fe(III) oxyhydroxide aggregates occur in the presence of exopolymeric organic carbon (i.e. organic substances excreted from microbial species (Toner et al., 2009)), it remains to be determined whether this is as a result of biotic or purely abiotic geochemical processes. Investigating the relative importance of the two remains an important topic for continuing deep-sea hydrothermal biogeochemical research.

5. Summary In this study, we have investigated dissolved and particulate organic carbon concentrations within an evolving hydrothermal plume. We conclude that organic carbon is entrained into buoyant plumes from chimney walls and adjacent areas of biologically rich diffuse flow but that there may be further addition of organic carbon during plume dispersal. Within and beneath hydrothermal plumes, we observe a relationship between POC and plume particulate matter—particularly Fe-oxyhydroxide phases. These close associations may result from scavenging of DOC from the surrounding plume-waters and/or in-situ microbial productivity (e.g. from active microbial oxidation of Fe(II) and Fe sulfides within the plume). The study of organic carbon cycling within hydrothermal plume environments is complicated by this potential for biological production and consumption. Consequently, we recommend that future studies should include a multidisciplinary approach to understand carbon cycling in deep-sea hydrothermal systems that encompasses the system as a whole, including dispersing hydrothermal plumes.

Acknowledgements We would like to thank Andy Gooday, Michael Bacon and three anonymous reviewers for their constructive comments. We thank the captain, crew and scientists on board AT15-6, AT15-12, AT15-13 and AT15-14 who contributed to the success of this work and to Chief Scientist, Jim Ledwell, who enabled SAB to participate in AT1514 (as part of the LADDER project), as well as Lauren Mullineaux the LADDER project coordinator who invited NLB to this cruise. We also thank Ryan Jackson for creation of the location map and to Tim Fischer for work up of the L-ADCP data. We thank Robert Head and Xi Pan for POC and DOC measurements and Steven Manganini, Maureen Auro and Scot Birdwhistell for sediment trap sample analysis. Support for vent fluid analysis was supported by the OCE-0327126, originally awarded to Karen Von Damm and subsequently managed by Julie Bryce. Support for sediment trap analysis was funded by OCE-0647948, awarded to CRG and OJR. SAB was funded by a NERC PhD studentship, NER/S/A/2004/12631. References Bailly-Bechet, M., Kerszberg, M., Gaill, F., Pradillon, F., 2008. A modeling approach of the influence of local hydrodynamic conditions on larval dispersal at hydrothermal vents. Journal of Theoretical Biology 255 (3), 320–331.

Baker, B.J., German, C.R., Elderfield, H., 1995. Hydrothermal plumes over spreading-center axes: Global distributions and geographical Interferences. In: Humphris, S.E., Zierenberg, R.A., Mullineaux, L.S., Thomson, R.E. (Eds.), Seafloor Hydrothermal Systems: Physical, Chemical, Biological, and Geological Interactions. American Geophysical Union, Washington, D.C, pp. 47–71. Baker, E.T., 1998. Patterns of event and chronic hydrothermal venting following a magmatic intrusion: new perspectives from the 1996 Gorda Ridge eruption. Deep-Sea Research 45, 2599–2618. Balistrieri, L.S., Murray, J.W., 1987. The influence of the major ions of seawater on the adsorption of simple organic-acids by goethite. Geochimica Et Cosmochimica Acta 51 (5), 1151–1160. Battley, E.H., 1998. The development of direct and indirect methods for the study of the thermodynamics of microbial growth. Thermochimica Acta 309 (1–2), 17–37. Bennett, S.A., Achterberg, E.P., Connelly, D.P., Statham, P.J., Fones, G.R., German, C.R., 2008. The distribution and stabilisation of dissolved Fe in deep-sea hydrothermal plumes. Earth and Planetary Science Letters 270, 157–167. ¨ Bennett, S.A., Rouxel, O.J., Schmidt, K., Garbe-Schonberg, D., Statham, P.J., German, C.R., 2009. Iron isotope fractionation in a buoyant hydrothermal plume, 5 deg S Mid-Atlantic Ridge. Geochimica Et Cosmochimica Acta 73, 5619–5634. Burd, B.J., Thomson, R.E., 1994. Hydrothermal venting at Endeavor Ridge—effect on zooplankton biomass throughout the water column. Deep-Sea Research Part I—Oceanographic Research Papers 41 (9), 1407–1423. Carlson, C.A., 2002. Production and removal processes. In: Hansell, D.A., Carlson, C.A. (Eds.), Biogeochemistry of Marine Dissolved Organic Matter. Academic Press. Carlson, C.A., Ducklow, H.W., 1995. Dissolved organic-carbon in the upper ocean of the central equatorial Pacific-Ocean, 1992—daily and finescale vertical variations. Deep-Sea Research Part II-Topical Studies in Oceanography 42 (2–3), 639–656. Comita, P.B., Gagosian, R.B., Williams, P.M., 1984. Suspended particulate organic material from hydrothermal vent waters at 21-degrees-N. Nature 307 (5950), 450–453. Comtet, T., Jollivet, D., Khripounoff, A., Segonzac, M., Dixon, D.R., 2000. Molecular and morphological identification of settlement-stage vent mussel larvae, Bathymodiolus azoricus (Bivalvia:Mytilidae), preserved in situ at active vent fields on the Mid-Atlantic Ridge. Limnology and Oceanography 45 (7), 1655–1661. Cowen, J.P., Bertram, M.A., Baker, E.T., Feely, R.A., Massoth, G.J., Summit, M., 1998. Geomicrobial transformation of manganese in Gorda Ridge event plumes. Deep-Sea Research Part II-Topical Studies in Oceanography 45 (12), 2713–2737. Cowen, J.P., Bertram, M.A., Wakeham, S.G., Thomson, R.E., Lavelle, J.W., Baker, E.T., Feely, R.A., 2001. Ascending and descending particle flux from hydrothermal plumes at Endeavour Segment, Juan de Fuca Ridge. Deep-Sea Research Part I-Oceanographic Research Papers 48 (4), 1093–1120. Cowen, J.P., Fornari, D.J., Shank, T.M., Love, B., Glazer, B., Treusch, A., Holmes, R.C., Soule, S.A., Baker, E.T., Tolstoy, M., Pomranin, K.R., 2007. Volcanic eruptions at East Pacific Rise near 9 degrees 50’N. EOS. Transactions, American Geophysical Union 8 (07), 81–83. Cowen, J.P., Massoth, G.J., Baker, E.T., 1986. Bacterial scavenging of Mn and Fe in a mid-field to far-field hydrothermal particle plume. Nature 322 (6075), 169–171. Cowen, J.P., Shackelford, R., McGee, D., Lam, P., Baker, E.T., Olson, E., 1999. Microbial biomass in the hydrothermal plumes associated with the 1998 axial volcano eruption. Geophysical Research Letters 26 (24), 3637–3640. Cowen, J.P., Wen, X.Y., Popp, B.N., 2002. Methane in aging hydrothermal plumes. Geochimica Et Cosmochimica Acta 66 (20), 3563–3571. Dick, G.J., Bradley, M.T., 2010. Microbial diversity and biogeochemistry of the Guaymas Basin deep-sea hydrothermal plume. Environmental Microbiology 12, 1190–1198. Dick, G.J., Clement, B.G., Webb, S.M., Fodrie, F.J., Bargar, J.R., Tebo, B.M., 2009. Enzymatic microbial Mn(II) oxidation and Mn biooxide production in the Guaymas Basin deep-sea hydrothermal plume. Geochimica Et Cosmochimica Acta 73 (21), 6517–6530. Elderfield, H., Schultz, A., 1996. Mid-ocean ridge hydrothermal fluxes and the chemical composition of the ocean. Annual Review of Earth and Planetary Sciences 24, 191–224. Feely, R.A., Trefry, J.H., Lebon, G.T., German, C.R., 1998. The relationship between P/Fe and V/Fe ratios in hydrothermal precipitates and dissolved phosphate in seawater. Geophysical Research Letters 25 (13), 2253–2256. German, C.R., Bourles, D.L., Brown, E.T., Hergt, J., Colley, S., Higgs, N.C., Ludford, E.M., Nelsen, T.A., Feely, R.A., Raisbeck, G., Yiou, F., 1997. Hydrothermal scavenging on the Juan de Fuca Ridge: Th-230(xs), Be-10, and REEs in ridgeflank sediments. Geochimica Et Cosmochimica Acta 61 (19), 4067–4078. German, C.R., Bowen, A., Coleman, M.L., Honig, D.L., Huber, J., Jakuba, M.V., Kinsey, J.C., Kurz, M.D., Leroy, S., McDermott, J., Mercier de Lepinay, B., Nakamura, K., Seewald, J.S., Smith, J.L., Sylva, S.P., Van Dover, C.L., Whitcomb, L.L., Yoerger, D.R., 2010. Diverse styles of submarine venting on the ultraslow spreading Mid-Cayman Rise. Proceedings of the National Academy of Sciences 107 (32), 14020–14025. German, C.R., Colley, S., Palmer, M.R., Khripounoff, A., Klinkhammer, G.P., 2002. Hydrothermal plume-particle fluxes at 13 degrees N on the East Pacific Rise. Deep-Sea Research Part I-Oceanographic Research Papers 49 (11), 1921–1940. German, C.R., Von Damm, K.L., 2004. Hydrothermal processes. In: Elderfield, H. (Ed.), The oceans and marine geochemistry. Elsevier-Pergamon, Oxford.

S.A. Bennett et al. / Deep-Sea Research I 58 (2011) 922–931

Gschwend, P.M., Schwarzenbach, R.P., 1992. Physical chemistry of organic compounds in the marine environment. Marine Chemistry 39 (1–3), 187–207. Haymon, R.M., Fornari, D.J., Vondamm, K.L., Lilley, M.D., Perfit, M.R., Edmond, J.M., Shanks, W.C., Lutz, R.A., Grebmeier, J.M., Carbotte, S., Wright, D., McLaughlin, E., Smith, M., Beedle, N., Olson, E., 1993. Volcanic-eruption of the midocean ridge along the East Pacific Rise crest at 9-degrees-45-52’N—Direct submersible observations of sea-floor phenomena associated with an eruption event in April, 1991. Earth and Planetary Science Letters 119 (1–2), 85–101. Hilton, J., Lishman, J.P., Mackness, S., Heaney, S.I., 1986. An automated-method for the analysis of particulate carbon and nitrogen in natural-waters. Hydrobiologia 141 (3), 269–271. Honjo, S., Dymond, J., Collier, R., Manganini, S.J., 1995. Export production of particles to the interior of the Equatorial Pacific Ocean during the 1992 EQPAC experiment. Deep-Sea Research Part II—Topical Studies in Oceanography 42 (2-3), 831–870. Huber, J.A., Butterfield, D.A., Baross, J.A., 2003. Bacterial diversity in a subseafloor habitat following a deep-sea volcanic eruption. Fems Microbiology Ecology 43 (3), 393–409. Jannasch, H.W., Mottl, M.J., 1985. Geomicrobiology of deep-sea hydrothermal vents. Science 229 (4715), 717–725. Juniper, S.K., Bird, D.F., Summit, M., Vong, M.P., Baker, E.T., 1998. Bacterial and viral abundances in hydrothermal event plumes over northern Gorda Ridge. DeepSea Research Part II-Topical Studies in Oceanography 45 (12), 2739–2749. Juniper, S.K., Martineu, P., Sarrazin, J., Gelinas, Y., 1995. Microbial mineral floc associated with nascent hydrothermal activity on coaxial segment, Juan-deFuca Ridge. Geophysical Research Letters 22 (2), 179–182. Karl, D.M., Knauer, G.A., Martin, J.H., Ward, B.B., 1984. Bacterial chemolithotrophy in the oceans is associated with sinking particles. Nature 309 (5963), 54–56. Karl, D.M., Taylor, G.T., Novitsky, J.A., Jannasch, H.W., Wirsen, C.O., Pace, N.R., Lane, D.J., Olsen, G.J., Giovannoni, S.J., 1988. A microbiological study of Guaymas Basin high-temperature hydrothermal vents. Deep-Sea Research Part a-Oceanographic Research Papers 35 (5), 777–791. Karl, D.M., Wirsen, C.O., Jannasch, H.W., 1980. Deep-sea primary production at the Galapagos hydrothermal vents. Science 207 (4437), 1345–1347. Kyte, F.T., Leinen, M., Heath, G.R., Zhou, L., 1993. Cenozoic sedimentation history of the central North Pacific—Inferences from the elemental geochemistry of core LL44-GPC3. Geochimica Et Cosmochimica Acta 57 (8), 1719–1740. Lam, P., Cowen, J.P., Jones, R.D., 2004. Autotrophic ammonia oxidation in a deepsea hydrothermal plume. FEMS Microbiology Ecology 47 (2), 191–206. Lam, P., Cowen, J.P., Popp, B.N., Jones, R.D., 2008. Microbial ammonia oxidation and enhanced nitrogen cycling in the Endeavour hydrothermal plume. Geochimica Et Cosmochimica Acta 72 (9), 2268–2286. Lang, S.Q., Butterfield, D.A., Lilley, M.D., Johnson, H.P., Hedges, J.I., 2006. Dissolved organic carbon in ridge-axis and ridge-flank hydrothermal systems. Geochimica Et Cosmochimica Acta 70 (15), 3830–3842. Laurent, M.C.Z., Gros, O., Brulport, J.P., Gaill, F., Le Bris, N., 2009. Sunken wood habitat for thiotrophic symbiosis in mangrove swamps. Marine Environmental Research 67 (2), 83–88. Le Bris, N., Govenar, B., Le Gall, C., Fisher, C.R., 2006. Variability of physicochemical conditions in 9 degrees 50’N EPR diffuse flow vent habitats. Marine Chemistry 98 (2–4), 167–182. Le Bris, N., Sarradin, P.M., Caprais, J.C., 2003. Contrasted sulphide chemistries in the environment of 13 degrees N EPR vent fauna. Deep-Sea Research Part I-Oceanographic Research Papers 50 (6), 737–747. Lupton, J.E., 1995. Hydrothermal plumes: near and far field, in: Seafloor hydrothermal systems: Physical, chemical, biological, and geological interactions. In: Humphris, S., Zierenberg, R., Mullineaux, L., Thomson, R. (Eds.), Geophysical Monograph, 91. American Geophysical Union, pp. 317–346. Luther, G.W., Rozan, T.F., Taillefert, M., Nuzzio, D.B., Di Meo, C., Shank, T.M., Lutz, R.A., Cary, S.C., 2001. Chemical speciation drives hydrothermal vent ecology. Nature 410 (6830), 813–816. McCollom, T.M., 2000. Geochemical constraints on primary productivity in submarine hydrothermal vent plumes. Deep-Sea Research Part I-Oceanographic Research Papers 47 (1), 85–101. McCollom, T.M., Shock, E.L., 1997. Geochemical constraints on chemolithoautotrophic metabolism by microorganisms in seafloor hydrothermal systems. Geochimica Et Cosmochimica Acta 61 (20), 4375–4391. Mullineaux, L.S., Adams, D.K., Mills, S.W., Beaulieu, S.E., 2010. Larvae from afar colonize deep-sea hydrothermal vents after a catastrophic eruption. Proceedings of the National Academy of Sciences of the United States of America 107 (17), 7829–7834.

931

Ortmann, A.C., Suttle, C.A., 2005. High abundances of viruses in a deep-sea hydrothermal vent system indicates viral mediated microbial mortality. Deep-Sea Research Part I-Oceanographic Research Papers 52 (8), 1515–1527. Pan, X., Sanders, R., Tappin, A.D., Worsfold, P.J., Achterberg, E.P., 2005. Simultaneous determination of dissolved organic carbon and total dissolved nitrogen on a coupled high-temperature combustion total organic carbon–nitrogen chemiluminescence detection (HTC TOC–NCD) system. Journal of Automated Methods & Management in Chemistry 4, 240–246. Rau, G.H., Hedges, J.I., 1979. C-13 depletion in a hydrothermal vent mussel—suggestion of a chemo-synthetic food source. Science 203 (4381), 648–649. Rickard, D., Luther, G.W., 2007. Chemistry of iron sulfides. Chemical Reviews 107 (2), 514–562. Roth, S.E., Dymond, J., 1989. Transport and settling of organic material in a deepsea hydrothermal plume: Evidence from particle flux measurements. Deep Sea Research Part A. Oceanographic Research Papers 36 (8), 1237–1254. Sarradin, P.M., Caprais, J.C., Riso, R., Kerouel, R., Aminot, A., 1999. Chemical environment of the hydrothermal mussel communities in the Lucky Strike and Menez Gwen vent fields, Mid-Atlantic ridge. Cahiers De Biologie Marine 40 (1), 93–104. Shank, T.M., Fornari, D.J., Von Damm, K.L., Lilley, M.D., Haymon, R.M., Lutz, R.A., 1998. Temporal and spatial patterns of biological community development at nascent deep-sea hydrothermal vents (9 degrees 500 N, East Pacific Rise). Deep-Sea Research Part II-Topical Studies in Oceanography 45 (1–3), 465–515. Straube, W.L., Deming, J.W., Somerville, C.C., Colwell, R.R., Baross, J.A., 1990. Particulate DNA in smoker fluids—Evidence for existence of microbial populations in hot hydrothermal systems. Applied and Environmental Microbiology 56 (5), 1440–1447. Sunamura, M., Higashi, Y., Miyako, C., Ishibashi, J., Maruyama, A., 2004. Two bacteria phylotypes are predominant in the Suiyo Seamount hydrothermal plume. Applied and Environmental Microbiology 70 (2), 1190–1198. Sylvan, J.B., Pyenson, B.C., Toner, B.M., Rouxel, O.J., Edwards, K.J., in review. Time series analysis of two hydrothermal plumes at 91N East Pacific Rise reveals distinct, heterogeneous bacterial populations. Geobiology. Tagliabue, A., Bopp, L., Dutay, J.C., Bowie, A.R., Chever, F., Jean-Baptiste, P., Bucciarelli, E., Lannuzel, D., Remenyi, T., Sarthou, G., Aumont, O., Gehlen, M., Jeandel, C., 2010. Hydrothermal contribution to the oceanic dissolved iron inventory. Nature Geoscience 3 (4), 252–256. Tolstoy, M., Cowen, J.P., Baker, E.T., Fornari, D.J., Rubin, K.H., Shank, T.M., Waldhauser, F., Bohnenstiehl, D.R., Forsyth, D.W., Holmes, R.C., Love, B., Perfit, M.R., Weekly, R.T., Soule, S.A., Glazer, B., 2006. A sea-floor spreading event captured by seismometers. Science 314 (5807), 1920–1922. Toner, B.M., Fakra, S.C., Manganini, S.J., Santelli, C.M., Marcus, M.A., Moffett, J.W., Rouxel, O., German, C.R., Edwards, K.J., 2009. Deep-sea hydrothermal iron(II) in stable association with organic carbon in plume particles. Nature Geoscience 2 (3), 197–201. Verardo, D.J., Froelich, P.N., McIntyre, A., 1990. Determination of organic-carbon and nitrogen in marine-sediments using the Carlo-Erba-Na-1500 analyzer papers. Deep-Sea Research Part a-Oceanographic Research 37 (1), 157–165. Vereshchaka, A.L., Vinogradov, G.M., 1999. Visual observations of the vertical distribution of plankton throughout the water column above Broken Spur vent field, Mid-Atlantic Ridge. Deep-Sea Research Part I-Oceanographic Research Papers 46 (9), 1615–1632. Von Damm, K.L., 1995a. Controls on the chemistry and temporal variability of seafloor hydrothermal fluids. In: Humphris, S.E., Zierenberg, R.A., Mullineaux, L.S., Thomson, R.E. (Eds.), Seafloor Hydrothermal Systems: physical, chemical, biological and geological interactions. Am. Geophys. Union, Washington DC, pp. 222–247. Von Damm, K.L., 2000. Chemistry of hydrothermal vent fluids from 9 degrees-10 degrees N, East Pacific Rise: Time zero, the immediate posteruptive period. Journal of Geophysical Research-Solid Earth 105 (B5), 11203–11222. Von Damm, K.L., Oosting, S.E., Kozlowski, R., Buttermore, L.G., Colodner, D.C., Edmonds, H.N., Edmond, J.M., Grebmeier, J.M., 1995b. Evolution of East Pacific Rise hydrothermal vent fluids following a volcanic-eruption. Nature 375 (6526), 47–50. Winn, C.D., Karl, D.M., Massoth, G.J., 1986. Microorganisms in deep-sea hydrothermal plumes. Nature 320 (6064), 744–746. Wu, J., Wells, M.L., Rember, R., 2011. Dissolved iron in the deep tropical– subtropical Pacific: Evidence for long-range transport of hydrothermal iron. Geochimica Et Cosmochimica Acta 75 (2), 460–468.